Jump to content

2 Drives disabled with red X. Appear uninstalled when array is started


Go to solution Solved by JorgeB,

Recommended Posts

I have two drives that had errors and while I procured two new drives and precleared them, they became disabled and now refuse to be assigned to the new drives that are now in the drive bays where the disabled drives were located. The two new drives have a preclear signature.

 

When I try to assign the drives, I select the drive in the “unassigned” drop down, the screen refreshes, and the drives remain unassigned. 

 

This started to occur two months ago and I have had my server off and on while I researched the issue and tried to resolve the issue through responses to other forum posts. I have exchanged the SATA cables to see whether that was the issue because I received a udma error before this all started, but it didn’t help. I noticed that the drives show up in the servers BIOS but when I start the unraid array the two drives, 5 & 6, show as “not installed”. The data for the drives still appear when I click on the devices, “drive 5” and “drive 6”. A red x appears beside each drive. I also click on the unassigned devices mount button for each drive but the buttons do nothing.

 

I am at a lost as to what to do now and I am worried that I will lose the data for the drives. Is there a way I can recover the data from parity and is there any solution for a problem like this situation that I now find myself? I have attached a system log, and diagnostics attached to this post. I also have screenshots of what I see on my unraid server.

tower_start_Button.png

Drive5 parity_Backup.jpg

Unraid_array_started.jpg

Unraid_Unassigned_drives.jpg

Unraid_unassigned_drive_options.jpg

Unraid_Array_stopped.jpg

tower-diagnostics-20240111-1710.zip tower-syslog-20240110-2325.zip

Link to comment

Are you following this procedure?

 

        https://docs.unraid.net/unraid-os/manual/storage-management/#normal-replacement

 

(Make sure that you are not missing step 6...)

 

Oh, yes.  You should not have lost any data as you have dual parity.  The data of those two missing disks is actually available as these two disks are being emulated using the parity disks and the remaining data disks. That means you can both read and write data to them!  When the disks are rebuilt, that emulated data will be written onto the new disks.)

Link to comment

Yes., I have tried firefox and chrome.  I will try logging directly into the gui without using a browser.

 

When I do #5 I assign one of the disks to one in the dropdown and the screen refreshes and the drive I try to assign the dropdown disk to remains unassigned.  I never see a checkbox.

 

I have a feeling that something is corrupted in the unassigned devices plugin or unraid.  I just can't seem to figure out what happened and how to fix it.  I have replaced drives in the past with no issue.  I have never seen a disabled disk warning before, or have "not installed" message after starting the array.

Link to comment

FYI,  I tried booting directly into the gui.  Not using the headless browser method from my laptop.  Same results

 

I shutdown my vms and plugins and turned them off so that they do not start on reboot. I clicked the safe mode checkbox and rebooted.  I tried to assign each device but both devices still refreshed and remained unassigned

safe mode.jpg

Link to comment

Okay,  I created a stock trial key and booted directly into the gui.  Not headless.  I was able to assign the two drives that I have not been able to assign with the licensed usb stick.   I have done nothing else and I am waiting for more guidance.  I am happy to see some progress.

Link to comment

That suggests something in your /config is causing the issue, you can backup the current flash drive first and then redo it and just restore the bare minimum, like the key, super.dat and the pools folder for the assignments, also copy the docker user templates folder, if all works you can then reconfigure the server or try restoring a few config files at a time from the backup to see if you can find the culprit.

Link to comment

Yes, I have a backup from april 2023.  Im trying to recover from that but I am having problem with 1 of my 2 lsi sas adapters now so I am working on that issue.  I will let the system rest for a few hours and try again. Sometimes when you shutdown a system and unplug the power something may reset.  I will let you know what I find out regarding the config.   I agree that it is probably corrupt.

Link to comment

The SAS adapter BIOS was corrupted.  To clear the BIOS I had to reset the sas card configuration to the default.  That was a new problem for me.

 

Just an FYI.  My parity drives were on the second card and when I booted up into unraid, my Parity disks were unassigned.  I reassign them to the proper drives.  I also assigned the two replacement drives to drive 5 & 6.  I am now getting a "too many wrong and missing drives" message near the start array button and the array won't start.

 

Do I just create a new config and accept a loss of data, or am I already screwed anyway?   

 

I attached 2 screenshots of what I see and a diagnostic.

wrong discs.jpg

start button.jpg

tower-diagnostics-20240124-0748.zip

Link to comment

Yes, I do still have the two original drives untouched since I removed them.  They were reporting data errors and unraid disabled the two drives a few days later on it's own.  maybe the config corruption occurred at that time and the drives are still usable?

