Why are there so many [xfs-xxx/dm-y] processes on my OCP worker node ?
Issue
- One of our worker nodes is showing a huge number of sleeping processes. I see a total of 3330. All the other nodes in the cluster have less than 600. These are docker processes and I'm not sure what is causing this.
- These processes are like below via
$ps -aux
:
root 71184 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-buf/dm-6]
root 71185 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-data/dm-6]
root 71186 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-conv/dm-6]
root 71187 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-cil/dm-6]
root 71188 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-reclaim/dm-]
root 71189 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-log/dm-6]
root 71191 0.0 0.0 0 0 ? S Jan08 0:00 [xfsaild/dm-6]
root 71343 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-buf/dm-7]
root 71344 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-data/dm-7]
root 71345 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-conv/dm-7]
root 71346 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-cil/dm-7]
root 71347 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-reclaim/dm-]
root 71348 0.0 0.0 0 0 ? S< Jan08 0:00 [xfs-log/dm-7]
root 71350 0.0 0.0 0 0 ? S Jan08 0:00 [xfsaild/dm-7]
- Please help to confirm whether it is normal.
Environment
- Red Hat Openshift Container Platform 3+
- RHEL 7+
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.