AIX 4.3 Stale LVs

Hi All,

I've inherited an old AIX 4.3 system with 1GB drives in it. Yesterday
I noticed some stale LVs. I googled the issue and ran some commands I
found in other results. Here is what the system looked like before I
started:

rootvg:
LV NAME TYPE LPs PPs PVs LV STATE MOUNT
POINT
hd5 boot 2 4 2 closed/syncd N/A
hd6 paging 128 256 2 open/stale N/A
hd8 jfslog 1 2 2 open/syncd N/A
hd4 jfs 2 4 2 open/stale /
hd2 jfs 146 292 2 open/stale /usr
hd9var jfs 1 2 2 open/stale /var
hd3 jfs 15 30 3 open/stale /tmp
hd1 jfs 1 2 2 open/syncd /home
lv01 jfs 1 1 1 closed/syncd N/A
lv00 jfs 5 10 1 open/syncd /local
lv02 jfs 360 360 2 open/syncd #

After running 'syncvg -v rootvg':
rootvg:
LV NAME TYPE LPs PPs PVs LV STATE MOUNT
POINT
hd5 boot 2 4 2 closed/syncd N/A
hd6 paging 128 256 2 open/stale N/A
hd8 jfslog 1 2 2 open/syncd N/A
hd4 jfs 2 4 2 open/syncd /
hd2 jfs 146 292 2 open/syncd /usr
hd9var jfs 1 2 2 open/syncd /var
hd3 jfs 15 30 3 open/stale /tmp
hd1 jfs 1 2 2 open/syncd /home
lv01 jfs 1 1 1 closed/syncd N/A
lv00 jfs 5 10 1 open/syncd /local
lv02 jfs 360 360 2 open/syncd #

Now only hd3 (/tmp) and paging space are stale

# syncvg -l hd6
0516-068 lresynclv: Unable to completely resynchronize volume. Run
diagnostics if neccessary.
0516-934 /usr/sbin/syncvg: Unable to synchronize logical volume hd6.

I see a missing device when I run this command:
# lsvg -p rootvg
rootvg:
PV_NAME PV STATE TOTAL PPs FREE PPs FREE
DISTRIBUTION
hdisk7 active 248 92
06..48..00..00..38
0516-304 lsvg: Unable to find device id 000048054d75e6b0 in the Device
Configuration Database.
000048054d75e6b0 missing 248 105
50..50..00..00..05
hdisk12 active 248 73
50..22..00..00..01
hdisk13 active 248 7
00..00..00..00..07
hdisk9 active 248 0
00..00..00..00..00

But this command shows all disks OK:
# lspv
hdisk7 00004805507bbe5d rootvg
hdisk9 0000145383ffebcf rootvg
hdisk12 0000145384007ccf rootvg
hdisk13 000014538402b614 rootvg

In more detail:
# lspv -l hdisk7
hdisk7:
LV NAME LPs PPs DISTRIBUTION MOUNT POINT
hd5 2 2 02..00..00..00..00 N/A
hd2 146 146 41..01..43..49..12 /usr
hd4 2 2 01..00..01..00..00 /
hd1 1 1 00..01..00..00..00 /home
hd8 1 1 00..00..01..00..00 N/A
hd9var 1 1 00..00..01..00..00 /var
hd3 3 3 00..00..03..00..00 /tmp

# lspv -l hdisk9
hdisk9:
LV NAME LPs PPs DISTRIBUTION MOUNT POINT
lv02 247 247 50..49..49..49..50 #
lv01 1 1 00..01..00..00..00 N/A

# lspv -l hdisk12
hdisk12:
LV NAME LPs PPs DISTRIBUTION MOUNT POINT
lv00 5 10 00..10..00..00..00 /local
hd2 146 146 00..01..47..49..49 /usr
hd5 2 2 00..02..00..00..00 N/A
hd8 1 1 00..01..00..00..00 N/A
hd1 1 1 00..01..00..00..00 /home
hd3 12 12 00..12..00..00..00 /tmp
hd9var 1 1 00..01..00..00..00 /var
hd4 2 2 00..00..02..00..00 /

# lspv -l hdisk13
hdisk13:
LV NAME LPs PPs DISTRIBUTION MOUNT POINT
lv02 113 113 50..05..00..15..43 #
hd6 128 128 00..45..49..34..00 N/A

So, 2 questions:
#1 How do I get the stale LVs to read open/synced
#2 How do I get the missing 000048054d75e6b0 out of the ODM?

