gluster/glusterd2

Continuous "0-epoll: Failed to dispatch handler" errors in brick logs

PrasadDesala opened this issue · 6 comments

PrasadDesala commented

Observed behavior

Brick logs are getting spammed with continuous "0-epoll: Failed to dispatch handler" error.

[2019-01-02 09:58:49.149423] E [MSGID: 101191] [event-epoll.c:759:event_dispatch_epoll_worker] 0-epoll: Failed to dispatch handler

The above events are continuously getting generated in the brick logs. On a brick log till now it has logged "4553" times and is getting increased.

Expected/desired behavior

No errors in brick logs.

Details on how to reproduce (minimal and precise)

  1. Create a 3 node gcs system using valgrind.
  2. Create 100 pvc.
  3. Enabled brick-mux.
  4. Stop/start the volume.
    Check brick logs.

Information about the environment:

Glusterd2 version used (e.g. v4.1.0 or master): v6.0-dev.94.git601ba61
Operating system used: Centos 7.6
Glusterd2 compiled from sources, as a package (rpm/deb), or container:
Using External ETCD: (yes/no, if yes ETCD version): yes; version 3.3.8
If container, which container image:
Using kubernetes, openshift, or direct install:
If kubernetes/openshift, is gluster running inside kubernetes/openshift or outside: Kubernetes
gpavinteractiv commented

I see something similar with a newly provisioned ovirt 4.3 cluster.
Lots of :

[2019-01-15 10:07:53.646643] E [MSGID: 101191] [event-epoll.c:671:event_dispatch_epoll_worker] 0-epoll: Failed to dispatch handler
atinmu commented

I see something similar with a newly provisioned ovirt 4.3 cluster.

What gluster version are you using? Are you using GD2 in any case? If not then it's unrelated.

@PrasadDesala Are you still seeing this happening? I tried to reproduce this couple of times but failed in replicating the problem.

gpavinteractiv commented

oVirt is using Gluster 5.2 on redhat 7.6 but you are right, that is not GD2. This bug report was one of the only result when googling the "0-epoll: Failed to dispatch handler" line

PrasadDesala commented

I see something similar with a newly provisioned ovirt 4.3 cluster.

What gluster version are you using? Are you using GD2 in any case? If not then it's unrelated.

@PrasadDesala Are you still seeing this happening? I tried to reproduce this couple of times but failed in replicating the problem.

Yes, I am still seeing this issue on the latest nightly build.
glusterd version: v6.0-dev.109.gitdfb2462
git SHA: dfb2462
go version: go1.11.2
go OS/arch: linux/amd64

emerzon commented

Seems to be same issue as reported here: https://bugzilla.redhat.com/show_bug.cgi?id=1651246
However there was no activity from devs there since November.