When updating to DP 9.04 in a StoreOnce environment, all gateway servers need to be updated to DP 9.04 as well.
A DP 9.04 Cell Manager wont work with 9.03 StoreOnce gateways.
If a customer reverts back to 9.03 setup a script needs to be executed to correct a schema.
Hello *
I'm running some test on DP9 prior to migrate the prod environment. Done a fresh install on a virtual environment I cannot log in the GUI and I get this error:
SSL peer certificate or SSH remote Key was not OK
I am using DP 7, When a backup object fails, I go to internal DB and right click on the failed job and click on remediate session, however it restarts the complete object.
I want this to restart from the value where it got failed. Suppose, on the first attempt, 30% backup was completed, so when i am restarting this, i want this to start from 30% not from starch.
When i click on resume session , it gives me below error.
Failed to load checkpoint informatrion. Sessioncannot be resumed and will be terminated.
Attaching error for the reference
Thanks,
Rohit Sharma
Hello there
I have Oracle 10G running on Win2008R2 x64. The documentation says: Oracle 10g w RMAN supported with DP 6.11. (http://h20566.www2.hpe.com/hpsc/doc/public/display?docId=emr_na-c01631170&lang=en-us&cc=us)
When i try to backup the database - session ends with error:
On target host in debug.log:
03.09.2015 3:25:02 Srdx.dll.32600.6332 ["/lib/cmn/win32io.c $Rev: 1477 $ $Date:: 2009-07-13 10:45:53":312] A.06.11 bDPWIN_00557
OB2_OpenFile: CreateFile() failed, errcode=3 [file=C:\Program Files\OmniBack\bin\drim\bin\drm.dll]
DP with the latest patches installed, what's wrong?
Thanks in advance.
Cell manager win 2008 r2 , dual core , 16gb , C drive of 30gb idb , omnidbcheck -extended is running for more than 12hrs, how do i reduce that time,need to shft my dp from C drive to E drive, hp saying not responsible if this prerequisite of extended command output is not shared with them. Cant keep my backups down for more than 12 hrs. Dp version 7.03 , part of migration to v 9.04 .
Suggest a solution.
Is it compulsory to run omnidbcheck -extended ?
This is on Windows Server 2008 R2, Data Protector 7.03 plus patches 107.
I've seen from a previous post that a hotfix is available for Linux. Is there one for Windows?
Jeremy
Hello there,
so our new StoreOnce is running, but we have some problems with backups over fibrechannel when source-side-deduplication is enabled.
If so we get this error message:
[Major] From: BSM@cellmanager.test.com "FCBACKUP" Time: 07.09.2015 07:21:55
Got error: "StoreOnce error: StoreOnce device offline, network error occurred or secure communication failed while contacting the StoreOnce device" when contacting "STOREONCE_DEVICE" B2D device!
So for what ever reason the cellmanager wants to connect to the StoreOnce over fibrechannel.
But our cellmanager has no fibrechannel and even if, this connection would make no sense...
The backup should be a direct connection from client to storeonce, why wants the cell manager to connect to the storeonce?
We use DP Version 9.04
Thanks in advance!
Hi guys,
we migrated from 2010 to Exchange 2013.
OS: Server 2012 R2
Until now i tested ALOT to get the Integration on this Box working.
The Problem we are facing seems to be related to the FQDN of the Exchangeserver.
We are making use of domain-name suffixes you can configure in Windows (System->ComputerName->More...)
So our machines are named "exch01.location.domain.local"
Our domain is "domain.local", location gets added manually via the described way above.
There are matching records in our DNS.
So what i found out is that....as soon as i rename the machine to "exch01.domain.local" the integration can be configured without any problem.
If i change the hostname back to our naming-convention DataProtector is doing some weird stuff.
First, when adding the machine under "Clients" it says "Machine identifies itself as exch01.location.mcl.local". (well...thats NOT wrong...and seems to be OK.)
Therefore the agents are getting installed without any problems.
But when creating a backup for that machine, right before configuration of the integration,
DP will prompt for logon-credentials .
This will happen over and over and over again.
So the User-Account for our Backups can't Authenticate.
(Account works on all other machines with these kind of names)
Diagnosing the debug.log file reveals the problem:
07.09.2015 13:38:44 OB2BAR_E2010_BAR.15320.11872 ["$PATH: integ/exch2010/ExchangeAgent.cpp $ $Rev: 41801 $ $Date:: 2014-03-26 16:29:59 +05":2465] A.09.00 b87
DbgStack: DbgFcnOut() function does not match function on stack!
ExitFcn="CExchangeAgent::CheckConfig", StackFcn="CPowerShellClient::CheckUserBelongsToGroup"
cmd:"C:\Program Files\OmniBack\bin\e2010_bar.exe" -chkconf
Function trace (currentLevel=2):
Level 2: CPowerShellClient::CheckUserBelongsToGroup(username = Administrator, domain= location.domain.local, group =Organization Management)
Level 1: CExchangeAgent::CheckConfig()
Level 0: CExchangeAgent::Initialize()
"CExchangeAgent::CheckConfig" found on stack, adjusted level to 1 (one or more DbgFcnOut() calls missing)
e2010.exe seems to have a problem with this domain suffix.