On Oct 24, 4:21 am, MJB <helios...@cox.net> wrote:[color=blue]
> Hi All,
>
> I've inherited an old AIX 4.3 system with 1GB drives in it. Yesterday
> I noticed some stale LVs. I googled the issue and ran some commands I
> found in other results. Here is what the system looked like before I
> started:
>
> rootvg:
> LV NAME TYPE LPs PPs PVs LV STATE MOUNT
> POINT
> hd5 boot 2 4 2 closed/syncd N/A
> hd6 paging 128 256 2 open/stale N/A
> hd8 jfslog 1 2 2 open/syncd N/A
> hd4 jfs 2 4 2 open/stale /
> hd2 jfs 146 292 2 open/stale /usr
> hd9var jfs 1 2 2 open/stale /var
> hd3 jfs 15 30 3 open/stale /tmp
> hd1 jfs 1 2 2 open/syncd /home
> lv01 jfs 1 1 1 closed/syncd N/A
> lv00 jfs 5 10 1 open/syncd /local
> lv02 jfs 360 360 2 open/syncd #
>
> After running 'syncvg -v rootvg':
> rootvg:
> LV NAME TYPE LPs PPs PVs LV STATE MOUNT
> POINT
> hd5 boot 2 4 2 closed/syncd N/A
> hd6 paging 128 256 2 open/stale N/A
> hd8 jfslog 1 2 2 open/syncd N/A
> hd4 jfs 2 4 2 open/syncd /
> hd2 jfs 146 292 2 open/syncd /usr
> hd9var jfs 1 2 2 open/syncd /var
> hd3 jfs 15 30 3 open/stale /tmp
> hd1 jfs 1 2 2 open/syncd /home
> lv01 jfs 1 1 1 closed/syncd N/A
> lv00 jfs 5 10 1 open/syncd /local
> lv02 jfs 360 360 2 open/syncd #
>
> Now only hd3 (/tmp) and paging space are stale
>
> # syncvg -l hd6
> 0516-068 lresynclv: Unable to completely resynchronize volume. Run
> diagnostics if neccessary.
> 0516-934 /usr/sbin/syncvg: Unable to synchronize logical volume hd6.
>
> I see a missing device when I run this command:
> # lsvg -p rootvg
> rootvg:
> PV_NAME PV STATE TOTAL PPs FREE PPs FREE
> DISTRIBUTION
> hdisk7 active 248 92
> 06..48..00..00..38
> 0516-304 lsvg: Unable to find device id 000048054d75e6b0 in the Device
> Configuration Database.
> 000048054d75e6b0 missing 248 105
> 50..50..00..00..05
> hdisk12 active 248 73
> 50..22..00..00..01
> hdisk13 active 248 7
> 00..00..00..00..07
> hdisk9 active 248 0
> 00..00..00..00..00
>
> But this command shows all disks OK:
> # lspv
> hdisk7 00004805507bbe5d rootvg
> hdisk9 0000145383ffebcf rootvg
> hdisk12 0000145384007ccf rootvg
> hdisk13 000014538402b614 rootvg
>
> In more detail:
> # lspv -l hdisk7
> hdisk7:
> LV NAME LPs PPs DISTRIBUTION MOUNT POINT
> hd5 2 2 02..00..00..00..00 N/A
> hd2 146 146 41..01..43..49..12 /usr
> hd4 2 2 01..00..01..00..00 /
> hd1 1 1 00..01..00..00..00 /home
> hd8 1 1 00..00..01..00..00 N/A
> hd9var 1 1 00..00..01..00..00 /var
> hd3 3 3 00..00..03..00..00 /tmp
>
> # lspv -l hdisk9
> hdisk9:
> LV NAME LPs PPs DISTRIBUTION MOUNT POINT
> lv02 247 247 50..49..49..49..50 #
> lv01 1 1 00..01..00..00..00 N/A
>
> # lspv -l hdisk12
> hdisk12:
> LV NAME LPs PPs DISTRIBUTION MOUNT POINT
> lv00 5 10 00..10..00..00..00 /local
> hd2 146 146 00..01..47..49..49 /usr
> hd5 2 2 00..02..00..00..00 N/A
> hd8 1 1 00..01..00..00..00 N/A
> hd1 1 1 00..01..00..00..00 /home
> hd3 12 12 00..12..00..00..00 /tmp
> hd9var 1 1 00..01..00..00..00 /var
> hd4 2 2 00..00..02..00..00 /
>
> # lspv -l hdisk13
> hdisk13:
> LV NAME LPs PPs DISTRIBUTION MOUNT POINT
> lv02 113 113 50..05..00..15..43 #
> hd6 128 128 00..45..49..34..00 N/A
>
> So, 2 questions:
> #1 How do I get the stale LVs to read open/synced
> #2 How do I get the missing 000048054d75e6b0 out of the ODM?[/color]

