<pbb>
flokli: arianvp: I was not 100% sure until now that the nscd memory-leak-ish issues are still there, but now I'm certain, I set a memory limit on the nscd systemd service and just saw a message on one server's console
<pbb>
"Memory cgroup out of memory: Killed process 486 (nscd) total-vm:1715076kB, anon-rss:201416kB, file-rss:0kB, shmem-rss:0kB, UID:64015, pgtables:1776kB, oom_score_adj:0"
<pbb>
Maybe we should set a generous memory limit on the process in the NixOS module to prevent it hanging up systems?
<pbb>
In this case I set a limit of 200MiB, which should most probably never get used
<flokli>
It shouldn't leak memory, tbh ;-)
<flokli>
The hosts positive cache still is active, but it's not cache entries, right?
hmpffff has joined #nixos-systemd
hmpffff_ has joined #nixos-systemd
hmpffff has quit [Ping timeout: 272 seconds]
hmpffff has joined #nixos-systemd
hmpffff_ has quit [Ping timeout: 256 seconds]
hmpffff has quit [Ping timeout: 260 seconds]
hmpffff has joined #nixos-systemd
hmpffff_ has joined #nixos-systemd
hmpffff__ has joined #nixos-systemd
hmpffff has quit [Ping timeout: 246 seconds]
hmpffff has joined #nixos-systemd
hmpffff_ has quit [Ping timeout: 272 seconds]
hmpffff__ has quit [Ping timeout: 256 seconds]
hmpffff_ has joined #nixos-systemd
hmpffff has quit [Ping timeout: 272 seconds]
hmpffff_ has quit [Ping timeout: 256 seconds]
hmpffff has joined #nixos-systemd
<flokli>
erp, wasn't quite awake. try a sudo /run/current-system/sw/bin/nscd -g | grep "number of cached"
<flokli>
speaking of, I wonder why I still see some cached values there. maybe we should also set enable-cache to "no". pbb, can you give this a try?