Re: [Mailman-Users] Trouble with some messages being scrubbed

2009-08-10 Thread Mark Sapiro
Scott Race wrote:
>
>Would the workaround you describe involve setting:
>
>convert_html_to_plaintext = 1
>and
>filter_mime_types = [add html/text]
>?


Not exactly. the workaround may be to just add text/html to
filter_mime_types, but it depends if there is anything in
pass_mime_types. If you have the default setting of

multipart/mixed
multipart/alternative
text/plain

in pass_mime types, you will already remove everything that isn't
text/plain if you just turn on content filtering. Also, in 2.1.5, even
if both filter_mime_types and pass_mime_types are empty, meaning
nothing gets removed, a multipart/alternative part will still get
replaced by it's first alternative (this is a setting -
collapse_alternatives - in later versions, but in 2.1.5 it is always
done).

Thus, to do the minimal filtering that will avoid this bug, set

filter_content = 1
filter_mime_types = []
pass_mime_types = []
convert_html_to_plaintext = 0


>One other question about upgrade/new install - I have read that upgrading an 
>existing installation is rather difficult and might involve downtime, so the 
>preferred method is to bring up a new server and move the lists over.  Is that 
>correct?


Upgrading from 2.1.5 to any more recent version is easy. It is much
easier than bringing up a new server and moving lists. There are
potential caveats regarding Python versions. See the FAQ at
. Other than that, it depends how your
original Mailman was installed. If from source, the upgrade process is
to run configure with the same options as before (can be found in
config.log if you still have the original) and make. Then if all went
OK, stop Mailman, backup your current installation just in case, run
make install and start Mailman. I've done this dozens of times on a
production system without problems. It only takes a few minutes.

Upgrading from a packege should also be easy, but the details depend on
the package.

-- 
Mark Sapiro The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://wiki.list.org/x/AgA3
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: http://wiki.list.org/x/QIA9


Re: [Mailman-Users] Trouble with some messages being scrubbed

2009-08-09 Thread Scott Race
Thanks for confirming Mark. I had read that this was a possible bug - so good 
to know that an upgrade/new install will likely cure it.

Would the workaround you describe involve setting:

convert_html_to_plaintext = 1
and
filter_mime_types = [add html/text]
?

One other question about upgrade/new install - I have read that upgrading an 
existing installation is rather difficult and might involve downtime, so the 
preferred method is to bring up a new server and move the lists over.  Is that 
correct?

Thanks again.
Scott

-Original Message-
From: Mark Sapiro [mailto:m...@msapiro.net] 
Sent: Sunday, August 09, 2009 9:30 PM
To: Scott Race; mailman-users@python.org
Subject: Re: [Mailman-Users] Trouble with some messages being scrubbed

Scott Race wrote:

>I am getting the following message in some of our posts, no content, just:
>
>Skipped content of type multipart/alternative-- next part 
>--


This is a bug in your 2.1.5 version, long since fixed.


>A non-text attachment was scrubbed...
>Name: not available
>Type: image/gif
>Size: 1994 bytes
>Desc: image003.gif
>There's a link to the attachment, image003.gif, which pulls up ok, but, no 
>content is being delivered.
>
>The list settings have filter_content set to No.


This is a scrubber issue. It is independent of content filtering except
in that if you filtered the text/html part out of the
multipart/alternative part leaving only the text/plain part, it would
probably appear OK in the archive.


>I read in another post that scrub_nondigest might be set to Yes, but I can't 
>find the setting for that.


Because it doesn't exist in 2.1.5, and if it were set in your case, it
would just break individual messages and mime digests as well as
archives and plain digests.

>Mailman 2.1.5.


If you are saying that this happens to individual messages and not to
just plain digests and archives, then this is not 'standard' 2.1.5,
and somehow Scrubber has been inserted into GLOBAL_PIPELINE either in
Defaults.py or mm_cfg.py or in a pipeline attribute of the list.

The real solution is to upgrade to a more recent Mailman that doesn't
have this Scrubber bug.

-- 
Mark Sapiro The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://wiki.list.org/x/AgA3
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: http://wiki.list.org/x/QIA9


Re: [Mailman-Users] Trouble with some messages being scrubbed

2009-08-09 Thread Mark Sapiro
Scott Race wrote:

>I am getting the following message in some of our posts, no content, just:
>
>Skipped content of type multipart/alternative-- next part 
>--


This is a bug in your 2.1.5 version, long since fixed.


>A non-text attachment was scrubbed...
>Name: not available
>Type: image/gif
>Size: 1994 bytes
>Desc: image003.gif
>There's a link to the attachment, image003.gif, which pulls up ok, but, no 
>content is being delivered.
>
>The list settings have filter_content set to No.


This is a scrubber issue. It is independent of content filtering except
in that if you filtered the text/html part out of the
multipart/alternative part leaving only the text/plain part, it would
probably appear OK in the archive.


>I read in another post that scrub_nondigest might be set to Yes, but I can't 
>find the setting for that.


Because it doesn't exist in 2.1.5, and if it were set in your case, it
would just break individual messages and mime digests as well as
archives and plain digests.

>Mailman 2.1.5.


If you are saying that this happens to individual messages and not to
just plain digests and archives, then this is not 'standard' 2.1.5,
and somehow Scrubber has been inserted into GLOBAL_PIPELINE either in
Defaults.py or mm_cfg.py or in a pipeline attribute of the list.

The real solution is to upgrade to a more recent Mailman that doesn't
have this Scrubber bug.

-- 
Mark Sapiro The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://wiki.list.org/x/AgA3
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: http://wiki.list.org/x/QIA9


[Mailman-Users] Trouble with some messages being scrubbed

2009-08-09 Thread Scott Race
I am getting the following message in some of our posts, no content, just:

Skipped content of type multipart/alternative-- next part 
--
A non-text attachment was scrubbed...
Name: not available
Type: image/gif
Size: 1994 bytes
Desc: image003.gif
There's a link to the attachment, image003.gif, which pulls up ok, but, no 
content is being delivered.

The list settings have filter_content set to No.

I read in another post that scrub_nondigest might be set to Yes, but I can't 
find the setting for that.  How would i find it, or where is the config file 
that holds this setting?   I looked and greped, couldn't find

Mailman 2.1.5.

Thanks!
Scott
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://wiki.list.org/x/AgA3
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: http://wiki.list.org/x/QIA9