I have a server that is a database master. It's supposed to replicate with about 75 other servers at various sites around the globe. About a month ago, after 2 new luns were added, the replication started failing. I am concerned that i/o to the san is being overloaded. Is there anything I can use to monitor the i/o through the fibre cards?

(It's a Hitachi SAN, and I am using Hitachi fibre cards - multipathing is enabled.)

# multipath -ll
mpath2 (360060e800429550000002955000003b9) dm-31 HITACHI,OPEN-V
[size=34G][features=0][hwhandler=0]
\_ round-robin 0 [prio=0][active]
\_ 0:0:0:2 sdc 8:32 [active][ready]
mpath1 (360060e8004295500000029550000039a) dm-30 HITACHI,OPEN-V
[size=34G][features=0][hwhandler=0]
\_ round-robin 0 [prio=0][active]
\_ 0:0:0:1 sdb 8:16 [active][ready]
mpath0 (360060e80042955000000295500000323) dm-29 HITACHI,OPEN-V*11
[size=374G][features=0][hwhandler=0]
\_ round-robin 0 [prio=0][active]
\_ 0:0:0:0 sda 8:0 [active][ready]

Any help is appreciated.

ETA: My multipath.conf:

defaults {
udev_dir /dev
polling_interval 10
selector "round-robin 0"
#path_grouping_policy multibus
path_grouping_policy failover
getuid_callout "/sbin/scsi_id -g -u -s /block/%n"
prio_callout /bin/true
path_checker readsector0
rr_min_io 100
rr_weight priorities
failback immediate
no_path_retry fail
user_friendly_name yes

iostat should monitor all the disk I/O including the ones presented via the SAN which are sda, sdb & sdc based on your output.

You may be able to get additional metrics by direct monitoring of your fibre switch(es).

However, it doesn't seem likely you'd be able to overload the I/O of fibre very easily so it may be another issue.

We do EMC with PowerPath and Navisphere here so I can't give you a lot of guidance on the Hitcachi equivalents. I would think that the Hitachi software you installed (if any) would have some tools for this as well.