without too onerous a lecture, this is why documentation and regular
backups are so valuable.
it looks like you have had a disk failure.
the AIX error log will hopefully tell you what happened, and when. try
the "errpt" command. Also look through the syslog (check /etc/
syslog.conf for setup)

run
lsdev -Cc disk
lscfg -vl hdisk\*
lslv -l lvname (hd6 etc; will probably fail)

hope you have some good backups :)
when was the last mksysb or other backups ?

seriously, you will probably be able to remove the half of the mirrors
that are on the problem PV, replace the failing disk (if that's the
problem) and remirror.
I had problems with AIX 4.3 and failing disks years back.
I wouldn't suggest removing ODM entries, it's very easy to corrupt the
VGDA on the PV's as it synchronises itself with the ODM and then
you'll really be in the do-do's.

On Oct 23, 3:26*pm, Henry <snogfest_hosebe...@yahoo.com> wrote:[color=blue]
> On Oct 24, 4:21 am, MJB <helios...@cox.net> wrote:
>
>
>
>
>[color=green]
> > Hi All,[/color]
>[color=green]
> > I've inherited an old AIX 4.3 system with 1GB drives in it. *Yesterday
> > I noticed some stale LVs. *I googled the issue and ran some commands I
> > found in other results. *Here is what the system looked like before I
> > started:[/color]
>[color=green]
> > rootvg:
> > LV NAME * * * * * * TYPE * * * LPs * PPs * PVs *LV STATE * * *MOUNT
> > POINT
> > hd5 * * * * * * * * boot * * * 2 * * 4 * * 2 * *closed/syncd *N/A
> > hd6 * * * * * * * * paging * * 128 * 256 * 2 * *open/stale * *N/A
> > hd8 * * * * * * * * jfslog * * 1 * * 2 * * 2 * *open/syncd * *N/A
> > hd4 * * * * * * * * jfs * * * *2 * * 4 * * 2 * *open/stale * */
> > hd2 * * * * * * * * jfs * * * *146 * 292 * 2 * *open/stale * */usr
> > hd9var * * * * * * *jfs * * * *1 * * 2 * * 2 * *open/stale * */var
> > hd3 * * * * * * * * jfs * * * *15 * *30 **3 * *open/stale * */tmp
> > hd1 * * * * * * * * jfs * * * *1 * * 2 * * 2 * *open/syncd * */home
> > lv01 * * * * * * * *jfs * * * *1 * * 1 * * 1 * *closed/syncd *N/A
> > lv00 * * * * * * * *jfs * * * *5 * * 10 **1 * *open/syncd * */local
> > lv02 * * * * * * * *jfs * * * *360 * 360 * 2 * *open/syncd * *#[/color]
>[color=green]
> > After running 'syncvg -v rootvg':
> > rootvg:
> > LV NAME * * * * * * TYPE * * * LPs * PPs * PVs *LV STATE * * *MOUNT
> > POINT
> > hd5 * * * * * * * * boot * * * 2 * * 4 * * 2 * *closed/syncd *N/A
> > hd6 * * * * * * * * paging * * 128 * 256 * 2 * *open/stale * *N/A
> > hd8 * * * * * * * * jfslog * * 1 * * 2 * * 2 * *open/syncd * *N/A
> > hd4 * * * * * * * * jfs * * * *2 * * 4 * * 2 * *open/syncd * */
> > hd2 * * * * * * * * jfs * * * *146 * 292 * 2 * *open/syncd * */usr
> > hd9var * * * * * * *jfs * * * *1 * * 2 * * 2 * *open/syncd * */var
> > hd3 * * * * * * * * jfs * * * *15 * *30 **3 * *open/stale * */tmp
> > hd1 * * * * * * * * jfs * * * *1 * * 2 * * 2 * *open/syncd * */home
> > lv01 * * * * * * * *jfs * * * *1 * * 1 * * 1 * *closed/syncd *N/A
> > lv00 * * * * * * * *jfs * * * *5 * * 10 **1 * *open/syncd * */local
> > lv02 * * * * * * * *jfs * * * *360 * 360 * 2 * *open/syncd * *#[/color]
>[color=green]
> > Now only hd3 (/tmp) and paging space are stale[/color]
>[color=green]
> > # syncvg -l hd6
> > 0516-068 lresynclv: Unable to completely resynchronize volume. Run
> > * * * * diagnostics if neccessary.
> > 0516-934 /usr/sbin/syncvg: Unable to synchronize logical volume hd6.[/color]
>[color=green]
> > I see a missing device when I run this command:
> > # lsvg -p rootvg
> > rootvg:
> > PV_NAME * * * * * PV STATE * *TOTAL PPs * FREE PPs * *FREE
> > DISTRIBUTION
