-
Notifications
You must be signed in to change notification settings - Fork 1.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
ZED constant config_sync #6667
Comments
As my system has been up several days now with no ZED, i looked at the pool status and iostats and found that the pool is once again spiking in I/O.
Im not finding anything wrong with the pool other than high I/O and starting zed up immediately begins spamming with config sync messages. Im unsure what to make of it and cant seem to find anything that is related via some searching. |
As well, several worker processes are spawning in conjunction with the disk IO
|
I had noticed on my system funtoo current zed constantly spiking the cpu after upgrade I will have to collect more data. |
i couldnt figure it out. after doing some debugs it looked like udev was constantly making the drive appear ONLINE even though that was its previous state. In the end i nuked the OS since all of my customizations are in docker and stored on the pools, and low and behold, the issue was solved. No more constant churn on the drives and the IO was back to idling with nothing going on. |
@firechicken86 thanks for following up and letting us know what was causing the load. Based on your description I'm going to close this issue since it sounds like there's too much we can do on the ZFS side. When udev generates an event we really do need to handle it. |
Sorry to post to a closed ticket but I am experiencing this issue. My system was automatically upgraded from RHEL 7.3 to 7.4, I removed and re-added zfs according to the docs. Same versions (slightly newer kernel) from what's listed at the top. I really don't want to blow away the system....so I'll try a few things.
seemed to fix it. |
Possible fix: #13729 |
System information
Describe the problem you're observing
after upgrading to centos 7.4 from 7.3 and removing/reinstalling the required ZFS repos and software packages, ZED continually creats event entries and writes to the pool continually.
Disabling ZED puts the pool activity to 0 and a scrub has yielded no errors so far. This is 1 pool of 3 but the only one that is showing this "issue"
Describe how to reproduce the problem
im unsure how to reporduce it from another system
Include any warning/errors/backtraces from the system logs
zpool iostat pool1 -v (while zed active):
zpool iostat pool1 -v (while zed inactive):
zpool history after shutting down zed:
The text was updated successfully, but these errors were encountered: