Configuraing new SSA disk

[cts123]

Hi,

I have bunch of SSA enclosure which are connected with one of our p series server running on AIX 5.1

There was a disk which was throwing error in errpt.

We had replaced the disk after moving the LV copy ( also after doing rmdev -dl) to another disk.

We have connected new disk in that SSA enclosure.

After cfgmgr we saw the disk is showing in system and if I do lspv, its showing this ( here it is hdisk11) .

/backup/export > lspv | grep -i none
hdisk9 0002f82a7c04eb7d None
hdisk11 none None

/usr/sbin/cluster > ssaxlate -l hdisk9
pdisk7

/usr/sbin/cluster > ssaxlate -l hdisk11
pdisk33

Here is the Pdisk location..

/usr/sbin/cluster > lscfg | grep -i pdisk
+ pdisk32 14-08-EA57-16-P SSA160 Physical Disk Drive (36400
+ pdisk31 14-08-EA57-01-P SSA160 Physical Disk Drive (36400
+ pdisk48 14-08-EA57-11-P SSA160 Physical Disk Drive (36400
+ pdisk24 14-08-EA57-03-P SSA160 Physical Disk Drive (36400
+ pdisk19 14-08-EA57-04-P SSA160 Physical Disk Drive (36400
+ pdisk30 14-08-EA57-12-P SSA160 Physical Disk Drive (36400
+ pdisk29 14-08-EA57-02-P SSA160 Physical Disk Drive (36400
+ pdisk2 14-08-EA57-05-P SSA160 Physical Disk Drive (36400
+ pdisk4 14-08-EA57-06-P SSA160 Physical Disk Drive (36400
+ pdisk5 14-08-EA57-07-P SSA160 Physical Disk Drive (36400
+ pdisk1 14-08-EA57-08-P SSA160 Physical Disk Drive (36400
+ pdisk3 14-08-EA57-09-P SSA160 Physical Disk Drive (36400
+ pdisk26 14-08-EA57-10-P SSA160 Physical Disk Drive (36400
+ pdisk7 14-08-EA57-13-P SSA160 Physical Disk Drive (36400
+ pdisk28 14-08-EA57-14-P SSA160 Physical Disk Drive (36400
+ pdisk6 14-08-EA57-15-P SSA160 Physical Disk Drive (36400
+ pdisk12 1j-08-7517-01-P SSA160 Physical Disk Drive (36400
+ pdisk14 1j-08-7517-02-P SSA160 Physical Disk Drive (36400
+ pdisk8 1j-08-7517-03-P SSA160 Physical Disk Drive (36400
+ pdisk0 1j-08-7517-04-P SSA160 Physical Disk Drive (36400
+ pdisk17 1j-08-7517-05-P SSA160 Physical Disk Drive (36400
+ pdisk22 1j-08-7517-06-P SSA160 Physical Disk Drive (36400
+ pdisk20 1j-08-7517-07-P SSA160 Physical Disk Drive (36400
+ pdisk15 1j-08-7517-08-P SSA160 Physical Disk Drive (36400
+ pdisk18 1j-08-7517-09-P SSA160 Physical Disk Drive (36400
+ pdisk21 1j-08-7517-10-P SSA160 Physical Disk Drive (36400
+ pdisk33 1j-08-7517-11-P SSA160 Physical Disk Drive (36400
+ pdisk23 1j-08-7517-12-P SSA160 Physical Disk Drive (36400
+ pdisk10 1j-08-7517-13-P SSA160 Physical Disk Drive (36400
+ pdisk16 1j-08-7517-14-P SSA160 Physical Disk Drive (36400
+ pdisk13 1j-08-7517-15-P SSA160 Physical Disk Drive (36400
+ pdisk11 1j-08-7517-16-P SSA160 Physical Disk Drive (36400


So my hdisk9 = pdisk7 =
pdisk7 14-08-EA57-13-P SSA160 Physical Disk Drive (36400


And hdisk11 =
pdisk33 1j-08-7517-11-P SSA160 Physical Disk Drive (36400

To assign new PVID, I was trying to do following ...But getting error.

/backup/export > chdev -l hdisk11 -a pv=yes
Method error (/usr/lib/methods/chgdisk):
0514-047 Cannot access a device.

And at the same time if I see errpt, its showing this ..


LABEL: SCSI_ERR10
IDENTIFIER: 0BA49C99

Date/Time: Mon Feb 12 10:53:36 EST
Sequence Number: 5413
Machine Id: 0002F82A4C00
Node Id: snswmspd
Class: H
Type: TEMP
Resource Name: scsi1
Resource Class: adapter
Resource Type: sym1010
Location: 1Z-09

Description
SCSI BUS ERROR

Probable Causes
CABLE
CABLE TERMINATOR
DEVICE
ADAPTER

Failure Causes
CABLE LOOSE OR DEFECTIVE
DEVICE
ADAPTER


Any idea ..? I wanted use the space of two remaining disk of the system, but look like its somethig else. This is the main production server of the Organization, hence taking downtime is again another pain.

