Netbackup 6.0 Ent Server with MP3 and Media Servers with SSO

[Thajwas__]

Hi Experts,

I have Master Server running NBU 6.0 and 4 Media Servers also running 6.0 and sharing S4 library drives with Master Server.

Incase i configure the storage devices on the server using GUI, all shared drives are configured as standalone.

now if i change the drive settings as robotic drives later and configure the storage units in the GUI, i get the generic SQL error after running
vmoprcmd -d

Further incase i delete all the configurations in admin console and do the configurations using commandline, it seems to be ok.and even vmupdate successfully inventorizes the library

but later when i issue backups, i keeps on loading the tapes and ultimately drives go down.

can anbody pls. suggest.

Thanks & Regards,
Thajwas

[Rakesh_Khandelw]

It looks like your drives are not configured correct on servers. Check your drive cofiguration against device mapping by running robtest, just to make sure you have assigned correct drive names/index numbers to drives on each host. If it's Solaris environment you can also run tpautoconf -t command to check the drive serial numbers and device number each serial number is associated on individual server .Once you have this information you can verify you configuration against it (assuming you can pull drive serial numbers from library console).

[Thajwas__]

Yes you are right sometimes this problem arises incase drives are not configured properly.

Ok when i do the configuration with command line, i actually verify the serial nos with tpautoconf -t and further verify with sgscan o/p as well.

Can it be possible that tpautoconf -t will not give the correct mapping. But then we are seeing multiple drives through a single fibre connection.

Can you pls. suggest further.

Thanks & Regards,
Thajwas

[Stumpr2]

Check to see if the windows servers have any hidden devices that may be causing issues. Several events can occur that contribute to the addition of ghost devices in the Windows 2000 Device Manager. Some of these include adding or removing hardware, changes to tape drives or tape libraries, or failure to use persistent binding, static indexing, or hard AL_Pa's on storage area network (SAN) equipment which
can allow changes to SCSI device path presentation of a device to the operating system. These ghost devices can retain Port, SCSI, Target, and logical unit number (LUN) information and conflict with the active devices being used by Windows 2000. Under these circumstances, hardware instability can result as well as other undesirable device operations.
Running the device discovery wizard, for example, may list 20 tape drives when only 10 tape drives physically exist. This information is being pulled from the operating system and can incorrectly associate tape device information with the VERITAS NetBackup (tm) hardware configuration.
You may also experience tape device drivers "disappear" or roll back to the previous drivers after a NetBackup tape device driver update. In this scenario, the drivers were actually applied to the ghost device and upon reboot, the active devices loaded the previous driver information.

[Dennis_Strom]

For me when all else fails delete all the devices then re-add them using the device configuration wizard. Do not use the shared drive wizard. Do this on all boxes starting with the master. With win boxes you probably have to reboot everything including the san.

[Thajwas__]

Hello Guys,

All the 4 Media Servers and Master Server are on Solaris. Also i tried the configuration with GUI and command line.

When i do with the command line , i take the o/p of tpconfig -t as reference for matching the drives, still the problem.

Even after device configuration, i cannot mount the tape using tpreq as well.

waiting for expert comments.

Thanks & Regards,
Thajwas

[Thajwas__]

Hello Guys,

I more thing i want to share about this case here is that even if i configure the devices via admin console and configure for just the master server,

all the drives are shown as standalone, an i have to manually drag them to robot.

but later once device configuration is done, i still cannot mount the tape inside the drive even if i try with tpreq command.

though inventory is successful.

Any expert comments are appreciated.

Thanks & Regards,
Thajwas

[Rakesh_Khandelw]

What kind of library (make/model) do you have and how is it connected to Master/Media servers?

What is OS running on servers?

[Thajwas__]

The Library model is C4 from Sun cascaded into three and thus toatal drives we have is 6

Alll the Media Server and Master servers are on Solaris. Master and few media Servers are on Solaris 10 and other on Solaris 9.

Thanks & Regards,
Thajwas

[Rakesh_Khandelw]

Are these native fibre drives or connected through some bridge?

[Thajwas__]

through SAN Switch OEM from SUN

[Thajwas__]

Hello ,

I managed to configure the drives on the Master Server and later manually configured drives on media servers as shared and controlled by master server.

But when i start the backups, few are successfull, but rest fail. And i see the below message in bptm logs.

Can anybody pls. suggest.

Thanks requestFailed: got gotCallback, failure status =
12:31:30.721 <2> packageMultiResourceRequestResult: totalNumberOfAllocations == 1
12:31:30.721 <2> packageMultiResourceRequestResult: retVal = 96
12:31:30.721 <16> RequestMultipleResources: MultiResReq.cpp:1429 resource request failed
12:31:30.721 <2> RequestMultipleResources: retVal = 96 emmStatus = 2005000
12:31:30.721 <2> RequestMultipleResources: returning
12:31:30.721 <4> nbjm_media_request: Error from RequestMultipleResources, Master server, returned error 96
12:31:30.722 <2> send_brm_msg: ERROR 96
12:31:32.730 <2> mpx_terminate_exit: EXITING with status 96
12:31:32.730 <2> cleanup: Detached from BPBRM shared memory

[Rakesh_Khandelw]

This may be helpful -

http://seer.support.veritas.com/docs/278555.htm

[Thajwas__]

Thanks, but i had already downloaded this and could not find anything relevant and helpful.

Actually the backups run and eachtime one drives goes down.

Pls. help

Regards,

set_job_details: LOG 1172027083 16 bptm 13205 error requesting media, TpErrno = Robot operation faile
d

11:04:43.516 <2> set_job_details: Done
11:04:43.561 <2> job_monitoring_exex: ACK disconnect
11:04:43.562 <2> job_disconnect: Disconnected
11:04:43.562 <2> db_error_add_to_file: dberrorq.c:midnite = 1171987200
11:04:43.607 <16> mount_open_media: error requesting media, TpErrno = Robot operation failed

[Dennis_Strom]

It is possible this is a problem with the robot. Sometime the robot gets out of calibration with the slots. Is it possible to check the library for any errors.