Public bug reported:
When you have a Redis database of dozens of gigabytes, it takes various
minutes to start. At that point, systemd's start timeout kills it and it
tries again. This doesn't recover.
Same for stop. It can take various minutes to save a database this big.
The service file needs a bigger TimeoutStartSec and TimeoutStopSec. I
don't know what value. The true problem is that systemd lacks an ability
for receiving 'I'm still starting/shutting down' signals from a process.
Absent that, very high timeouts seems the only real fix.
Was observed in Ubuntu24.04, but affects many. Our database file is 17
GB on disk, 50 GB in memory.
** Affects: redis (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2068542
Title:
Redis start is impossible with big DB because of start timeout
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/redis/+bug/2068542/+subscriptions
--
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
