On 04/22/2011 03:30 PM, Laurent Bigonville wrote: > > > size=500G features='1 queue_if_no_path' hwhandler='1 rdac' wp=rw >>> > > |-+- policy='round-robin 0' prio=12 status=active >>> > > | |- 0:0:1:1 sdb 8:16 active ready running >>> > > | `- 1:0:1:1 sdd 8:48 active ready running >>> > > `-+- policy='round-robin 0' prio=2 status=enabled >>> > > |- 0:0:0:1 sda 8:0 active ghost running >>> > > `- 1:0:0:1 sdc 8:32 active ghost running Okay!! I never got to this one because I never worked with this target. Thank you. It is always good to know more. :-) The secondary paths (or in this case, the ghost paths) are, by design, supposed to fail on I/O requests. But they should respond to read_capacity and other commands. More details as described in the source code itself:
* PATH_GHOST:
* - Use: Only hp_sw and rdac
* - Description: Indicates a "passive/standby" path on
active/passive HP
* arrays. These paths will return valid answers to certain SCSI
commands
* (tur, read_capacity, inquiry, start_stop), but will fail I/O
commands.
* The path needs an initialization command to be sent to it in
order for
* I/Os to succeed.
*
So going back to this bug reports actual question. If the multipathed
map is working, there's no bug. You could ignore these scsi errors. This
is similar to a faults scenario. At faults, during the recovery timeo,
scsi throws a lot of error messages, which should be ignored.
--
Ritesh Raj Sarraf
RESEARCHUT - http://www.researchut.com
"Necessity is the mother of invention."
signature.asc
Description: OpenPGP digital signature
