Start a Conversation

This post is more than 5 years old

Solved!

Go to Solution

4001

March 6th, 2008 01:00

figure out disk identity from aix servers

I got a mail from a SA ..hez getting hardware error from wrongly assigned disks to two aix servers ( unix platform ) one server is connected to clarrion and the other oen to symmetrix, now he wants me to deallocate them ...
he dosen't know the decimal lunid and has just posted the o/p of the
cmd .... lscfg -vl hdisk4..
how do i figure out the decimal lun ID of the disks and remove them

2 Intern

 • 

20.4K Posts

March 6th, 2008 07:00

like Stefano mentioned already ...LUNZ are special devices ..really just place holders. You will usually see them on systems where there is no real lun at address 0. In your case ..if this AIX host had four zones to Clariion ..and that particular storage group did not have a lun at Hostid 0 ..then you will see these four ghost luns. Since they are not real luns ..i am not surprised you are not getting anything from lsattr.

2 Intern

 • 

20.4K Posts

March 6th, 2008 04:00

download inq utility, to identify Clariion LUN:

inq -clar_wwn

you will see something like this:

-------------------------------------------------------------------------------------------------
CLARiiON Device     Array Serial #   SP IP Address      LUN      WWN (all 32 hex digits required)
-------------------------------------------------------------------------------------------------
/dev/rhdiskpower22  APM12345678901   B  10.1.8.27       0x0061   60060160b0a1160038fb12bc026ddc11
/dev/rhdiskpower23  APM12345678901   A  10.1.8.26       0x0060   60060160b0a1160036e5a1ab006ddc11
/dev/rhdiskpower24  APM12345678901   A  10.1.8.26       0x00f8   60060160b0a1160059953a13aa06db11
/dev/rhdiskpower25  APM12345678901   A  10.1.8.26       0x00fc   60060160b0a1160015e4ec41aa06db11
/dev/rhdiskpower26  APM12345678901   A  10.1.8.26       0x0100   60060160b0a11600e942ea5caa06db11


from this ouput if you take for example last LUN 0x0100 and convert from hex to decimal ..that will corresponde to lun 256 in Navisphere.

for Symmetrix run:

inq -sym_wwn

and output should look something like this:

Symmetrix Device    Symm Serial #  Device #  WWN
-----------------------------------------------------------------------------
/dev/rhdiskpower0   987654321098   0862      60060480987654321098424f42383632
/dev/rhdiskpower1   987654321098   055C      60060480987654321098424f42353543
/dev/rhdiskpower2   987654321098   055D      60060480987654321098424f42353544
/dev/rhdiskpower3   987654321098   055E      60060480987654321098424f42353545
/dev/rhdiskpower4   987654321098   055F      60060480987654321098424f42353546
/dev/rhdiskpower5   987654321098   0560      60060480987654321098424f42353630


Third column is your symmetrix device id

2 Intern

 • 

2.8K Posts

March 6th, 2008 05:00

Dynamox RIA posted the same question in the Open System area of the forums .. The other thread have been removed, so please keep on updating this one.

Message was edited by:
Stefano Del Corno

2 Intern

 • 

2.8K Posts

March 6th, 2008 06:00

LUNZ is a Clariion specific feature.. Please ask details in the Clariion area of the forums .. Here we can talk about Symms and DMX .. but please open a new thread in the Clariion area if you want details on LUNZ. I think Mr Dynamox can give you more details.

7 Posts

March 6th, 2008 06:00

hi thr ..
Thanx for the reply ...

the syminq o/p gave us the lun id .. but looks like the hdisks he wants to be removed from both the servers have all 0s as their value

/dev/rhdisk4 :DGC :LUNZ :0207 :00000000 : FAILED
/dev/rhdisk5 :DGC :LUNZ :0207 :00000000 : FAILED
/dev/rhdisk6 :DGC :LUNZ :0207 :00000000 : FAILED
/dev/rhdisk7 :DGC :LUNZ :0207 :00000000 : FAILED






... moreover when he did a lsattr -El hdisk4 and lsattr -El hdisk78 he got the values as

lun_id 0x0 Logical Unit Number ID FALSE


does this mean that these are ghost devices ...

2 Intern

 • 

2.8K Posts

March 6th, 2008 06:00

In the following example the disk (hdisk0) is at lun 81 ...

lscfg -vpl hdisk0
hdisk0 U9113.550.65FE44A-V6-C2-T1-L810000000000 Virtual SCSI Disk Drive

147 Posts

March 6th, 2008 19:00

Failed means that a scsi inquiry failed against the device, commonly because of an exisitng scsi reservation

in this case, it looks like its device 000 which is usually the vcmdb, which is a dummy device you dont have to worry about

7 Posts

March 6th, 2008 19:00

this is the syminq o/p of the server connected to symm

/dev/rhdisk4 :EMC :SYMMETRIX :5670 :5100000000 : FAILED

/dev/rhdisk78 :EMC :SYMMETRIX :5670 :5100000000 : FAILED


lsattr -El hdisk4

lun_id 0x0 Logical Unit Number ID False

147 Posts

March 6th, 2008 19:00

LUNZ appears when the server connects to a Clariion and hasn't been allocated to a storage group yet.

So just unzone it from the Clariion, then LUNZ won't be presented from the Clariion anymore.

2 Intern

 • 

20.4K Posts

March 6th, 2008 19:00

syminq should not choke on VCM device ...my syminq shows VCM just fine without throwing up any errors.

7 Posts

March 6th, 2008 20:00

the syminq o/p is already in the post

2 Intern

 • 

20.4K Posts

March 6th, 2008 20:00

can you also ask you SA to run symcfg discover and then sympd list

7 Posts

March 6th, 2008 20:00

hi dynamox ...

looks like wht u inferred is true .. all those luns are ghost luns ..

In case of clariion ... LUNz should disappear once a LUN zero is bound, or when Storage Group access has been attained

Thanx for ur help .. but wht can a aix SA do to get rid of the errors cropping up ..

7 Posts

March 6th, 2008 20:00

ok wil do that right away .

2 Intern

 • 

20.4K Posts

March 6th, 2008 20:00

can you have your SA run syminq -symmids ...post output here but do erase last two digits in the ID column from your output, that's your symm id ..we don't need to see that :)
No Events found!

Top