[dm-devel] Multipath not balancing I/O across paths evenly.

Devin delldudedevin at gmail.com
Wed Jul 10 20:52:27 UTC 2013


We started to notice that on our iSCSI hosts that are using Multipath that
they seem to be sending most of the traffic across one port even though
they have 4 paths for each mpath device. We are using a Hitachi SAN.

Here is my multipath configuration:
defaults {
        user_friendly_names yes
}

devices {
        device {
                vendor                   "HITACHI"
                product                  "DF.*"
                path_grouping_policy  multibus
                path_checker              tur
                getuid_callout            "/sbin/scsi_id -g -u -s /block/%n"
                no_path_retry             1
                features                  "1 queue_if_no_path"
                }
}


The output of multipath looks like:

mpath2 (360060e80101af190058baee900000007) dm-6 HITACHI,DF600F
size=1.5T features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=0 status=active
  |- 8:0:0:1   sdg  8:96   active ready running
  |- 9:0:0:1   sdh  8:112  active ready running
  |- 7:0:0:1   sdf  8:80   active ready running
  `- 10:0:0:1  sdi  8:128  active ready running
mpath0 (360060e80101af190058baee900000005) dm-5 HITACHI,DF600F
size=1.0T features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=0 status=active
  |- 9:0:0:0   sdd  8:48   active ready running
  |- 8:0:0:0   sdc  8:32   active ready running
  |- 10:0:0:0  sde  8:64   active ready running
  `- 7:0:0:0   sdb  8:16   active ready running

However when I do a tcpdump and count the number of packets to each of the
IQN's they are not equal, not even close.


IQN1: 19,579

IQN2: 12,778

IQN3: 70,674

IQN4: 67


Is there anything I need to do in order to ensure that Multipath balances
more evenly across all the paths???

Any help would be appreciated.


Devin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/dm-devel/attachments/20130710/14e58a85/attachment.htm>


More information about the dm-devel mailing list