Unsolved

This post is more than 5 years old

2 Intern

 • 

1.3K Posts

2066

December 13th, 2008 08:00

Loosing connection to both paths(SPA/SPB)

HP-UX 11.11
EMC powermt for PowerPath (c) Version 3.0.3
Fail over mode=1 at Clariion

As you can see only single HBA/Basisfailover
eurdbs3d:/root# powermt display dev=c33t3d4
CLARiiON ID=xxxxxxxxxxxxxx
Logical device ID=60060160F1731100799E3EEE2543DB11
state=alive; policy=BasicFailover; priority=0; queued-IOs=0
Owner: default=SP A, current=SP A
==============================================================================
---------------- Host --------------- - Stor - -- I/O Path - -- Stats ---
### HW Path I/O Paths Interf. Mode State Q-IOs Errors
==============================================================================
33 0/0/12/1/0.97.4.19.0.3.4 c33t3d4 SP A0 active alive 0 1
34 0/0/12/1/0.97.6.19.0.3.4 c34t3d4 SP B0 active alive 0 1

IO timeout set to 180 as usaul. Both paths are in VG.
eurdbs3d:/dev/rdsk# pvdisplay /dev/dsk/c34t3d4
Device file path "/dev/dsk/c34t3d4" is an alternate path
to the Physical Volume. Using Primary Link "/dev/dsk/c33t3d4".
--- Physical volumes ---
PV Name /dev/dsk/c33t3d4
PV Name /dev/dsk/c34t3d4 Alternate Link
VG Name /dev/vg42
PV Status available
Allocatable yes
VGDA 2
Cur LV 1
PE Size (Mbytes) 32
Total PE 2156
Free PE 262
Allocated PE 1894
Stale PE 0
IO Timeout (Seconds) 180
Autoswitch On
Proactive Polling On

Here is the error .As per the logs LUN is NOT switching(may be switching but not able to use the other path) between SPs rather loosing the connection to both SPs at the same time

Dec 1 14:04:21 eurdbs3d vmunix: LVM: VG 64 0x2a0000: PVLink 31 0x213400 Failed! The PV is not accessible.
Dec 1 14:04:21 eurdbs3d vmunix: LVM: VG 64 0x2a0000: PVLink 31 0x223400 Failed! The PV is not accessible.
Dec 1 14:04:26 eurdbs3d vmunix: LVM: VG 64 0x2a0000: PVLink 31 0x213400 Recovered.
Dec 1 14:04:26 eurdbs3d vmunix: LVM: VG 64 0x2a0000: PVLink 31 0x223400 Recovered.
Dec 1 14:06:29 eurdbs3d vmunix: LVM: VG 64 0x2a0000: PVLink 31 0x213400 Failed! The PV is not accessible.
Dec 1 14:06:29 eurdbs3d vmunix: LVM: VG 64 0x2a0000: PVLink 31 0x223400 Failed! The PV is not accessible.
Dec 1 14:06:34 eurdbs3d vmunix: LVM: VG 64 0x2a0000: PVLink 31 0x213400 Recovered.
Dec 1 14:06:34 eurdbs3d vmunix: LVM: VG 64 0x2a0000: PVLink 31 0x223400 Recovered.

I have another LUN coming from same array/HBA/VG which is not reporting any error. Is there any LUN level property/settings which related to failover/pathswitcing?

As you can see there is license for Clarrion (single HBA)
eurdbs3d:/root# emcpreg -list

Key BJIY-D2YM-VKCU-F2HA-TW9H-IN6M
Product: PowerPath
Capabilities: Symmetrix

what is your thoughts?

11 Legend

 • 

20.4K Posts

 • 

87.4K Points

December 14th, 2008 12:00

this LUN in question ..in its properties you don't have "Auto assignment Enabled" checked ?

2 Intern

 • 

265 Posts

December 14th, 2008 15:00

chk emc68926 &
Proactive polling should be turned off.

2 Intern

 • 

265 Posts

December 14th, 2008 15:00

ive faced a similar problem ( but with dm-multipath native linux tool and with eva4400)
os keep losing native devices for about 3-4 second
solutions was changing some parameter in dm-multipath
path checker was changed from readsector 0 to tur
tur based on scsci command to check if lun is accessible
. allow I/O queuing ,and ive changed number of path retry
but i think firstly as dynamox suggested , check auto assignment option for ur lun

341 Posts

December 15th, 2008 01:00

Hi Santhosh,

Can you confirm that the /etc/Navisphere/agent.config file is setup correctly as per emc59539

the most important part is setting initiator type to "no AutoTrespass" by commenting out the following entry in the agent.config file:

# OptionsSupported AutoTrespass

Stop and start the Navisphere host agent

See also solutions emc66214, emc66215, emc66216, and emc65749.

Conor

2 Intern

 • 

1.3K Posts

December 15th, 2008 09:00

auto-trespass is already disabled.

