Jump to content

parity check triggered on reboot

Go to solution Solved by JorgeB,

Recommended Posts

I was getting crc count error warnings for one of my disks (sdi in attached diagnostics), so I was moving it around to different drive bays (so as to use different SATA ports and cables), but somehow its always the same disk which seems to show CRC errors. Nothing was wrong with data / filesystems during this, except for the latest reboot. After the reboot, a parity check started automatically


1. can I know what caused this? I assume unclean shutdown, timeout on something exceeded, but I can't see what

2. the trouble disk was on a btrfs pool of 2 HDDs. Why did the array get affected?

3. Anything else I am missing? is there another underlying issue


I upgraded from 6.11.5 to 6.12.1 yesterday, and it went smooth including importing a zfs pool. I don't believe this is probably the culprit, but still worth a mention


the log is full of disk errors, but I cannot pinpoint which disk. I am assuming sdi based on creeping CRC errors


diagnostics attached



Link to comment
6 hours ago, JorgeB said:

Correct, there should be diags saved in the flash drive /logs folder from that shutdown that might show the issue.

nice. I didn't know a diag is automatically saved in such a case. Attached the one from shutdown. seems the troubled btrfs pool drive had problems unmounting

diags attached



I am still not clear why this should mark the array dirty. a bit scary if issues with a cache pool would affect array operations


so what should be my next steps? I am not very familiar with btrfs recovery. I ran a scrub before reboot and it seemed OK, not sure how I see the same issue that the logs see. If the disk needs replacing, how do I replace it?


Link to comment
13 minutes ago, JorgeB said:

The issues with tsdb2 disk start write from boot, and are constantly spamming the log, looks more like a connection issue, replace cables for that disk or swap it to a different controller.

OK, I will swap things out once the current parity check finishes. I have also increased the shutdown timeout in disk settings for now to avoid this for next shutdown


I did move the drive from motherboard SATA to my LSI controller when the crc errors first showed up. The drive bays use different cables too. I will also examine the drive side connectors this time when I swap it. I have one more spare slot where I can try to put the disk


If this continues, is there a way I can downgrade tsdb to a single drive pool temporarily? would like to take the disk out and test it outside the system if swapping cables etc doesnt work

Link to comment
10 hours ago, JorgeB said:

It's not using the LSI controller, it's connected to the Intel SCU controller, that's the secondary Intel SATA/SAS controller

arrrrgggh. sorry, and thanks for spotting this. I mistakenly moved the other 2TB drive to the LSI controller. no wonder the crc errors didn't stop. 

I have a spare port on the motherboard SATA, so will rewire to that and report back

Link to comment

I have switched the affected drive to the other motherboard SATA controller, so far so good


started a scrub on the tsdb pool, and its reporting some errors. I did not check the fix errors 

Scrub started:    Sat Jun 24 11:12:36 2023
Status:           running
Duration:         0:13:38
Time left:        3:06:22
ETA:              Sat Jun 24 14:32:36 2023
Total to scrub:   2.48TiB
Bytes scrubbed:   173.21GiB  (6.82%)
Rate:             216.83MiB/s
Error summary:    verify=2 csum=120266
  Corrected:      0
  Uncorrectable:  0
  Unverified:     0


how do I go about fixing the errors? should I run with fix errors checkbox? need to wait for initial scrube to finish, or just cancel and do the fix steps?

EDIT: started a correcting scrub 

Edited by apandey
updated on scrub
Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Create New...