On 07/23/2010 03:21 PM, Anil wrote:
I removed zero-copy to eliminate possible bug in my block layer code.
But still I see these bellow messages and that too continually. I'm
just allocating pages, filling up bio, using bio_add_page and then
calling generic_make_request. And for each IO that happens I see the
icsi connection getting dropped.
I've tried with various queue properties for the block device (but, i
dont think that makes any sense in my case, as i'm directly reading
from scsi device and writing to another scsi device, my block device
properties are not playing any role in there, as my block device sits
on top of both those devices). Should I be tweaking any underlying
device queue properties?
Unfortunately, we've started seeing these errors on SLES 11 kernels
too. Still the below errors occur only on xen machines, and it doesnt
matter if the disks that are syncing data are iscsi or xvd's. Kindly
give your advice.
Thanks,
Anil P.
Jul 24 01:36:29 DS-1-29-SLES10-SP2-64bit-Xen iscsid: connection1:0 is
operational after recovery (1 attempts)
You should have a error message before this. What is it? Give me the
beginning. Don't just cut it off.
Also can you replicate this with a upstream kernel?
--
You received this message because you are subscribed to the Google Groups
"open-iscsi" group.
To post to this group, send email to open-is...@googlegroups.com.
To unsubscribe from this group, send email to
open-iscsi+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/open-iscsi?hl=en.