# grep -i tres agent.config
# auto-trespass when it starts up. To enable auto-trespass on Agent
# OptionsSupported AutoTrespass
# will be to disable auto-trespass.

Already gave a try with proactive polling diabled even though the primus details does not really match with our envioronment

2 Intern

 • 

1.3K Posts

December 15th, 2008 11:00

auto asighnment is disabled too

341 Posts

December 16th, 2008 01:00

Can you try enabling auto-assignment on the LUN and retest?

2 Intern

 • 

1.3K Posts

December 16th, 2008 07:00

i am not enabling the auto asighnment as my working LUn has the same setting.
The strange thing is that when tried tress passing the LUN it return success status. But effectively no change and the LUN still stick to the same SP. why it could be?

let me know if you any thing strange. ?

LOGICAL UNIT NUMBER 45
Prefetch size (blocks) = 0
Prefetch multiplier = 4
Segment size (blocks) = 0
Segment multiplier = 4
Maximum prefetch (blocks) = 4096
Prefetch Disable Size (blocks) = 4097
Prefetch idle count = 40

Variable length prefetching YES
Prefetched data retained YES

Read cache configured according to
specified parameters.

Total Hard Errors: 0
Total Soft Errors: 0
Total Queue Length: 511313114
Name LUN 45
Minimum latency reads N/A


Read Histogram[0] 2
Read Histogram[1] 287
Read Histogram[2] 355
Read Histogram[3] 47209
Read Histogram[4] 319988519
Read Histogram[5] 5250023
Read Histogram[6] 10855643
Read Histogram[7] 31392342
Read Histogram[8] 2257985
Read Histogram[9] 6523019
Read Histogram overflows 0

Write Histogram[0] 0
Write Histogram[1] 182
Write Histogram[2] 344
Write Histogram[3] 3606852
Write Histogram[4] 14511428
Write Histogram[5] 3598578
Write Histogram[6] 2171197
Write Histogram[7] 717521
Write Histogram[8] 836232
Write Histogram[9] 185537
Write Histogram overflows 0

Read Requests: 376315384
Write Requests: 25627871
Blocks read: 1791312442
Blocks written: 1032833740
Read cache hits: 258643780
Read cache misses: N/A
Prefetched blocks: 1670304624
Unused prefetched blocks: 1893785558
Write cache hits: 2869029
Forced flushes: 73114
Read Hit Ratio: N/A
Write Hit Ratio: N/A
RAID Type: RAID5
RAIDGroup ID: 2
State: Bound
Stripe Crossing: 57954008
Element Size: 128
Current owner: SP A
Offset: 0
Auto-trespass: DISABLED
Auto-assign: DISABLED
Write cache: ENABLED
Read cache: ENABLED
Idle Threshold: 0
Idle Delay Time: 20
Write Aside Size: 2048
Default Owner: SP B
Rebuild Priority: ASAP
Verify Priority: ASAP
Prct Reads Forced Flushed: 0
Prct Writes Forced Flushed: 0
Prct Rebuilt: 100
Prct Bound: 100
LUN Capacity(Megabytes): 138105
LUN Capacity(Blocks): 282839040
UID: 60:06:01:60:F1:73:11:00:A6:9E:6D:22:C3:E5:DA:11
Bus 0 Enclosure 1 Disk 0 Queue Length: 627018027
Bus 0 Enclosure 1 Disk 1 Queue Length: 604945012
Bus 0 Enclosure 1 Disk 2 Queue Length: 517805969
Bus 0 Enclosure 1 Disk 3 Queue Length: 589108731
Bus 0 Enclosure 1 Disk 4 Queue Length: 491836429
Bus 0 Enclosure 1 Disk 5 Queue Length: 495230649
Bus 0 Enclosure 1 Disk 6 Queue Length: 524849429
Bus 0 Enclosure 1 Disk 7 Queue Length: 499729222
Bus 0 Enclosure 1 Disk 8 Queue Length: 474984733
Bus 0 Enclosure 1 Disk 9 Queue Length: 484793717
Bus 0 Enclosure 1 Disk 10 Queue Length: 479735172
Bus 0 Enclosure 1 Disk 11 Queue Length: 476891039
Bus 0 Enclosure 1 Disk 12 Queue Length: 477240770
Bus 0 Enclosure 1 Disk 13 Queue Length: 497170234
Bus 0 Enclosure 1 Disk 14 Queue Length: 520140978
Bus 0 Enclosure 1 Disk 0 Hard Read Errors: 0

2 Intern

 • 

1.3K Posts

December 19th, 2008 06:00

can the initiatortype be changed online? i am thinking of swiching to "HP autotresspass" after disabling/removing the powerpath for testing purpose.

11 Legend

 • 

20.4K Posts

 • 

87.4K Points

December 19th, 2008 13:00

all solutions that i have seen require host reboot.

2 Intern

 • 

1.3K Posts

October 4th, 2009 04:00

hosts are migrated and moved to PP 5.1(SP1) and with two HBA.
No Events found!

Top