SRT restoration issue Netbackup 7.5 OS solx86_64 solaris 10 x86_64
Hi support,
i am facing a problem deyails given below,
root@Master # /usr/openv/netbackup/bin/bmrprep -restore -srt Oraclesrt -client Oracle-abc -config current -logging -quickformat
[Error] V-126-3 'bmrprep' could not complete the requested operation.
Please see logs for additional information.
[Info] V-126-43 Cleaning up. Please wait...
root@Master# /usr/openv/netbackup/bin/bmrprep -restore -srt Oraclesrt -client Oracle-abc -config current -logging -quickformat
[Error] V-126-3 'bmrprep' could not complete the requested operation.
Please see logs for additional information.
[Info] V-126-43 Cleaning up. Please wait...
[1]+ Done jnbSA (wd: ~)
(wd now: /usr/openv/netbackup/bin)
i also try to delete the srt but fail.
root@Master # ./bmrsrtadm
Select one of the following options:
1. Create a new Shared Resource Tree.
2. Create a new CD image based Shared Resource Tree.
3. Copy an existing Shared Resource Tree to a new location.
4. Import a Shared Resource Tree.
5. Modify an existing Shared Resource Tree.
6. Delete an existing Shared Resource Tree.
7. List Shared Resource Trees available on this server.
8. Quit.
Enter your selection (1-8) [1] : 6
Enter the name of an existing SRT : Oraclesrt
You have chosen to delete the "Oraclesrt" SRT from the server.
Are you sure you want to delete this SRT? (y/n) : y
Removing SRT "Oraclesrt" from server -- please stand by ...
V-128-749 Operation failed due to network problem.
[Error] V-125-61 Cannot remove SRT from server.
[Error] V-125-443 SRT modification did not succeed. Check the logs.
urgent response is highly Appreciated
.
Comments 14 Comments • Jump to latest comment
hi,
did you had a chance to look into to below T/N
http://www.symantec.com/business/support/index?pag...
http://www.symantec.com/business/support/index?pag...
Could you run the list SRTs option and show us the output please
Is that SRT on the server you are running the command from?
Could you vxlogview commands against ODI's 119, 125 and 126 for the period of you trying your commands and post the outputs as attachments please
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
hi Nagalla,
yup i already read the links but no success
meanwhile i remove the srt via bmrsrtadm and also GUI, removed the policy make new srt and policy for backup. now i get the error change
root@Master # /usr/openv/netbackup/bin/bmrprep -restore -srt Solaris10u9 -client Oracle -config current
[Error] V-126-3 'bmrprep' could not complete the requested operation.
Please see logs for additional information.
The Shared Resource Tree "Solaris10u9" could not be shared by the Boot Server hosting it. Please verify that the required configuration in the Boot Server is correct. The nfs daemons and configuration files need to be properly working for this operation to work.
[Info] V-126-43 Cleaning up. Please wait...
Hi Mark,
i run the command on Master server.
can u explain the "ODI's 119, 125 and 126 for the period of you trying your commands and post the outputs as attachments please" then i collect the logs and definitly attach for reference
waiting ur reply
Sorry - that was a typo - should have been OID's (operator ID's)
So if you have just made changes then you can run:
/usr/openv/netbackup/bin/vxlogview -o 119 -t 00:01:00>/tmp/oid119.txt
This will output the logs for process ID 199 for a period of the last hour - you can then post the text file produced on here
Do the same for the others (125 and 126) - although you may well spot what is wrong when you look at the outputs
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
hi all
please review the attachment
hi Mark,
please find the attached file
The logs are saying that the SRT may be set for exclusive use of client "(none)" - not very helpful!
Take a look at this link to see if the SRT is locked out from anything using it:
http://www.symantec.com/docs/HOWTO45871
However - it also says the following and i would look at the communications issue first:
Could not contact the Bare Metal Restore Boot Server daemon (bmrbd) on host 'Master-NTC'.
Make sure that the host is reachable and that the daemon is running, and retry the operation.
So this does need to be checked - or you are running the commands from the wrong location?
It goes on to say :
Configuration 'current' does not belong to client '172.16.72.212' - so there is something else wrong here as well.
Finally it says Allocation of SRT 'Solaris10u9' to client 'Oracle-NTC' failed, rc=1.