> > hdisk7 * * * * * *active * * *248 * * * * 92
> > 06..48..00..00..38
> > 0516-304 lsvg: Unable to find device id 000048054d75e6b0 in the Device
> > * * * * Configuration Database.
> > 000048054d75e6b0 *missing * * 248 * * * * 105
> > 50..50..00..00..05
> > hdisk12 * * * * * active * * *248 * * * * 73
> > 50..22..00..00..01
> > hdisk13 * * * * * active * * *248 * * * * 7
> > 00..00..00..00..07
> > hdisk9 * * * * * *active * * *248 * * * * 0
> > 00..00..00..00..00[/color]
>[color=green]
> > But this command shows all disks OK:
> > # lspv
> > hdisk7 * * * * 00004805507bbe5d * *rootvg
> > hdisk9 * * * * 0000145383ffebcf * *rootvg
> > hdisk12 * * * *0000145384007ccf * *rootvg
> > hdisk13 * * * *000014538402b614 * *rootvg[/color]
>[color=green]
> > In more detail:
> > # lspv -l hdisk7
> > hdisk7:
> > LV NAME * * * * * * * LPs * PPs * DISTRIBUTION * * * * *MOUNT POINT
> > hd5 * * * * * * * * * 2 * * 2 * * 02..00..00...00..00 * *N/A
> > hd2 * * * * * * * * * 146 * 146 * 41..01..43..49...12 * */usr
> > hd4 * * * * * * * * * 2 * * 2 * * 01..00..01...00..00 * */
> > hd1 * * * * * * * * * 1 * * 1 * * 00..01..00...00..00 * */home
> > hd8 * * * * * * * * * 1 * * 1 * * 00..00..01...00..00 * *N/A
> > hd9var * * * * * * * *1 * * 1 * * 00..00..01..00..00 * */var
> > hd3 * * * * * * * * * 3 * * 3 * * 00..00..03...00..00 * */tmp[/color]
>[color=green]
> > # lspv -l hdisk9
> > hdisk9:
> > LV NAME * * * * * * * LPs * PPs * DISTRIBUTION * * * * *MOUNT POINT
> > lv02 * * * * * * * * *247 * 247 * 50..49..49..49...50 * *#
> > lv01 * * * * * * * * *1 * * 1 * * 00..01..00...00..00 * *N/A[/color]
>[color=green]
> > # lspv -l hdisk12
> > hdisk12:
> > LV NAME * * * * * * * LPs * PPs * DISTRIBUTION * * * * *MOUNT POINT
> > lv00 * * * * * * * * *5 * * 10 * *00..10..00...00..00 * */local
> > hd2 * * * * * * * * * 146 * 146 * 00..01..47..49...49 * */usr
> > hd5 * * * * * * * * * 2 * * 2 * * 00..02..00...00..00 * *N/A
> > hd8 * * * * * * * * * 1 * * 1 * * 00..01..00...00..00 * *N/A
> > hd1 * * * * * * * * * 1 * * 1 * * 00..01..00...00..00 * */home
> > hd3 * * * * * * * * * 12 * *12 * *00..12..00...00..00 * */tmp
> > hd9var * * * * * * * *1 * * 1 * * 00..01..00..00..00 * */var
> > hd4 * * * * * * * * * 2 * * 2 * * 00..00..02...00..00 * */[/color]
>[color=green]
> > # lspv -l hdisk13
> > hdisk13:
> > LV NAME * * * * * * * LPs * PPs * DISTRIBUTION * * * * *MOUNT POINT
> > lv02 * * * * * * * * *113 * 113 * 50..05..00..15...43 * *#
> > hd6 * * * * * * * * * 128 * 128 * 00..45..49..34...00 * *N/A[/color]
>[color=green]
> > So, 2 questions:
> > #1 How do I get the stale LVs to read open/synced
> > #2 How do I get the missing 000048054d75e6b0 *out of the ODM?[/color]
>
> without too onerous a lecture, this is why documentation and regular
> backups are so valuable.
> it looks like you have had a disk failure.
> the AIX error log will hopefully tell you what happened, and when. try
> the "errpt" command. Also look through the syslog (check /etc/
> syslog.conf for setup)
>
> run
> lsdev -Cc disk
> lscfg -vl hdisk\*
> lslv -l lvname (hd6 etc; will probably fail)
>
> hope you have some good backups :)
> when was the last mksysb or other backups ?
>
> seriously, you will probably be able to remove the half of the mirrors
> that are on the problem PV, replace the failing disk (if that's the
> problem) and remirror.
> I had problems with AIX 4.3 and failing disks years back.
> I wouldn't suggest removing ODM entries, it's very easy to corrupt the
> VGDA on the PV's as it synchronises itself with the ODM and then
> you'll really be in the do-do's.- Hide quoted text -
>
> - Show quoted text -[/color]

