From frysinsp at jmu.edu  Wed Jul  2 13:32:50 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Wed, 02 Jul 2008 13:32:50 -0400
Subject: [Mulberry-discuss] Going off-line
Message-ID: <FFD701E2DF266B5C94C53D0F@steven-frysingers-ibook-g4.local>

Since I installed the latest version of Mulberry (on a G4 iBook under 10.4) 
I've been plagued with it dropping the inbox connection. I'll be typing a 
reply, and when I hit send it tells me it sent it, but that no local 
process could occur (and then it blanks out my inbox queue). If I reconnect 
to the inbox, I can send it again and my local files are updates (although 
sometimes it takes 3 tries). Throughout, my addressees get multiple copies.

Is there any solution to this problem?

Thanks!
Steve

********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From Hagedorn at uni-koeln.de  Wed Jul  2 14:00:05 2008
From: Hagedorn at uni-koeln.de (Sebastian Hagedorn)
Date: Wed, 02 Jul 2008 20:00:05 +0200
Subject: [Mulberry-discuss] Going off-line
In-Reply-To: <FFD701E2DF266B5C94C53D0F@steven-frysingers-ibook-g4.local>
References: <FFD701E2DF266B5C94C53D0F@steven-frysingers-ibook-g4.local>
Message-ID: <52E7E3146BE07021535EA972@G5>

-- "Dr. Steven P. Frysinger" <frysinsp at jmu.edu> is rumored to have mumbled 
on 2. Juli 2008 13:32:50 -0400 regarding [Mulberry-discuss] Going off-line:

> Since I installed the latest version of Mulberry (on a G4 iBook under
> 10.4) I've been plagued with it dropping the inbox connection. I'll be
> typing a reply, and when I hit send it tells me it sent it, but that no
> local process could occur (and then it blanks out my inbox queue). If I
> reconnect to the inbox, I can send it again and my local files are
> updates (although sometimes it takes 3 tries). Throughout, my addressees
> get multiple copies.
>
> Is there any solution to this problem?

Not sure. Sounds like you have flaky networking. How do you access the 
Internet? Is there a NAT router involved?
--
Sebastian Hagedorn - RZKR-R1 (Flachbau), Zi. 18, Robert-Koch-Str. 10
Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
Universit?t zu K?ln / Cologne University - Tel. +49-221-478-5587
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 194 bytes
Desc: not available
Url : http://lists.mulberrymail.com/pipermail/mulberry-discuss/attachments/20080702/33abc935/attachment.bin 

From frysinsp at jmu.edu  Wed Jul  2 14:06:45 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Wed, 02 Jul 2008 14:06:45 -0400
Subject: [Mulberry-discuss] Going off-line
In-Reply-To: <52E7E3146BE07021535EA972@G5>
References: <FFD701E2DF266B5C94C53D0F@steven-frysingers-ibook-g4.local>
	<52E7E3146BE07021535EA972@G5>
Message-ID: <5868326EC6C36863328B2F20@steven-frysingers-ibook-g4.local>

I'm presently working in Germany, and it would be tempting to blame my 
Deutsche Telekom DSL connection (with a wireless router) in my apartment. 
But the problem also occurs in my university office, where I connect to 
their network via Ethernet. Could it be that my IMAP server at my home 
university is dropping me? (Hmmm...now that I've got a couple other IMAP 
servers in my set up maybe I'll do some experiments to see if I get dropped 
by them as well.)

Thanks!
Steve

--On July 2, 2008 8:00:05 PM +0200 Sebastian Hagedorn 
<Hagedorn at uni-koeln.de> wrote:

> -- "Dr. Steven P. Frysinger" <frysinsp at jmu.edu> is rumored to have
> mumbled on 2. Juli 2008 13:32:50 -0400 regarding [Mulberry-discuss] Going
> off-line:
>
>> Since I installed the latest version of Mulberry (on a G4 iBook under
>> 10.4) I've been plagued with it dropping the inbox connection. I'll be
>> typing a reply, and when I hit send it tells me it sent it, but that no
>> local process could occur (and then it blanks out my inbox queue). If I
>> reconnect to the inbox, I can send it again and my local files are
>> updates (although sometimes it takes 3 tries). Throughout, my addressees
>> get multiple copies.
>>
>> Is there any solution to this problem?
>
> Not sure. Sounds like you have flaky networking. How do you access the
> Internet? Is there a NAT router involved?
> --
> Sebastian Hagedorn - RZKR-R1 (Flachbau), Zi. 18, Robert-Koch-Str. 10
> Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
> Universit?t zu K?ln / Cologne University - Tel. +49-221-478-5587



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From frysinsp at jmu.edu  Wed Jul  2 14:57:11 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Wed, 02 Jul 2008 14:57:11 -0400
Subject: [Mulberry-discuss] Further to dropped connection
Message-ID: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>

I just did the experiment. I was connected to three different IMAP systems 
simultaneously, and they were eventually all dropped together, so it seems 
this is a Mulberry issue, not a server problem.

Any ideas gratefully accepted!
Steve

********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From daboo at mulberrymail.com  Wed Jul  2 14:58:05 2008
From: daboo at mulberrymail.com (Cyrus Daboo)
Date: Wed, 02 Jul 2008 14:58:05 -0400
Subject: [Mulberry-discuss] Going off-line
In-Reply-To: <5868326EC6C36863328B2F20@steven-frysingers-ibook-g4.local>
References: <FFD701E2DF266B5C94C53D0F@steven-frysingers-ibook-g4.local>
	<52E7E3146BE07021535EA972@G5>
	<5868326EC6C36863328B2F20@steven-frysingers-ibook-g4.local>
Message-ID: <85F110198BCD0605A60ED00F@caldav.corp.apple.com>

Hi,

--On July 2, 2008 2:06:45 PM -0400 "Dr. Steven P. Frysinger" 
<frysinsp at jmu.edu> wrote:

> I'm presently working in Germany, and it would be tempting to blame my
> Deutsche Telekom DSL connection (with a wireless router) in my apartment.
> But the problem also occurs in my university office, where I connect to
> their network via Ethernet. Could it be that my IMAP server at my home
> university is dropping me? (Hmmm...now that I've got a couple other IMAP
> servers in my set up maybe I'll do some experiments to see if I get
> dropped by them as well.)

What polling interval do you have set for new mail checking? You might want 
to try dialing that down to the minimum of 1 minute and see if that helps. 
Also check whether you have any local software that might be interfering 
with the network connection (e.g. firewall, anti-virus).

-- 
Cyrus Daboo


From srb at umich.edu  Wed Jul  2 15:22:28 2008
From: srb at umich.edu (Steve Burling)
Date: Wed, 02 Jul 2008 15:22:28 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
Message-ID: <A25FC203A8793E138DB42DB2@srb.local>

--On July 2, 2008 2:57:11 PM -0400 "Dr. Steven P. Frysinger" 
<frysinsp at jmu.edu> wrote:

> I just did the experiment. I was connected to three different IMAP
> systems simultaneously, and they were eventually all dropped together, so
> it seems this is a Mulberry issue, not a server problem.

To which I reply:

I am routinely connected to two IMAP servers (one a Cyrus server, and one 
Exchange), all day every day, from either my office on gig ethernet, or at 
home wirelessly.  The *only* time I've seen this behavior is when I'm 
traveling, at one particular hotel where I stay a lot, where their net 
seems to reset on a very short interval (of inactivity?).  At that hotel, 
it happens whether I connect wirelessly or plug my own wireless access 
point into their hard-wired network.

It's a pain in the ass, but I think that it's a networking issue, and not 
an issue with Mulberry, except to the extent that Mulberry's recovery from 
networking issues has always felt a bit fragile.

-- 
Steve Burling                                    <mailto:srb at umich.edu>
University of Michigan, ICPSR                    Voice: +1 734 615.3779
330 Packard Street                               FAX:   +1 734 647.8700
Ann Arbor, MI 48104-2910

From frysinsp at jmu.edu  Wed Jul  2 15:48:03 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Wed, 02 Jul 2008 15:48:03 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <A25FC203A8793E138DB42DB2@srb.local>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local>
Message-ID: <B04C942CAD7A74B730E8AC3C@steven-frysingers-ibook-g4.local>

I suppose it could be coincidental, but it happens from two different 
locations (although they might be using the same or similar network 
protocols). Also, it never happened with the older version of Mulberry in 
these same locations (though that might be the recovery issue you 
mentioned).

Thanks!
Steve

--On July 2, 2008 3:22:28 PM -0400 Steve Burling <srb at umich.edu> wrote:

> --On July 2, 2008 2:57:11 PM -0400 "Dr. Steven P. Frysinger"
> <frysinsp at jmu.edu> wrote:
>
>> I just did the experiment. I was connected to three different IMAP
>> systems simultaneously, and they were eventually all dropped together, so
>> it seems this is a Mulberry issue, not a server problem.
>
> To which I reply:
>
> I am routinely connected to two IMAP servers (one a Cyrus server, and one
> Exchange), all day every day, from either my office on gig ethernet, or
> at home wirelessly.  The *only* time I've seen this behavior is when I'm
> traveling, at one particular hotel where I stay a lot, where their net
> seems to reset on a very short interval (of inactivity?).  At that hotel,
> it happens whether I connect wirelessly or plug my own wireless access
> point into their hard-wired network.
>
> It's a pain in the ass, but I think that it's a networking issue, and not
> an issue with Mulberry, except to the extent that Mulberry's recovery
> from networking issues has always felt a bit fragile.
>
> --
> Steve Burling                                    <mailto:srb at umich.edu>
> University of Michigan, ICPSR                    Voice: +1 734 615.3779
> 330 Packard Street                               FAX:   +1 734 647.8700
> Ann Arbor, MI 48104-2910



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From Hagedorn at uni-koeln.de  Wed Jul  2 15:56:24 2008
From: Hagedorn at uni-koeln.de (Sebastian Hagedorn)
Date: Wed, 02 Jul 2008 21:56:24 +0200
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <A25FC203A8793E138DB42DB2@srb.local>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local>
Message-ID: <A597CD4CB3EE4672283E6DB9@G5>

Hi,

-- Steve Burling <srb at umich.edu> is rumored to have mumbled on 2. Juli 2008 
15:22:28 -0400 regarding Re: [Mulberry-discuss] Further to dropped 
connection:

> --On July 2, 2008 2:57:11 PM -0400 "Dr. Steven P. Frysinger"
> <frysinsp at jmu.edu> wrote:
>
>> I just did the experiment. I was connected to three different IMAP
>> systems simultaneously, and they were eventually all dropped together, so
>> it seems this is a Mulberry issue, not a server problem.

nobody said it's a server problem. I still believe it's a networking 
problem. This mail you sent via T-Online. I know for a fact that DSL 
connections in Germany are dropped at regular intervals so that you don't 
have a semi-static IP address. That explains what you see for the home 
situation. Is it really the same at the University? If it really is, 
perhaps their networking is flaky? I'm one of the network administrators 
for Cologne University, so please believe me when I tell you it's not out 
of the question ;-)

> It's a pain in the ass, but I think that it's a networking issue, and not
> an issue with Mulberry, except to the extent that Mulberry's recovery
> from networking issues has always felt a bit fragile.

And more so in the latest versions ... auto-recovery used to work pretty 
well, but it doesn't anymore. I quit and restart Mulberry whenever my cable 
modem connection gets dropped, jut to be safe.
--
Sebastian Hagedorn - RZKR-R1 (Flachbau), Zi. 18, Robert-Koch-Str. 10
Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
Universit?t zu K?ln / Cologne University - Tel. +49-221-478-5587
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 194 bytes
Desc: not available
Url : http://lists.mulberrymail.com/pipermail/mulberry-discuss/attachments/20080702/82104e54/attachment.bin 

From frysinsp at jmu.edu  Wed Jul  2 16:09:39 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Wed, 02 Jul 2008 16:09:39 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <A597CD4CB3EE4672283E6DB9@G5>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
Message-ID: <51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>

Yeah, I just wanted to rule my JMU server out - they do plenty of flaky 
things, so that was a possibility (though they're usually only badly 
behaved on the SMTP side).

I don't know what my local German university's network is based on, but it 
does happen there too. (Of course they have their own silliness - I have to 
change SMTP servers when I'm in their building... :)

I changed my Alert frequency to 1 minute (from 5), which is I think what 
Cyrus was suggesting, and so far so good. Up until now I've been getting 
dropped pretty regularly, so maybe this will get ahead of the time-out 
"features" you're talking about.

Thanks!
Steve

--On July 2, 2008 9:56:24 PM +0200 Sebastian Hagedorn 
<Hagedorn at uni-koeln.de> wrote:

> Hi,
>
> -- Steve Burling <srb at umich.edu> is rumored to have mumbled on 2. Juli
> 2008 15:22:28 -0400 regarding Re: [Mulberry-discuss] Further to dropped
> connection:
>
>> --On July 2, 2008 2:57:11 PM -0400 "Dr. Steven P. Frysinger"
>> <frysinsp at jmu.edu> wrote:
>>
>>> I just did the experiment. I was connected to three different IMAP
>>> systems simultaneously, and they were eventually all dropped together,
>>> so it seems this is a Mulberry issue, not a server problem.
>
> nobody said it's a server problem. I still believe it's a networking
> problem. This mail you sent via T-Online. I know for a fact that DSL
> connections in Germany are dropped at regular intervals so that you don't
> have a semi-static IP address. That explains what you see for the home
> situation. Is it really the same at the University? If it really is,
> perhaps their networking is flaky? I'm one of the network administrators
> for Cologne University, so please believe me when I tell you it's not out
> of the question ;-)
>
>> It's a pain in the ass, but I think that it's a networking issue, and not
>> an issue with Mulberry, except to the extent that Mulberry's recovery
>> from networking issues has always felt a bit fragile.
>
> And more so in the latest versions ... auto-recovery used to work pretty
> well, but it doesn't anymore. I quit and restart Mulberry whenever my
> cable modem connection gets dropped, jut to be safe.
> --
> Sebastian Hagedorn - RZKR-R1 (Flachbau), Zi. 18, Robert-Koch-Str. 10
> Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
> Universit?t zu K?ln / Cologne University - Tel. +49-221-478-5587



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From frysinsp at jmu.edu  Wed Jul  2 16:10:02 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Wed, 02 Jul 2008 16:10:02 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <A597CD4CB3EE4672283E6DB9@G5>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
Message-ID: <FE1F9BD3F523A8A574A5D45E@steven-frysingers-ibook-g4.local>

Yeah, I just wanted to rule my JMU server out - they do plenty of flaky 
things, so that was a possibility (though they're usually only badly 
behaved on the SMTP side).

