[Bugs] [Bug 1695099] The number of glusterfs processes keeps increasing, using all available resources

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1695099



--- Comment #2 from Christian Ihle <christian.ihle at drift.oslo.kommune.no> ---
I have been experimenting with setting "max_inflight_operations" to 1 in
Heketi, as mentioned in https://github.com/heketi/heketi/issues/1439
Example of how to configure this:
https://github.com/heketi/heketi/blob/8417f25f474b0b16e1936a66f9b63bcedfba6e4c/tests/functional/TestSmokeTest/config/heketi.json

I am not able to reproduce the issue anymore when the value is set to 1.
The number of glusterfs-processes varies between 0 and 2 during volume changes,
but always settles on 1 single process afterwards.

This seems to be an easy workaround, but hopefully the bug will be fixed so I
can revert back to concurrent Heketi again.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.