High gang, I had a power outtage today, I do not have a UPS yet, and when I brough the server back up the last drive I put in red balled. I checked the cable and power and they seem to be fine. There is not a lot of data on the drive since it has only been in a few weeks and passed 3 passes with the preclear script.

 

I am on v 4.7 and syslog is attached.

 

Thank you for your assistance.

 

John

syslog-2012-05-07.txt

Smart Report

Smartctl.txt

Any Opinions on this? I am not sure if I should do the trust my array procedure or not.

Any Opinions on this? I am not sure if I should do the trust my array procedure or not.

A "red" indicator ONLY occurs when a write to the disk fails.

 

Therefore, it is guaranteed that the disk is not correct...

It could be as innocent as a time-stamp of when the disk was last mounted, or it could be as important as your most precious photo of a family member.  It could be part of the data structures used for the journaling file system, or part of the directory hierarchy structure.

 

It is your call...  If you are tempted to use the "trust" procedure than I feel you should at the very least

ALSO perform a non-correcting parity check to make sure all is OK

and

ALSO perform a reiserfsck file-system check on all of the data disks in your array.

 

Joe L.

I am open to suggestions as I am a linux noob and the smartctl looks good from what I know. there is nothing irreplacable on that drive, I am more worried about the integrity of the rest of the array.

See here: http://lime-technology.com/wiki/index.php/Troubleshooting#Hard_drive_failures

Thanks, I am poking around in there now, I just don't understand the log and the smartctl well enough to decide on what direction to go in, Is the drive actually failing or was it a trainsent problem due to the power outage. Looks like reiserfsck --check is my first step if I understand this correctly.

 

I just don't want to leave the array running too long in this condition if I am doing things in the wrong order or the wrong thing all together.

The SMART report has no line feeds and is very hard to read. It looks ok from what I can make out.

The SMART report has no line feeds and is very hard to read. It looks ok from what I can make out.

 

Thank you,  do I  reenable the drive and run reiserfsck or do I run that before doing the trust my array first?

 

EDIT, Sorry stupid question as drives have to be unmounted to run this. I am on my 4th 12 hr shift in a row so my thought process is not what it should be

 

Reiserfsck is running, so far no coruption on the disabled disk others are testing.

 

 

Did you read this section: http://lime-technology.com/wiki/index.php/Troubleshooting#What_do_I_do_if_I_get_a_red_ball_next_to_a_hard_disk.3F

Did you read this section: http://lime-technology.com/wiki/index.php/Troubleshooting#What_do_I_do_if_I_get_a_red_ball_next_to_a_hard_disk.3F

 

Yes and I got the impression from JoeL's reply that trust my array may not have been the best course of action and that more info was needed.

 

Ultimatly I ASSUME it comes down to trust my array, or rebuild from parity and I don't know enough about this to know which is best to trust.

 

As I stated eariler nothing on the red balled drive is irreplacable, it is the integrity of the array as a whole I am concerned with.

 

Thank you again for the help !!

 

 

Or am I mis reading this and it is rebuilding from parity either way? And the question is if the physical drive is to be trusted or not?

 

I would rebuilt the red drive. This is the safest option.

Thank you, I will do that when the current reiserfsck is finished...or when I get back home from work in the morning, which ever comes first... 

Looks like everything is fine.

 

Thanks for all of the help !! Just in from work going to pass out now.....