Issue Migration taking exponentially long because of listmng

[tkalfaoglu]

Today I am migrating an Onyx server to another new Onyx Server.

The main holdup is the listmng processing. Mailman lists with thousands of members are being added to the new server one by one.
# ps ax|grep listmng
33004 ? S 0:00 /usr/local/psa/admin/bin/listmng add-member bla_bla bla.info [email protected]

two seconds later, same list, another user.
No wonder the migration has been going on for 8 hours now!

This needs to be done in bulk -- not one user at a time. This might be alright for small mailman installations but I have many lists with TENS OF THOUSANDS users.

 

[tkalfaoglu]

Furthermore, it sends a "[email protected] was added to the list" to the admin, AND "welcome to the list!" to [email protected] to EACH member.
I don't think these are necessary/useful during a migration.

Finally, the THIRD glitch with mailman: NOQUEUE: reject: RCPT from localhost[::1]: 454 4.7.1 <[email protected]>: Relay access denied;
for every mail being sent out. Apparently 127.0.0.1 does not get automatically whitelisted when PLESK is installed.

 

[tkalfaoglu]

What I did, is that I hacked into add_members program of mailman and DISABLED both mails from getting sent out..
I will restore it when the migration is complete.. I am also going to whitelist 127.0.0.1 now using the PLESK interface. I didn't do that before because I did not want tens of thousands of "welcome to the list" emails being sent out.

 

[tkalfaoglu]

Followup: the lists migration never went correctly. So I followed PLESK guides to manually copy the lists over.
Likewise DNS migration also caused problems, and it was not until I clicked on APPLY TEMPLATE at the new server for each domain that those domains started working..