I don't know what my local German university's network is based on, but it 
does happen there too. (Of course they have their own silliness - I have to 
change SMTP servers when I'm in their building... :)

I changed my Alert frequency to 1 minute (from 5), which is I think what 
Cyrus was suggesting, and so far so good. Up until now I've been getting 
dropped pretty regularly, so maybe this will get ahead of the time-out 
"features" you're talking about.

Thanks!
Steve

--On July 2, 2008 9:56:24 PM +0200 Sebastian Hagedorn 
<Hagedorn at uni-koeln.de> wrote:

> Hi,
>
> -- Steve Burling <srb at umich.edu> is rumored to have mumbled on 2. Juli
> 2008 15:22:28 -0400 regarding Re: [Mulberry-discuss] Further to dropped
> connection:
>
>> --On July 2, 2008 2:57:11 PM -0400 "Dr. Steven P. Frysinger"
>> <frysinsp at jmu.edu> wrote:
>>
>>> I just did the experiment. I was connected to three different IMAP
>>> systems simultaneously, and they were eventually all dropped together,
>>> so it seems this is a Mulberry issue, not a server problem.
>
> nobody said it's a server problem. I still believe it's a networking
> problem. This mail you sent via T-Online. I know for a fact that DSL
> connections in Germany are dropped at regular intervals so that you don't
> have a semi-static IP address. That explains what you see for the home
> situation. Is it really the same at the University? If it really is,
> perhaps their networking is flaky? I'm one of the network administrators
> for Cologne University, so please believe me when I tell you it's not out
> of the question ;-)
>
>> It's a pain in the ass, but I think that it's a networking issue, and not
>> an issue with Mulberry, except to the extent that Mulberry's recovery
>> from networking issues has always felt a bit fragile.
>
> And more so in the latest versions ... auto-recovery used to work pretty
> well, but it doesn't anymore. I quit and restart Mulberry whenever my
> cable modem connection gets dropped, jut to be safe.
> --
> Sebastian Hagedorn - RZKR-R1 (Flachbau), Zi. 18, Robert-Koch-Str. 10
> Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
> Universit?t zu K?ln / Cologne University - Tel. +49-221-478-5587



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From frysinsp at jmu.edu  Wed Jul  2 16:10:56 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Wed, 02 Jul 2008 16:10:56 -0400
Subject: [Mulberry-discuss] Further to dropped connection
Message-ID: <E3A35D15B7C2E9C516CADBE2@steven-frysingers-ibook-g4.local>

[I guess not - just did it again...]

Yeah, I just wanted to rule my JMU server out - they do plenty of flaky 
things, so that was a possibility (though they're usually only badly 
behaved on the SMTP side).

I don't know what my local German university's network is based on, but it 
does happen there too. (Of course they have their own silliness - I have to 
change SMTP servers when I'm in their building... :)

I changed my Alert frequency to 1 minute (from 5), which is I think what 
Cyrus was suggesting, and so far so good. Up until now I've been getting 
dropped pretty regularly, so maybe this will get ahead of the time-out 
"features" you're talking about.

Thanks!
Steve

--On July 2, 2008 9:56:24 PM +0200 Sebastian Hagedorn 
<Hagedorn at uni-koeln.de> wrote:

> Hi,
>
> -- Steve Burling <srb at umich.edu> is rumored to have mumbled on 2. Juli
> 2008 15:22:28 -0400 regarding Re: [Mulberry-discuss] Further to dropped
> connection:
>
>> --On July 2, 2008 2:57:11 PM -0400 "Dr. Steven P. Frysinger"
>> <frysinsp at jmu.edu> wrote:
>>
>>> I just did the experiment. I was connected to three different IMAP
>>> systems simultaneously, and they were eventually all dropped together,
>>> so it seems this is a Mulberry issue, not a server problem.
>
> nobody said it's a server problem. I still believe it's a networking
> problem. This mail you sent via T-Online. I know for a fact that DSL
> connections in Germany are dropped at regular intervals so that you don't
> have a semi-static IP address. That explains what you see for the home
> situation. Is it really the same at the University? If it really is,
> perhaps their networking is flaky? I'm one of the network administrators
> for Cologne University, so please believe me when I tell you it's not out
> of the question ;-)
>
>> It's a pain in the ass, but I think that it's a networking issue, and not
>> an issue with Mulberry, except to the extent that Mulberry's recovery
>> from networking issues has always felt a bit fragile.
>
> And more so in the latest versions ... auto-recovery used to work pretty
> well, but it doesn't anymore. I quit and restart Mulberry whenever my
> cable modem connection gets dropped, jut to be safe.
> --
> Sebastian Hagedorn - RZKR-R1 (Flachbau), Zi. 18, Robert-Koch-Str. 10
> Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
> Universit?t zu K?ln / Cologne University - Tel. +49-221-478-5587



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From iane at sussex.ac.uk  Thu Jul  3 06:33:49 2008
From: iane at sussex.ac.uk (Ian Eiloart)
Date: Thu, 03 Jul 2008 11:33:49 +0100
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
Message-ID: <9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>



--On 2 July 2008 16:09:39 -0400 "Dr. Steven P. Frysinger" 
<frysinsp at jmu.edu> wrote:

>
> I don't know what my local German university's network is based on, but
> it does happen there too. (Of course they have their own silliness - I
> have to change SMTP servers when I'm in their building... :)
>

That's JMU's fault, as far as I can tell. They use Mirapoint Messaging 
Server, which doesn't seem to be able to listen on the alternate port 587. 
I say "seem" because Mirapoint's online documentation is not helpful on 
this point, but I've found messages on the web that support my claim.

It's recognised best practice for institutions to block outbound 
connections on port 25, so that your machine can't send spam. Your email 
provider *should* provide a facility that allows you to send only 
*authenticated* email. See <http://www.ietf.org/rfc/rfc5068.txt> section 3.1

-- 
Ian Eiloart
IT Services, University of Sussex
x3148

From frysinsp at jmu.edu  Thu Jul  3 07:44:23 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Thu, 03 Jul 2008 07:44:23 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
Message-ID: <9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>

Yeah, without being up on the port details I suspected JMU too, but my 
colleagues and students at HTW confirm that they can only use HTW's SMTP 
server to get out - and then only while they're in the building. So if they 
take a laptop back and forth they're forced to switch back and forth (or 
else use a webmail client). But who wants to use something besides 
Mulberry?!? :) The problem apparently starts at UNI- Saarbr?cken, with 
whom they share a network.

This is further substantiated by the fact that I can't get out of HTW with 
my other two SMTP accounts either.

Note, that this doesn't mean JMU isn't ALSO blocking. They do seem to have 
shutdown forwarding from my other accounts, and can't seem to figure out 
how they did it.

Everytime I think I need to learn more about network details, I find 
something else to do... :)

Steve

--On July 3, 2008 11:33:49 AM +0100 Ian Eiloart <iane at sussex.ac.uk> wrote:

>
>
> --On 2 July 2008 16:09:39 -0400 "Dr. Steven P. Frysinger"
> <frysinsp at jmu.edu> wrote:
>
>>
>> I don't know what my local German university's network is based on, but
>> it does happen there too. (Of course they have their own silliness - I
>> have to change SMTP servers when I'm in their building... :)
>>
>
> That's JMU's fault, as far as I can tell. They use Mirapoint Messaging
> Server, which doesn't seem to be able to listen on the alternate port
> 587. I say "seem" because Mirapoint's online documentation is not helpful
> on this point, but I've found messages on the web that support my claim.
>
> It's recognised best practice for institutions to block outbound
> connections on port 25, so that your machine can't send spam. Your email
> provider *should* provide a facility that allows you to send only
> *authenticated* email. See <http://www.ietf.org/rfc/rfc5068.txt> section
> 3.1
>
> --
> Ian Eiloart
> IT Services, University of Sussex
> x3148



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From brian at reid.org  Thu Jul  3 08:46:35 2008
From: brian at reid.org (Brian Reid)
Date: Thu, 03 Jul 2008 05:46:35 -0700
Subject: [Mulberry-discuss] Going off-line
In-Reply-To: <FFD701E2DF266B5C94C53D0F@steven-frysingers-ibook-g4.local>
References: <FFD701E2DF266B5C94C53D0F@steven-frysingers-ibook-g4.local>
Message-ID: <CA837B76B37A944040727A14@visibletrout.134lan.net>

I've had that problem from time to time. For me, it is always caused by my having another IMAP session open on the same mailbox somewhere else. If I find and close the other IMAP session, the behavior that you describe goes away.


> Since I installed the latest version of Mulberry (on a G4 iBook under 10.4) I've been plagued with it dropping the inbox connection. I'll be typing a reply, and when I hit send it tells me it sent it, but that no local process could occur (and then it blanks out my inbox queue). If I reconnect to the inbox, I can send it again and my local files are updates (although sometimes it takes 3 tries). Throughout, my addressees get multiple copies.
>
> Is there any solution to this problem?
>
> Thanks!
> Steve
>
> ********************************************
> Steven P. Frysinger, Ph.D.
>
> Professor,
>   Integrated Science and Technology Department
>   Computer Science Department
> Director,
>   Environmental Information Systems Program
>   Environmental Management Program
>   Center for Environment, Health and Safety
> James Madison University
> College of Integrated Science and Technology
> ISAT/CS Room 309
> 701 Carrier Drive - MSC 4102
> Harrisonburg, Virginia 22807
> United States of America
> Tel: 540/568-2710
> Fax: 540/568-2768
> Net: frysinsp at jmu.edu
> Web: http://www.isat.jmu.edu/faculty/frysinger.html
>
> ---------------------
>
> Gastprofessor,
>     Fachbereiche GIS und Maschinenbau
> Environmental Informatics Group (EIG)
> Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
> Goebenstra?e 40
> 66117 Saarbr?cken
> Germany
> Cell:  +49 (0)175 955 2303
> Tel:   +49 (0)681 876 5665
> Net:   Steven.Frysinger at enviromatics.net



From Hagedorn at uni-koeln.de  Thu Jul  3 08:51:18 2008
From: Hagedorn at uni-koeln.de (Sebastian Hagedorn)
Date: Thu, 03 Jul 2008 14:51:18 +0200
Subject: [Mulberry-discuss] Going off-line
In-Reply-To: <CA837B76B37A944040727A14@visibletrout.134lan.net>
References: <FFD701E2DF266B5C94C53D0F@steven-frysingers-ibook-g4.local>
	<CA837B76B37A944040727A14@visibletrout.134lan.net>
Message-ID: <373D02F289E5A5B15C506111@tyrion.rrz.uni-koeln.de>

--On 3. Juli 2008 05:46:35 -0700 Brian Reid <brian at reid.org> wrote:

> I've had that problem from time to time. For me, it is always caused by
> my having another IMAP session open on the same mailbox somewhere else.

That's only an issue with few servers these days. Most IMAP servers can 
handle mutliple connections per mailbox. But I agree that in principle this 
could be a cause of disconnects. In this particular case I don't believe 
that it is, though.
-- 
     .:.Sebastian Hagedorn - RZKR-R1 (Geb?ude 52), Zimmer 18.:.
Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
.:.Universit?t zu K?ln / Cologne University - ? +49-221-478-5587.:.
                   .:.:.:.Skype: shagedorn.:.:.:.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 186 bytes
Desc: not available
Url : http://lists.mulberrymail.com/pipermail/mulberry-discuss/attachments/20080703/bf9e43c8/attachment.bin 

From frysinsp at jmu.edu  Thu Jul  3 08:54:09 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Thu, 03 Jul 2008 08:54:09 -0400
Subject: [Mulberry-discuss] Going off-line
In-Reply-To: <CA837B76B37A944040727A14@visibletrout.134lan.net>
References: <FFD701E2DF266B5C94C53D0F@steven-frysingers-ibook-g4.local>
	<CA837B76B37A944040727A14@visibletrout.134lan.net>
Message-ID: <ED382837EF635291B6C8C30A@steven-frysingers-ibook-g4.local>

Yes, I've seen this with the older version when I forgot to close Mulberry 
at home and then accessed it at work. But in this case I'm only accessing 
each IMAP mailbox once.

