Server Fault Stack Exchange
Server Fault Stack Exchange

Questions tagged [oom-killer]

oom-killer Out-Of-Memory killer in the Linux kernel

A part of the Linux memory management that will sacrifice one or more processes in order to free up memory for the system when all else fails.

42 questions
3
votes
4 answers

mysqld crashing almost every hour - mysqld invoked oom-killer

I am running wordpress on a centOS digital ocean droplet my mysql keeps crashing almost every hour -- with an out of memory error manually restarting mysqld brings it back up via sudo service mysqld restart Does anyone know what could be causing…
Alicia R
  • 31
  • 1
  • 6
2
votes
0 answers

What abbreviations in OOM Killer memory statistics report means?

When OOM Killer or kernel reports memory state it uses the next abbreviations Node 0 DMA: 26*4kB (M) 53*8kB (UM) 33*16kB (ME) 23*32kB (UME) 6*64kB (ME) 7*128kB (UME) 1*256kB (M) 2*512kB (ME) 0*1024kB 0*2048kB 0*4096kB = 4352kB Node 0 DMA32: 803*4kB…
Raf
  • 173
  • 7
2
votes
1 answer

Why total-vm is is different than show in table?

I am running into oom_killer activity and cannot undesrand dmesg output [Fri Jan 11 02:11:48 2019] lowmem_reserve[]: 0 2811 64249 64249 [Fri Jan 11 02:11:48 2019] Node 0 DMA32 free:248704kB min:2956kB low:3692kB high:4432kB active_anon:2031516kB…
Antoniossss
  • 123
  • 1
  • 7
2
votes
3 answers

Stop a process (script) from invoking oom killer

I had a very unfortunate situation, where a bash script that contained a subtle error gone wild and took all available memory and then started killing other tasks (production apps!) to get even more. How can I future-proof invocations of this and…
Imaskar
  • 123
  • 5
2
votes
1 answer

Understanding OOM killer logs

I run some processes inside docker container and I use memory limitation for this container. Sometimes some processes inside docker container got killed by OOM killer. I see that in syslog file: beam.smp invoked oom-killer: gfp_mask=0xd0, order=0,…
sergeyz
  • 223
  • 1
  • 3
  • 8
1
vote
0 answers

Linux out of memory killer kills process when there is more than enough memory available

We are running a 64-Bit Ubuntu and our 32GB of physical memory is split into 3 zones (DMA: 16MB, DMA32: 4GB and Normal: 30GB). According to the dmesg logs printed below our system ran out of free memory on the Normal zone. The logs show that free…
Christof Länzlinger
  • 11
  • 1
1
vote
0 answers

Processes is getting killed randomly on REdHat

We have around java virtual machines with limited memory, and MongoDB. Deployed processes are randomly getting killed/terminated abruptly in time. No matter java process, nodejs or mongodb process. We tried OOM configuration changes also which did…
user135968
  • 11
  • 1
1
vote
1 answer

OOM score adjustements

On Ubuntu 12.04/Java 1.7 I have been reading/searching about OOM killer and generally understood what "proc" man page says especially oom_adj part. It all started because my important "MQ listeners" are getting killed. We have a master listener…
user871199
  • 185
  • 5
1
vote
1 answer

On SmartOS how to tell if process was killed for OOM

A daemon process of mine gets killed almost immediately, as soon as it starts receiving requests. The application log has absolutely no error output, even when I try to run it in foreground. I have a suspicion that the OS might be killing it for…
fulv
  • 113
  • 5
1
vote
1 answer

Interpreting OOM killer logs when triggered on a Java thread

When Linux OOM Killer interrupts a process, the kernel logs usually provide enough information about the culprit's memory consumption (even it is not killed eventually). For example, when snmpd process becomes an OOM trigger, its memory state can be…
Toparvion
  • 113
  • 3
0
votes
2 answers

Kubernetes DaemonSet Pod exceeds memory limit but is not killed

We use kubernetes (specifically: openshift) to run our infrastructure in our team. One of the daemonsets, fluentd, is currently causing a lot of trouble, frequently taking full nodes down with huge CPU, memory and disk I/O requirements (like really,…
aspyct
  • 360
  • 1
  • 6
  • 21
0
votes
1 answer

What are the specific column headers for the "Mem-Info" dump after "oom-killer"

I'm running a Debian 8 Jessie development server. 16GB mem. I've looked at the other posts about oom-killer, but none are really pertinent to what I'm asking here I'm trying to track down what process caused an oom-killer event. There's a memory…
BurningKrome
  • 525
  • 2
  • 12
  • 22
0
votes
0 answers

Strange numbers in Oom killer log

First, the log: [Wed aug 22 01:11:31 2018] java invoked oom-killer: gfp_mask=0x14280ca(GFP_HIGHUSER_MOVABLE|__GFP_ZERO), nodemask=(null), order=0, oom_score_adj=0 [Wed aug 22 01:11:31 2018] java cpuset=/ mems_allowed=0-1 [Wed aug 22 01:11:31 2018]…
Imaskar
  • 123
  • 5
0
votes
1 answer

Number of httpd processes and prefork.c

Hi I got OOM killer problems which caused: [mpm_prefork:error] [pid 18103] (12)Cannot allocate memory: AH00159: fork: Unable to fork new process I had to restart the server to get it working. In /var/log/messages near the OOM killer messages it…
Luke Wenke
  • 103
  • 4
0
votes
1 answer

VmRSS used to only about 25%, yet oom-killer strikes

I have a dedicated MySQL server equipped with 128 GB RAM. MySQL recently gets killed by the oom-killer, although MySQL is configured to use 95 GB in the worst case. In my research I came across this: # cat /proc/11895/status Name: mysqld State: S…
tombom
  • 105
  • 7
1
2
3