For a week now, I have been having all kinds of issues that I just cannot figure out what is going on. This Unraid server has been running for at least 10 years without any major issues. But now, all of a sudden, I am getting all kinds of problems.

It started with my Plex installation returning "transcoding errors" so I couldn't play certain files that needed transcoding (eac3 codec). So I reset the codec library. That didnt help.

But then I noticed that I couldn't delete some files in my shares. They were coming back as locked or read only. I rebooted the server and now I could delete the files I needed. But, then I noticed that my array gained like a couple of TB of space. And I did notice a few subdirectories were gone without me deleting them. At the same time, now all of my containers in the Docker were gone. I reinstalled them with using the template and they were back. Everything was good (except for the missing data), for a while. Unraid was transcoding and all was well for a few hours. Then started getting the transcoding issue again, rebooted, and the containers were gone again. Rinse, repeat.

So I found some threads suggesting that the USB drive may be the problem. So I created a replacement USB drive yesterday, reinstalled the containers, let it run all night but this morning, transcoder errors, reboot, and no containers.

I am at a loss on what to do next. Can anyone suggest a next step? I have attached the server log from the reboot this morning.

Update, if i reinstall the containers, and after they all start, and then reboot, the containers are in tact and working. So it seems that something happens over time (couple to a few hours) causing the containers to crash and disappear.

Sep 19 082319 unRaid kernel Linux v.txt

Edited September 19, 2025Sep 19 by Hawat

When the issue happens again, post the complete diagnostics before rebooting.

Ok, it happened again overnight/this morning. This morning, I tried watching a show through plex and got an error. So i downloaded the diagnostics (posted below) and rebooted. After the reboot, there are no docker containers. It is empty. Also, I added 2 subdirectories yesterday to the array each with files within them. They are gone as well.

unraid-diagnostics-20250921-0835.zip

Edited September 21, 2025Sep 21 by Hawat

Sep 21 00:52:43 unRaid kernel: REISERFS error (device md4p1): vs-4080 reiserfsfree_block: block 40119350: bit already cleared

Sep 21 00:52:43 unRaid kernel: REISERFS (device md4p1): Remounting filesystem read-only

Check filesystem on disk4.

P.S. Reiser will not be available on the next Unraid release (after 7.2), so you should get around to converting those filesystems.

So I ran a parity check a couple of days ago and had no errors reported. But are you saying that this disk 4 is throwing an error and causing my system to be in read only mode (from what I am reading in what was posted)? If so, I will replace it and let's see what happens. OK, Disk 4 replaced and now let's hope for the best during the rebuild. And, yes, I am aware of the filesystem change. I was just about to start that process when this problem developed. Hopefully I can get to that soon.

.

Edited September 22, 2025Sep 22 by Hawat

While my array is rebuilding itself, can I reinstall a container?  Or should I wait until the rebuild is complete?

Parity check has nothing to do with filesystem issues, see the link above to fix it.

Thank you and understand. I just replaced the disk and it is in rebuild mode right now. So, can I reinstall my containers while the rebuild is going on or should I wait until it is complete?

Edited September 22, 2025Sep 22 by Hawat

Did you fix the file system? Rebuilding the disk won't help with that.

I very sorry, but maybe I don't understand. Under the read check in the FAQ, it says to replace the disk if it shows errors. I just got in front of this and replaced the disk. No where in this section of the FAQ does it say anything about fixing the filesystem.

Are you saying that when this replacement disk gets done being rebuilt I might have the same problem?

Under read checks it says this:

When to use a read check

Consider running a read check in the following situations:

• No parity devices assigned: When your array lacks parity protection.

• Verifying disk integrity: To confirm the integrity of each disk without making any corrections.

• Post-hardware issue resolution: After fixing hardware issues, to ensure all disks are healthy.

What to do if errors are found

If your read check reports errors, here are some steps to follow:

1. Review SMART reports: Check the SMART reports for the affected disks to understand their health status.

2. Inspect connections: Ensure that physical connections and cables are secure.

3. Consider replacements: If a disk shows persistent errors, think about replacing it.

4. Backup critical data: Move important data off any disks reporting errors as soon as possible to prevent data loss.

There wasn't any issue with disk4, just its filesystem. The link I posted sends you to the right section on the docs, if you are now rebuilding the disk instead, finish that, then check filesystem.

