Multipath not working in AIX 5.3
Hi
I am facing a problem in multipathing in AIX 5.3. My AIX box connected having two FC HBAs and both are connected with two FC SAN switches. I have assigned 3 Luns to AIX If i issue lsdev -Cc disk the Luns shows exeactly 3 mean Multipathing working. But Ironacly If i switched off one SAN switch and when I issue df -g or ls to mounted storage for checking. The above command get hang and If i issue cfgmgr it also get stuck. Please help why this problem is coming.. RGD Charanjit Singh |
What does the error report say?
What about "mount", "lslv", "lsvg" and friends? Just because "lsdev" is OK, doesn't mean everything else is. |
what reserve policy do you have on the luns?
what fast fail policy do you have on the luns? what dynamic tracking policy do you have on the luns? what it the back end storage type, firmware, device driver, host attachment script, etc.? |
Storage is IBM DS 4700 and from storage manager if I see its showing me this error in event log:
logical drive not on preferred path due to adt rdac failover - event 4011 Please help why the is error message is coming how to rectify it. And this error message comes when I switched off one SAN switch for checking multipath failover. |
...
|
You can check your MPIO paths (without doing something drastic like turning off a SAN switch) by using the lspath command. Run this command against your 3 LUNs(hdisk2, 3 and 4 i'm assuming?) to check that there are paths coming from both HBA's.
lspath -H -l hdisk# You might also want to check the zoning on your switches. |
lspath
Quote:
it shows me something like this # lspath -l hdisk4 fcs0 path Enable fcs1 path Enable Message comes means path is coming from both fc HBA isn't it. If I switch off one switch then issue this same command again. It shows me like this fcs0 path Enable fcs1 path Failed But Ironacly if I issued ls or try to make any folder its got stucked. By the suggetion of IBM experts I have updated AIX TL and SP and installed the SDDPCM driver. Now after updating and instaing driver issue is it is faling over and failing back but takes a time of 5-10 mins. I have asked my storage engg. for rechecking zones it said all settings are OK from his end. Please suggest where is the loophole. Thx & RGD Charanjit |
what does:
lsattr -El hdisk4 show? Sounds like you do not have dynamic tracking and fast fail enabled... |
Quote:
Thx Charanjit |
All times are GMT -5. The time now is 12:05 PM. |