make_net_recovery list_expander failure

[MikeL_4]

We have been running make_net_recovery on a server now for well over a year. The server using Veritas Storage Foundation for the rootdg.

This week the make_net_recovery failed with the following messages:

ERROR: There is a problem with entirely including device
'/dev/vx/dsk/rootdg'. Check the existence of the device 'inc_entire'.

** 0 - The Volume Group or Filesystem is Not included in the
System Recovery Archive
** 1 - The Volume Group or Filesystem is Partially included in the
System Recovery Archive
** 2 - The Volume Group or Filesystem is Fully included in the
System Recovery Archive

ERROR: The list_expander command failed. This could be due to a problem with
the -x options specified - Check the existence of the included
device(s).

I am using the following command for the make_net_recovery:

make_net_recovery -v -x inc_entire=/dev/vx/dsk/rootdg -s cdcign1-8 -P s -a cdcign1-8:/var/opt/i
gnite/recovery/archives/osglbhp2

[MikeL_4]

ps: I can also do the list_expander command manually and it works ok:

/opt/ignite/lbin/list_expander

[MikeL_4]

Found the problem, when I did a vxprint I received the following:

=>vxprint -ht
vxvm:vxprint: ERROR: IPC failure: Configuration daemon is not accessible
=>ps -efx |grep vxconfig
root 296 1 0 10:46:43 ? 0:00 /usr/sbin/vxconfigd -k -m enable

I did notice in the server startup log the following messages:

vxvm:vxconfigd: ERROR: Disk group rootdg: Cannot recover temp database:
Duplicate record in configuration
Consider use of "vxconfigd -x cleartempdir" [see vxconfigd(1M)].
vxvm:vxconfigd: ERROR: enable failed: Error in disk group configuration copies
Duplicate record in configuration; transactions are disabled.


VxVM general startup ................................................ vxvm:vxconfigd: ERROR: Disk group rootdg: Cannot recover temp database:
Duplicate record in configuration
Consider use of "vxconfigd -x cleartempdir" [see vxconfigd(1M)].
vxvm:vxconfigd: ERROR: enable failed: Error in disk group configuration copies
Duplicate record in configuration; transactions are disabled. OK

Performed the following to resolve:

1. Stop and restart the VxVM configuration daemon:

# vxdctl stop
# vxconfigd -x cleartempdir

2. Create one or more dummy volumes in the disk group in which the error is occurring, until the file /etc/vx/tempdb/ increases. For example,

# vxassist -g rootdg make dummy1 100m
# vxassist -g rootdg make dummy2 100m

3. Remove the dummy volumes, such as:

# vxedit -g rootdg -rf rm dummy1
# vxedit -g rootdg -rf rm dummy2

4. Again, stop and restart the VxVM configuration daemon:

# vxdctl stop
# vxconfigd -x cleartempdir

The cause of the problem was introduced in patches: PHKL_36995 and PHCO_36994 and will only occur after a server reboot...

[MikeL_4]

See my last posting for cause and resolution.... Patches to fix the patches that introduced the issue have not been released by HP yet..