How do I get statistics for our HBAs?
We are using PowerPath 5.5 on 64-bit RHEL 5.6 (kernel 2.6.18-238.5.1.el5) with QLogic 2460 HBAs.
Using:
powermt display
only reports:
Symmetrix logical device count=0
CLARiiON logical device count=2
Hitachi logical device count=0
HP xp logical device count=0
Ess logical device count=0
Invista logical device count=0
==============================================================================
----- Host Bus Adapters --------- ------ I/O Paths ----- ------ Stats ------
### HW Path Summary Total Dead IO/Sec Q-IOs Errors
==============================================================================
1 qla2xxx optimal 4 0 - 0 0
2 qla2xxx optimal 4 0 - 0 0
Why are the Stats not accounted for?
Is there any method by which I can get throughput and latency for the HBAs?
TIA
you can try "powermt display dev=all every=10" ..or powermt display every=10"
Thanks for the response - powermt display every=10 does give me IO/Sec after the first sample.
How does this equate to throughput or latency?
Is the IO a Linux IO (512-byte) or a standard FC IO (I think 2048 bytes)?
I'm trying to determine the load on these HBAs...
io size depends on what your application is doing, probably need to look on the array side.
Thanks.
This is an Oracle DB using ASM.
I would expect the latency to be near 0 (on the order of ~2ms) if the Q-IOs are near or at 0, correct?
to me latency = response time , 2ms is awesome
Very fast SAN - too much for this environment at the current time.
Looking at potentially removing this DB from this SAN - to what type of storage is the question.
That's why I'm trying to get these stats...
Thanks to your help, this seems to work very well:
powermt display every=5 | grep qla | awk '{now=strftime("%d-%b-%Y,%T"); print now " " $0}'