all of the logical volume in the volume group have stale mirror!!!

[John Poff · Accepted solution]

Hello Francesca,

To determine if the problem is the controller or a disk drive, I would see if you can access one of the other disks on that same controller. If another disk has errors, it is probably the controller, but if the other disks seem ok it is probably just a bad disk. Judging by the errors in your syslog, my guess is that it is a bad controller. Deciphering the dev: field yields:

c14t4d0
c14t5d0
c14t6d0

So my guess is that it is pretty unlikely that all those disks died at the same time, and it probably is a controller. Time to call HP.

Here is a link to a thread that explains how to decipher the dev: entry in those syslog error messages:

http://forums.itrc.hp.com/cm/QuestionAnswer/1,,0x31627bb04b5cd611abdb0090277a778c,00.html

JP

[Sean OB_1]

If all of your mirrors are stale that indicates the controller is bad.

Check your logs for I/O errors, if you are seeing more than one drive having errors then it is almost certainly the controller.

[Dietmar Konermann]

Other candidates include cables, terminators, connectors, etc. The usual suspects.

[MANOJ SRIVASTAVA]

I think the porblems lie in the conmtroller or the conenction with the disks which make them offline etc while workin you need to repalce or tightern the conenctions , the same disables the mirroring bcoz the PV is not availbe erratically , what I will do is to log a H/W call with HP and get the controller changed and then watch .


Manoj Srivastava

[Joaquin Gil de Vergara]

run a

# vgsync vg00

if disks are correct it resynchronize the vg

[Francesca Caruso]

Thank you!
I'll try to access the other disks on the controller and I'll let you know.

Thanks!!!

Francesca

[Francesca Caruso]

vgsync worked! So, I must only see if there is a problems on the other disk on the controller or not.

Thanks

Francesca