osd crashed

[cola16]

The disk that is causing this problem is fine when SMART scanned.

When I do dmesg | grep ‘ sd ’, I get nothing but a message that the device is connected.



What should I check?

 

[gurubert]

Remove this OSD and redeploy it. There may be just a bit error on the disk.

 

[cola16]

There may be just a bit error on the disk.

I redeployed last night.
This morning I found similar errors.


When I ran journalctl -eu ceph-osd@14,
last night I could see errors related to IO errors.
Today, I cannot find the errors from the attached files.

I have osd crash twice.
I couldn't see any hardware errors in s.m.a.r.t and dmesg, so I asked to get if there was anything else I could check.

 

[gurubert]

I would replace the disk now.