Authentication should be at "domain.local" and NOT "location.domain.local".
Is there any way modify this kind of behavior???
Seems like e2010.exe is taking everything behind the hostname as a authentication-domain.
So if our suffix would be machinename.location.country.planet.universe.local it would try to authenticate on that.
Looks like a bug to me??
I kinda gave up on this by now.
(added location.domain.local users in Cell, added a new authentication domain in our AD + a matching BackupUser with sufficient rights, modified several Textfiles (Config\Server\Integ\Config\E2010, cell_server......etc.), used omniinetpasswd.exe to add several users......
Maybe theres somebody around that faced the same problem in his environment.
We cant be the only ones using Dns-suffixes an trying to backup Exchange 2013??
help :(
Hello All,
Could soemone help me with the this query.......
compatiability matrix for DP 7.0 for windows and solaris servers.....
Hi Folks,
This is very common issue we trobleshoot while integrating Data Protector with Oracle supported version or when someone modified the Oracle users from the database.
What you need to know?
1-The user you are entering for the integration must exist in the Data Protecter userlist as Admin.
2-The following attribute must be added to the user you're using to integrate, from Oracle.
Example:
Lets say that your username is kumar
So the output of the following command
SQL> select * from v$pwfile_users;
Should be:
USERNAME SYSDB SYSOP
----------------------------------------
KUMAR TRUE TRUE
SYSTEM FALSE TRUE
If Kumar has NOT SYSDB Attribute they you get the message:
ORA-01017: invalid username/password; logon denied
In order to grant SYSDB attributes to KUMAR you must run the following command:
***You must be logged with as SYSDB user
SQL> grant sysdba to KUMAR;
After this your should be able to integrate or to get your backups running again.
If you are completely sure that you are using the right username and password and everything is in place then log a case with support.
Dont forget to provide Kudos if you find this topic useful.
Thanks,
Walter.
Hello folks!
This problem is becuase the original cell manager is down and the customer couldn't delete the client.
To re-import the client into a new Cell Manager you have to delete the content of the file or registry on the client that contains the information related to the previous cell manager that this client belongs to.
On Windows this information is in the registry on HKEY_LOCAL_MACHINE\SOFTWARE\Hewlett-Packard\OpenView\OmniBackII\Site\CellServer.
On Unix and Linux the file is in /etc/opt/omni/client/cell_server file
If you need to force several clients to perform this action you can run the following command
omnicc -update_all [-force_cs]
Have a nice day!
Walter.
Important "NOTE" to remember, Data Protector does not support CBT based backups on NFS datastores. A requirement prior to the backup, is to confirm that the virtual machine disks are residing on VMFS datastores. Secondly, the disks for the virtual machine cannot be independent.
Hello Group,
The team here have a DP9.04 cell manager running on Windows 2012R2. The same machine acts as a backup host. The system is patched at level 9.04.
I am unable to restore ESXi vm disks to a local directory on the backup host. The vm was backed without any issues via VCenter from an ESXi 5.0 cluster. I can select the VM version for restore via the GUI.
I have even tried backing up and restoring a test VM from an ESXi 4.1 host to the local directory. Works fine - all files appear, so not a permissions issue.
From the log trace below, you will see that the restore session starts, then aborts almost immediately with "Restore Failed".
Has anyone here seen this kind of problem ?
[Normal] From: RSM@svreohbck01.private.dom "" Time: 08/09/2015 12:52:57
Restore session 2015/09/08-32 started.
[Major] From: VEPALIB_VMWARE@svreohbck01.private.dom "/Head Office - Manhattan Bldg" Time: 08/09/2015 12:53:00
Restore failed.
[Normal] From: RSM@svreohbck01.private.dom "" Time: 08/09/2015 12:53:39
OB2BAR application on "svreohbck01.private.dom" disconnected.
[Normal] From: RSM@svreohbck01.private.dom "" Time: 08/09/2015 12:53:39
Restore Statistics:
Session Queuing Time (hours) 0.01
-------------------------------------------
Completed Disk Agents ........ 0
Failed Disk Agents ........... 0
Aborted Disk Agents .......... 0
-------------------------------------------
Disk Agents Total ........... 0
===========================================
Completed Media Agents ....... 0
Failed Media Agents .......... 0
Aborted Media Agents ......... 0
-------------------------------------------
Media Agents Total .......... 0
===========================================
Mbytes Total ................. 0 MB
Used Media Total ............. 0
Disk Agent Errors Total ...... 0
Since upgrading the cell from 9.02 to 9.03, customer was not able to enable ECC on some AIX Systems.
But It worked fine in 9.02
Customer could enable ECC on only 3 AIX Systems, the remaining AIX clienst failed to be encrypted..
Checking the cell_info on the CS showed that all those AIX systems which could not be encrypted ddid not have the -ts_core package listed.
Those 3 AIX systems that can be encrypted do have the –ts-core package installed.