Steve

--On July 3, 2008 5:46:35 AM -0700 Brian Reid <brian at reid.org> wrote:

> I've had that problem from time to time. For me, it is always caused by
> my having another IMAP session open on the same mailbox somewhere else.
> If I find and close the other IMAP session, the behavior that you
> describe goes away.
>
>
>> Since I installed the latest version of Mulberry (on a G4 iBook under
>> 10.4) I've been plagued with it dropping the inbox connection. I'll be
>> typing a reply, and when I hit send it tells me it sent it, but that no
>> local process could occur (and then it blanks out my inbox queue). If I
>> reconnect to the inbox, I can send it again and my local files are
>> updates (although sometimes it takes 3 tries). Throughout, my addressees
>> get multiple copies.
>>
>> Is there any solution to this problem?
>>
>> Thanks!
>> Steve
>>
>> ********************************************
>> Steven P. Frysinger, Ph.D.
>>
>> Professor,
>>   Integrated Science and Technology Department
>>   Computer Science Department
>> Director,
>>   Environmental Information Systems Program
>>   Environmental Management Program
>>   Center for Environment, Health and Safety
>> James Madison University
>> College of Integrated Science and Technology
>> ISAT/CS Room 309
>> 701 Carrier Drive - MSC 4102
>> Harrisonburg, Virginia 22807
>> United States of America
>> Tel: 540/568-2710
>> Fax: 540/568-2768
>> Net: frysinsp at jmu.edu
>> Web: http://www.isat.jmu.edu/faculty/frysinger.html
>>
>> ---------------------
>>
>> Gastprofessor,
>>     Fachbereiche GIS und Maschinenbau
>> Environmental Informatics Group (EIG)
>> Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
>> Goebenstra?e 40
>> 66117 Saarbr?cken
>> Germany
>> Cell:  +49 (0)175 955 2303
>> Tel:   +49 (0)681 876 5665
>> Net:   Steven.Frysinger at enviromatics.net
>
>



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From Hagedorn at uni-koeln.de  Thu Jul  3 09:00:59 2008
From: Hagedorn at uni-koeln.de (Sebastian Hagedorn)
Date: Thu, 03 Jul 2008 15:00:59 +0200
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
Message-ID: <CEB53F70299C3BFB452BFF3D@tyrion.rrz.uni-koeln.de>

--On 3. Juli 2008 07:44:23 -0400 "Dr. Steven P. Frysinger" 
<frysinsp at jmu.edu> wrote:

> Yeah, without being up on the port details I suspected JMU too, but my
> colleagues and students at HTW confirm that they can only use HTW's SMTP
> server to get out - and then only while they're in the building. So if
> they take a laptop back and forth they're forced to switch back and forth
> (or else use a webmail client).

We used to have a setup like that for a while ;-)

> But who wants to use something besides
> Mulberry?!? :) The problem apparently starts at UNI- Saarbr?cken, with
> whom they share a network.
>
> This is further substantiated by the fact that I can't get out of HTW
> with my other two SMTP accounts either.

Using port 25 or port 587? I think we have established that port 25 is 
blocked. That's basically standard policy for universities. However, port 
587 should *not* be blocked. So why don't you try using that? You need to 
write ":587" after the SMTP server name in your prefererences to do that ...
-- 
     .:.Sebastian Hagedorn - RZKR-R1 (Geb?ude 52), Zimmer 18.:.
Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
.:.Universit?t zu K?ln / Cologne University - ? +49-221-478-5587.:.
                   .:.:.:.Skype: shagedorn.:.:.:.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 186 bytes
Desc: not available
Url : http://lists.mulberrymail.com/pipermail/mulberry-discuss/attachments/20080703/70105538/attachment.bin 

From frysinsp at jmu.edu  Thu Jul  3 09:05:04 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Thu, 03 Jul 2008 09:05:04 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <CEB53F70299C3BFB452BFF3D@tyrion.rrz.uni-koeln.de>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<CEB53F70299C3BFB452BFF3D@tyrion.rrz.uni-koeln.de>
Message-ID: <7F6779215EAA82F1A558CA01@steven-frysingers-ibook-g4.local>

--On July 3, 2008 3:00:59 PM +0200 Sebastian Hagedorn 
<Hagedorn at uni-koeln.de> wrote:

> Using port 25 or port 587? I think we have established that port 25 is
> blocked. That's basically standard policy for universities. However, port
> 587 should *not* be blocked. So why don't you try using that? You need to
> write ":587" after the SMTP server name in your prefererences to do that
> ...

Okay, I'll give this a try when I get to the office tomorrow. But since I'm 
presumably using the default port (25?) now, why does it work from my 
apartment (like right now) and not from the German campus?

Again, I'm unenlightened about ports and their management, so forgive me if 
this is a dumb question.

Steve


********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From Hagedorn at uni-koeln.de  Thu Jul  3 09:09:49 2008
From: Hagedorn at uni-koeln.de (Sebastian Hagedorn)
Date: Thu, 03 Jul 2008 15:09:49 +0200
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <7F6779215EAA82F1A558CA01@steven-frysingers-ibook-g4.local>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<CEB53F70299C3BFB452BFF3D@tyrion.rrz.uni-koeln.de>
	<7F6779215EAA82F1A558CA01@steven-frysingers-ibook-g4.local>
Message-ID: <6EFEF115134F17AC3C30C08C@tyrion.rrz.uni-koeln.de>

--On 3. Juli 2008 09:05:04 -0400 "Dr. Steven P. Frysinger" 
<frysinsp at jmu.edu> wrote:

> --On July 3, 2008 3:00:59 PM +0200 Sebastian Hagedorn
> <Hagedorn at uni-koeln.de> wrote:
>
>> Using port 25 or port 587? I think we have established that port 25 is
>> blocked. That's basically standard policy for universities. However, port
>> 587 should *not* be blocked. So why don't you try using that? You need to
>> write ":587" after the SMTP server name in your prefererences to do that
>> ...
>
> Okay, I'll give this a try when I get to the office tomorrow. But since
> I'm presumably using the default port (25?) now, why does it work from my
> apartment (like right now) and not from the German campus?

Easy: the university blocks port 25, your provider does not. That's par for 
the course.
-- 
     .:.Sebastian Hagedorn - RZKR-R1 (Geb?ude 52), Zimmer 18.:.
Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
.:.Universit?t zu K?ln / Cologne University - ? +49-221-478-5587.:.
                   .:.:.:.Skype: shagedorn.:.:.:.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 186 bytes
Desc: not available
Url : http://lists.mulberrymail.com/pipermail/mulberry-discuss/attachments/20080703/84955f28/attachment.bin 

From iane at sussex.ac.uk  Thu Jul  3 09:39:54 2008
From: iane at sussex.ac.uk (Ian Eiloart)
Date: Thu, 03 Jul 2008 14:39:54 +0100
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
Message-ID: <8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>



--On 3 July 2008 07:44:23 -0400 "Dr. Steven P. Frysinger" 
<frysinsp at jmu.edu> wrote:

> Yeah, without being up on the port details I suspected JMU too, but my
> colleagues and students at HTW confirm that they can only use HTW's SMTP
> server to get out - and then only while they're in the building. So if
> they take a laptop back and forth they're forced to switch back and forth
> (or else use a webmail client). But who wants to use something besides
> Mulberry?!? :) The problem apparently starts at UNI- Saarbr?cken, with
> whom they share a network.
>
> This is further substantiated by the fact that I can't get out of HTW
> with my other two SMTP accounts either.
>
> Note, that this doesn't mean JMU isn't ALSO blocking. They do seem to
> have shutdown forwarding from my other accounts, and can't seem to figure
> out how they did it.

No, you misunderstand. Every site SHOULD block port 25, because that's the 
port used to deliver unauthenticated email to other sites. Only the mail 
servers on a site should be able to do that. HTW is doing that, and that 
means that machines on their site are far less likely to become spam 
sources. Good for them!

Where you have an email account (eg at JMU) you should be permitted to 
connect on port 587. JMU simply aren't offering that facility. They require 
you to connect on port 25. They ought to be aware that this is no longer 
possible from a very large number of sites.

>
> Everytime I think I need to learn more about network details, I find
> something else to do... :)
>
> Steve
>
> --On July 3, 2008 11:33:49 AM +0100 Ian Eiloart <iane at sussex.ac.uk> wrote:
>
>>
>>
>> --On 2 July 2008 16:09:39 -0400 "Dr. Steven P. Frysinger"
>> <frysinsp at jmu.edu> wrote:
>>
>>>
>>> I don't know what my local German university's network is based on, but
>>> it does happen there too. (Of course they have their own silliness - I
>>> have to change SMTP servers when I'm in their building... :)
>>>
>>
>> That's JMU's fault, as far as I can tell. They use Mirapoint Messaging
>> Server, which doesn't seem to be able to listen on the alternate port
>> 587. I say "seem" because Mirapoint's online documentation is not helpful
>> on this point, but I've found messages on the web that support my claim.
>>
>> It's recognised best practice for institutions to block outbound
>> connections on port 25, so that your machine can't send spam. Your email
>> provider *should* provide a facility that allows you to send only
>> *authenticated* email. See <http://www.ietf.org/rfc/rfc5068.txt> section
>> 3.1
>>
>> --
>> Ian Eiloart
>> IT Services, University of Sussex
>> x3148
>
>
>
> ********************************************
> Steven P. Frysinger, Ph.D.
>
> Professor,
>   Integrated Science and Technology Department
>   Computer Science Department
> Director,
>   Environmental Information Systems Program
>   Environmental Management Program
>   Center for Environment, Health and Safety
> James Madison University
> College of Integrated Science and Technology
> ISAT/CS Room 309
> 701 Carrier Drive - MSC 4102
> Harrisonburg, Virginia 22807
> United States of America
> Tel: 540/568-2710
> Fax: 540/568-2768
> Net: frysinsp at jmu.edu
> Web: http://www.isat.jmu.edu/faculty/frysinger.html
>
> ---------------------
>
> Gastprofessor,
>     Fachbereiche GIS und Maschinenbau
> Environmental Informatics Group (EIG)
> Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
> Goebenstra?e 40
> 66117 Saarbr?cken
> Germany
> Cell:  +49 (0)175 955 2303
> Tel:   +49 (0)681 876 5665
> Net:   Steven.Frysinger at enviromatics.net
>



-- 
Ian Eiloart
IT Services, University of Sussex
x3148

From dave at scocca.org  Thu Jul  3 09:50:00 2008
From: dave at scocca.org (Dave Scocca)
Date: Thu, 03 Jul 2008 09:50:00 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
Message-ID: <90D74E496B0961DE492A7D9C@[10.8.2.178]>



--On 7/3/2008 11:33 AM +0100 Ian Eiloart wrote:

>> I don't know what my local German university's network is based on, but
>> it does happen there too. (Of course they have their own silliness - I
>> have to change SMTP servers when I'm in their building... :)
>
> That's JMU's fault, as far as I can tell. They use Mirapoint Messaging
> Server, which doesn't seem to be able to listen on the alternate port
> 587. I say "seem" because Mirapoint's online documentation is not helpful
> on this point, but I've found messages on the web that support my claim.

I use IMAP Partners, who use a Mirapoint server, and that server IS 
configured for an alternate SMTP port--so it's not an inherent limitation 
of the server.  (They use 2525 rather than 587--I do not know whether the 
nonstandard choice of port is Mirapoint's or IMAP Partners'.)

Dave

From frysinsp at jmu.edu  Thu Jul  3 10:14:58 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Thu, 03 Jul 2008 10:14:58 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
Message-ID: <F082366BC769FC785D3C8EFE@[192.168.2.101]>

If this is the case, it may not be JMU's fault - I'm not specifying a port, 
so I guess I'm trying 25 by default. Sebastian suggested overriding that, 
and I'll try that in the morning when I get to the office.

In the meanwhile, further to the issue of getting dropped:

My connections had been dropping about every 10-15 minutes. About an hour 
ago I shut down Mulberry 4 and ran 3, bringing in my latest preferences (so 
they're configured the same). Haven't been dropped yet. Okay, can't prove 
the null hypothesis, but if it doesn't drop me in the next few hours, I 
think that will point to an issue in version 4. Maybe, as someone 
previously posted, it handles server disconnects a little less robustly.

I'll let you know how the experiment goes either way.

Steve

--On Thursday, July 3, 2008 2:39 PM +0100 Ian Eiloart <iane at sussex.ac.uk> 
wrote:

>
>
> --On 3 July 2008 07:44:23 -0400 "Dr. Steven P. Frysinger"
> <frysinsp at jmu.edu> wrote:
>
>> Yeah, without being up on the port details I suspected JMU too, but my
>> colleagues and students at HTW confirm that they can only use HTW's SMTP
>> server to get out - and then only while they're in the building. So if
>> they take a laptop back and forth they're forced to switch back and forth
>> (or else use a webmail client). But who wants to use something besides
>> Mulberry?!? :) The problem apparently starts at UNI- Saarbr?cken, with
>> whom they share a network.
>>
>> This is further substantiated by the fact that I can't get out of HTW
>> with my other two SMTP accounts either.
>>
>> Note, that this doesn't mean JMU isn't ALSO blocking. They do seem to
>> have shutdown forwarding from my other accounts, and can't seem to figure
>> out how they did it.
>
> No, you misunderstand. Every site SHOULD block port 25, because that's
> the port used to deliver unauthenticated email to other sites. Only the
> mail servers on a site should be able to do that. HTW is doing that, and
> that means that machines on their site are far less likely to become spam
> sources. Good for them!
>
> Where you have an email account (eg at JMU) you should be permitted to
> connect on port 587. JMU simply aren't offering that facility. They
> require you to connect on port 25. They ought to be aware that this is no
> longer possible from a very large number of sites.
>
>>
>> Everytime I think I need to learn more about network details, I find
>> something else to do... :)
>>
>> Steve
>>
>> --On July 3, 2008 11:33:49 AM +0100 Ian Eiloart <iane at sussex.ac.uk>
>> wrote:
>>
>>>
>>>
>>> --On 2 July 2008 16:09:39 -0400 "Dr. Steven P. Frysinger"
>>> <frysinsp at jmu.edu> wrote:
>>>
>>>>
>>>> I don't know what my local German university's network is based on, but
>>>> it does happen there too. (Of course they have their own silliness - I
>>>> have to change SMTP servers when I'm in their building... :)
>>>>
>>>
>>> That's JMU's fault, as far as I can tell. They use Mirapoint Messaging
>>> Server, which doesn't seem to be able to listen on the alternate port
>>> 587. I say "seem" because Mirapoint's online documentation is not
>>> helpful on this point, but I've found messages on the web that support
>>> my claim.
>>>
>>> It's recognised best practice for institutions to block outbound
>>> connections on port 25, so that your machine can't send spam. Your email
>>> provider *should* provide a facility that allows you to send only
>>> *authenticated* email. See <http://www.ietf.org/rfc/rfc5068.txt> section
>>> 3.1
>>>
>>> --
>>> Ian Eiloart
>>> IT Services, University of Sussex
>>> x3148
>>
>>
>>
>> ********************************************
>> Steven P. Frysinger, Ph.D.
>>
>> Professor,
>>   Integrated Science and Technology Department
>>   Computer Science Department
>> Director,
>>   Environmental Information Systems Program
>>   Environmental Management Program
>>   Center for Environment, Health and Safety
>> James Madison University
>> College of Integrated Science and Technology
>> ISAT/CS Room 309
>> 701 Carrier Drive - MSC 4102
>> Harrisonburg, Virginia 22807
>> United States of America
>> Tel: 540/568-2710
>> Fax: 540/568-2768
>> Net: frysinsp at jmu.edu
>> Web: http://www.isat.jmu.edu/faculty/frysinger.html
>>
>> ---------------------
>>
>> Gastprofessor,
>>     Fachbereiche GIS und Maschinenbau
>> Environmental Informatics Group (EIG)
>> Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
>> Goebenstra?e 40
>> 66117 Saarbr?cken
>> Germany
>> Cell:  +49 (0)175 955 2303
>> Tel:   +49 (0)681 876 5665
>> Net:   Steven.Frysinger at enviromatics.net
>>
>
>
>
> --
> Ian Eiloart
> IT Services, University of Sussex
> x3148



