Cache drive went read-only suddenly (5.0.6)

[chickensoup]

Hey guys,

 

So a reboot seems to have fixed this but I thought I'd dump my log since it was really weird and kind of out of the blue.

 

I was copying an MKV across to my unraid box and as I shut down a VM my desktop was running (which shares the same network adapter on my desktop) as if almost perfectly timed my copy bombed and as far as I can tell my cache drive became read-only. I was left with a 0byte file on my cache drive and no longer able to write. I was able to delete the file from the cache drive then rebooted the server then everything seems to be good again.

 

Any ideas on what might have caused this? Is my cache drive on it's way out?

 

Syslog (from webui) + cache drive smart report attached. 

 

syslog_tail.txt

smartctl.txt

[chickensoup]

Just to add to this- it seems that my cache drive writes while working again have significantly decreased in write speed. Previously I was getting ~100MB/Sec and now I get around 30MB/Sec. The data is being written to the cache drive (not the array) according to my mapped share and the unraid webUI (# writes).

 

 

[dgaschk]

Attach entire syslog. See here: http://lime-technology.com/forum/index.php?topic=9880.0

[chickensoup]

Well, I don't know what caused it but I'm going to have to call this solved again for now.

 

Yesterday I realized the BIOS date was wrong on my server (mobo battery is dying). After correcting it I shut everything down and today it starts boot-looping. I removed the flash, re-ran make_bootable and checked over the BIOS settings again. I noticed the controllers were in IDE Mode instead of AHCI.. probably since the last time the BIOS reset. I re-configured it all, rebooted and all looks good now. Over 100MB/Sec writes again.

 

Bad luck, bad flash, bad bios reset.. who knows. Fixed

[chickensoup]

I spoke too soon once again :'(

 

After an hour or so of heavy copying, back to 28-30MB/Sec and cache drive is at 46 degrees. I have a spare certified repaired drive here (same model) I might try as a replacement in the interim and see what happens.

 

Syslog attached.

syslog.txt

[dgaschk]

At what time did the slowdown occur?

[chickensoup]

At what time did the slowdown occur?

 

I'm not sure exactly, I've swapped out the drive for now as it got even hotter (47 degrees) and was performing badly- the replacement seems to be OK but I haven't had time to test it properly yet, job interview/study/running around.

 

I've moved the replacement to a cooler position as well, I'll run a few independent tests on the drive I pulled out (and get it hot first) and if it looks OK maybe I'll keep it for something else. It's probably just getting tired.