How to know when a path was upped

[Brais]

Hello,

  I would like to know where (logfile) can i check when a drive path was upped.

Thank you.

[Mark_Solutions]

On Windows it writes to the application event log

As it is an EMM operation you would need to look in the EMM logs

vxlogview -o nbemm -t 03:00:00

will show the last 3 hours of EMM logs

Hope this helps

[mph999]

... also in the lltid log ..

/usr/openv/volmgr/debug/ltid ..

Example ...

13:12:48.752 [16326] <4> emmlib_UpdateMachineState: (0) Machine < womble >, Previous MachineStateBits < 14 >

13:14:42.890 [16326] <4> LtidProcCmd: Pid=22349, Data.Pid=22349, Type=100, Param1=0, Param2=0, LongParam=2118485824

13:14:43.900 [16326] <4> LtidProcCmd: Pid=22349, Data.Pid=22349, Type=126, Param1=1, Param2=1, LongParam=0

13:14:43.915 [16326] <6> DownDrivePath: DOWNing path /dev/rmt/1cbn for HP.C5713A.000 on host womble

13:14:43.916 [16326] <6> WriteEntry: Updating drive HP.C5713A.000 at path /dev/rmt/1cbn on attach host

13:14:43.916 [16326] <2> update_drive: (0) Inputs: MediaServer = < womble >, NdmpHost = <  >, DriveName = < HP.C5713A.000 >, DrivePath = < /dev/rmt

/1cbn >

Regards,

Martin

[Yasuhisa_Ishika]

On Unix, check syslog file. Default syslog file on Solaris is /var/adm/messages, /var/log/syslog on Linux, /var/admsyslog on HP-UX(if I remember right). By default, no syslog on AIX.

[Marianne]

Please add VERBOSE entry to vm.conf on ALL media servers and restart Device Management service/daemon.

This will start to log device-level actions/issues to Syslog on Unix/Linux media servers and to Event Viewer on Windows media servers.

You will find every DOWN and UP action in these logs. It will also tell you if EMM or 'Operator' has performed these actions.

Example:
/var/adm/messages on Solaris
/var/log/messages on Linux
Event Viewer Application log on Windows.
 

[Brais]

Thanks to all for your replies.

please could you give me an example when the drive goes UP instead of DOWNing

Thanks in advance.

[Mark_Solutions]

NetBackup itself (EMM) will down a drive when it has an issue (three read / write / write protect failures in a specified period)

A drive is generally only "UP'd" during a service restart or if Manaually Up'd by a NetBackup Admin

#EDIT#

Can I ask why you need to know - do you have a specific issue?

[Brais]

There were some drives with issues, which were put down in purporse.

Someone put them up again, thus backups failed and tapes were frozen.

I would like to know the exact time when they were upped, to try to guess who did it.

Regards

[Marianne]

NBU will not UP drives automatically. Always UP'ed by Operator.

I will dig and see if I can find logs in my 'archives' with relevant examples.

[Mark_Solutions]

Are you on Windows or Unix?

[Brais]

Unix, Solaris 9

[Mark_Solutions]

In that case look to see if you already have any of the logs in place as suggested for Unix in the above answers - I will do more digging to see if i can think of anywhere else it will be logged

My suggestion of the EMM earlier will not give you what you want so please ignore that one

[Marianne]

Example of drive being UP'ed - message in Windows Event Viewer Application log after VERBOSE entry was added to vm.conf and services restarted:

2011/03/17    05:22:16    NetBackup Device Manager    Information    None    2735    N/A    Mediasrv1   Operator/EMM server has UP'ed drive QUANTUM.SDLT320.000 (device 0)

I could not find a Unix/Linux example, so I have just added VERBOSE entry to vm.conf on our in-house RHEL lab server and restarted NBU. Then manually down'ed and up'ed the drive.

Entries in /var/log/messages:

Feb 27 10:27:39 softserv01 ltid[26632]: Operator/EMM server has DOWN'ed drive IBM.ULTRIUM-TD3.000 (device 0)
...
Feb 27 10:28:05 softserv01 ltid[26632]: Operator/EMM server has UP'ed drive IBM.ULTRIUM-TD3.000 (device 0)

[Brais]

Thanks for your time.

This shows what I was looking for.

Sad VERBOSE was not enabled at the time the issue happened.

[Marianne]

Add VERBOSE entry for 'next time'!