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From Hagedorn at uni-koeln.de  Thu Jul  3 10:27:02 2008
From: Hagedorn at uni-koeln.de (Sebastian Hagedorn)
Date: Thu, 03 Jul 2008 16:27:02 +0200
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <F082366BC769FC785D3C8EFE@[192.168.2.101]>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
	<F082366BC769FC785D3C8EFE@[192.168.2.101]>
Message-ID: <1B98B6B850B45D486F161F40@tyrion.rrz.uni-koeln.de>

--On 3. Juli 2008 10:14:58 -0400 "Dr. Steven P. Frysinger" 
<frysinsp at jmu.edu> wrote:

> In the meanwhile, further to the issue of getting dropped:
>
> My connections had been dropping about every 10-15 minutes.

That's way too often, even for DSL. Those shouldn't occur more often than 
two or three times per day.

> About an hour
> ago I shut down Mulberry 4 and ran 3, bringing in my latest preferences
> (so they're configured the same). Haven't been dropped yet. Okay, can't
> prove the null hypothesis, but if it doesn't drop me in the next few
> hours, I think that will point to an issue in version 4. Maybe, as
> someone previously posted, it handles server disconnects a little less
> robustly.

I don't understand why there should be so many disconnects to begin with! 
And it's not like everybody has this problem, either. If you're at all 
Terminal-savvy, could you try this:

Open the Terminal and write this command:

ping servername

Substitute the actual server name for servername, like this:

% ping smtp.uni-koeln.de

PING msa.rrz.uni-koeln.de (134.95.19.54): 56 data bytes
64 bytes from 134.95.19.54: icmp_seq=0 ttl=61 time=0.522 ms
64 bytes from 134.95.19.54: icmp_seq=1 ttl=61 time=0.451 ms
64 bytes from 134.95.19.54: icmp_seq=2 ttl=61 time=0.480 ms
...

Assuming the ping works at all (it might get blocked by a firewall), do you 
get intermittent dropped packages? If ping does not work, could you try 
traceroute instead?

BTW, if you don't want to use the Terminal, you could use Network Utility 
in the Utilities folder instead (it's in Applications). Both Ping and 
Traceroute are available there as well.
-- 
     .:.Sebastian Hagedorn - RZKR-R1 (Geb?ude 52), Zimmer 18.:.
Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
.:.Universit?t zu K?ln / Cologne University - ? +49-221-478-5587.:.
                   .:.:.:.Skype: shagedorn.:.:.:.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 186 bytes
Desc: not available
Url : http://lists.mulberrymail.com/pipermail/mulberry-discuss/attachments/20080703/5bb82447/attachment.bin 

From iane at sussex.ac.uk  Thu Jul  3 10:37:12 2008
From: iane at sussex.ac.uk (Ian Eiloart)
Date: Thu, 03 Jul 2008 15:37:12 +0100
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <F082366BC769FC785D3C8EFE@[192.168.2.101]>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
	<F082366BC769FC785D3C8EFE@[192.168.2.101]>
Message-ID: <3B830E3C91123EAEF61AD81D@lewes.staff.uscs.susx.ac.uk>



--On 3 July 2008 10:14:58 -0400 "Dr. Steven P. Frysinger" 
<frysinsp at jmu.edu> wrote:

> If this is the case, it may not be JMU's fault - I'm not specifying a
> port, so I guess I'm trying 25 by default. Sebastian suggested overriding
> that, and I'll try that in the morning when I get to the office.

No, it REALLY is JMU's fault. 25 is the default for most email clients, but 
some modern email clients  (eg the iPhone's mail application) default to 
587, and most are capable of connecting on 587.

JMU should be listening on 587, but they aren't. I know because I tried to 
connect, and there really is nothing listening there. Could be a temporary 
condition, but JMU's online documentation doesn't mention port 587, AND it 
explicitly says to connect on port 25, AND they're using software that 
doesn't seem to even have the capability to listen there.

RFC 5068 was published in November 2007, defining the practice as best 
common practice. However, rfc2476 -  which defined the message submission 
service - was published in December 1998. In part, this was because 
Internet Service Providers were already blocking port 25. Some sites opted 
for a variety of alternate port numbers to get around this (ours used 225, 
I think) and it was a relief that this was standardised. rfc4409 superceded 
2476 in April 2006.

> In the meanwhile, further to the issue of getting dropped:
>
> My connections had been dropping about every 10-15 minutes. About an hour
> ago I shut down Mulberry 4 and ran 3, bringing in my latest preferences
> (so they're configured the same). Haven't been dropped yet. Okay, can't
> prove the null hypothesis, but if it doesn't drop me in the next few
> hours, I think that will point to an issue in version 4. Maybe, as
> someone previously posted, it handles server disconnects a little less
> robustly.
>
> I'll let you know how the experiment goes either way.
>
> Steve
>
> --On Thursday, July 3, 2008 2:39 PM +0100 Ian Eiloart <iane at sussex.ac.uk>
> wrote:
>
>>
>>
>> --On 3 July 2008 07:44:23 -0400 "Dr. Steven P. Frysinger"
>> <frysinsp at jmu.edu> wrote:
>>
>>> Yeah, without being up on the port details I suspected JMU too, but my
>>> colleagues and students at HTW confirm that they can only use HTW's SMTP
>>> server to get out - and then only while they're in the building. So if
>>> they take a laptop back and forth they're forced to switch back and
>>> forth (or else use a webmail client). But who wants to use something
>>> besides Mulberry?!? :) The problem apparently starts at UNI-
>>> Saarbr?cken, with whom they share a network.
>>>
>>> This is further substantiated by the fact that I can't get out of HTW
>>> with my other two SMTP accounts either.
>>>
>>> Note, that this doesn't mean JMU isn't ALSO blocking. They do seem to
>>> have shutdown forwarding from my other accounts, and can't seem to
>>> figure out how they did it.
>>
>> No, you misunderstand. Every site SHOULD block port 25, because that's
>> the port used to deliver unauthenticated email to other sites. Only the
>> mail servers on a site should be able to do that. HTW is doing that, and
>> that means that machines on their site are far less likely to become spam
>> sources. Good for them!
>>
>> Where you have an email account (eg at JMU) you should be permitted to
>> connect on port 587. JMU simply aren't offering that facility. They
>> require you to connect on port 25. They ought to be aware that this is no
>> longer possible from a very large number of sites.
>>
>>>
>>> Everytime I think I need to learn more about network details, I find
>>> something else to do... :)
>>>
>>> Steve
>>>
>>> --On July 3, 2008 11:33:49 AM +0100 Ian Eiloart <iane at sussex.ac.uk>
>>> wrote:
>>>
>>>>
>>>>
>>>> --On 2 July 2008 16:09:39 -0400 "Dr. Steven P. Frysinger"
>>>> <frysinsp at jmu.edu> wrote:
>>>>
>>>>>
>>>>> I don't know what my local German university's network is based on,
>>>>> but it does happen there too. (Of course they have their own
>>>>> silliness - I have to change SMTP servers when I'm in their
>>>>> building... :)
>>>>>
>>>>
>>>> That's JMU's fault, as far as I can tell. They use Mirapoint Messaging
>>>> Server, which doesn't seem to be able to listen on the alternate port
>>>> 587. I say "seem" because Mirapoint's online documentation is not
>>>> helpful on this point, but I've found messages on the web that support
>>>> my claim.
>>>>
>>>> It's recognised best practice for institutions to block outbound
>>>> connections on port 25, so that your machine can't send spam. Your
>>>> email provider *should* provide a facility that allows you to send only
>>>> *authenticated* email. See <http://www.ietf.org/rfc/rfc5068.txt>
>>>> section 3.1
>>>>
>>>> --
>>>> Ian Eiloart
>>>> IT Services, University of Sussex
>>>> x3148
>>>
>>>
>>>
>>> ********************************************
>>> Steven P. Frysinger, Ph.D.
>>>
>>> Professor,
>>>   Integrated Science and Technology Department
>>>   Computer Science Department
>>> Director,
>>>   Environmental Information Systems Program
>>>   Environmental Management Program
>>>   Center for Environment, Health and Safety
>>> James Madison University
>>> College of Integrated Science and Technology
>>> ISAT/CS Room 309
>>> 701 Carrier Drive - MSC 4102
>>> Harrisonburg, Virginia 22807
>>> United States of America
>>> Tel: 540/568-2710
>>> Fax: 540/568-2768
>>> Net: frysinsp at jmu.edu
>>> Web: http://www.isat.jmu.edu/faculty/frysinger.html
>>>
>>> ---------------------
>>>
>>> Gastprofessor,
>>>     Fachbereiche GIS und Maschinenbau
>>> Environmental Informatics Group (EIG)
>>> Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
>>> Goebenstra?e 40
>>> 66117 Saarbr?cken
>>> Germany
>>> Cell:  +49 (0)175 955 2303
>>> Tel:   +49 (0)681 876 5665
>>> Net:   Steven.Frysinger at enviromatics.net
>>>
>>
>>
>>
>> --
>> Ian Eiloart
>> IT Services, University of Sussex
>> x3148
>
>
>
> ********************************************
> Steven P. Frysinger, Ph.D.
>
> Professor,
>   Integrated Science and Technology Department
>   Computer Science Department
> Director,
>   Environmental Information Systems Program
>   Environmental Management Program
>   Center for Environment, Health and Safety
> James Madison University
> College of Integrated Science and Technology
> ISAT/CS Room 309
> 701 Carrier Drive - MSC 4102
> Harrisonburg, Virginia 22807
> United States of America
> Tel: 540/568-2710
> Fax: 540/568-2768
> Net: frysinsp at jmu.edu
> Web: http://www.isat.jmu.edu/faculty/frysinger.html
>
> ---------------------
>
> Gastprofessor,
>     Fachbereiche GIS und Maschinenbau
> Environmental Informatics Group (EIG)
> Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
> Goebenstra?e 40
> 66117 Saarbr?cken
> Germany
> Cell:  +49 (0)175 955 2303
> Tel:   +49 (0)681 876 5665
> Net:   Steven.Frysinger at enviromatics.net
>



-- 
Ian Eiloart
IT Services, University of Sussex
x3148

From frysinsp at jmu.edu  Thu Jul  3 10:37:11 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Thu, 03 Jul 2008 10:37:11 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <1B98B6B850B45D486F161F40@tyrion.rrz.uni-koeln.de>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
	<F082366BC769FC785D3C8EFE@[192.168.2.101]>
	<1B98B6B850B45D486F161F40@tyrion.rrz.uni-koeln.de>
Message-ID: <1094D9B7CD8AC7849164FBCB@[192.168.2.101]>

Can't get through to JMU at all, but I'm hitting both the SMTP and IMAP 
servers of my other two accounts without any glitches.

If it is a matter of Mulberry 4 getting tripped up by one of my hosts being 
skittish, I guess I wouldn't be surprised if it was JMU.

