Need suggestion how to do multiple dangerous actions

[Plamen]

Hello,

 

My unraid has the following drives:

- 2, 3, 3, 8, 14

- parity 14

- cache 256gb

 

Use case:

Cache has windows vm share with cache only option. I also have Plex running with all scanning and analyzing for intros enabled.

 

Problem:

I just lost one of the 3tb. It has 1tb of data written on it.

 

The goal is to only buy 1 new hdd. And maybe 1 SSD as well.

 

Ideally I would like to replace the parity with 18tb, and use the old 14tb parity drive to replace the 3tb that is dead.

Since the 3tb is successfully "emulated" I consider moving it's data to the other drives, removing it completely from the system. Than since the parity is not valid any more, replace that parity drive with the new one, and rebuild only once.

 

^ I would need confirmation if this is possible and correct, and maybe a suggestion which part of the guide for replacing parity I should follow. 

 

And my second question is about appdata and similar folders with higher rate of read/writes. I consider getting a large SSD ( 1 or 2tb ) as cache drive, and use the "preffered" option. The failed drive is actually my second drive housing appdata that has failed, it may be just a coincidence, however it's possible activity is enough to kill non-ssd.

 

^ Will this reduce read/writes to the array itself and only sync during the "mover" action?

[trurl]

On mobile now so can't explain details. Don't do anything without further advice. 

 

There is a built-in procedure called Parity Swap just for this situation. 

 

Will follow up in a few hours if nobody else does. 

 

In the meantime, 

 

Go to Tools-Diagnostics and attach the complete Diagnostics ZIP file to your NEXT post in this thread. 

[Plamen]

Here is the requested diagnostics. I do tinker a lot with the settings, and try different things. You may see some weird shares configurations.

 

No worries, I am not doing any actions, since I haven't bought the HDD yet. I need to know if this will work the way I think it may, or I will just get another 14tb. Also the store website says delivery time is 8 days, so it will take a while until a get the new HDD.

plasmennas-diagnostics-20210126-1241.zip

Edited January 26, 2021 by Plamen

[itimpi]

Just as a check point are you sure the ‘failed’ drive really has failed?   It is more common for drives to be disabled because a write to them failed due to an external factor rather than the drive actually failing.  Knowing the answer to this might affect whether it is recommended you go with the Parity Swap procedure or not.
 

it is also worth pointing out that the 6.9.0 release supports multiple cache pools so if you have 2 SSD’s you could easily have one dedicated to VM use and the other to Docker use with either/both/none doing actual caching of writes to shares.

[trurl]

SMART for disabled disk2 looks OK and since you rebooted after it became disabled nothing to see in syslog about that.

 

Check connections then run an extended SMART test on it.

[Plamen]

Did an extended SMART test, and it showed no issues.

 

I followed the "How do hard drives become disabled?" guide to re-enable the drive. Now it's rebuilding.

 

It turns out that you were correct - it was an issue with connection, specifically the power cable. I was using SATA splitter from 1 to 4. Replaced it by adding one more SATA to the PSU. Originally used splitter for less cables in the case.

 

( This is the second drive that failed. When the first one did, and I reconnected a brand new drive. I listened with stethoscope, and it was making bad sounds, so I assumed it was bad. When I connected the new drive, it started making absolutely the same noise, so I  assumed faulty connector. I have 4 power connectors on that cable, but I need 3, so I just moved the drive to another one, and it hasn't failed since then. Now with the second drive failing I noticed it's also connected to the same splitter cable, so I replaced the entire cable. )

 

Now a re-build is running on the 3tb drive, and I started pre-clear on the drive that I previously replaced, that was probably working as well.

 

----

 

Does anyone has a suggestion about the second part of my question about cache drives - does "Prefer" keep second copy of the data on to the array, protecting the data, while still keeping it in the cache on the SSD for quick access:

Quote

Prefer indicates that all new files and subdirectories should be written to the Cache disk/pool, provided enough free space exists on the Cache disk/pool.
If there is insufficient space on the Cache disk/pool, then new files and directories are created on the array.
When the mover is invoked, files and subdirectories are transferred off the array and onto the Cache disk/pool.

 

[itimpi]

10 minutes ago, Plamen said:

Did an extended SMART test, and it showed no issues.

 

I followed the "How do hard drives become disabled?" guide to re-enable the drive. Now it's rebuilding.

 

Does anyone has a suggestion about the second part of my question about cache drives - does "Prefer" keep second copy of the data on to the array, protecting the data, while still keeping it in the cache on the SSD for quick access:

 

No, Unraid never keeps copies of a file in 2 places intentionally.

 

some people have exploited the fact that if a file DOES exist in 2 places then Unraid preferentially finds and uses the copy on the cache.   You can then have user supplied scripts that operate a bit like mover to periodically copy the cache copy of the file over-writing the one on the array.

[Plamen]

Thank you all for the help. I will keep researching about duplicating data after I get new ssd (maybe the CA plugin that does just that). For now appdata is on the array itself.

Edited January 28, 2021 by Plamen

[trurl]

CA Backup plugin will backup appdata to the array so you can keep appdata on cache where it belongs. domains and system shares should be kept on cache also. Dockers and VMs will perform better and won't keep array spunup since these files are always open.