I really do appreciate the help, and I'm sorry, but I am not seeing any reference to the "Check Filesystem" on the page you linked to me. When I click the link it takes me to "Checking Array Configuration" which includes "Parity Check" and "Read Check". I did a search for "file system" and came across this: https://docs.unraid.net/unraid-os/using-unraid-to/manage-storage/file-systems/#checking-a-file-system. I assume you meant this section? Also, there is no mention of my way outdated reiserfs in the article. It appears that I handle it like XFS?

Checking array devices

Regularly checking your Unraid array is crucial for maintaining data integrity and identifying potential issues before they result in data loss. You can initiate a check using the Check button under Array Operations. Depending on your array's configuration, this button allows you to conduct either a parity check or a read check.

For convenience, you can schedule these checks to run automatically at intervals that suit you by navigating to Settings → Scheduler. It's advisable to perform automated, non-correcting checks on a monthly or quarterly basis to ensure the ongoing health of your data.

Parity & Read checks

• Parity checks

• Read checks

You're right, sorry, the docs site was updated recently by LT and the automatic link is not working correctly; see here:

https://docs.unraid.net/unraid-os/using-unraid-to/manage-storage/file-systems/#checking-a-file-system

No problem. Man, i thought I was just blind. Thank you so much for the help. Once this finishes (sometime overnight) I will run the file system check and will see what that turns up.

Ok, so I started the filesystem check 2 hours ago. I refreshed the WEB GUI tab to see if there was any progress but page wont refresh - the notification in the tab is just spinning. Is this normal? The FAQ states: 7. Monitor progress: Monitor the progress in the output box. Use Refresh if required.

So I assume this page should refresh. Am I stuck? Or wait it out? Do you have any idea on how long this should take?

Does the status change if you open a new GUI page?

Its crazy it just refreshed after about 10 minutes. I got the response back which i attached to below. The key things, I think are this:

6 found corruptions can be fixed when running with --fix-fixable

###########

reiserfsck finished at Tue Sep 23 07:53:45 2025

###########

bad_indirect_item: block 34244486: The item (67985 82359 0x21d58001 IND (1), len 4048, location 48 entry count 0, fsck need 0, format new) has the bad pointer (122) to the block (34381981), which is in tree already

bad_indirect_item: block 34244486: The item (67985 82359 0x21d58001 IND (1), len 4048, location 48 entry count 0, fsck need 0, format new) has the bad pointer (286) to the block (34382145), which is in tree already

bad_indirect_item: block 34244486: The item (67985 82359 0x21d58001 IND (1), len 4048, location 48 entry count 0, fsck need 0, format new) has the bad pointer (476) to the block (34382335), which is in tree already

bad_indirect_item: block 39911562: The item (67990 82355 0x5e66001 IND (1), len 4048, location 48 entry count 0, fsck need 0, format new) has the bad pointer (542) to the block (40119350), which is in tree already

vpf-10640: The on-disk and the correct bitmaps differs.

So, not sure how to or where to run this --fix-fixable command. Any help here? The FAQ doesnt cover reiserfs any more. So there is no mention of what to do.

reiserfsck 3.6.27.txt

To the right side of the Check button, there's a box to enter any options; remove the default and enter --fix-fixable then click Check

Ok, started the fix about an hour ago and i'm in spinning wait on the refresh of the page again. It seems to be done. There is no activity on the server but I cannot get a refresh. Even if I open another tab. It will open to the Main screen, I can see all the drives, but when I click on Disk 4, I go into wait mode. Suggestions?

You should always be able to refresh the browser during a check; try a different browser.

Opened in Edge as well. Same thing . I can get to every tab (Dashboard, Users, Docker, etc) but once I click on Disk 4 on the Main tab, it goes into neverland

Do you think it is safe to stop the array out of maintenance mode and re-launch in normal mode? All the drives are off. The main tab shows this:

Edited September 23, 2025Sep 23 by Hawat

11 hours ago, Hawat said:

Do you think it is safe to stop the array out of maintenance mode and re-launch in normal mode?

It should be, but before stopping, open a terminal window and type reiserfsck --check /dev/md4p1 and post the output from that.

JorgeB - very sorry its been a couple of very busy days. But just to catch up, i did end up rebooting the server but did so before you posted about the opening a terminal window to pull the log. But since then, everything seems to be working just fine, so fixing the filesystem appears to have worked. I can put files on the array and the mover does its job with no errors. Also, I can reboot and the docker containers are right where they are supposed to be. So, thank you very much for all the help with this. I really appreciate it.