Steve

--On Thursday, July 3, 2008 4:27 PM +0200 Sebastian Hagedorn 
<Hagedorn at uni-koeln.de> wrote:

> --On 3. Juli 2008 10:14:58 -0400 "Dr. Steven P. Frysinger"
> <frysinsp at jmu.edu> wrote:
>
>> In the meanwhile, further to the issue of getting dropped:
>>
>> My connections had been dropping about every 10-15 minutes.
>
> That's way too often, even for DSL. Those shouldn't occur more often than
> two or three times per day.
>
>> About an hour
>> ago I shut down Mulberry 4 and ran 3, bringing in my latest preferences
>> (so they're configured the same). Haven't been dropped yet. Okay, can't
>> prove the null hypothesis, but if it doesn't drop me in the next few
>> hours, I think that will point to an issue in version 4. Maybe, as
>> someone previously posted, it handles server disconnects a little less
>> robustly.
>
> I don't understand why there should be so many disconnects to begin with!
> And it's not like everybody has this problem, either. If you're at all
> Terminal-savvy, could you try this:
>
> Open the Terminal and write this command:
>
> ping servername
>
> Substitute the actual server name for servername, like this:
>
> % ping smtp.uni-koeln.de
>
> PING msa.rrz.uni-koeln.de (134.95.19.54): 56 data bytes
> 64 bytes from 134.95.19.54: icmp_seq=0 ttl=61 time=0.522 ms
> 64 bytes from 134.95.19.54: icmp_seq=1 ttl=61 time=0.451 ms
> 64 bytes from 134.95.19.54: icmp_seq=2 ttl=61 time=0.480 ms
> ...
>
> Assuming the ping works at all (it might get blocked by a firewall), do
> you get intermittent dropped packages? If ping does not work, could you
> try traceroute instead?
>
> BTW, if you don't want to use the Terminal, you could use Network Utility
> in the Utilities folder instead (it's in Applications). Both Ping and
> Traceroute are available there as well.
> --
>      .:.Sebastian Hagedorn - RZKR-R1 (Geb?ude 52), Zimmer 18.:.
> Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
> .:.Universit?t zu K?ln / Cologne University - ? +49-221-478-5587.:.
>                    .:.:.:.Skype: shagedorn.:.:.:.



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From iane at sussex.ac.uk  Thu Jul  3 10:38:50 2008
From: iane at sussex.ac.uk (Ian Eiloart)
Date: Thu, 03 Jul 2008 15:38:50 +0100
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <90D74E496B0961DE492A7D9C@[10.8.2.178]>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<90D74E496B0961DE492A7D9C@[10.8.2.178]>
Message-ID: <7C22CAC4602CA7791E7DEFEF@lewes.staff.uscs.susx.ac.uk>



--On 3 July 2008 09:50:00 -0400 Dave Scocca <dave at scocca.org> wrote:

>
>
> --On 7/3/2008 11:33 AM +0100 Ian Eiloart wrote:
>
>>> I don't know what my local German university's network is based on, but
>>> it does happen there too. (Of course they have their own silliness - I
>>> have to change SMTP servers when I'm in their building... :)
>>
>> That's JMU's fault, as far as I can tell. They use Mirapoint Messaging
>> Server, which doesn't seem to be able to listen on the alternate port
>> 587. I say "seem" because Mirapoint's online documentation is not helpful
>> on this point, but I've found messages on the web that support my claim.
>
> I use IMAP Partners, who use a Mirapoint server, and that server IS
> configured for an alternate SMTP port--so it's not an inherent limitation
> of the server.  (They use 2525 rather than 587--I do not know whether the
> nonstandard choice of port is Mirapoint's or IMAP Partners'.)

This could be done by redirection of port 2525 traffic to port 587, before 
it reaches the server. I've seen that offered as a solution.

> Dave



-- 
Ian Eiloart
IT Services, University of Sussex
x3148

From frysinsp at jmu.edu  Thu Jul  3 10:45:40 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Thu, 03 Jul 2008 10:45:40 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <3B830E3C91123EAEF61AD81D@lewes.staff.uscs.susx.ac.uk>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
	<F082366BC769FC785D3C8EFE@[192.168.2.101]>
	<3B830E3C91123EAEF61AD81D@lewes.staff.uscs.susx.ac.uk>
Message-ID: <5DE0B89E5A5CD5E60499381C@[192.168.2.101]>

Okay, I've raised that issue to the JMU IT folks to see what they say.

BTW, Mulberry 3 is still up and running, no drops.

Steve

--On Thursday, July 3, 2008 3:37 PM +0100 Ian Eiloart <iane at sussex.ac.uk> 
wrote:

>
>
> --On 3 July 2008 10:14:58 -0400 "Dr. Steven P. Frysinger"
> <frysinsp at jmu.edu> wrote:
>
>> If this is the case, it may not be JMU's fault - I'm not specifying a
>> port, so I guess I'm trying 25 by default. Sebastian suggested overriding
>> that, and I'll try that in the morning when I get to the office.
>
> No, it REALLY is JMU's fault. 25 is the default for most email clients,
> but some modern email clients  (eg the iPhone's mail application) default
> to 587, and most are capable of connecting on 587.
>
> JMU should be listening on 587, but they aren't. I know because I tried
> to connect, and there really is nothing listening there. Could be a
> temporary condition, but JMU's online documentation doesn't mention port
> 587, AND it explicitly says to connect on port 25, AND they're using
> software that doesn't seem to even have the capability to listen there.
>
> RFC 5068 was published in November 2007, defining the practice as best
> common practice. However, rfc2476 -  which defined the message submission
> service - was published in December 1998. In part, this was because
> Internet Service Providers were already blocking port 25. Some sites
> opted for a variety of alternate port numbers to get around this (ours
> used 225, I think) and it was a relief that this was standardised.
> rfc4409 superceded 2476 in April 2006.
>
>> In the meanwhile, further to the issue of getting dropped:
>>
>> My connections had been dropping about every 10-15 minutes. About an hour
>> ago I shut down Mulberry 4 and ran 3, bringing in my latest preferences
>> (so they're configured the same). Haven't been dropped yet. Okay, can't
>> prove the null hypothesis, but if it doesn't drop me in the next few
>> hours, I think that will point to an issue in version 4. Maybe, as
>> someone previously posted, it handles server disconnects a little less
>> robustly.
>>
>> I'll let you know how the experiment goes either way.
>>
>> Steve
>>
>> --On Thursday, July 3, 2008 2:39 PM +0100 Ian Eiloart <iane at sussex.ac.uk>
>> wrote:
>>
>>>
>>>
>>> --On 3 July 2008 07:44:23 -0400 "Dr. Steven P. Frysinger"
>>> <frysinsp at jmu.edu> wrote:
>>>
>>>> Yeah, without being up on the port details I suspected JMU too, but my
>>>> colleagues and students at HTW confirm that they can only use HTW's
>>>> SMTP server to get out - and then only while they're in the building.
>>>> So if they take a laptop back and forth they're forced to switch back
>>>> and forth (or else use a webmail client). But who wants to use
>>>> something besides Mulberry?!? :) The problem apparently starts at UNI-
>>>> Saarbr?cken, with whom they share a network.
>>>>
>>>> This is further substantiated by the fact that I can't get out of HTW
>>>> with my other two SMTP accounts either.
>>>>
>>>> Note, that this doesn't mean JMU isn't ALSO blocking. They do seem to
>>>> have shutdown forwarding from my other accounts, and can't seem to
>>>> figure out how they did it.
>>>
>>> No, you misunderstand. Every site SHOULD block port 25, because that's
>>> the port used to deliver unauthenticated email to other sites. Only the
>>> mail servers on a site should be able to do that. HTW is doing that, and
>>> that means that machines on their site are far less likely to become
>>> spam sources. Good for them!
>>>
>>> Where you have an email account (eg at JMU) you should be permitted to
>>> connect on port 587. JMU simply aren't offering that facility. They
>>> require you to connect on port 25. They ought to be aware that this is
>>> no longer possible from a very large number of sites.
>>>
>>>>
>>>> Everytime I think I need to learn more about network details, I find
>>>> something else to do... :)
>>>>
>>>> Steve
>>>>
>>>> --On July 3, 2008 11:33:49 AM +0100 Ian Eiloart <iane at sussex.ac.uk>
>>>> wrote:
>>>>
>>>>>
>>>>>
>>>>> --On 2 July 2008 16:09:39 -0400 "Dr. Steven P. Frysinger"
>>>>> <frysinsp at jmu.edu> wrote:
>>>>>
>>>>>>
>>>>>> I don't know what my local German university's network is based on,
>>>>>> but it does happen there too. (Of course they have their own
>>>>>> silliness - I have to change SMTP servers when I'm in their
>>>>>> building... :)
>>>>>>
>>>>>
>>>>> That's JMU's fault, as far as I can tell. They use Mirapoint Messaging
>>>>> Server, which doesn't seem to be able to listen on the alternate port
>>>>> 587. I say "seem" because Mirapoint's online documentation is not
>>>>> helpful on this point, but I've found messages on the web that support
>>>>> my claim.
>>>>>
>>>>> It's recognised best practice for institutions to block outbound
>>>>> connections on port 25, so that your machine can't send spam. Your
>>>>> email provider *should* provide a facility that allows you to send
>>>>> only *authenticated* email. See <http://www.ietf.org/rfc/rfc5068.txt>
>>>>> section 3.1
>>>>>
>>>>> --
>>>>> Ian Eiloart
>>>>> IT Services, University of Sussex
>>>>> x3148
>>>>
>>>>
>>>>
>>>> ********************************************
>>>> Steven P. Frysinger, Ph.D.
>>>>
>>>> Professor,
>>>>   Integrated Science and Technology Department
>>>>   Computer Science Department
>>>> Director,
>>>>   Environmental Information Systems Program
>>>>   Environmental Management Program
>>>>   Center for Environment, Health and Safety
>>>> James Madison University
>>>> College of Integrated Science and Technology
>>>> ISAT/CS Room 309
>>>> 701 Carrier Drive - MSC 4102
>>>> Harrisonburg, Virginia 22807
>>>> United States of America
>>>> Tel: 540/568-2710
>>>> Fax: 540/568-2768
>>>> Net: frysinsp at jmu.edu
>>>> Web: http://www.isat.jmu.edu/faculty/frysinger.html
>>>>
>>>> ---------------------
>>>>
>>>> Gastprofessor,
>>>>     Fachbereiche GIS und Maschinenbau
>>>> Environmental Informatics Group (EIG)
>>>> Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
>>>> Goebenstra?e 40
>>>> 66117 Saarbr?cken
>>>> Germany
>>>> Cell:  +49 (0)175 955 2303
>>>> Tel:   +49 (0)681 876 5665
>>>> Net:   Steven.Frysinger at enviromatics.net
>>>>
>>>
>>>
>>>
>>> --
>>> Ian Eiloart
>>> IT Services, University of Sussex
>>> x3148
>>
>>
>>
>> ********************************************
>> Steven P. Frysinger, Ph.D.
>>
>> Professor,
>>   Integrated Science and Technology Department
>>   Computer Science Department
>> Director,
>>   Environmental Information Systems Program
>>   Environmental Management Program
>>   Center for Environment, Health and Safety
>> James Madison University
>> College of Integrated Science and Technology
>> ISAT/CS Room 309
>> 701 Carrier Drive - MSC 4102
>> Harrisonburg, Virginia 22807
>> United States of America
>> Tel: 540/568-2710
>> Fax: 540/568-2768
>> Net: frysinsp at jmu.edu
>> Web: http://www.isat.jmu.edu/faculty/frysinger.html
>>
>> ---------------------
>>
>> Gastprofessor,
>>     Fachbereiche GIS und Maschinenbau
>> Environmental Informatics Group (EIG)
>> Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
>> Goebenstra?e 40
>> 66117 Saarbr?cken
>> Germany
>> Cell:  +49 (0)175 955 2303
>> Tel:   +49 (0)681 876 5665
>> Net:   Steven.Frysinger at enviromatics.net
>>
>
>
>
> --
> Ian Eiloart
> IT Services, University of Sussex
> x3148



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From daboo at mulberrymail.com  Thu Jul  3 16:01:38 2008
From: daboo at mulberrymail.com (Cyrus Daboo)
Date: Thu, 03 Jul 2008 16:01:38 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <F082366BC769FC785D3C8EFE@[192.168.2.101]>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
	<F082366BC769FC785D3C8EFE@[192.168.2.101]>
Message-ID: <7965B99D2D953BD410552B97@caldav.corp.apple.com>

Hi Dr.,

--On July 3, 2008 10:14:58 AM -0400 "Dr. Steven P. Frysinger" 
<frysinsp at jmu.edu> wrote:

> My connections had been dropping about every 10-15 minutes. About an hour
> ago I shut down Mulberry 4 and ran 3, bringing in my latest preferences
> (so they're configured the same). Haven't been dropped yet. Okay, can't
> prove the null hypothesis, but if it doesn't drop me in the next few
> hours, I think that will point to an issue in version 4. Maybe, as
> someone previously posted, it handles server disconnects a little less
> robustly.
>
> I'll let you know how the experiment goes either way.

Turn on protocol logging for Errors using 
<http://trac.mulberrymail.com/mulberry/wiki/logging> (will work in both v3 
and v4). Run both v3 and v4 one after the other for a period of time. Then 
send the log files. Maybe we can determine that the same disconnects are 
happening to both v3 and v4 and that therefore it is a regression in v4 
that is causing your problems.