Need some suggestion. I just manually plugged out the faulty disk and put back the new disk. Do I need to run any SSA utility for this new disk.

Thanks ..

Regards,

-Sam

 

[DukeSSD]

chdev only works if the disk HAS a pvid.

Create a vg on the disk or add and remove it from a vg to assign a vg.

If you get errors then use diag and certify the disk to check it is OK.

The scsi_err10 is a bit odd for a SSA disk but SSA does use SCSI as a protocol so may be nothing to worry about, then again it is logged against location 1Z-09 which is not an SSA adapter so you may have another problem on a scsi bus / device.

After the remove disk, rmdev, cfgmgr there shouldn't be anything else to do except create a vg or add the disk to a vg to get a pvid and then off you go.

 

[cts123]

Hi DukeSSD,

Thats not the case. I can remember I did couple of times to assign PVID for a new disk without assigning that to a new VG.

But I tried your process too, to extend the existing VG with that disk hdisk11, and getting following ...


Before command completion, additional instructions may appear below.

0516-024 extendvg: Unable to open physical volume.
Either PV was not configured or could not be opened. Run
diagnostics.
0516-792 extendvg: Unable to extend volume group.
0516-1254 extendvg: Changing the PVID in the ODM.
Method error (/usr/lib/methods/chgdisk):
0514-047 Cannot access a device.

0516-024 extendvg: Unable to open physical volume.
Either PV was not configured or could not be opened. Run
diagnostics.
0516-792 extendvg: Unable to extend volume group.
0516-792 extendvg: Unable to extend volume group.


Regards,

-Sam

 

[p5wizard]

You said in OP you unplugged old/bad disk and replugged replacement disk.

Did you first rmdev pdisk/hdisk pair and afterward cfgmgr to discover the new disk?


HTH,

p5wizard

 

[cts123]

Now ..I am doing Certify disk ..through diag option and..will let you know ..when I am done..

CERTIFY DISK 106
Certifying:-
pdisk33 Serial# 94D4610F In Location 1j-08-7517-11-P

Certify is [ 11 ] % Complete

[>>>>> ]

Please stand by.

Regards,

-Samir

 

[cts123]

p5wizard,

yes ..I did that. I did rmdev for pdisk/hdisk pair before entering the new disk. And I did cfgmgr too. Let see how this certify process goes ...will let you know. I am doing it ..when all the applications are running on that machine.

Regards,

-Sam

 

[cts123]

Here is the result...

TESTING COMPLETE on Tue Apr 24 11:17:02 EDT 2007 801010

No trouble was found.

The resources tested were:

- pdisk33 1j-08-7517-11-P SSA160 Physical Disk Drive (36400 MB)

Use Enter to continue.


/home/smukhod > lspv | grep None
hdisk9 0002f82a7c04eb7d None
hdisk11 none None

/home/smukhod > set -o vi

/home/smukhod > ssaxlate -l pdisk33
hdisk11

/home/smukhod > chdev -l hdisk11 -a pv=yes
Method error (/usr/lib/methods/chgdisk):
0514-047 Cannot access a device.


No idea ..whats going on ..!!

Regards,

-Samir

 

[DukeSSD]

Saw something like this before.
Had a 3rd party hardware support company supply a disk, probably from ebay, from a shark or similar, and we could not get the disk to work.
It would certify 100%
Eventually we paid IBM to sort it out.
Chap turns up with a disk, hot swaps it and it configured fine and worked as expected.
Where did you get the disk?
Can you send it back?
From the errors and google I think the disk is bad.

 

[p5wizard]

If that replacement disk comes from a Shark (or any other RAID group), you can probably get it to work by reformatting it first.

diag -> service aids -> ssa service aids -> format (if memory serves me right.)

But then again, if it configures as a pdisk and a hdisk, that shouldn't be the problem. But I guess it can't hurt to try...


HTH,

p5wizard

 

[cts123]

DukeSSD and p5wizard,

Thanks to both of you. I will try the format option after sometime and will let you know the status.

By the way the HDD we received from IBM only. If I see the format option doesn't work, it looks like we have to ask for another replacement of the disk.

Thanks again.

Regards,

-Sam

 

[p5wizard]

Another thing to check is the firmware level of your SSA adapter(s) and other SSA disks.

Some combinations of microcode just don't work...


HTH,

p5wizard