Thomas Gunn Posted February 9 Share Posted February 9 Hello. Have an array ZFS data drive that's showing as 'Device is disabled, Contents emulated'. Tried looking up what to do for file checking but that section in the guide (https://docs.unraid.net/unraid-os/manual/storage-management/#zfs) isn't updated yet. Didn't find anything else that seemed to help when searching around so far. Stopped array, put into maintenance mode, short smart test didn't show any errors, long test is currently running. What is the best course of action? Thanks Side question: Since most of these drives are in a USB bay (not recommended I'm assuming), if this happened to 3+ drives at the same time would I be screwed for rebuilding the array? New to Unraid, if anything looks funny from the logs. 7om-unraid-diagnostics-20240209-0032.zip 7om-unraid-diagnostics-20240209-1216.zip syslog-20240209-003245.txt syslog-20240208-195012.txt Quote Link to comment
trurl Posted February 9 Share Posted February 9 Most likely disk5 disconnected, which is not uncommon for USB, and likely to occur again since there is nothing you can do about it. If you were using SATA/SAS connections then not only would it be much less likely, but you could try to troubleshoot it by changing cables, etc. Start the array in normal (not maintenance) mode and post new diagnostics so we can see if emulated disk5 is mounting. Quote Link to comment
Thomas Gunn Posted February 9 Author Share Posted February 9 Will do once the long test is done. The disconnection occurred when trying to add a Thunderbolt hub to the system that turned out to be a POS (it seems), so I know how (-ish) it disconnected. Quote Link to comment
itimpi Posted February 9 Share Posted February 9 As long as the emulated drive is mounting OK then you should follow the standard procedure for replacing a disabled drive as documented here inthe online documentation accessible via the ‘Manual’ link at the bottom of the GUI or the DOCS link at the top of each forum page. Quote Link to comment
trurl Posted February 9 Share Posted February 9 You were trying to lookup the documentation for repairing the filesystem of an unmountable disk. Disabled and unmountable are 2 different and independent conditions, requiring different fixes. A disk can be enabled and unmountable, or disabled and mountable, or both disabled and also unmountable. All we know so far is that the disk is disabled. Quote Link to comment
trurl Posted February 9 Share Posted February 9 2 minutes ago, itimpi said: As long as the emulated drive is mounting OK 24 minutes ago, trurl said: Start the array in normal (not maintenance) mode and post new diagnostics so we can see if emulated disk5 is mounting. Quote Link to comment
Thomas Gunn Posted February 10 Author Share Posted February 10 Long test finished without error. Array started in normal mode, diagnostics taken after and attached. 7om-unraid-diagnostics-20240209-2249.zip 7om-unraid-smart-20240209-2240.zip Quote Link to comment
trurl Posted February 10 Share Posted February 10 Disk5 passed extended SMART self-test. Emulated disk5 is mounted, but empty or nearly so. Is that expected? Probably hadn't got past highwater on disk2 yet so it probably is since all of the later disks are similar. Quote Link to comment
itimpi Posted February 10 Share Posted February 10 @Thomas Gunn FYI: Diagnostics includes SMART results for all drives so normally no need to include it separately. Quote Link to comment
Thomas Gunn Posted February 10 Author Share Posted February 10 @trurl No it was pretty full due to emptying Drive 8 for having bad sectors and getting ready to replace it. @itimpi Thanks for the tip. Quote Link to comment
Solution trurl Posted February 11 Solution Share Posted February 11 9 hours ago, Thomas Gunn said: No it was pretty full due to emptying Drive 8 for having bad sectors and getting ready to replace it. My bad, haven't gotten used to looking at df results for ZFS. Should be OK to rebuild on top of the same disk. Quote Link to comment
Thomas Gunn Posted February 11 Author Share Posted February 11 Will do Is there anywhere that tells what exactly it is that UNRAID doesn't like about the drive and why it's disabled? Quote Link to comment
trurl Posted February 11 Share Posted February 11 On 2/9/2024 at 4:45 PM, trurl said: Most likely disk5 disconnected, which is not uncommon for USB, and likely to occur again Quote Link to comment
Thomas Gunn Posted February 11 Author Share Posted February 11 Is rebuilding the only way to fix a disabled drive? This looks like the instructions I need to follow. Just want to make sure since this is the first time I've had to do this with a drive that I hadn't cleared and replaced. https://docs.unraid.net/unraid-os/manual/storage-management/#rebuilding-a-drive-onto-itself Quote Link to comment
