oom-killer killed processes in 3scale containers
Issue
oom-killer
killed processes in 3scale containers.-
The memory limits of 3scale containers were the default value.
-
Is there any specific reason for the OOM? Is it a common issue or a specific issue of this process?
2019/05/28 17:46:51 114,"May 28 17:46:51 EXAMPLE kernel: ruby invoked oom-killer: gfp_mask=0xd0, order=0, oom_score_adj=976" 2019/05/28 17:46:51 131,May 28 17:46:51 EXAMPLE kernel: ruby cpuset=docker-f683ae6b1ce6751dcb08f97b00631f06f8be69fb0e97cc15787801d6d6181b68.scope mems_allowed=0
-
-
The memory usage of the
system-developer
container in system-app-6-xxxxx suddenly bloats from around 82.7% to 93% in 5 minutes. On the other hand, the other 2 system-developer containers in system-app-6-yyyyy and system-app-6-zzzzz keep around 80% memory usage. Why memory usage of the container is so high?
Environment
- Red Hat 3scale API Management
- 2.x On-Premises
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.