!coeAONBrWyDJnYMbMi:nixos.org

NixOS System Operations

700 Members
About system administration for running NixOS systems in production. Declaratively manage your operations. | Room recommendations: #networking:nixos.org190 Servers

Load older messages


SenderMessageTime
5 Oct 2024
@ma27:nicht-so.sexyma27
In reply to @3von:matrix.org
What is IIRC?
if I recall correctly.
11:21:00
@scrumplex:duckhub.ioScrumplexThere are also some other secret providers like KeePassXC11:21:02
@3von:matrix.org3vonOh I c. OK activating keyring now11:21:45
@3von:matrix.org3vonThat worked like a charm. Thanks all11:23:37
@philip4g:matrix.orgPhilip

Hello all. Can anyone suggest why my NixOS system is periodically killing processes with an 'out of memory' status, like this:

Oct 02 10:52:15 pmwhome kernel: Out of memory: Killed process 3118 (dbus-daemon) total-vm:5420kB, anon-rss:0kB, file-rss:2348kB, shmem-rss:0kB, UID:1000 pgtables:48kB oom_score_adj:200
Oct 04 22:55:19 pmwhome kernel: Out of memory: Killed process 2355 (dbus-daemon) total-vm:5420kB, anon-rss:0kB, file-rss:1416kB, shmem-rss:0kB, UID:1000 pgtables:48kB oom_score_adj:200
Oct 05 08:01:44 home kernel: Out of memory: Killed process 2216 (systemd) total-vm:20340kB, anon-rss:0kB, file-rss:1896kB, shmem-rss:0kB, UID:1000 pgtables:80kB oom_score_adj:100

Meanwhile, I have a huge amount of memory:

$ free -m
               total        used        free      shared  buff/cache   available
Mem:           64048        3573       18849          10       41626       59752
Swap:         953851        1525      952326
15:33:47
@philip4g:matrix.orgPhilip *

Hello all. Can anyone suggest why my NixOS system is periodically killing processes with an 'out of memory' status, like this:

Oct 02 10:52:15 home kernel: Out of memory: Killed process 3118 (dbus-daemon) total-vm:5420kB, anon-rss:0kB, file-rss:2348kB, shmem-rss:0kB, UID:1000 pgtables:48kB oom_score_adj:200
Oct 04 22:55:19 home kernel: Out of memory: Killed process 2355 (dbus-daemon) total-vm:5420kB, anon-rss:0kB, file-rss:1416kB, shmem-rss:0kB, UID:1000 pgtables:48kB oom_score_adj:200
Oct 05 08:01:44 home kernel: Out of memory: Killed process 2216 (systemd) total-vm:20340kB, anon-rss:0kB, file-rss:1896kB, shmem-rss:0kB, UID:1000 pgtables:80kB oom_score_adj:100

Meanwhile, I have a huge amount of memory:

$ free -m
               total        used        free      shared  buff/cache   available
Mem:           64048        3573       18849          10       41626       59752
Swap:         953851        1525      952326
15:34:05
@k900:0upti.meK900Well presumably you don't have memory at the time 15:35:11
@k900:0upti.meK900Also, swap doesn't really count 15:35:19
@philip4g:matrix.orgPhilipThe kernel would consider it out of memory without using swap? Or what do you mean by doesn't count?15:39:23
@k900:0upti.meK900The kernel would consider it an out of memory condition if it can't find a page to swap out 15:41:05
@k900:0upti.meK900Which is why you can't just throw swap at your OOM problems and expect them to go away 15:42:23
@elvishjerricco:matrix.orgElvishJerricco Philip: Huh, that's extremely odd; @rosscomputerguy:matrix.org was having a very similar issue in #users:nixos.org the other day 15:43:30
@elvishjerricco:matrix.orgElvishJerriccoOOMs on processes that weren't using much memory when the total used wasn't very high15:43:46
@elvishjerricco:matrix.orgElvishJerriccoI think they found out it was some kernel tuning they had done that was the problem. But maybe that was wrong and the problem is somewhere in nixos?15:44:44
@elvishjerricco:matrix.orgElvishJerricco (also, that is an unbelievable amount of swap. Absolutely a waste lol) 15:47:56

There are no newer messages yet.


Back to Room ListRoom Version: 10