UNSOLVED
LUN not ready intervention required
Hi EMC,
I have an EMC clariion connected to a clustered pair of sun x4270 (x86) servers running Solaris 10.
I am installing Oracle 10.2g on my EMC and I have no problem when I install oracle on my first server, but when I extend the Oracle to the second server, I get the following erros in my dmesg on my first server:
May 12 14:32:06 vs5a scsi: [ID 107833 kern.warning] WARNING: /pci@0,0/pci8086,340c@5/pci1077,171@0,1/fp@0,0/disk@w500601604be01df1,2 (sd16):
May 12 14:32:06 vs5a Error for Command: read(10) Error Level: Fatal
May 12 14:32:06 vs5a scsi: [ID 107833 kern.notice] Requested Block: 6673 Error Block: 6673
May 12 14:32:06 vs5a scsi: [ID 107833 kern.notice] Vendor: DGC Serial Number: 00000027ACCL
May 12 14:32:06 vs5a scsi: [ID 107833 kern.notice] Sense Key: Not Ready
May 12 14:32:06 vs5a scsi: [ID 107833 kern.notice] ASC: 0x4 (LUN not ready intervention required), ASCQ: 0x3, FRU: 0x0
May 12 14:32:06 vs5a scsi: [ID 107833 kern.warning] WARNING: /pci@0,0/pci8086,340c@5/pci1077,171@0/fp@0,0/disk@w500601614be01df1,2 (sd15):
May 12 14:32:06 vs5a Error for Command: read(10) Error Level: Fatal
May 12 14:32:06 vs5a scsi: [ID 107833 kern.notice] Requested Block: 6673 Error Block: 6673
May 12 14:32:06 vs5a scsi: [ID 107833 kern.notice] Vendor: DGC Serial Number: 00000027ACCL
May 12 14:32:06 vs5a scsi: [ID 107833 kern.notice] Sense Key: Not Ready
May 12 14:32:06 vs5a scsi: [ID 107833 kern.notice] ASC: 0x4 (LUN not ready intervention required), ASCQ: 0x3, FRU: 0x0
May 12 14:32:07 vs5a emcp: [ID 801593 kern.notice] Info: Unable to follow volume 6006016037102B00D0846FED3C7AE011 to SPA
May 12 14:32:07 vs5a emcp: [ID 801593 kern.notice] Info: Assigned volume 6006016037102B00D0846FED3C7AE011 to SPA
I get a similar message from my second server but with SPB instead of SPA.
Can you provide some guidance towards a solution please?
Thanks,
Kerry
856
0
Posted May 12th, 2011 12:00
I have moved this discussion to the Clariion Support Forum to ensure it gets to the right audience.
856
0
Posted May 12th, 2011 12:00
how's everything zoned ? Each HBA is zoned to SPA and SPB ? What kind of multipathing software are you running, what is the failover mode set on Clariion? What do you mean by "extend Oracle" ?
856
0
Posted May 13th, 2011 03:00
Hi dynamox,
Each sun server is set to communicate to either SP and the only software I have on it the servers is solaris and oracle, I'm very new to emc and I'm not sure what kind of multi-pathing im using exactly as my set up is done with from a script made by our software developers. Is there someway I can check to veirfy? The Array failover mode is currently set to 1, and by extend oracle, I mean I installed the oracle instance on my other server (this is the point when the error occured)
The odd thing is that I've used the same procedure for 3 other installs with the exact same configuration and hardware and not had any errors...
Please let me know if I can provide more info to clarify the situation.
Thanks,
Kerry856
0
Posted May 21st, 2011 05:00
You are using PowerPath as multi-pathing solution. Apparently your first server does not have access to SPB and your second server does not have access to SPA so each tries constantly to trespass LUN to itself.
You either need to connect each server to both SPA and SPB (preferred) or at least make sure they are connected to the same SP.
856
0
Posted May 22nd, 2011 16:00
Hi Kerry,
As you use the empc driver you have Powerpath as multipathing software
Can you provide the output of the following command :
powermt display dev=all
Here is the settings mutipathing settings for Solaris/clariion:
FailoverMode 4 is only for Flare > 24 [Preferred for Solaris 10]
856
0
Posted June 12th, 2011 09:00
an easy way to check fail over mode is `navicli -h x.x.x.x port -list failovermode`
Stephanie_P
1 Rookie
•
180 Posts
856
0
Posted May 12th, 2011 09:00
Hi Kerry,
The team is going to move this question over to the Support Forums Community where your question can be addressed.
Thanks,
Stephanie