Red Hat Bugzilla – Bug 1299708
[Nimble Storage] Unnecessary wait on stand-by paths for TUR in rescan-scsi-bus.sh
Last modified: 2017-09-20 22:34:07 EDT
Description of problem:
rescan-scsi-bus.sh has testonline() function which issues TUR and retry 8 times with an interval of 1 second sleep. This is causing a huge delay in scanning large number of volumes if stand-by paths are present. TUR on stand-by paths return 2/4/0b check condition. This condition needs to be handled in rescan-scsi-bus.sh and avoid looping in testonline().
Also, udevadm settle calls are hanging and taking long time to complete.
while test $RC = 2 -o $RC = 6 && test $ctr -le 2; do
if test $RC = 2 -a "$RMB" != "1"; then echo -n "."; let LN+=1; sleep 1
else usleep 20000; fi
sg_turs /dev/$SGDEV >/dev/null 2>&1
mpathje (215df250f8212599b6c9ce9001e319f94_1) dm-512 Nimble,Server
size=1.0G features='1 queue_if_no_path' hwhandler='1 alua' wp=rw
|-+- policy='round-robin 0' prio=50 status=active
| |- 6:0:4:78 sdcf 69:48 active ready running
| `- 7:0:5:78 sdzq 131:576 active ready running
`-+- policy='round-robin 0' prio=1 status=enabled
|- 6:0:7:78 sdka 65:480 active ghost running
`- 7:0:0:78 sdrv 134:400 active ghost running
[root@hitdev-rhel67 ~]# sg_turs -v /dev/sdka
test unit ready cdb: 00 00 00 00 00 00
test unit ready: Fixed format, current; Sense key: Not Ready
Additional sense: Logical unit not accessible, target port in standby state
Info fld=0x0 
device not ready
I've had a brief discussion with upstream developers on this issue. They are looking into this as well.
Can you provide any update on this?. Was the change made?.
David left Red Hat - and we are in a process of searching for the new maintainer of sg3_utils package. Unfortunately - I don't see any clear pointer to David's discussion with upstream - based on lkml and recent activity, it seems sg3_utils is now maintained at https://github.com/hreinecke/sg3_utils ...
the sg3_utils testing has transfer to storage-QE team. so I will take the bug for