Thanks for the info. We do have current backups but no mksysb for
this system. errpt talks about hdisk12 but when I run diags on
hdisk12 (as well as the others) it comes back clean. Also lspv shows
hdisk12 as available.

How do I determine what device is associated with 000048054d75e6b0
from the following:

0516-304 lsvg: Unable to find device id 000048054d75e6b0 in the
Device Configuration Database

Is that a disk? If so which hdisk#? All I can tell is that 'it' is
missing but I do not know that 'it' is

Thanks again

> Thanks for the info. We do have current backups but no mksysb for[color=blue]
> this system. errpt talks about hdisk12 but when I run diags on
> hdisk12 (as well as the others) it comes back clean. Also lspv shows
> hdisk12 as available.
>
> How do I determine what device is associated with 000048054d75e6b0
> from the following:
>
> 0516-304 lsvg: Unable to find device id 000048054d75e6b0 in the
> Device Configuration Database[/color]

odmget -q "attribute LIKE pvid AND value LIKE 000048054d75e6b0*" CuAt

if there is nothing in the ODM, perhaps this command can solve your
problem :

reducevg rootvg 000048054d75e6b0

Pierre-yves

On Oct 29, 12:31*pm, Pierre-Yves Fontaniere <p...@cc.in2p3.fr> wrote:[color=blue][color=green]
> > Thanks for the info. *We do have current backups but no mksysb for
> > this system. *errpt talks about hdisk12 but when I run diags on
> > hdisk12 (as well as the others) it comes back clean. *Also lspv shows
> > hdisk12 as available.[/color]
>[color=green]
> > How do I determine what device is associated with 000048054d75e6b0
> > from the following:[/color]
>[color=green]
> > * * *0516-304 lsvg: Unable to find device id 000048054d75e6b0 in the
> > Device Configuration Database[/color]
>
> odmget -q "attribute LIKE pvid AND value LIKE 000048054d75e6b0*" CuAt
>
> if there is nothing in the ODM, perhaps this command can solve your
> problem :
>
> reducevg rootvg 000048054d75e6b0
>
> Pierre-yves[/color]

I ran the odmget command and got nothing in return. So I ran the
reducevg command and got:

0516-016 ldeletepv: Cannot delete physical volume with allocated
partitions. Use either migratepv to move the
partitions or
reducevg with the -d option to delete the
partitions.
0516-884 reducevg: Unable to remove physical volume
000048054d75e6b0.

Not sure I want to use the 'reducevg -d' command since I'm still not
sure which hdisk 000048054d75e6b0 is.

Thanks for the commands though

Perhaps try issuing an rmlvcopy command for the stale LVs and see if
it will let you by specifying the PVID. Here is an example using your
values:

# rmlvcopy hd3 1 000048054d75e6b0
# rmlvcopy hd6 1 000048054d75e6b0

You can also list any LVs that have partitions on that mystery drive
the same way:

# lspv -l 000048054d75e6b0

If you can list the LVs that it thinks are still on that PV, simply
rmlvcopy those and then reducevg when no partitions are left.

And when I looked for the information in the ODM for my hdisk0, maybe
I did something wrong, but the LIKE option did not work for me, but I
was able to find my hard disks by providing trailing zeros for the
value. Try this:

# odmget -q "value = 000048054d75e6b00000000000000000" CuAt

My server gave two entries: Attribute = pvid and Attribute = pv

Check for known physical disks
$ lsdev -Cc disk

Check which device is using which disk for devices within your rootvg
$ lslv -l DEVX.

