Hello Wonderful Unraid folks,

 

I was having an issue with AppData Backup Failing to Run, and noticed my Cache Drive was not mountable. I restarted unraid and the drive came back momentarily and then off again it went. I shut my server down, disconnected and reconnected the drive and now certain dockers are not functioning.

 

Cna you help me get my system back up and running?

 

Diagnostics attached.

tangosierra-diagnostics-20241204-1137.zip

Restarted Unraid, Cache Drive is unavailable agian.

 

Another restart and the cache drive is back.. The following containers aren't running:

 

I was able to get the docker containers to fire back up by doing a force update on them. 

 

I still have about 200gb on the Cache Drive, I can shut the system off and bring it back online and move some of the data, however I am not sure this is good for the system as a whole. Any help would be greatly appreciated.

 

Current Unraid Version 6.12.13

Array (2) 12TB Parity

(1) 12TB Data

(3) 4TB Data

 

(1) 2TB NVME for AppData

(1) 2TB NVME for VMs

(1) 2TB SSD for "Cache"

 

Mapping per below:

 

 

I am happy to use one of the NVME drives ILO of the SSD. Just need to know what my best course of action is here.

 

Thanks!

There are issues with the flash drive:

 

Dec  4 11:28:51 TangoSierra kernel: sd 0:0:0:0: [sda] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=0s
Dec  4 11:28:51 TangoSierra kernel: sd 0:0:0:0: [sda] tag#0 Sense Key : 0x3 [current] 
Dec  4 11:28:51 TangoSierra kernel: sd 0:0:0:0: [sda] tag#0 ASC=0x11 ASCQ=0x0 
Dec  4 11:28:51 TangoSierra kernel: sd 0:0:0:0: [sda] tag#0 CDB: opcode=0x28 28 00 00 00 0e 88 00 00 b8 00
Dec  4 11:28:51 TangoSierra kernel: critical medium error, dev sda, sector 3720 op 0x0:(READ) flags 0x80700 phys_seg 184 prio class 2
Dec  4 11:28:51 TangoSierra emhttpd: error: get_fs_sizes, 7096: Input/output error (5): statfs: /boot
Dec  4 11:28:51 TangoSierra kernel: FAT-fs (sda1): FAT read failed (blocknr 3688)

 

You should fix that first

Thanks Jorge, what do you think that cause is? A faulty flash drive or maybe a different port? This drive was installed less than a year ago. Attached are the updated logs with the system running for about 16 hours.

tangosierra-diagnostics-20241205-0626.zip

Edited December 5, 20241 yr by crspangenberg

"critical medium error" usually means a media error, so most likely a bad flash 

 

 

Sounds good, thank you Jorge, I will go pickup a new flash drive today and get that provisioned. Appreciate your help as always.

So how screwed am I, for whatever reason the cloud based flash backup is no longer active and I can't generate a backup from there (even though I have been able to in the past). The USB drive is now unreadable... I have a picture of the drive assignments but that is about it...

Edited December 5, 20241 yr by crspangenberg

For the devices you just need to reassign the as they, use the screenshot, did you manage to copy anything from the flash? You just need the config folder.

No, unfortunately the flash drive went belly up. Can any of this information be gathered from the diagnostics that were run on the server before this happened?

Some info/files can be gotten from the diags, but you can just assign the devices and then reconfigure the rest, like network, server name, etc.

Thank you @JorgeB , anything I should be careful of as I configure this thing? When I create the new USB, should I utilize the existing OS Version? Or should I update to the latest?

Use the latest stable or 7.0.0-rc1

@JorgeB, that cache drive that started this whole thing isn't being recognized, before I start the array, should I replace the drive with a different ssd? Does it matter the size?

 

In reality, I don't necessarily need the drive, could I use one of the existing NVME drives, and adjust the shares accordingly?

45 minutes ago, crspangenberg said:

could I use one of the existing NVME drives, and adjust the shares accordingly?

You can.

@JorgeB Okay, array is back up.

 

Shares are now showing errors:

 

 

 

I see in my community application backup there is a flash back from a few days ago. 

 

 

Would my best course of action be to replace the SSD with a drive of the same size as a cache pool. And then restore from this backup?

 

 

 

I assume cache no longer exists? If yes, you just need to correct the shares and click apply.

@JorgeB, correct Cache doesn't exist. Changing the shares does fix the issue, however Inoticed that my Plugins are now missing as well. If I restore from the backup and add in a new Cache drive will I cause any issues? I really would like to be back to where I was before this all started occuring. 

 

I've made so many changes to the system over the years that I really don't know where to begin in getting everything back to where it was. 

Correct the share primary and secondary storage and clicking apply must fix the shares, you may need to just make a dummy change, the important part is to click apply.

 

For the plugins, you should be able to restore them from the /config/plugins backup, but copy just the ones you need, you may have some extra stuff there, and some folders there are stock.

@JorgeB sorry for the confusion.

 

But if I restore from the backup that I have and add in a new cache drive, will everything go back to normal? All plugins / drives / etc.

 

No more modifications.

@JorgeB,

 

Please disregard previous post. I copied over the plugins folder and everything seems to be working as intended. I just need to transfer the license over and we should be good to go. Thank you so much for your help. Is there anything else I should do to make sure this is function as it should?

Should be all.