-- 
Cyrus Daboo


From frysinsp at jmu.edu  Thu Jul  3 17:24:23 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Thu, 03 Jul 2008 17:24:23 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <7965B99D2D953BD410552B97@caldav.corp.apple.com>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
	<F082366BC769FC785D3C8EFE@[192.168.2.101]>
	<7965B99D2D953BD410552B97@caldav.corp.apple.com>
Message-ID: <0547A08307E541996CE0AB9F@[192.168.2.101]>

Okay, I've just started logging v3. BTW, it's been up and running for about 
8 hours now without dropping the connection. I'd be tempted to stick with 
it, but I was just reminded how nice it is to have longer file names... :)

Steve

--On Thursday, July 3, 2008 4:01 PM -0400 Cyrus Daboo 
<daboo at mulberrymail.com> wrote:

> Hi Dr.,
>
> --On July 3, 2008 10:14:58 AM -0400 "Dr. Steven P. Frysinger"
> <frysinsp at jmu.edu> wrote:
>
>> My connections had been dropping about every 10-15 minutes. About an hour
>> ago I shut down Mulberry 4 and ran 3, bringing in my latest preferences
>> (so they're configured the same). Haven't been dropped yet. Okay, can't
>> prove the null hypothesis, but if it doesn't drop me in the next few
>> hours, I think that will point to an issue in version 4. Maybe, as
>> someone previously posted, it handles server disconnects a little less
>> robustly.
>>
>> I'll let you know how the experiment goes either way.
>
> Turn on protocol logging for Errors using
> <http://trac.mulberrymail.com/mulberry/wiki/logging> (will work in both
> v3 and v4). Run both v3 and v4 one after the other for a period of time.
> Then send the log files. Maybe we can determine that the same disconnects
> are happening to both v3 and v4 and that therefore it is a regression in
> v4 that is causing your problems.
>
> --
> Cyrus Daboo
>



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From frysinsp at jmu.edu  Fri Jul  4 03:09:22 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Fri, 04 Jul 2008 03:09:22 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <BB861976C4EFC815859F714C@ninevah-2.local>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
	<F082366BC769FC785D3C8EFE@[192.168.2.101]>
	<7965B99D2D953BD410552B97@caldav.corp.apple.com>
	<6B4135F2C351F038152B241F@steven-frysingers-ibook-g4.local>
	<BB861976C4EFC815859F714C@ninevah-2.local>
Message-ID: <1FFECF8E57B8BD253FDA4F1D@steven-frysingers-ibook-g4.local>

I'm putting this out to the list because others following this might be 
interested. I've asked our IT folks if they've set up the IMAP server to 
drop the connection on a regular basis. It occurs to me that one thing I 
used to notice in v3 is that while the connection didn't go away, on a 
regular basis (10 mins?) the message body of the current email message 
would blank out. My hunch is that this was an artifact of how v3 handled 
the glitch.

Thanks to all for your help in trying to understand this! Unless JMU tells 
me they'll stop doing it (!) I think it's something to live with for awhile.

Steve

--On July 3, 2008 9:18:21 PM -0400 Cyrus Daboo <daboo at mulberrymail.com> 
wrote:

> Hi Dr.,
>
> --On July 3, 2008 6:37:40 PM -0400 "Dr. Steven P. Frysinger"
> <frysinsp at jmu.edu> wrote:
>
>> Okay, here are the two log files. The v3 file was close to an hour
>> without incident (at least visible to me, though there are clearly events
>> in the file). The v4 file is only about 15 minutes long, because that's
>> how long it took to disconnect, so I figured you'd have what you need.
>> But if not let me know.
>
> Well the v3 log file shows that you are being forcibly disconnected about
> every ten minutes. There appears to be some difference in the code paths
> for where these exceptions are being detected and handled. So at this
> point there is clearly a failure in v4 to recover from a "transient"
> network problem. But the fundamental cause here is the 10 disconnects. I
> can't give you any obvious pointers on that. You might want to pull up
> the console.app and look at the system log around about the same time
> that you see the exceptions being thrown in the Mulberry log. That may
> provide some hint as to what is happening.
>
> --
> Cyrus Daboo
>



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From alex at alex.org.uk  Fri Jul  4 04:44:55 2008
From: alex at alex.org.uk (Alex Bligh)
Date: Fri, 04 Jul 2008 09:44:55 +0100
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <1FFECF8E57B8BD253FDA4F1D@steven-frysingers-ibook-g4.local>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
	<F082366BC769FC785D3C8EFE@[192.168.2.101]>
	<7965B99D2D953BD410552B97@caldav.corp.apple.com>
	<6B4135F2C351F038152B241F@steven-frysingers-ibook-g4.local>
	<BB861976C4EFC815859F714C@ninevah-2.local>
	<1FFECF8E57B8BD253FDA4F1D@steven-frysingers-ibook-g4.local>
Message-ID: <7B816246895F9F646127662A@Ximines.local>

Cyrus Daboo wrote:
> Well the v3 log file shows that you are being forcibly disconnected about
> every ten minutes. There appears to be some difference in the code paths
> for where these exceptions are being detected and handled. So at this
> point there is clearly a failure in v4 to recover from a "transient"
> network problem. But the fundamental cause here is the 10 disconnects. I
> can't give you any obvious pointers on that. You might want to pull up
> the console.app and look at the system log around about the same time
> that you see the exceptions being thrown in the Mulberry log. That may
> provide some hint as to what is happening.

Mulberry 4 on the Mac (which the OP is using) is particularly bad at
recovering from network glitches in a way that Mulberry 3 on the PC
and (I think) Mulberry 4 on the PC never was.

However, I've seen the "message sent but not sent" (normally meaning
the message has gone out by SMTP but the file to IMAP's INBOX.Sent
folder did not work so the message did not actually disappear from
screen and a login/logout is necessary to fix) on both Mac and PC
versions of Mulberry 4. I've never put any effort into replicating
it.

Alex

From daboo at mulberrymail.com  Fri Jul  4 14:52:39 2008
From: daboo at mulberrymail.com (Cyrus Daboo)
Date: Fri, 04 Jul 2008 14:52:39 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <1FFECF8E57B8BD253FDA4F1D@steven-frysingers-ibook-g4.local>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
	<F082366BC769FC785D3C8EFE@[192.168.2.101]>
	<7965B99D2D953BD410552B97@caldav.corp.apple.com>
	<6B4135F2C351F038152B241F@steven-frysingers-ibook-g4.local>
	<BB861976C4EFC815859F714C@ninevah-2.local>
	<1FFECF8E57B8BD253FDA4F1D@steven-frysingers-ibook-g4.local>
Message-ID: <C0DAAAE4605A346C3E9E61C2@ninevah-2.local>

Hi,

--On July 4, 2008 3:09:22 AM -0400 "Dr. Steven P. Frysinger" 
<frysinsp at jmu.edu> wrote:

> I'm putting this out to the list because others following this might be
> interested. I've asked our IT folks if they've set up the IMAP server to
> drop the connection on a regular basis. It occurs to me that one thing I
> used to notice in v3 is that while the connection didn't go away, on a
> regular basis (10 mins?) the message body of the current email message
> would blank out. My hunch is that this was an artifact of how v3 handled
> the glitch.
>
> Thanks to all for your help in trying to understand this! Unless JMU
> tells me they'll stop doing it (!) I think it's something to live with
> for awhile.

FYI I found the problem with Mulberry not auto-recovering from these as 
well as it used to: <http://trac.mulberrymail.com/mulberry/ticket/308>.

Its now been fixed on the 4.1 branch. However, there is no publicly 
available build of that right now. I have been using that branch for a 
number of enhancements and fixes mostly related to personal things that I 
need to do (caldav/cardav mostly). This branch only has changes for OS X. I 
can make an build of that available.

At some point I should pull down some of the fixes I have done there to the 
4.0 trunk and then push out 4.0.9 release. At this point though I am really 
not sure when that will happen. There are a number of outstanding bugs in 
trac that need to be addressed. Unfortunately my time to work on that is 
limited.

-- 
Cyrus Daboo


From frysinsp at jmu.edu  Fri Jul  4 15:35:41 2008
From: frysinsp at jmu.edu (Dr. Steven P. Frysinger)
Date: Fri, 04 Jul 2008 15:35:41 -0400
Subject: [Mulberry-discuss] Further to dropped connection
In-Reply-To: <C0DAAAE4605A346C3E9E61C2@ninevah-2.local>
References: <A513556DE4FC02D6989F1AE0@steven-frysingers-ibook-g4.local>
	<A25FC203A8793E138DB42DB2@srb.local> <A597CD4CB3EE4672283E6DB9@G5>
	<51A577772150A211A9464A4C@steven-frysingers-ibook-g4.local>
	<9DF79045007BE3523C5CDBA2@lewes.staff.uscs.susx.ac.uk>
	<9E10E7B797B3DAAD3856DA16@steven-frysingers-ibook-g4.local>
	<8EBC35B6FE9E9FA6C010C715@lewes.staff.uscs.susx.ac.uk>
	<F082366BC769FC785D3C8EFE@[192.168.2.101]>
	<7965B99D2D953BD410552B97@caldav.corp.apple.com>
	<6B4135F2C351F038152B241F@steven-frysingers-ibook-g4.local>
	<BB861976C4EFC815859F714C@ninevah-2.local>
	<1FFECF8E57B8BD253FDA4F1D@steven-frysingers-ibook-g4.local>
	<C0DAAAE4605A346C3E9E61C2@ninevah-2.local>
Message-ID: <FAEA5E99CFFBE02A616D8BAC@steven-frysingers-ibook-g4.local>

Very cool! Let me know if you need a beta tester...

Steve

--On July 4, 2008 2:52:39 PM -0400 Cyrus Daboo <daboo at mulberrymail.com> 
wrote:

> Hi,
>
> --On July 4, 2008 3:09:22 AM -0400 "Dr. Steven P. Frysinger"
> <frysinsp at jmu.edu> wrote:
>
>> I'm putting this out to the list because others following this might be
>> interested. I've asked our IT folks if they've set up the IMAP server to
>> drop the connection on a regular basis. It occurs to me that one thing I
>> used to notice in v3 is that while the connection didn't go away, on a
>> regular basis (10 mins?) the message body of the current email message
>> would blank out. My hunch is that this was an artifact of how v3 handled
>> the glitch.
>>
>> Thanks to all for your help in trying to understand this! Unless JMU
>> tells me they'll stop doing it (!) I think it's something to live with
>> for awhile.
>
> FYI I found the problem with Mulberry not auto-recovering from these as
> well as it used to: <http://trac.mulberrymail.com/mulberry/ticket/308>.
>
> Its now been fixed on the 4.1 branch. However, there is no publicly
> available build of that right now. I have been using that branch for a
> number of enhancements and fixes mostly related to personal things that I
> need to do (caldav/cardav mostly). This branch only has changes for OS X.
> I can make an build of that available.
>
> At some point I should pull down some of the fixes I have done there to
> the 4.0 trunk and then push out 4.0.9 release. At this point though I am
> really not sure when that will happen. There are a number of outstanding
> bugs in trac that need to be addressed. Unfortunately my time to work on
> that is limited.
>
> --
> Cyrus Daboo
>



********************************************
Steven P. Frysinger, Ph.D.

Professor,
  Integrated Science and Technology Department
  Computer Science Department
Director,
  Environmental Information Systems Program
  Environmental Management Program
  Center for Environment, Health and Safety
James Madison University
College of Integrated Science and Technology
ISAT/CS Room 309
701 Carrier Drive - MSC 4102
Harrisonburg, Virginia 22807
United States of America
Tel: 540/568-2710
Fax: 540/568-2768
Net: frysinsp at jmu.edu
Web: http://www.isat.jmu.edu/faculty/frysinger.html

---------------------

Gastprofessor,
    Fachbereiche GIS und Maschinenbau
Environmental Informatics Group (EIG)
Hochschule f?r Technik und Wirtschaft (HTW) des Saarlandes
Goebenstra?e 40
66117 Saarbr?cken
Germany
Cell:  +49 (0)175 955 2303
Tel:   +49 (0)681 876 5665
Net:   Steven.Frysinger at enviromatics.net


From franki at aerodyne.com  Thu Jul 10 15:06:34 2008
From: franki at aerodyne.com (Frank J. Iannarilli)
Date: Thu, 10 Jul 2008 15:06:34 -0400
Subject: [Mulberry-discuss] S/MIME certificate error and workaround? (Cyrus?)
Message-ID: <EE4A4E3DBF231245F7B2FCAC@[192.168.2.3]>

Hi Cyrus (and All),


I'm struggling to get Mulberry (4.0.8 Win32/WinXP) to handle an x509 
certificate from a government colleague.  Another colleague has no trouble 
having that cert work using Thunderbird. I've made some effort the past 2 
days to uncover the root cause so as to stick with Mulberry.  I think I see 
what's going on, and the question, likely to Cyrus, is whether there's an 
easy certificate modification (e.g., using openssl utility) or a "quick" 
doable fix to Mulberry's S/MIME plugin code.


Here's the problem rundown and data:

  1. Upon receiving an email containing this colleague's signature, I first 
deleted any older certificates and also his contact info from the address 
book (as suggested by government support folks). Then upon re-starting 
Mulberry, I open that email and accept/save his certificate as well as 
ingest his email into the address book.

  2. Now on attempting to send him back an encrypted email, upon my 
signing/encrypting, and hitting SEND, Mulberry replies with:
   S/MIME Plugin Error: could not find certificate for: hisemail

Just so you know we're not chasing phantoms or dumb RTFM problems, I've 
been for some time successfully exchanging encrypted emails with other 
government colleagues, including the guy in question here when employing an 
earlier certificate of his.

  3. I dumped his .p7s file and someone else's .p7s file (one that works in 
Mulberry) to see what differences exist, using:
   openssl pkcs7 -print_certs -inform DER -in smime.p7s -text

What I think is the essential difference is that, in the non-working guy's 
PKCS7, he actually has *two*  V3 certificates appearing back-to-back (each 
with a different serial number)  The first one gives his CN (common name), 
without the emailaddress=, and fails to include the "Subject Alternative 
Name" option field that would provide his email address.  The second 
certificate has the same CN, but does provide the aforementioned "Subject 
Alternative Name" (with email). The working-guy's PKCS7 provides 1 
certificate, containing the "Subject Alternative Name" with email.

So my surmise is that Mulberry's function GetEmail(X509* cert, cdstrvect& 
email)(in CSMIMEPluginDLL.*) is getting spoofed by the above situation.

Any feedback appreciated!

