Skip to content
View in the app

A better way to browse. Learn more.

Unraid

A full-screen app on your home screen with push notifications, badges and more.

To install this app on iOS and iPadOS
  1. Tap the Share icon in Safari
  2. Scroll the menu and tap Add to Home Screen.
  3. Tap Add in the top-right corner.
To install this app on Android
  1. Tap the 3-dot menu (⋮) in the top-right corner of the browser.
  2. Tap Add to Home screen or Install app.
  3. Confirm by tapping Install.

BTRFS Cache Scrub Call Traces After Upgrade From rc8q to rc10b

Featured Replies

Hey guys...woke up this morning to a new one :)

My fix common problems plugin told me that call traces were found on my server...looking into the diagnostic log it appears the below section is the relevant part.  There's more than this but it appears to just endlessly repeat these same basic issues.  Can anyone smarter than me tell me what these mean or what is most likely going on, and if there's anything I can do to fix it?  I know I don't have the newest hardware, but I haven't had these issues ever before and Unraid has always been super stable in the past.  Thank you!!

Edited by d.ohlin

Always post the complete diagnostics, the culprit is usually before the trace, in this case likely a corrupt docker image.

 

 

  • Author

@johnnie.black My fault...that was dumb of me that makes total sense now that I think about it.

 

I'm hoping it's not a corrupt docker image in this case as I don't run as docker images on my server but what do I know haha :)

 

Attached is the full .zip diagnostics file, hopefully that's got everything needed.  Let me know if I can do anything else.  Thanks!

 

Edited by d.ohlin

The trace is related to btrfs but strangely I don't see nothing on the logs about which filesystem is causing the issues, and you may not have dockers but there's a docker image, try running a scrub on all 3 btrfs filesystem in use: docker, libvirt (VMs) and cache.

  • Author

@johnnie.black Sorry...noob question time :(

 

How do I do a manual scrub on those 3 filesystems?

 

EDIT: I found the scrub option under the main cache drive (I have 2) and clicked that...still unsure on the docker and libvirt ones though.

 

EDIT2: The cache scrub above completed with the following:


scrub status for f36f83a5-045e-43f6-a2a5-74bde4367eaa
    scrub started at Mon Oct 30 11:11:08 2017 and finished after 00:00:24
    total bytes scrubbed: 6.01GiB with 0 errors

 

However, running the above definitely causes the system to produce call traces when I click the "Log" option in the upper right hand corner...and after rebooting and running a cache scrub again it appears I can reliably reproduce the call traces this way :(


I also did a "--readonly" filesystem check on cache which gave the following:

 

checking extents
checking free space cache
checking fs roots
checking csums
checking root refs
Checking filesystem on /dev/sdf1
UUID: f36f83a5-045e-43f6-a2a5-74bde4367eaa
found 69873664 bytes used, no error found
total csum bytes: 67260
total tree bytes: 212992
total fs tree bytes: 32768
total extent tree bytes: 16384
btree space waste bytes: 121360
file data blocks allocated: 69660672
 referenced 69484544

 

I upgraded to 6.4.0_rc10b from 6.4.0_rc8q...I never got these errors in 8q.  Is there anything that changed between then and now that might be contributing to here?

 


EDIT 3: So I just reverted back to rc8q and can confirm I can now run a scrub on my cache successfully without triggering the call traces...so that seems to rule out hardware issues and it's definitely something between 8q and 10b (although I'm not sure if it was between 8q and 9f, or 9f and 10b).  I can't even find a download for 9f to be able to manually upgrade and determine which update the issue lies in.  Now what? :(

Edited by d.ohlin

Not sure also, first time I see that issue, call trace without an associate btrfs error/warning, you can try to re-format your cache:

 

backups all data, format, restore, and see if the issue continues.

  • Author

Hmmmm...alright I'm officially perplexed.  In case anyone happens to find themselves running into anything similar, here's exactly the steps I took.

 

So I wasn't completely sure on the best way to format the cache drive (in my case, drives - 2), so I did the following:

 

1. Stopped the array

2. Unassigned both cache devices

3. Re-assigned the cache devices as add'l. members of the main storage pool

4. Re-started the array (which obviously began clearing these 2 news disks)

5. Cancelling the clearing operation (pretty much immediately) and re-stopped the array

6. Re-assigned both previous cache disks as cache devices again

7. Re-started the array

 

At that point it gave me the "unmountable" note and advised me to format them.  I of course did so, and afterwards I was able to successfully run 3 scrubs on my cache pool (with and without the 'fix errors' box checked) without error.  So I'm definitely confused as to what might have been going on before.

 

Regardless, a little bit of up time should tell whether I'm truly good or not as previously the errors had happened within <24 hours of system uptime.  Fingers crossed!

  • Author

Just figured I'd come back to give an update on this as so often people don't post any confirmation of whether their solutions ended up working :)

 

It would appear that I'm good to go (knock on wood!) as I'm at ~50 hours of uptime so far with no issues, and I just now confirmed again that I can run a scrub on the cache drive without triggering any log errors.  Thanks again for all the help!

Edited by d.ohlin

Archived

This topic is now archived and is closed to further replies.

Account

Navigation

Search

Search

Configure browser push notifications

Chrome (Android)
  1. Tap the lock icon next to the address bar.
  2. Tap Permissions → Notifications.
  3. Adjust your preference.
Chrome (Desktop)
  1. Click the padlock icon in the address bar.
  2. Select Site settings.
  3. Find Notifications and adjust your preference.