1

After a few days, docker fails with errors

docker[8927]: level=error msg="Error running exec in container: execcantrun: Cannot run exec command 34a26 in container 4972e576: [8] System error: fork/exec /proc/self/exe: cannot allocate memory\n" 

Docker 1.9.1 (upgrade is not an option)

USER       PID %CPU %MEM    VSZ     RSS    TTY      STAT START   TIME COMMAND 
root         34763  9.3    0.8    10680004   535208   ?      Ssl  Mar13 1334:17 /usr/bin/docker daemon -H fd:// --insecure-registry=0.0.0.0/0 --log-level=warn --ip-forward=false  

See VSZ is almost 10G

free -m

             total       used       free     shared    buffers     cached 
Mem:         63012      60128       2883        816          1       6775 
-/+ buffers/cache:      53351       9660 
Swap:         8191       7581        610  

If I restart docker service, the issue goes away until reoccurrence after a few days.

After docker restart, VSZ is around 2G

root      27144  2.4  0.0 1786272 44432 ?       Ssl  17:44   4:47 /usr/bin/docker daemon -H fd:// --insecure-registry=0.0.0.0/0 --log-level=warn --ip-forward=false  

Anyone seen it or suggestion?

0

The issue is related to https://github.com/docker/docker/issues/8539 which is resolved in docker 1.10 or later. There was some issue with docker log driver that resulted in buffer problems specifically for flushing a container's STDOUT.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.