[SOLVED] How do I prevent a full cache drive from causing mayhem?

[02ranger]

Hi! So far my Unraid build has been solid (because Unraid is awesome) but last night my cache drive filled up for the first time and more data was queued to be downloaded and as a result Docker just did not handle it well. I experienced the execution error described in this forum post. By the time I woke up this morning the mover had done it's thing and the cache drive was no longer full but the damage had been done. Luckily no lasting damage, one reboot later and all is well again. 

 

My question is, why did this happen in the first place? The share that was being written to has "Use cache pool" set to "yes" which I thought meant it would start writing directly to the array if the cache drive was full. Is there something I can do to prevent this from happening again, other than just increasing the size of my cache pool or being more aware of the size of incoming data? Is there a way to "reserve" X amount of space on the cache drive? Do I have an incorrect understanding of that "Use cache pool" setting?

 

Thanks for whatever assistance you can provide!

 

Edit: I may have misrepresented but I don't really think the issue came down to the downloaded data trying to write to the cache drive, although maybe it was. I feel like the downloaded data probably filled the cache drive and then other processes that also use the cache drive were unable to write to the full drive, but my understanding of Unraid and Docker isn't detailed enough to know for sure.

Edited May 2, 2021 by 02ranger
SOLVED

[trurl]

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

[02ranger]

Diagnostics file is attached.

unraid-diagnostics-20210502-0955.zip

[itimpi]

Did you have a Minimum Free Space value set for the cache?     This is required to indicate to unRaid the threshold at which it should stop trying to use the cache and write any new files directly to the array.    It is recommended that this is set to be larger than the largest file you expect to write.  Once unRaid has selected a drive for a new file it will give an error if the file does not subsequently fit.

[02ranger]

I have not. I see the minimum free space setting per share but I guess I'm not following exactly how that will help prevent this from occurring again in the future. Does unraid also apply this minimum free space to the cache drive for each individual share? I guess I interpreted that as specifically pertaining to the array so no single drive in the array ran out of space in the middle of writing a file, but never thought it would apply to the cache drive.

[itimpi]

15 minutes ago, 02ranger said:

I have not. I see the minimum free space setting per share but I guess I'm not following exactly how that will help prevent this from occurring again in the future. Does unraid also apply this minimum free space to the cache drive for each individual share? I guess I interpreted that as specifically pertaining to the array so no single drive in the array ran out of space in the middle of writing a file, but never thought it would apply to the cache drive.

Yes - there is a setting for the cache drive.

 

if running 6.8.3 then it is under Settings - Global Share Settings

If running 6.9.2 then click on the pool name to set it for that pool.

[02ranger]

OK, thanks for the help!! I think that should prevent this from happening to me again, so long as I set that minimum free space to a large enough value to cover whatever individual files I download. Might be time to increase the size of my cache pool since a large enough free space value eats about a quarter of my available cache space....