Check if problem causing devices are mirrored or not.
$ lsvg -l rootvg
$ lslv DEVX; lslv -l DEVX

In case they are mirrored run a rmlvcopy
$ /usr/sbin/rmlvcopy MYLV 1 PVID
Example: /usr/sbin/rmlvcopy hd3 1 000048054d75e6b0
Example: /usr/sbin/rmlvcopy hd6 1 000048054d75e6b0

Remove the offending PVID
$ reducevg rootvg PVID

In case a devices has pv allocated on the missing PVID and it is NOT
mirrored the device must be recreated.
From your output i assume tmp is not mirrored.
1) Login via console:
2) Stop all processes which are not needed.
3) Check which process is using /tmp
$ fuser -xuc /tmp
or
$ lsof /tmp
4) kill all processes using /tmp
5) recreate /tmp - The hd4 should have at least 128MB free since since
it will be the temporary tmp target

hth
Hajo


Reference:

Technote:
[url]http://www-01.ibm.com/support/docview.wss?uid=isg3T1000426[/url]

AIX 4.3 documentation
[url]http://publib16.boulder.ibm.com/pseries/en_US/infocenter/base/aix43.htm[/url]

AIX Version 4.3 Problem Solving Guide and Reference
[url]http://publib16.boulder.ibm.com/pseries/en_US/infocenter/base/43_docs/aixprob/prbslvgd/toc.htm[/url]

MJB a écrit :[color=blue]
> On Oct 29, 12:31 pm, Pierre-Yves Fontaniere <p...@cc.in2p3.fr> wrote:[color=green][color=darkred]
>>> Thanks for the info. We do have current backups but no mksysb for
>>> this system. errpt talks about hdisk12 but when I run diags on
>>> hdisk12 (as well as the others) it comes back clean. Also lspv shows
>>> hdisk12 as available.
>>> How do I determine what device is associated with 000048054d75e6b0
>>> from the following:
>>> 0516-304 lsvg: Unable to find device id 000048054d75e6b0 in the
>>> Device Configuration Database[/color]
>> odmget -q "attribute LIKE pvid AND value LIKE 000048054d75e6b0*" CuAt
>>
>> if there is nothing in the ODM, perhaps this command can solve your
>> problem :
>>
>> reducevg rootvg 000048054d75e6b0
>>
>> Pierre-yves[/color]
>
> I ran the odmget command and got nothing in return. So I ran the
> reducevg command and got:
>
> 0516-016 ldeletepv: Cannot delete physical volume with allocated
> partitions. Use either migratepv to move the
> partitions or
> reducevg with the -d option to delete the
> partitions.
> 0516-884 reducevg: Unable to remove physical volume
> 000048054d75e6b0.
>
> Not sure I want to use the 'reducevg -d' command since I'm still not
> sure which hdisk is.
>
> Thanks for the commands though
>[/color]
I think your disk with pvid 000048054d75e6b0 is just a "ghost" of an old
disk which has been removed.
Anyway, if you look at partitions stale :

hd3 : /tmp and hd6 (paging) ... that's not a big deal to loose them.

/tmp is entirely on hdisk7 and hdisk12
hdisk7:
hd3 3 3 00..00..03..00..00 /tmp
hdisk12:
hd3 12 12 00..12..00..00..00 /tmp

corresponding to the result of lsvg -l :
hd3 jfs 15 30 3 open/stale /tmp

you can confirm with :
lslv -l hd3

and hd6 is all on hdisk13
hdisk13:
hd6 128 128 00..45..49..34..00 N/A

So i think you can safely break the mirror with rmlvcopy and delete
entries of the pvid 000048054d75e6b0 with reducevg.
But it's true that it's a risky operation. It's never too late to make a
mksysb.

At the end, don't forget to recreate copies of /tmp and hd6 on disks
with free PPs. ;)

Pierre-yves

On Oct 30, 4:57*am, Pierre-Yves Fontaniere <p...@cc.in2p3.fr> wrote:[color=blue]
> MJB a écrit :[/color]

To All,

Thanks for all the help and ideas so far. I agree that the missing
device 000048054d75e6b0 is a ghost of an older disk. I've tried all
the above ideas and every time I get some various error code with the
following description:

Unable to find device id 000048054d75e6b0 in the
Device Configuration Database

I also agree that hd3 and hd6 do not NEED to be mirrored but am just
looking to clean things up a bit. This is and OLD server and has
limited users on it.