Cache drive upgrade, which procedure to follow?

[Yarok]

Hi everybody,

 

I'd like to swap my 250Gb cache for a 1Tb drive. 

 

I've seen several methods here and on the wiki but they're either 6.2 and above or 6.1 and below. Thing is, i run 6.1.3 so i'm not sure which procedure to follow. 

 

V6.4 is just out (as of today in fact) so is an upgrade recommended to have the Use cache drive: prefer/replace options available?

 

Thanks for the help!

Edited January 21, 2018 by Yarok

[JorgeB]

Upgrade to v6.4 and use this one:

 

https://lime-technology.com/forums/topic/46802-faq-for-unraid-v6/?do=findComment&comment=511923

 

[Yarok]

Thank you Johnnie!

[Yarok]

Okay so i gave the step by step a try after upgrading to 6.4 but i'm stuck a some point. More specifically:

 

- Click on Shares and change to "Yes" all cache shares with "Use cache disk:" set to "Only" or "Prefer"

 

When in Global Share Settings / Cache Settings, the Use cache disk only has a Yes or No setting, but no Only or Prefer.

 

 

 

Any idea what i should do next?

[JonathanM]

8 minutes ago, Yarok said:

When in Global Share Settings / Cache Settings,

There's your problem. For this operation you set the cache preference on a per share basis, not globally. Go to the Shares tab, and click on the share(s) you need to change.

[Yarok]

Hmm, not sure what went wrong here...

 

I wanted to change the cache preference in the Shares tab like you've mentioned, Jonathan, but it seems my 250Gb is now unmountable for lack of a file system (yet btrfs seems to be detected!?).

 

 

Any way i can mount the drive back without formatting and losing its content?

Edited January 21, 2018 by Yarok

[JorgeB]

Is the cache filesystem btrfs? Also post your diagnostics: Tools -> Diagnostics

[Yarok]

Yes, the filesystem was btrfs.I've attached the diagnostics.deimos-diagnostics-20180121-2216.zip

[JorgeB]

It's complaining of a missing device:

 

Jan 21 20:30:28 DEIMOS kernel: BTRFS warning (device sdi1): devid 2 uuid c1780e17-92fd-4249-925f-80793524bf85 is missing

Try again after changing the cache slots to 2, but leave the second cache slot empty.

[Yarok]

allright that did the trick. I've started a backup of my cache on another drive (there are files there that do not belong on a particular share, as i'm using the cache for Transmission and such).

 

When i come back home later today i'll simply stop the array, switch the drive to the 1TB and restore the backup folders on the new cache.

 

Does this sounds good? Is there something else i need to take care of?

[JorgeB]

1 hour ago, Yarok said:

there are files there that do not belong on a particular share, as i'm using the cache for Transmission and such

Any files outside a share, i.e., on the cache root will need to me moved manually.

[Yarok]

Back home, trying to unmount the 250Gb and mount the 1Tb, but:

 

- if i mount the 1tb and set the number of cache slots to 1, the status is "Unmountable: Unsupported partition layout". When i hit Format, nothing happens (the screen sort of flashes like something should happen, then gives me the same status);

- if i mount the 1tb and set the number of cache slots to 2, the status is "Unmountable: no pool uuid". Same behaviour when i try to Format, nothing happens;

 

I still can put the system back to the old state by setting 2 slots and mounting the old 250Gb.

 

 

[JorgeB]

Please post your diags after adding the new cache device and trying to format it.

[Yarok]

Here's the diagnostic when the 1Tb is mounted and only one slot is configured. The status on the main screen is Unmountable: No file system, this time.

 

(and again, thank you Johnnie!)

 

deimos-diagnostics-20180122-2017.zip

Edited January 22, 2018 by Yarok

[JorgeB]

ATA errors ending up with the disk being disable:

 

Quote

Jan 22 20:13:52 DEIMOS kernel: ata7.00: status: { DRDY }
Jan 22 20:13:52 DEIMOS kernel: ata7.00: failed command: WRITE FPDMA QUEUED
Jan 22 20:13:52 DEIMOS kernel: ata7.00: cmd 61/40:98:f0:62:70/05:00:74:00:00/40 tag 19 ncq dma 688128 out
Jan 22 20:13:52 DEIMOS kernel:         res 40/00:5c:40:00:00/00:00:00:00:00/40 Emask 0x10 (ATA bus error)
Jan 22 20:13:52 DEIMOS kernel: ata7.00: status: { DRDY }
Jan 22 20:13:52 DEIMOS kernel: ata7: hard resetting link
Jan 22 20:14:02 DEIMOS kernel: ata7: softreset failed (1st FIS failed)
Jan 22 20:14:02 DEIMOS kernel: ata7: hard resetting link
Jan 22 20:14:12 DEIMOS kernel: ata7: softreset failed (1st FIS failed)
Jan 22 20:14:12 DEIMOS kernel: ata7: hard resetting link
Jan 22 20:14:47 DEIMOS kernel: ata7: softreset failed (1st FIS failed)
Jan 22 20:14:47 DEIMOS kernel: ata7: limiting SATA link speed to 1.5 Gbps
Jan 22 20:14:47 DEIMOS kernel: ata7: hard resetting link
Jan 22 20:14:52 DEIMOS kernel: ata7: softreset failed (1st FIS failed)
Jan 22 20:14:52 DEIMOS kernel: ata7: reset failed, giving up
Jan 22 20:14:52 DEIMOS kernel: ata7.00: disabled

 

Check cable, it might also be a bad disk, since it's disabled there's no SMART report.

[Yarok]

Well Johnnie, it seems that did it, yet again. Not on the first try (had to reboot a few times), and while i swapped the sata cable, i disconnected the other Gb drive completely. After that i was able to Format correctly.

 

Copied my Cache backup from this morning, updated Transmission (als had a message saying that the Docker image was not valid anymore because of the upgrade to v6.4), and now everything seems to run as it should!!

 

So again, thanks a lot for your help