I was having issues with Docker, so I decided to update to 6.3.3. Upgrade seemed to work fine. But now I can't start Docker. I've attached my syslog, which has a bunch of errors, but I'm not sure what they mean. Thanks for your help.

unraid-syslog-20170405-1304.zip

Your sig says you are still on a very old v5rc, which of course is wrong since you were having docker issues before upgrading to 6.3.3. So, what version of unRAID did you just update from? Did you read the release notes for 6.3.3?

 

I have not looked at your syslog. For v6, unless we specifically ask for just a syslog, you should always go to Tools - Diagnostics and post the complete diagnostics zip, which contains syslog(s) and many other things which might help diagnose your problem.

Okay, sorry didn't know about the diagnostics. I've attached the them. And I upgraded from 6.3.2. And I only just started noticing that one of my docker apps wasn't working (Sonarr) last week. I just figured I'd try to update to see if it would fix it.

unraid-diagnostics-20170405-1319.zip

Your cache drive is basically pooched

Quote

197 Current_Pending_Sector  0x0032   192   192   000    Old_age   Always       -      1463

 

Because of that, you've got tons of this:

 

pr  5 13:10:07 unraid kernel: ata6.00: status: { DRDY ERR }

Apr  5 13:10:07 unraid kernel: ata6.00: error: { UNC }
Apr  5 13:10:07 unraid kernel: ata6.00: configured for UDMA/133
Apr  5 13:10:07 unraid kernel: sd 6:0:0:0: [sdd] tag#3 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08
Apr  5 13:10:07 unraid kernel: sd 6:0:0:0: [sdd] tag#3 Sense Key : 0x3 [current] 
Apr  5 13:10:07 unraid kernel: sd 6:0:0:0: [sdd] tag#3 ASC=0x11 ASCQ=0x4 
Apr  5 13:10:07 unraid kernel: sd 6:0:0:0: [sdd] tag#3 CDB: opcode=0x28 28 00 4b f8 00 c8 00 00 08 00
Apr  5 13:10:07 unraid kernel: blk_update_request: I/O error, dev sdd, sector 1274544331
Apr  5 13:10:07 unraid kernel: ata6: EH complete

 

Which culminates with the actual problem with docker:

Apr  5 13:19:29 unraid root: ERROR: unable to resize '/var/lib/docker': Read-only file system

Squid already answered your problem, but since you seemed unaware that you had a bad cache disk, I have to ask: Do you have Notifications setup?

 

For your sig, in the forum, if you click on the dropdown next to your name in the upper right corner and select Account Settings and then select Signature on the left, you will get to a page that will let edit your sig, and also you can turn on viewing signatures.

 

 

I swapped out my cache drive with something I had, and re-added the docker containers and that seemed to work fine. It still looked like I got some errors on that drive though. I'll buy a new cache drive soon and replace it.

 

Thanks for your help!

Since I was messing with my box already yesterday, I tried adding a second parity drive. After the parity built, I couldn't start docker again. I tried deleting the docker  image and starting over, but no luck there. Any help is appreciated. Attached diagnostics again. Thanks!

unraid-diagnostics-20170406-1634.zip

Your replacement cache is worse than the other one (and it was very bad).

  5 Reallocated_Sector_Ct   0x0033   041   041   140    Pre-fail  Always   FAILING_NOW 1265

Do you test your disks before using them?

 

Do you have Notifications setup? Even if you don't, the Dashboard should be showing you that you have a problem.

Thanks for looking at this. I ran preclear on all these disks, 3 passes, but it was probably over two years ago. I would have used this 2TB disk as a replacement if one of my array disks failed. I'm kind of glad this happened. Would it make sense to get an SSD for cache? I'll shop around for some backup drives for my array, too.

I also never looked at the dashboard because I always just assumed that a green dot on the main screen meant it was okay. On the dashboard, the cache drive has a yellow triangle warning icon there. Just checked my alert settings and they weren't working... probably since I turned on 2-step verification on my Google account. It's working now, but which alerts should I be making sure I get?

1 hour ago, pmpj218 said:

which alerts should I be making sure I get?

Turn on Help for an explanation of Notice, Warning, Alert. I get emails for all three plus daily Array Status.

Is there a reason I can't check the email boxes next to Notices, Warnings, and Alerts? I would think I want those via email, but the check boxes are disabled.

Nevermind, I can't read. I enabled the alerts.

I find myself in this same dark hole of docker not starting.  Attached are my diagnostics.

 

Extended smart tests on all drives turns up nothing (although the next day one of the drives that is very old has no record of the smart report, not sure if that is "normal."). Then I revisit later and it does see the test...weird...

Actually what i think happened here is that I kicked off smart extended on multiple drives at once, and the system can't keep them straight when you do that.  As I revisit the thing now, it sees the test again for that drive as completed without error and a different drive as test interrupted.  I think it is severely confused, so I'm kicking off the one that was supposedly interrupted again.

 

When I first saw the error, I decided to restart the thing. Upon reboot it started a parity check.  After 8 hours it finished successfully, corrected tens of errors supposedly.  Restarted again, still no docker.  I thought that was really odd, so I kicked off another parity check, but not correcting errors.  After just a few minutes it reported finding 17 errors.  I immediately cancelled and thought "oh no, my parity might be bad."

 

Then the smart tests that surfaced nothing.  Smart attributes on the ancient WDC Velociraptor show this that I believe is more likely not the drive but a cable or memory?

199 UDMA CRC error count, flag 0x003e, value 200, worst 200, threshold 000, type Old age, updated Always, failed Never, raw value 31

 

I've gone ahead and ordered a couple drives to be ready to act but also to move away from ReiserFS if I can figure out how to do that.

 

Any help is appreciated.

backup-diagnostics-20220315-2240.zip

3 hours ago, zaker said:

I find myself in this same dark hole of docker not starting. 

 

Mar 15 22:31:44 backup kernel: REISERFS (device md2): Remounting filesystem read-only

 

Check filesystem on disk2, once that's fixed you should convert to xfs.

5 hours ago, zaker said:

move away from ReiserFS if I can figure out how to do that.

Short answer, copy the files elsewhere, format to XFS or BTRFS, copy files back.

 

There is NO way to switch file systems in Unraid without losing the data that is on the drive.

 

Take a look at the sticky thread in the general support area for some strategies, my favorite is either a new additional drive as large or larger than any existing data drive, or if you have enough cumulative free space to empty your largest data drive by copying files to other drives.

 

Either way, copying is much faster than moving for a bunch of reasons, plus you can verify the copy is good before formatting the ReiserFS source drive.

 

It's possible you could get away with copying the data from the read only disk2 to another disk if you have enough space, then reformat that disk after the data on that disk is safely copied.

And you really want your appdata and docker.img on fast pool (cache) so they will perform better and not keep array disks spunup since these files are always open.

Is it normal for the parity drive to report no smart test result when it is done? B/c that has happened twice now....

1 minute ago, zaker said:

Is it normal for the parity drive to report no smart test result when it is done? B/c that has happened twice now....

No. Try disabling spin down and try again.