I use mailman with kubernetees, and I have four pods.
I currently use persistentVolumeClaim for two folders in the mailman configuration "cache" and "messages". I share these folders, as mailman needs to be consistent for these two folders, but I'm wondering if I should share other folders as well, like locks and such.
There are all the folders in mailman archives cache data etc lists locks logs master.pid messages queue templates
I ask this question because I often have a problem when creating a mailing list. I think I have a problem with the locks, but I'm not sure if the problem is due to the fact that the locks folder is not shared.
If you want more details, here's the trace I sometimes get, when someone tries to create a mailing list:
File "falcon/app.py", line 365, in falcon.app.App.__call__ File "/usr/lib/python3.12/site-packages/mailman/rest/lists.py", line 322, in on_post mlist = create_list(**validator(request)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.12/site-packages/mailman/app/lifecycle.py", line 102, in create_list call_name(config.mta.incoming).create(mlist) File "/usr/lib/python3.12/site-packages/mailman/mta/postfix.py", line 96, in create self.regenerate() File "/usr/lib/python3.12/site-packages/mailman/mta/postfix.py", line 106, in regenerate with Lock(lock_file): ^^^^^^^^^^^^^^^ File "/usr/lib/python3.12/site-packages/flufl/lock/_lockfile.py", line 467, in __exit__ self.unlock() File "/usr/lib/python3.12/site-packages/flufl/lock/_lockfile.py", line 417, in unlock raise NotLockedError('Already unlocked') flufl.lock._lockfile.NotLockedError: Already unlocked
When I check the lock folder, I have these two files, and I'm wondering if I should delete these two files : -rw-rw---- 2 mailman nogroup 97 Feb 6 2025 master.lck -rw-rw---- 2 mailman nogroup 97 Feb 6 2025 master.lck|mailman-core-deployment-5f59557b66-bkkf4|16|4268829230172994873
Do you have any answer for me ?