Check communications and where it is all running from first
Hope this helps
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Hi Mark,
i change it the SRT to specfic the Oracleclient machine but no success
can u helpme after re-view this error logs
/usr/openv/logs/bmrd/*.log
0,51216,119,119,1,1357112198623,4693,1,0:,54:The specified boot server Master-NTC still hosts SRTs.,27:bmrd.cpp:DeleteBootServer(),1
0,51216,128,119,2,1357112560339,5108,1,0:,26:Replaced object in m_vSrt.,31:CBmr::readXml(xml, len, filter),1
0,51216,128,119,3,1357112560926,5108,1,0:,39:Failure pulling client custom tar file.,41:methods.cpp:ProcessClientCustomSrtFiles(),1
0,51216,128,119,4,1357112561145,5108,1,0:,29:Replaced object in m_vConfig.,31:CBmr::readXml(xml, len, filter),1
0,51216,128,119,5,1357112561216,5108,1,0:,30:Replaced object in m_vNetwork.,31:CBmr::readXml(xml, len, filter),1
0,51216,128,119,6,1357112562382,5108,1,0:,52:Child process pid=5301 exited with non-zero status=1,37:bmrSystemCommon.cpp:BmrSystemCommon(),1
1,51216,128,119,7,1357112562383,5108,1,0:,0:,33:methodsPlat.cpp:VerifyDhcpdPlat(),1,(967|)
0,51216,128,119,8,1357112562383,5108,1,0:,20:DHCPD is not running,37:methodsPlat.cpp:ConfigureDHCPMacros(),1
The errors in the 119 log are as follows:
BMR Master Server can not function. Please check if BMR Master Server Database has been setup and whether the Database Service is running.
Operation failed due to network problem.
Cannot query the BMR database for existing boot servers.
Either the BMR database has not been setup,
or the database service is not functioning,
or the bmrd on the BMR master server is not running.
Fix these problems and retry.
You need to check that everything is actually Ok first then try again. This may need some hosts file entries or it could even be a disk space issue if / is out of disk space as the tftpp directory cannot write any files.
It does go on to say:
Failed to configure DHCP macros, output = /export/srt/VMSRT already has an entry in /etc/dfs/dfstab.
However, /export/srt/VMSRT must be shared read-only with root access.
Use ro and anon=0 for /export/srt/VMSRT. The /etc/dfs/dfstab file must be
fixed and /export/srt/VMSRT shared before 0:c:29:3a:52:2f can boot.
cleaning up preexisting install client "0:c:29:3a:52:2f"
To disable 0:c:29:3a:52:2f in the DHCP server,
remove the entry with Client ID 01000C293A522F
If not already configured, enable PXE boot by creating
a macro named 01000C293A522F with:
Boot server IP (BootSrvA) : 172.16.72.210
Boot file (BootFile) : 01000C293A522
So it all points to something not being setup correctly here - work through the basics of the setup first to make sure it is all OK before trying to do anything else
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
May be taking this off on a tangent but why on a Solaris box would you choose to use BMR rather than creating flash archives of your critical servers? Flash archives are an extremely simple way to achieve bare metal restores on Solaris servers in literally one step. It takes a complete image of your system state at the time the flash archive is taken, including all installed and configured 3rd party software and device drivers. Store these archives somewhere on your network and if you ever have a complete failure you can simply run the Solaris installation DVD and choose to install from a flash archive and your ENTIRE system is returned to a production state in one easy step... Simply restore your latest Oracle Backup and your off and running. The only time you would ever re-create the flash archive is if you have applied a recommended patch cluster or additional software..
The beauty of flash archives is that its platform independent. You could take a flash archive of a server which for arguments sake , say an M4000 and do the flash archive restore to an x86 server and vice versa. All drivers for that server will be installed and a new path_to_inst file will be created for the devices on the new hardware... A MUCH simpler solution for Solaris than BMR... The only fit I see for BMR is on the Windows platform it was originally designed for..
Check the man page on the command "flarcreate"
Jeff Foglietta
Practice Lead - Data Protection and High Availability
Under the KISS principle, the same technology should be applied to all systems in an environment. That way, in theory, the same skills are required to recover any system. All of us know the difference between theory and practice. Since BMR is a cross-platform solution, management prefers it to having separate technologies and separate training for each operating system type. Also, use of Flash Archive is predicated on a "properly installed" system. Unfortunately, there are too many ways to install a system that would complicate the use of Flash Archive, and reduce its utility.
Steve Baker
Once again we realize there are many ways to skin a cat. That said, With all due respect regarding the tail end of your post Steve I must disagree. You state the "use of flash archives are predicated on a properly installed system" as if this is a chore... Sure , ALL flash archives are created from a properly running system.. Thats the idea! And when it is restored it will once again be a properly running installed system on every level including device drivers and apps.
"too many ways to install a system that would complicate the use of flash archives, and reduce its utility???
I'm not entirely sure what this means, however I do know that I disagree and have not shared your experience in this regard. Flash Archives are much quicker and simpler to manage on Solaris systems when it comes to bare metal. Should i lose a Solaris system I'm going right to my flash archive.. Why? Because the system is fully restored in one step, application binaries and all... So all I need to restore is my data if necessary. If my data resides on a raid protected storage array, even better... One step... Done!
Like I said.. different means to an common end.. You have your way, I have mine.. I like my way better.
Regards
Jeff Foglietta
Practice Lead - Data Protection and High Availability
hi mark,
i reconfigure all the things from scrath and also configure the dhcp server.
now its working no issue with srt restoration its working now.
thanx
Good stuff - if you feel that any suggestion has helped solve your issue dont forget to close the thread by using the "Mark as solution" option which helps others searching for similar issues in the future
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Would you like to reply?
Login or Register to post your comment.