[Mailman-Users] Manually setting digest version and/or number and other questions

[Mark Sapiro]

On 02/10/2014 04:29 PM, George Kasica wrote:
> I've got an existing list that I just moved to mailman and I'm seeing the digests start over obviously at v1 n1. Is there any way to set those values somehow so I don't end up duplicating things?


Use bin/withlist or bin/config_list or the script at 
<http://www.msapiro.net/scripts/set_attributes> to set the list's volume 
attribute to the desired volume and the list's next_digest_number to the 
desired number.


> Also related to that when a user lipoma at the archive by date, thread, etc. all the mail is there. If he downloads the compressed file he's telling me it only has about 4-5 messages. Haven't had time to
> Investigate that one for sure yet but was curious what the archive process is like. Is it run once a day or??


When the .txt.gz files are updated depends on the setting of 
GZIP_ARCHIVE_TXT_FILES (default = No) and the running by cron of 
Mailman's cron/nightly_gzip.

If GZIP_ARCHIVE_TXT_FILES is set to yes in mm_cfg.py, the 
archives/private/LISTNAME/<period>.txt file will be gzipped with every 
list post. This is very inefficient. Our official advice in Defaults.py is

# Set this to Yes to enable gzipping of the downloadable archive .txt file.
# Note that this is /extremely/ inefficient, so an alternative is to just
# collect the messages in the associated .txt file and run a cron job every
# night to generate the txt.gz file.  See cron/nightly_gzip for details.
GZIP_ARCHIVE_TXT_FILES = No

My recommendation is to remove the cron/nightly_gzip entry from 
Mailman's crontab, and remove all the 
archives/private/LISTNAME/<period>.txt.gz files. Then the archive index 
will link to the .txt files which are always up to date. This actually 
saves space because the .txt.gz files are in addition to the .txt files 
which are always there. The only possible saving with the .gz files is 
in the serving of a smaller file from the web, but some web servers will 
decompress the file before sending it anyway, and the saving if any is 
minimal.


> Lastly is there a way to scrub either email or user name out of the digest/archive process to prevent harvesting?


Why are digest's a concern more than posts?

The only control is ARCHIVER_OBSCURES_EMAILADDRS which only replaces '@' 
with ' at ' in the archives and which is Yes by default anyway.

If you or your users are really concerned about harvesting, a better 
solution is to make the archive private so it is only accessible to list 
members.

-- 
Mark Sapiro <mark at msapiro.net>        The highway is for gamblers,
San Francisco Bay Area, California    better use your sense - B. Dylan