 

How should I proceed?  Should I shut shutdown the unraid server after unassigning the two replacement drives and then put the old drives back into their old location?  Then boot up hoping for the best?

Link to comment
46 minutes ago, jkwaterman said:

Do I just create a new config and accept a loss of data, or am I already screwed anyway?   

 

NO!!  At this point, it appears that disks 5 and 6 are being emulated and all of their contents are available.  Be patient and work through the questions and steps that you are being asked about and how to proceed.   (In fact, those old disks may well be intact with all  of their data safe!   There are are a lot of things that can cause a disk to become disable that are not related to the actual disk itself!)

Link to comment
1 hour ago, jkwaterman said:

Just an FYI.  My parity drives were on the second card and when I booted up into unraid, my Parity disks were unassigned.  I reassign them to the proper drives.  I also assigned the two replacement drives to drive 5 & 6.  I am now getting a "too many wrong and missing drives" message near the start array button and the array won't start.

Strange that the parity disks would be unassigned because of that, they should just show up as missing until available again.

 

I agree that using the old disks, if they are healthy, is probably the best solution, but if you have a backup of the flash drive when the disks were disabled, you can restore super.dat to get the array back as it was before, do that the disabled disks can be emulated/rebuilt if needed.

Link to comment

Hey guys, I'm just chiming in here as jkwaterman is a friend of mine, and we've already been chatting via email - I sent him here for expert advice.  I'm super happy to see JorgeB, trurl and Frank1940 are helping out - you guys are sharp so I know he's in good hands.

 

I read through everything, and I do have a few thoughts.  Everything you guys are suggesting is pretty much a match for what I've advised via email as well, so we're all already on the same page.

 

Restoring the super.dat from his Apr 2023 backup is a great idea, but I think that only applies if he didn't change any drives between the backup and before the drives failed out.  If we send him down this path, I think he first needs to confirm he didn't upgrade/swap/add any drives post backup, and also he should have a new backup of his current (bad) config, in case this goes sideways and he wants to get back to the current state.  I wanted to point this out since I didn't see anyone ask this particular question.

 

I also strongly agree with trying to use the original failed drives, and that he should perform SMART tests to validate the drives are okay before re-using them.

 

One thing I'm not sure about is if he uses the old drives, should he use the Trust Parity feature (I assume that's still a feature, been a decade since I last did this).  I'm imagining that he's got two paths forward with the old drives.  He could recreate the array config using all the original drives, and do a Trust Parity so it won't be rebuilt, and then immediately swap out the two suspect drives and rebuild onto the replacements.  Basically, with this approach he's using the GUI to recreate the pre-failed drive config state, and then manually failing/upgrading the drives.  Otherwise, he could again recreate the array config using all the original drives, but don't Trust Parity and instead rebuild new Parity via the data on the suspect drives.  This second approach sounds slightly riskier, as we're trusting the suspect drives to survive the parity rebuild, and unfortunately we don't know the nature of the errors that started this whole fiasco.

 

I know for a fact that he has started the array numerous times in disk emulation mode, so data could have been written to the array.  Additionally we are both users of the My Movies software, which has a habit of updating local movie data from online web contributions that other users continually submit, and this metadata in turn gets written to the array.  It's probably safe to assume that My Movies was running at some point during disk emulation mode, so that the current parity data no longer matches the data on the failed drives.  I just wanted to point this out, so that we all know to only either trust the parity data, or trust the suspect drive data, but expect the two data sources to be slightly out of sync with each other.  Note that the updates from My Movies are trivial and will automatically be reapplied if he reverts to the old drive data, so no risk of data loss there if he reverts to them.

 

One question I had myself is:  Is it possible to manually fix the drive config, via text editing, so that the parity drives are re-added to the array in a trusted state, but the 2 failed drives are still shown as missing/wrong/replaced?  I was thinking there was a way to accomplish this via text file edits, but I really don't know.

 

 

1 hour ago, Frank1940 said:

An off-the-wall question.  You have 19+ disks in your server.  Does it have a single-rail power supply that can provide 45-50 amps of +12volts for those drives? 

 

6 minutes ago, jkwaterman said:

i have a Seasonic FOCUS SSR-750FM 750W 80+ Gold power supply.  I've been using it for 3 - 4 years.

 

I helped with his server build.  This power supply has 62A on +12V if I'm not mistaken.

 

 

Thanks for helping jkwaterman out, guys, I know we both really appreciate it!!!

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.

Guest
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...