trurl Posted February 11 Share Posted February 11 That is the correct documentation to rebuild onto the same disk. The alternative is to New Config and rebuild parity instead, but that will lose anything written to the emulated disk and doesn't save any time. Something has to be rebuilt to get the array back in sync. Another alternative if you insist on USB is to not have any parity disk. That way there is nothing to get out-of-sync when a drive disconnects, so nothing needs to be rebuilt. Of course, nothing can be rebuilt either. USB enclosures that support UASP might avoid some of the problems associated with USB such as no SMART reports, inconsistent or not unique identifiers, incorrect disk sizes. But I'm not sure it will prevent disconnects. Quote Link to comment
Thomas Gunn Posted February 12 Author Share Posted February 12 Thanks for the help. It's rebuilding now. Quote Link to comment
trurl Posted February 12 Share Posted February 12 On 2/10/2024 at 10:02 PM, Thomas Gunn said: hadn't cleared and replaced Just thought I would make this a "teaching moment". The only scenario where Unraid requires a clear disk is when adding it to a new data slot in an array that already has valid parity. This is so parity will remain valid, since a clear disk is all zeros and so has no effect on parity. In this scenario, Unraid will clear the disk if it hasn't been precleared, before it will let you format it so it is ready to be accessed. No point in clearing a disk to replace another disk since the replacement is going to be completely overwritten by the rebuild. Preclear is really just for giving a new disk a good workout before trusting it in the array. Quote Link to comment
Thomas Gunn Posted February 14 Author Share Posted February 14 I'm not used to parity yet so I was getting all the files off the drive while it was able to be read. But yes, thank you. While disk 5 is rebuilding, the drive that I wanted to remove before [disk 8] has a tonne of reallocated and reported uncorrect sectors (currently 61k errors @ 6 of 16 TB rebuilt). No errors are reported on the rebuilding process. This is affecting the rebuild time substantially; which I can live with as long it gets done right. Will these bad sectors affect the files rebuilt for disk 5? Quote Link to comment
trurl Posted February 14 Share Posted February 14 Since you have dual parity you should be able to rebuild disk5 without disk8. Which of your disks have SMART ( 👎 ) warnings on the Dashboard page? Quote Link to comment
trurl Posted February 14 Share Posted February 14 Didn't notice any other SMART problems for other disks. You really should have mentioned disk8 from the outset. Probably it has been going bad for a while now. Do you have Notifications setup to alert you immediately by email or other agent as soon as a problem is detected? Do you have another disk you can rebuild disk8 onto? Quote Link to comment
Thomas Gunn Posted February 15 Author Share Posted February 15 None are showing smart warnings right now, but disk 8 was before I'm pretty sure. I have 2 new drives but they're 18TB, so was going to use them for new parity drives after all of this got sorted out and use the 16TB parities to replace the 8TB and keep one as a precleared spare. The 8TB that's in there now actually replaced a different 8TB that was having a few issues. Seems like this one is even worse. I even did a full badblocks run with 4 passes of writes (0 01 10 11, I believe) on it before adding it and extended test to make sure it was decent. Didn't show any errors then. Yes have alerts setup immediately via PushOver. Since posting earlier, the error count has skyrocketed from 61 000 to 67 000 000 @ 6.7 of 16TB rebuilt, and now disk 9 is showing about the same as what disk 8 does. Tried to do a short SMART test via GUI on disk 9 but it won't run it. Didn't try via CLI yet. Disk 8 and 9 are in the same USB enclosure. It's an 8 bay enclosure, but only shows up as 4 paired controllers (or something like that when I had it on windows). 8 and 9 are on the same controller, even though it says in the notes: 1-1 for disk 9 and 1-8 for disk 8 (it was swapped to 1-2 when doing some troubleshooting and didn't move it back). I'm guessing that's why it's showing that, but don't know why exactly. Also, disk 5 is now not showing any writing to it while the rebuild progresses now and is skipping along at 62MB/s, when it was usually 30MB/s before. What would you suggest or what else can I provide for details? 7om-unraid-diagnostics-20240214-2024.zip Quote Link to comment
Thomas Gunn Posted February 15 Author Share Posted February 15 Just noticed that parity is showing all those errors as well now on the dashboard, whereas when it was 61 000 parity didn't show any errors I'm pretty sure. Quote Link to comment
Thomas Gunn Posted February 15 Author Share Posted February 15 And the log is full. Did the diagnostics even give you the right thing you needed? Quote Link to comment
Recommended Posts
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.