I didnt get any response to my first query, so I'm asking again. Apologies if you replied and I missed it - please re-send.
I have a system with about 50 SCSI drives spread across 5 buses, mostly with 1 ext3 filesystem per drive. After a crash I decided to accept the boot-time offer to fsck them. About half a dozen proceeded to be checked simultaneously, but thereafter the rest went off one by one.
Is this a bug? Should fsck be able to handle this many in parallel? Or should it do things in smaller parallel chunks?
I have to confess that I'm running RHEL4 and my e2fsck version is still 1.35, 28-Feb-2004. Maybe someone knows that this has been fixed in a later version.
My fstab has a '2' in the 6th position for all filesystems except root, which has a '1' as instructed in 'man fstab'
Cheers, Terry.
T. Horsnell wrote:
I didnt get any response to my first query, so I'm asking again. Apologies if you replied and I missed it - please re-send.
I have a system with about 50 SCSI drives spread across 5 buses, mostly with 1 ext3 filesystem per drive. After a crash I decided to accept the boot-time offer to fsck them. About half a dozen proceeded to be checked simultaneously, but thereafter the rest went off one by one.
Is this a bug? Should fsck be able to handle this many in parallel? Or should it do things in smaller parallel chunks?
It is my understanding that fsck will only check 1 drive on a bus at a time. I see this with my IDE drives. You said that you only see about 1/2 a dozen start up. Are you sure this isn't *5* (one on each SCSI bus)? And then when a drive completes, another on the same bus is then scheduled?
I have to confess that I'm running RHEL4 and my e2fsck version is still 1.35, 28-Feb-2004. Maybe someone knows that this has been fixed in a later version.
Isn't this list for Fedora Core questions?
My fstab has a '2' in the 6th position for all filesystems except root, which has a '1' as instructed in 'man fstab'