Regards,
FRank


Frank J. Iannarilli   franki at aerodyne.com
Aerodyne Research, Inc., 45 Manning Road, Billerica, MA 01821 USA
www.aerodyne.com/cosr/cosr.html

From gessel at blackrosetech.com  Wed Jul 16 16:25:53 2008
From: gessel at blackrosetech.com (David Gessel)
Date: Wed, 16 Jul 2008 13:25:53 -0700
Subject: [Mulberry-discuss] Firefox as handler for text/html
Message-ID: <ACA3AA914C8385D7F7DBC4F6@410C30E69E457113BFD56209>

I tried to configure click actions for text/html to open with firefox, but 
I get an error:

16 bit MS-DOS Subsystem
C:\blah\UNAME~4.EXE
The NTVDM CPU has encountered an illegal instruction
CS:0DED IP:0196 OP: 63 73 73 3e 0d Choose 'Close' to terminate the 
application

Is there a trick to getting it work?

From pwilson at apnic.net  Wed Jul 16 19:45:42 2008
From: pwilson at apnic.net (Paul Wilson)
Date: Thu, 17 Jul 2008 09:45:42 +1000
Subject: [Mulberry-discuss] Firefox as handler for text/html
In-Reply-To: <ACA3AA914C8385D7F7DBC4F6@410C30E69E457113BFD56209>
References: <ACA3AA914C8385D7F7DBC4F6@410C30E69E457113BFD56209>
Message-ID: <269D96D50F2B53E4E0F27474@wav24.apnic.net>

I'd like to know how to set this up on Mac OS as well.

Paul

--On 16 July 2008 1:25:53 PM -0700 David Gessel <gessel at blackrosetech.com> 
wrote:

> I tried to configure click actions for text/html to open with firefox,
> but I get an error:
>
> 16 bit MS-DOS Subsystem
> C:\blah\UNAME~4.EXE
> The NTVDM CPU has encountered an illegal instruction
> CS:0DED IP:0196 OP: 63 73 73 3e 0d Choose 'Close' to terminate the
> application
>
> Is there a trick to getting it work?



From gessel at blackrosetech.com  Sun Jul 20 18:41:38 2008
From: gessel at blackrosetech.com (David Gessel)
Date: Sun, 20 Jul 2008 15:41:38 -0700
Subject: [Mulberry-discuss] Offline/sync enhancement sugggestion
Message-ID: <569783EA4D15D19373288957@bernadinism>

I use mulberry on my laptop and travel frequently which means connecting 
frequently over horrible connections, disconnecting frequently and 
sometimes in a rush and sometimes unexpectedly.

I've found that "connected" mode is hard to manage on the fly because it 
can hang on slow connections at inopportune moments and because of frequent 
user errors of not manually disconnecting before breaking the network 
connection and thus losing access to unsynchronized work.

But "disconnected" mode is tedious because the "synchronize" command is 
buried in the UI and sometimes similar user errors result in not 
synchronizing and messages I thought had gone out don't go until, say, I 
get connected at the next destination.

It would make my life easier if:

- The "Check" button in disconnected mode would execute the last "update 
and disconnect" options used saving digging through a couple of menu items.

- There was a "synchronize all mailboxes" button on the menu bar in 
connected mode.

- There was an "idle time" timer that would "update and disconnect" in 
disconnect mode after some user-specifiable idle time (I understand that 
background synchronization is difficult, though that would be even better).

- The same "idle time" timer would "synchronize all mailboxes" in connected 
mode.

I've also found that some data isn't synchronized between the local and 
server copies - the symptom is that I searched for a message I was pretty 
sure I'd received and couldn't find it in disconnected mode, but could in 
connected mode despite repeated synchronization of the mailbox.  A complete 
message by message resync or "forced local mail cache rebuild" would also 
be helpful.

-David

From mulberrymail.com at alker.net  Thu Jul 24 00:03:01 2008
From: mulberrymail.com at alker.net (Ken Alker)
Date: Wed, 23 Jul 2008 21:03:01 -0700
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
Message-ID: <35311DE191350C86F4F2AA3F@[10.0.0.198]>

We are switching from a Cyrus server to Zimbra, but I want to continue 
using Mulberry as my email client.  I've been using it for several years, 
and I like it very much.  Unfortunately, we are running into a problem that 
is making Mulberry unusable with Zimbra.

Any time I open a mailbox that was last sorted by anything other than the 
message number, I get a Warning window stating: "Some information needed to 
sort all the messages in unavailable.  Do you want to download the missing 
message information and then do a full sort, or just sort the currently 
available messages?".

Choosing "Full Sort" can take several minutes to run on a large box (mine 
are often 5,000-20,000 messages in length).  If I only had to deal with 
this once in a great moon, it would not be a big deal, but any time the 
mailbox is closed and then reopened this same Warning window pops back up 
forcing yet another "Full Sort".  If a "Current Sort" had been selected 
first, then trying to sort by another column would result in the Warning 
dialog reappearing.  Once a "Full Sort" is done on any given column within 
the mailbox, the Warning window does not appear again when sorting by 
another header (unless, the mailbox is closed and re-opened).

The process that is taking a long time after selecting "Full Sort" isn't 
the sort itself, it is "Fetching message information..." (per the status 
information at the bottom left in Mulberry that includes an incrementing 
counter).

Periodically, my Inbox simply begins re-fetching message information with 
no warning, causing the long delay.  I've not pinned this down to any 
particular periodicity, but it seems to happen about once per hour.  I 
don't know if this is due to something happening on the server, or possibly 
the appearance of new email in my Inbox, or something else.

Performing a search occasionally causes re-fetching.

When using Mulberry with the Cyrus server, this Warning would pop up from 
time to time, but it was extremely rare.  In fact, I think it would only 
require a sort the first time you sorted a box that had never been sorted 
previously, and then (almost) never pop up again.

As an aside, we recently installed NGINX (proxy) in front of our current 
Cyrus server and I re-pointed Mulberry to use NGINX and found that I had 
the same problem.  We are not using NGINX with Zimbra (even though it comes 
with it).

Does anyone know why this might be happening?

Has anyone else experienced this problem?

Is anyone else who has large mailboxes and is using Mulberry with Zimbra 
NOT has this problem?

If no one has a quick answer to this question, I'd sure some guidance in 
debugging the problem.  I have acces to the servers as well as the client.

Thank you!
Ken

/**************************************************************
 Ken Alker         ken at impulse.net            ham radio: KA6SDU
 Impulse Internet Services               http://www.impulse.net
 Santa Barbara,  San Luis Obispo,  Ventura, Los Angeles, Orange
 T-3 / T-1 / ADSL / ISDN / 56K / web hosting / wireless / co-lo
***************************************************************/

From quanah at fast-mail.org  Thu Jul 24 00:21:23 2008
From: quanah at fast-mail.org (Quanah Gibson-Mount)
Date: Wed, 23 Jul 2008 21:21:23 -0700
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
Message-ID: <D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>

--On Wednesday, July 23, 2008 9:03 PM -0700 Ken Alker 
<mulberrymail.com at alker.net> wrote:

> Does anyone know why this might be happening?

I'll raise some awareness on it.

> Has anyone else experienced this problem?

I just did a quick test.  Mailbox with 500 messages it didn't occur, with 
1000, it did.  Sorting that 1000 entry mailbox in other ways continued to 
work fine until closing that tab and reopening it, and then the message 
came up again.

> Is anyone else who has large mailboxes and is using Mulberry with Zimbra
> NOT has this problem?

I always see it on large mailboxes with Zimbra, but it hasn't been 
particularly slow for me.  But I rarely sort by anything other than order 
received.

> If no one has a quick answer to this question, I'd sure some guidance in
> debugging the problem.  I have acces to the servers as well as the client.

I have direct access to the engineers since I'm a Zimbra employee. Let me 
know if you find something, as I can easily pass it on. ;)

--Quanah



From quanah at fast-mail.org  Thu Jul 24 00:25:45 2008
From: quanah at fast-mail.org (Quanah Gibson-Mount)
Date: Wed, 23 Jul 2008 21:25:45 -0700
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
Message-ID: <404C6A69D0E16AB287CF7DE0@[192.168.1.199]>

--On Wednesday, July 23, 2008 9:21 PM -0700 Quanah Gibson-Mount 
<quanah at fast-mail.org> wrote:

> --On Wednesday, July 23, 2008 9:03 PM -0700 Ken Alker
> <mulberrymail.com at alker.net> wrote:
>
>> Does anyone know why this might be happening?
>
> I'll raise some awareness on it.
>
>> Has anyone else experienced this problem?
>
> I just did a quick test.  Mailbox with 500 messages it didn't occur, with
> 1000, it did.  Sorting that 1000 entry mailbox in other ways continued to
> work fine until closing that tab and reopening it, and then the message
> came up again.

Actually, I believe it occurs with any number of messages, it just doesn't 
throw a warning until I exceed 1000 messages or so.

--Quanah

From digant at stanford.edu  Thu Jul 24 01:57:53 2008
From: digant at stanford.edu (Digant C Kasundra)
Date: Wed, 23 Jul 2008 22:57:53 -0700
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
Message-ID: <0015FC8339F4CF6A87815014@[192.168.0.178]>

--On Wednesday, July 23, 2008 9:21 PM -0700 Quanah Gibson-Mount 
<quanah at fast-mail.org> wrote:

>> If no one has a quick answer to this question, I'd sure some guidance in
>> debugging the problem.  I have acces to the servers as well as the
>> client.
>
> I have direct access to the engineers since I'm a Zimbra employee. Let me
> know if you find something, as I can easily pass it on. ;)
>

I've noticed that there are times when I do search (for instance, if I pick 
Correspondant from the quick search drop down and search for someone), i 
don't get all the messages.  I've noticed it not show me messages that I 
know are there and I can see when I do the same search on the Zimbra 
webapp.  I'll do more testing to see if I can craft a repeatable scenario.



From quanah at fast-mail.org  Thu Jul 24 02:00:39 2008
From: quanah at fast-mail.org (Quanah Gibson-Mount)
Date: Wed, 23 Jul 2008 23:00:39 -0700
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <0015FC8339F4CF6A87815014@[192.168.0.178]>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
	<0015FC8339F4CF6A87815014@[192.168.0.178]>
Message-ID: <06B22589D42B9E54B438E798@[192.168.1.199]>

--On Wednesday, July 23, 2008 10:57 PM -0700 Digant C Kasundra 
<digant at stanford.edu> wrote:

> --On Wednesday, July 23, 2008 9:21 PM -0700 Quanah Gibson-Mount
> <quanah at fast-mail.org> wrote:
>
>>> If no one has a quick answer to this question, I'd sure some guidance in
>>> debugging the problem.  I have acces to the servers as well as the
>>> client.
>>
>> I have direct access to the engineers since I'm a Zimbra employee. Let me
>> know if you find something, as I can easily pass it on. ;)
>>
>
> I've noticed that there are times when I do search (for instance, if I
> pick Correspondant from the quick search drop down and search for
> someone), i don't get all the messages.  I've noticed it not show me
> messages that I know are there and I can see when I do the same search on
> the Zimbra webapp.  I'll do more testing to see if I can craft a
> repeatable scenario.

The way it does searches internally is definitely pickier than cyrus-sasl, 
I've noticed that.

--Quanah





From Hagedorn at uni-koeln.de  Thu Jul 24 04:36:02 2008
From: Hagedorn at uni-koeln.de (Sebastian Hagedorn)
Date: Thu, 24 Jul 2008 10:36:02 +0200
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
Message-ID: <E5DCB521D55BD2992F570216@tyrion.rrz.uni-koeln.de>

Hi,

--On 23. Juli 2008 21:03:01 -0700 Ken Alker <mulberrymail.com at alker.net> 
wrote:

> We are switching from a Cyrus server to Zimbra, but I want to continue
> using Mulberry as my email client.  I've been using it for several years,
> and I like it very much.  Unfortunately, we are running into a problem
> that is making Mulberry unusable with Zimbra.
>
> Any time I open a mailbox that was last sorted by anything other than the
> message number, I get a Warning window stating: "Some information needed
> to sort all the messages in unavailable.  Do you want to download the
> missing message information and then do a full sort, or just sort the
> currently available messages?".

you can disable that warning in the preferences:

Preferences -> Mailbox -> Advanced -> Message Display Cache

If you set that to "Unlimited Total" you won't be warned again.

> Choosing "Full Sort" can take several minutes to run on a large box (mine
> are often 5,000-20,000 messages in length).

That indicates that Zimbra doesn't support the SORT extension. Could you do 
the following, please?

Select the server in Mulberry and click on the Details icon (a bubble with 
an 'i' inside). Copy the Capability list and post it here.

> Does anyone know why this might be happening?

My guess is that Zimbra doesn't allow server-side sorting. So Mulberry has 
to download all the relevant headers to be able to sort.
-- 
     .:.Sebastian Hagedorn - RZKR-R1 (Geb?ude 52), Zimmer 18.:.
Zentrum f?r angewandte Informatik - Universit?tsweiter Service RRZK
.:.Universit?t zu K?ln / Cologne University - ? +49-221-478-5587.:.
                   .:.:.:.Skype: shagedorn.:.:.:.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 186 bytes
Desc: not available
Url : http://lists.mulberrymail.com/pipermail/mulberry-discuss/attachments/20080724/125e1e38/attachment-0001.bin 

From alex at alex.org.uk  Thu Jul 24 04:43:41 2008
From: alex at alex.org.uk (Alex Bligh)
Date: Thu, 24 Jul 2008 09:43:41 +0100
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <E5DCB521D55BD2992F570216@tyrion.rrz.uni-koeln.de>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<E5DCB521D55BD2992F570216@tyrion.rrz.uni-koeln.de>
Message-ID: <99886FF871B6CB299A3551FD@Ximines.local>



--On 24 July 2008 10:36:02 +0200 Sebastian Hagedorn <Hagedorn at uni-koeln.de> 
wrote:

> My guess is that Zimbra doesn't allow server-side sorting. So Mulberry
> has to download all the relevant headers to be able to sort.

Gosh, that would be an end to my plan to migrate to Zimbra. Please do
post the capability list.

