Jump to content

Suddenly have permissions issues on my cache drive


Recommended Posts

What device is the cache drive?

 

The syslog shows that there appears to be reiserfs file system corruption on /dev/sdf1.  That could explain your problems.  If that is the cache disk then the recovery action is to make use of reiserfsck. 

Link to comment

What device is the cache drive?

 

The syslog shows that there appears to be reiserfs file system corruption on /dev/sdf1.  That could explain your problems.  If that is the cache disk then the recovery action is to make use of reiserfsck.

Yup, that's the one.  Just found the guide on the wiki: http://lime-technology.com/wiki/index.php/Check_Disk_Filesystems

 

Is there any chance of Data loss?  Is there a real reason why this would randomly happen?

Link to comment

Yup, that's the one.  Just found the guide on the wiki: http://lime-technology.com/wiki/index.php/Check_Disk_Filesystems

 

Is there any chance of Data loss?  Is there a real reason why this would randomly happen?

This normally happens when a write to the disk has failed for some reason. 

 

In terms of data loss, if it happens it is normally restricted to the file that was being written at the time the write failed, and even then it is normally recovered successfully from the reiserfs transaction logs.  Why this should happen could be a wide variety of reasons ranging from problems with the cables, unscheduled power loss, forced power off, etc.

 

You should start by stopping the array and then running the command:

reiserfsck --check /dev/sdf1

.  This will both confirm whether you have a reiserfs error, and give recommendations on how to proceed to fix it.    If in any doubt you should check back here before taking suggested recovery action as if you make an error at that point it can result in data loss.

Link to comment

In terms of data loss, if it happens it is normally restricted to the file that was being written at the time the write failed, and even then it is normally recovered successfully from the reiserfs transaction logs.  Why this should happen could be a wide variety of reasons ranging from problems with the cables, unscheduled power loss, forced power off, etc.

 

You should start by stopping the array and then running the command:

reiserfsck --check /dev/sdf1

.  This will both confirm whether you have a reiserfs error, and give recommendations on how to proceed to fix it.    If in any doubt you should check back here before taking suggested recovery action as if you make an error at that point it can result in data loss.

 

It was aborted with the error

 

"bread:  Cannot read the block:  (Input/output error)"

 

 

Link to comment

It was aborted with the error

 

"bread:  Cannot read the block:  (Input/output error)"

That suggest that the device might be physically failing, in which case recovery options are limited and you may need to use your backups to get data back. 

 

The other possibility is that the drive has dropped offline for some reason.  In which case you should power off; check the cables; and reboot the system as that sometimes gets it back.  It is also worth checking that the drive can be seen at the BIOS level.

Link to comment

It was aborted with the error

 

"bread:  Cannot read the block:  (Input/output error)"

That suggest that the device might be physically failing, in which case recovery options are limited and you may need to use your backups to get data back. 

 

The other possibility is that the drive has dropped offline for some reason.  In which case you should power off; check the cables; and reboot the system as that sometimes gets it back.  It is also worth checking that the drive can be seen at the BIOS level.

 

I can read everything off it fine, I just can't write to it.  I'm copying a few files off it and onto the main server right now (I had to disable usage of the cache disk, so it's going slow).  Is there a way to just format the drive and try again?

Link to comment

I can read everything off it fine, I just can't write to it.  I'm copying a few files off it and onto the main server right now (I had to disable usage of the cache disk, so it's going slow).  Is there a way to just format the drive and try again?

If you can get the data off the drive, then you can try putting it through a pre-clear cycle to see what happens.

Link to comment

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...