[dm-devel] Mulipath / DM-Multipath / Hitachi iSCSI SAN

[devin]

Dear Users - 

I am running a Hitachi AMS2500 with 10GB iSCSI SAN,
and I am running RedHat Enterprise 5.7 x86_64bit. I installed Multipath
and I am seeing strange issues with the box. Basically what appears to
be happening is some of my LUNS seem to go into an enabled/enabled
status and I/O just queue's up and then eventually times out. I am not
sure why this is happening, it is almost like a LUN goes offline for a
bit and then I/O back's up and causes major issues. I noticed in the


/var/log/messages 

Feb 23 17:59:55 ash-db-3 multipathd: mpath22: load
table [0 2147483648 multipath 1 queue_if_no_path 0 2 1 round-robin 0 2 1
70:0 1000 69:240 1000 round-robin 0 2 1 70:32 1000 70:16 1000] 
Feb 23
17:59:55 ash-db-3 multipathd: dm-51: add map (uevent) 
Feb 23 17:59:55
ash-db-3 multipathd: dm-51: devmap already registered 
Feb 23 18:01:51
ash-db-3 kernel: sd 6:0:0:16: timing out command, waited 7s
Feb 23
18:01:51 ash-db-3 multipathd: sdbs: hds prio: SCSI error 
Feb 23
18:01:55 ash-db-3 multipathd: mpath17: load table [0 2147483648
multipath 1 queue_if_no_path 0 2 1 round-robin 0 2 1 68:96 1000 68:80
1000 round-robin 0 2 1 68:48 1000 68:64 1000] 
Feb 23 18:02:09 ash-db-3
multipathd: dm-37: add map (uevent) 
Feb 23 18:02:09 ash-db-3
multipathd: dm-37: devmap already registered 

Also if i just Grep for
"timing out" i get: 

Feb 23 17:09:21 ash-db-3 kernel: sd 6:0:0:2:
timing out command, waited 7s
Feb 23 17:11:33 ash-db-3 kernel: sd
8:0:0:10: timing out command, waited 7s
Feb 23 17:18:46 ash-db-3 kernel:
sd 7:0:0:15: timing out command, waited 7s
Feb 23 17:22:44 ash-db-3
kernel: sd 8:0:0:2: timing out command, waited 7s
Feb 23 17:27:54
ash-db-3 kernel: sd 6:0:0:10: timing out command, waited 7s
Feb 23
17:34:42 ash-db-3 kernel: sd 8:0:0:2: timing out command, waited 7s
Feb
23 17:35:12 ash-db-3 kernel: sd 8:0:0:17: timing out command, waited
7s
Feb 23 17:37:01 ash-db-3 kernel: sd 6:0:0:2: timing out command,
waited 7s
Feb 23 17:37:17 ash-db-3 kernel: sd 6:0:0:5: timing out
command, waited 7s
Feb 23 17:39:15 ash-db-3 kernel: sd 6:0:0:16: timing
out command, waited 7s
Feb 23 17:44:11 ash-db-3 kernel: sd 6:0:0:2:
timing out command, waited 7s
Feb 23 17:46:36 ash-db-3 kernel: sd
8:0:0:2: timing out command, waited 7s
Feb 23 17:48:30 ash-db-3 kernel:
sd 6:0:0:7: timing out command, waited 7s
Feb 23 17:51:42 ash-db-3
kernel: sd 6:0:0:16: timing out command, waited 7s
Feb 23 17:56:53
ash-db-3 kernel: sd 6:0:0:15: timing out command, waited 7s
Feb 23
17:59:08 ash-db-3 kernel: sd 6:0:0:2: timing out command, waited 7s
Feb
23 17:59:36 ash-db-3 kernel: sd 6:0:0:23: timing out command, waited
7s
Feb 23 18:01:51 ash-db-3 kernel: sd 6:0:0:16: timing out command,
waited 7s 

then when I look at "multipath -ll" i see for instance:


mpath12 (360060e80100b05f00530286700000008) dm-7
HITACHI,DF600F
size=1.0T features='1 queue_if_no_path' hwhandler='0'
wp=rw
|-+- policy='round-robin 0' prio=1 status=enabled
| |- 8:0:0:2 sdo
8:224 active ready running
| `- 6:0:0:2 sdm 8:192 active ready
running
`-+- policy='round-robin 0' prio=0 status=enabled
 |- 7:0:0:2
sdl 8:176 active ready running
 `- 5:0:0:2 sdn 8:208 active ready
running
mpath5 (360060e80100b05f0053028670000000f) dm-29
HITACHI,DF600F
size=1.0T features='1 queue_if_no_path' hwhandler='0'
wp=rw
|-+- policy='round-robin 0' prio=1 status=active
| |- 5:0:0:9 sdaq
66:160 active ready running
| `- 7:0:0:9 sdao 66:128 active ready
running
`-+- policy='round-robin 0' prio=0 status=enabled
 |- 6:0:0:9
sdan 66:112 active ready running
 `- 8:0:0:9 sdap 66:144 active ready
running 

So I wonder if somehow I am getting bit by the "1
queue_if_no_path", however not sure why some LUNS are alwasy
ready/enabled, and then some go enabled/enabled. If anyone can provide
assistance I would be forever grateful!! 

Thanks Much. 

Devin Acosta 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/dm-devel/attachments/20120223/42d4e7a2/attachment.htm>