Example:
failed --> -host "server507.domain" -os "ibm rs6000 aix-7.1" -core A.09.00 -da A.09.00
worked --> -host "server508.domain" -os "ibm rs6000 aix-7.1" -encryption 1 -core A.09.00 -da A.09.00 -ts_core A.09.00
In DP 8.00 new package TS_CORE was introduced and all non-DP software moved from CORE to TS_CORE packet.
Libcrypto and libssl libraries were moved to ts_core package as part of DP 8.0 changes.
These libs are needed by Inet if ECC communication is enabled.
Solution:
To bring TS-CORE onto the involved clients a new installation was necessary, updating CORE was not sufficient.
After successful installation of -ts_core ECC worked fine as before.
To help Data Protector work correctly with your configuration, remember to install a client using the canonical name of the client. Then import the other names or IP addresses for the client as "Type" "Virtual Host" in the "import Client" window. Otherwise Data Protector will consider them all separate hosts.
Hello there,
a few times in this week we saw that our Cell Manager has an 100% CPU utilization.
I could determine that this came from the "get_info" command. The process list shows:
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
5538 root 20 0 107m 952 476 R 74.5 0.0 681:23.06 get_info
5511 root 20 0 107m 952 476 R 49.6 0.0 678:48.51 get_info
5503 root 20 0 107m 944 476 R 49.2 0.0 677:45.77 get_info
The detailed list shows:
root 5538 5537 64 Sep09 ? 11:21:34 /bin/ksh /opt/omni/sbin/utilns/get_info -getversions ssp_test_info
Where does this process came from? How can I stop it?
At least we dont start it with awereness.
Thanks
Hey there,
so I noticed today that one of our VTL Library is almost full (Backup started a mount request).
So I investigated a bit and saw that all Tapes are declared as full, but there is just little data on it.
We use an StoreOnce VTL and the library is almost new. We have 6 libraries and every other one is perfectly fine.
But maybe this one is more special:
1. Normally we use a mirror for backups. Its already configured in the backup but OpenVMS has problems with that (case is open at HP). So we disabled the divces. Backups are fine (they just say at the beginning they dont have any mirror devices). I dont believe thats relavant, but who knows.
2. And that could be maybe a thing: At our old "real" VTL we used a diffrent block size (63 kB), otherwise OpenVMS backups
would fail. Now we use the default block size (256). And the backups run fine. Is it possible that this could cause this?
Any other ideas?
Thanks!
I executing a DR test and I did succesfully restored the Cell Manager computer to another hardware.
Now I trying to restore the other servers, which are running on vmware hosts.
The EADR process boots successfuly from the prepared ISO image, I modified the recovery.srd file with the data from the recovered Cell Manager. (dev, decioctl, devaddr, physloc, storname parameters).
Phase 2 starts successfully.
The restore session connects to the RMA running on CM. Loads the first tape and restores the first session. Then I get a message "Completed Disk Agent for minint-xxxxxx:\\?\Volume{aaaaaaaa-aaaa-aaa-aaa-aaaaaaaaa}. After that nothing happens. The other 2 session is pending. The tape console tells that the Tape is Reading...
The 1st and 3rd session are on the same media, the 2nd is on a different one.
1st session is \\?\Volume{aaaaaaaa-aaaa-aaa-aaa-aaaaaaaaa} the system reserved partition.
2nd session is Drive C:
3rd session is Configuration.
All the servers are Windows 2008 servers.
1. I have a Vsphere 5.5 vmhost that I have presented 4 VTL Tapes drives to. Please see the following list:
lsscsi -d
[2:0:0:0] disk VMware Virtual disk 1.0 /dev/sda [8:0]
[33:0:0:0] tape HP Ultrium 6-SCSI ED61 /dev/st0 [9:0]
[33:0:1:0] tape HP Ultrium 6-SCSI ED61 /dev/st1 [9:1]
[33:0:2:0] tape HP Ultrium 6-SCSI ED61 /dev/st2 [9:2]
[33:0:3:0] tape HP Ultrium 6-SCSI ED61 /dev/st3 [9:3]
2. I have installed the a DP 9.02 Media Agent the following pieces to the vmhost:
'
-key core -desc "Core Component" -nlsset 159 -nlsId 12262 -version A.09.00
-key ts_core -desc "Core Technology Stack" -nlsset 159 -nlsId 12288 -version A.09.00
-key da -desc "Disk Agent" -nlsset 159 -nlsId 12162 -flags 0x3 -ntpath "" -uxpath "" -version A.09.00
-key ma -desc "General Media Agent" -nlsset 159 -nlsId 12163 -flags 0x10003 -ntpath "" -uxpath "" -version A.09.00.
3. The problem is when I run "/opt/omni/lbin/devbra -dev /dev" it hangs and never returns any results so I am not able to configure the drives on my 9.02 Linux cell manage..
Hi,
For testing purposes before I have my actual storage, I would like to backup to a folder on a local drive of my cell manager server.
Is it possible to do this in DP? what would the steps be as when I attempt this the drive never shows as a choice of backup device when I go to make a backup.
Thanks