Weird RAID problem with no solution on my TS-453A

[bonanzabonanza]

hello,
I'm in a "no exit" situation with my nas

configuration is easy
3 disks RAID 5 + 1 disk (maybe my fault marked ad available and not spare - seems not important)
all 3Tb WD red

yesterday 1 disk was marked in a warning state (not defunct) - SMART paramaters over quota. suggest to replace disk.

should be a not dangerous situation

after this, using file sharing become almost impossible. minutes to copy a few Kb file and minutes to browse small directories

web interface nearly unusable. 5 minutes to log-in and more to see the main menu
logs not accessible - helpdesk not accessible

wasting a lot of time I've been able to reach storage pool page where the 3 disk are in use (one with warning - as already said).
immediate prompt to perform a file system check.
I've done and waited 18 hours ca.
at the end no errors but again the same prompt.
during that time nothing is possibile:
- no way to mark disk as defective
- no way to add the 4th disk as spare
- no way to execute any of the 'manage' commands

so, this seems to be an infinite loop with no exit and no action possible

the only thing I suppose possible is to shutdown nas, remove warning disk and hope that RAID survives let me a chance to add the spare disk
but I have to do this in a "not clean filesystem" state

any other idea?

thank you
p.

[dolbyman]

no..the way it works it to hotswap the busted drive

you cannot modify a degraded raid..thats why all other options are greyed out

[bonanzabonanza]

no..the way it works it to hotswap the busted drive

you cannot modify a degraded raid..thats why all other options are greyed out

so, I have to remove the 'warned' disk?

[dolbyman]

pull it out and replace it with a new one

[bonanzabonanza]

pull it out and replace it with a new one

it seems that nothings has exploded!

do you think that I have to keep finger crossed for 6 hours?

[dolbyman]

you can always login via ssh and check the rebuild progress via

Code: Select all

cat /proc/mdstat

[bonanzabonanza]

you can always login via ssh and check the rebuild progress via

Code: Select all

cat /proc/mdstat

immediately after disk removal, web interface has come back alive
it's friday afternoon, I've put rebuild as priority.

only doubt: even if rebuilding is asking me to perform a file check.
I suppose it's better to postpone at the end of rebuild

thank you so much!