Alex

From D.Nash at its.utexas.edu  Thu Jul 24 09:35:50 2008
From: D.Nash at its.utexas.edu (Donald Nash)
Date: Thu, 24 Jul 2008 08:35:50 -0500
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <E5DCB521D55BD2992F570216@tyrion.rrz.uni-koeln.de>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<E5DCB521D55BD2992F570216@tyrion.rrz.uni-koeln.de>
Message-ID: <8D5864073E73D2759665A285@cpe-70-116-77-75.austin.res.rr.com>

--On July 24, 2008 10:36:02 AM +0200 Sebastian Hagedorn 
<Hagedorn at uni-koeln.de> wrote:

> That indicates that Zimbra doesn't support the SORT extension.

Yep, that's it.  I went to <http://bugzilla.zimbra.com/> and searched for 
"IMAP SORT".  I found a bug entitled "Provide IMAP SORT capability," and it 
is still outstanding.  See here:

    <http://bugzilla.zimbra.com/show_bug.cgi?id=12491>.

-- 
Donald L. Nash, <D.Nash at its.utexas.edu>
Information Technology Services, The University of Texas at Austin

From daboo at mulberrymail.com  Thu Jul 24 09:46:39 2008
From: daboo at mulberrymail.com (Cyrus Daboo)
Date: Thu, 24 Jul 2008 09:46:39 -0400
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
Message-ID: <17518839F02DC04E72BBF7B1@caldav.corp.apple.com>

Hi Quanah,

--On July 23, 2008 9:21:23 PM -0700 Quanah Gibson-Mount 
<quanah at fast-mail.org> wrote:

>> If no one has a quick answer to this question, I'd sure some guidance in
>> debugging the problem.  I have acces to the servers as well as the
>> client.
>
> I have direct access to the engineers since I'm a Zimbra employee. Let me
> know if you find something, as I can easily pass it on. ;)

Please ask them to support the SORT and THREAD extensions as defined in 
RFC5256. That will allow clients to use server-side sorting/threading which 
will be much more efficient for non-caching clients like Mulberry and 
webmail solutions.

-- 
Cyrus Daboo


From quanah at fast-mail.org  Thu Jul 24 11:13:45 2008
From: quanah at fast-mail.org (Quanah Gibson-Mount)
Date: Thu, 24 Jul 2008 08:13:45 -0700
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <17518839F02DC04E72BBF7B1@caldav.corp.apple.com>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
	<17518839F02DC04E72BBF7B1@caldav.corp.apple.com>
Message-ID: <CB667F7C904078C7FCDE4606@[192.168.1.199]>

--On Thursday, July 24, 2008 9:46 AM -0400 Cyrus Daboo 
<daboo at mulberrymail.com> wrote:

> Hi Quanah,
>
> --On July 23, 2008 9:21:23 PM -0700 Quanah Gibson-Mount
> <quanah at fast-mail.org> wrote:
>
>>> If no one has a quick answer to this question, I'd sure some guidance in
>>> debugging the problem.  I have acces to the servers as well as the
>>> client.
>>
>> I have direct access to the engineers since I'm a Zimbra employee. Let me
>> know if you find something, as I can easily pass it on. ;)
>
> Please ask them to support the SORT and THREAD extensions as defined in
> RFC5256. That will allow clients to use server-side sorting/threading
> which will be much more efficient for non-caching clients like Mulberry
> and webmail solutions.

Yeah, I'll pass this on.

--Quanah




From quanah at fast-mail.org  Thu Jul 24 12:46:58 2008
From: quanah at fast-mail.org (Quanah Gibson-Mount)
Date: Thu, 24 Jul 2008 09:46:58 -0700
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <CB667F7C904078C7FCDE4606@[192.168.1.199]>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
	<17518839F02DC04E72BBF7B1@caldav.corp.apple.com>
	<CB667F7C904078C7FCDE4606@[192.168.1.199]>
Message-ID: <59F29BDFE43DC5AC53C74262@[192.168.1.199]>

--On Thursday, July 24, 2008 8:13 AM -0700 Quanah Gibson-Mount 
<quanah at fast-mail.org> wrote:

>> Please ask them to support the SORT and THREAD extensions as defined in
>> RFC5256. That will allow clients to use server-side sorting/threading
>> which will be much more efficient for non-caching clients like Mulberry
>> and webmail solutions.

One problem appears to be on the Mulberry side -- It fails to update the 
capabilities list after authing.  For example, what Mulberry reports as my 
capability list with Zimbra is:

IMAP4rev1 ID LITERAL+ SASL-IR AUTH=PLAIN

But if you do this manually, you see the capabilities are different after 
login is successful:

a01 capability
* CAPABILITY IMAP4rev1 ID LITERAL+ SASL-IR AUTH=PLAIN
a01 OK completed
a02 login XXXXX YYYYYYYYY
a02 OK [CAPABILITY IMAP4rev1 ACL BINARY CATENATE CHILDREN CONDSTORE ENABLE 
ESEARCH ID IDLE LIST-EXTENDED LITERAL+ LOGIN-REFERRALS MULTIAPPEND 
NAMESPACE QRESYNC QUOTA RIGHTS=ektx SASL-IR SEARCHRES UIDPLUS UNSELECT 
WITHIN] LOGIN completed
a03 capability
* CAPABILITY IMAP4rev1 ACL BINARY CATENATE CHILDREN CONDSTORE ENABLE 
ESEARCH ID IDLE LIST-EXTENDED LITERAL+ LOGIN-REFERRALS MULTIAPPEND 
NAMESPACE QRESYNC QUOTA RIGHTS=ektx SASL-IR SEARCHRES UIDPLUS UNSELECT 
WITHIN
a03 OK CAPABILITY completed

--Quanah



From quanah at fast-mail.org  Thu Jul 24 14:17:51 2008
From: quanah at fast-mail.org (Quanah Gibson-Mount)
Date: Thu, 24 Jul 2008 11:17:51 -0700
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <59F29BDFE43DC5AC53C74262@[192.168.1.199]>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
	<17518839F02DC04E72BBF7B1@caldav.corp.apple.com>
	<CB667F7C904078C7FCDE4606@[192.168.1.199]>
	<59F29BDFE43DC5AC53C74262@[192.168.1.199]>
Message-ID: <3891C578B461F2B8E28F6B72@[192.168.1.199]>

--On Thursday, July 24, 2008 9:46 AM -0700 Quanah Gibson-Mount 
<quanah at fast-mail.org> wrote:

> --On Thursday, July 24, 2008 8:13 AM -0700 Quanah Gibson-Mount
> <quanah at fast-mail.org> wrote:
>
>>> Please ask them to support the SORT and THREAD extensions as defined in
>>> RFC5256. That will allow clients to use server-side sorting/threading
>>> which will be much more efficient for non-caching clients like Mulberry
>>> and webmail solutions.

Also, is Mulberry happy with THREAD=ORDEREDSUBJECT, or does it also require 
THREAD=REFERENCES?

--Quanah

From daboo at mulberrymail.com  Thu Jul 24 21:37:04 2008
From: daboo at mulberrymail.com (Cyrus Daboo)
Date: Thu, 24 Jul 2008 21:37:04 -0400
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <3891C578B461F2B8E28F6B72@[192.168.1.199]>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
	<17518839F02DC04E72BBF7B1@caldav.corp.apple.com>
	<CB667F7C904078C7FCDE4606@[192.168.1.199]>
	<59F29BDFE43DC5AC53C74262@[192.168.1.199]>
	<3891C578B461F2B8E28F6B72@[192.168.1.199]>
Message-ID: <73561396EBC65A7E73631251@ninevah.local>

Hi Quanah,

--On July 24, 2008 11:17:51 AM -0700 Quanah Gibson-Mount 
<quanah at fast-mail.org> wrote:

>>>> Please ask them to support the SORT and THREAD extensions as defined in
>>>> RFC5256. That will allow clients to use server-side sorting/threading
>>>> which will be much more efficient for non-caching clients like Mulberry
>>>> and webmail solutions.
>
> Also, is Mulberry happy with THREAD=ORDEREDSUBJECT, or does it also
> require THREAD=REFERENCES?

I think it does use ORDEREDSUBJECT if that is the only one present, but 
really people expect the REFERENCES behavior.

-- 
Cyrus Daboo


From iane at sussex.ac.uk  Fri Jul 25 05:43:52 2008
From: iane at sussex.ac.uk (Ian Eiloart)
Date: Fri, 25 Jul 2008 10:43:52 +0100
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <59F29BDFE43DC5AC53C74262@[192.168.1.199]>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
	<17518839F02DC04E72BBF7B1@caldav.corp.apple.com>
	<CB667F7C904078C7FCDE4606@[192.168.1.199]>
	<59F29BDFE43DC5AC53C74262@[192.168.1.199]>
Message-ID: <48DF788E77DACF24C4D30E76@lewes.staff.uscs.susx.ac.uk>



--On 24 July 2008 09:46:58 -0700 Quanah Gibson-Mount <quanah at fast-mail.org> 
wrote:

> --On Thursday, July 24, 2008 8:13 AM -0700 Quanah Gibson-Mount
> <quanah at fast-mail.org> wrote:
>
>>> Please ask them to support the SORT and THREAD extensions as defined in
>>> RFC5256. That will allow clients to use server-side sorting/threading
>>> which will be much more efficient for non-caching clients like Mulberry
>>> and webmail solutions.
>
> One problem appears to be on the Mulberry side -- It fails to update the
> capabilities list after authing.

Yes, we found this problem when we deployed Perdition IMAP proxy, with two 
different back ends - during a transition from UoW IMAP to Cyrus IMAP.

We had to publish only the IMAP extensions that both back ends supported. 
Mainly because Mulberry (our officially supported client) didn't check 
CAPABILITY after authentication, and would crash if some initially 
advertised extensions were not actually available. If I remember rightly, 
attempting to modify ACLs on mailboxes caused the problem.


-- 
Ian Eiloart
IT Services, University of Sussex
x3148

From quanah at fast-mail.org  Fri Jul 25 14:45:44 2008
From: quanah at fast-mail.org (Quanah Gibson-Mount)
Date: Fri, 25 Jul 2008 11:45:44 -0700
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <48DF788E77DACF24C4D30E76@lewes.staff.uscs.susx.ac.uk>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<D8608E6BD8E4F84D9E36CE5C@[192.168.1.199]>
	<17518839F02DC04E72BBF7B1@caldav.corp.apple.com>
	<CB667F7C904078C7FCDE4606@[192.168.1.199]>
	<59F29BDFE43DC5AC53C74262@[192.168.1.199]>
	<48DF788E77DACF24C4D30E76@lewes.staff.uscs.susx.ac.uk>
Message-ID: <C65CEA6C07B7B66CB97DDBF3@[192.168.1.199]>

We've hit the same issue with Thunderbird as well, just as a side note.

--Quanah

--On Friday, July 25, 2008 10:43 AM +0100 Ian Eiloart <iane at sussex.ac.uk> 
wrote:

>
>
> --On 24 July 2008 09:46:58 -0700 Quanah Gibson-Mount
> <quanah at fast-mail.org> wrote:
>
>> --On Thursday, July 24, 2008 8:13 AM -0700 Quanah Gibson-Mount
>> <quanah at fast-mail.org> wrote:
>>
>>>> Please ask them to support the SORT and THREAD extensions as defined in
>>>> RFC5256. That will allow clients to use server-side sorting/threading
>>>> which will be much more efficient for non-caching clients like Mulberry
>>>> and webmail solutions.
>>
>> One problem appears to be on the Mulberry side -- It fails to update the
>> capabilities list after authing.
>
> Yes, we found this problem when we deployed Perdition IMAP proxy, with
> two different back ends - during a transition from UoW IMAP to Cyrus IMAP.
>
> We had to publish only the IMAP extensions that both back ends supported.
> Mainly because Mulberry (our officially supported client) didn't check
> CAPABILITY after authentication, and would crash if some initially
> advertised extensions were not actually available. If I remember rightly,
> attempting to modify ACLs on mailboxes caused the problem.
>
>
> --
> Ian Eiloart
> IT Services, University of Sussex
> x3148





From quanah at fast-mail.org  Mon Jul 28 14:26:30 2008
From: quanah at fast-mail.org (Quanah Gibson-Mount)
Date: Mon, 28 Jul 2008 11:26:30 -0700
Subject: [Mulberry-discuss] using Mulberry with Zimbra;
 constant requirement to perform Full Sort
In-Reply-To: <8D5864073E73D2759665A285@cpe-70-116-77-75.austin.res.rr.com>
References: <35311DE191350C86F4F2AA3F@[10.0.0.198]>
	<E5DCB521D55BD2992F570216@tyrion.rrz.uni-koeln.de>
	<8D5864073E73D2759665A285@cpe-70-116-77-75.austin.res.rr.com>
Message-ID: <D64DB36B3BAF8170E864DC36@[192.168.1.199]>

--On Thursday, July 24, 2008 8:35 AM -0500 Donald Nash 
<D.Nash at its.utexas.edu> wrote:

> --On July 24, 2008 10:36:02 AM +0200 Sebastian Hagedorn
> <Hagedorn at uni-koeln.de> wrote:
>
>> That indicates that Zimbra doesn't support the SORT extension.
>
> Yep, that's it.  I went to <http://bugzilla.zimbra.com/> and searched for
> "IMAP SORT".  I found a bug entitled "Provide IMAP SORT capability," and
> it is still outstanding.  See here:
>
>     <http://bugzilla.zimbra.com/show_bug.cgi?id=12491>.

<http://bugzilla.zimbra.com/show_bug.cgi?id=30285>

for THREAD=ORDEREDSUBJECT tracking.

Both will be in the ZCS 5.5 release.

--Quanah