KasparsB, thank you for your reply.
In fact, since the installer is an MSI, there is no way to run the executable as administrator since the MSI package itself only extracts the installation files and runs the setup automatically, without giving me the option to change security parameters. I gave full read/write permissions to the user who accesses the file system (trustedinstaller).
Make sure that your Core and Cell SErver patches are up to date, for you , this would mean upgrading to DP 7.03. If you still have this problem, you would need to open a case
However, ther may be other members of the Forum who have real-world experience with this condition and that may be able to give an answer from a different perspective
Remove the Oracle configuration file
UNIX Cell Manager
/etc/opt/omni/server/Integ/config/Oracle8
Windows 2003 Cell Manager
Program Files -> Omniback -> COnfig -> Server -> Integ -> Config -> Oracle8
Windows 2008 Cell Manager
ProgramData -> Omniback -> COnfig -> Server -> Integ -> Config -> Oracle8
The configuration file will be named [instancename]%server name without the [ ] marks
This will mean that you are starting off fresh, creating a new Configuration file
Bob,
Thanks for your post. I found the files and deleted them(windows cell manager) that DP created for each of our 2 unix nodes and that cleared the form when I fired DP up and began adding the job. However I still get the same error. I tried choosing the use control file option vs the rman catalog option but it still gives me the rman invalid username and password error. Again I verified that the username and password works and here me showing my work:
ddcpayp1:10g2:ddcdb1:/opt/oracle>rman target / catalog rman/xxxxxxxx@ddcadmp
Recovery Manager: Release 10.2.0.4.0 - Production on Tue Apr 1 12:59:20 2014
Copyright (c) 1982, 2007, Oracle. All rights reserved.
connected to target database: DDCPAYP (DBID=3032698228)
connected to recovery catalog database
RMAN> exit
Recovery Manager complete.
ddcpayp1:10g2:ddcdb1:/opt/oracle>sqlplus dp_backup
SQL*Plus: Release 10.2.0.4.0 - Production on Tue Apr 1 13:00:27 2014
Copyright (c) 1982, 2007, Oracle. All Rights Reserved.
Enter password:
Connected to:
Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - 64bit Production
With the Partitioning, Real Application Clusters, OLAP, Data Mining
and Real Application Testing options
SQL>
Any other ideas(that one was great btw)?
Sincerely,
Dave
Hi,
just open a backup specification that contains such Windows server, in the Source tab expand the the server, then expand the CONFIGURATION: subtree - you will see what goes there. It is a collection of a number of different configuration databases and stuff that makes up the identity of a Windows OS installation. It's here because, in Windows, most of this information is held in locked files and traditionally couldn't be saved reliably by just pulling all files from the file system.
That being said, it's a really good question if backing up CONFIGURATION: makes any sense today with C: being backed up with the full might of a VSS snapshot, including all the metadata writers we see to jump in since 2008(R2). More specifically, I would ask if pulling Profiles this way is still of any use, given you will then back them up twice anyway. I tend to uncheck Profiles in CONFIGURATION: for a while now, but would really like to know if leaving out the entire tree would create any problems on a EADR of a hosed machine. Anything of relevance should be in the VSS snapshot in at least a crash consistent state - ususally better due to VSS writer integration file system quiesce support.
Never heard any clear words about it, though.
HTH anyway,
Andre.
Andre,
Thanks for the info.
I am trying to figure out what "QuotaInformation" backs up. Can you shed any light on this for me?
Thanks!
Re,
just guessing from the name of it: Sounds like metadata describing file system/disk quotas for users. It's a classic feature in multi-user environments to control how much disk space an individual user can allocate, so none of them can fill up your entire file system and cause problems for everyone else.
Cf: http://en.wikipedia.org/wiki/Disk_quota
HTH,
Andre.
Have you tried connecting in the manner I first suggested?
sqlplus /nolog
connect username/password@service as sysdba
This is the way that Data Protector tries to log in to the Oracle database. You cannot use any o fthe know shortcuts, like
connect / as sysdba
or
connect username/password@service
If you can commect this way
connect username/password@service as sysdba
and yet, cannot connect using the DP GUI to configure the instance, you will probably need to open a case with the response center
Yes I can connect(I previouly verified sysdba was granted on both nodes to dp_backup user). I will open a ticket then. Thanks for your help Bob!
NONE:10g2:ddcdb1:/opt/oracle/etc>sqlplus /nolog
SQL*Plus: Release 10.2.0.4.0 - Production on Tue Apr 1 15:10:09 2014
Copyright (c) 1982, 2007, Oracle. All Rights Reserved.
SQL> connect dp_backup/xxxxxxxx@ddcpayp1 as sysdba
Connected.
SQL>
Dave
Normally speaking, I woudl recommend that you install DP 7 on your existing cell manager, and do your upgrade. However, a check of the DP 7 support matrix shows no support for W2003 32-bit (only 64-bit)
However, DP 6.2 is supported on 32-bit, Therefore, this should be a multi-step operation. Get teh 6.2 Installatio bits
INstall DP 6.2 on your existing system according to the Installation and licensing Guide for 6.2
I did not see any kind of IDB conversion involved in this upgrade. I recommend that, at this time, you install the DP 6.21 patch bundle, available from
Run a few backups to be sure that the upgrade went successfully, then, install DP 6.2 on your new call manager. Run a few local backups (on the cell manager system), then upgrade it to DP 7
It appears that there is no IDB conversion for this version either. I advise that, at this time, you upgrade to the DP 7.03 patch bundle
Now, having said all of this, there is a direct upgrade path from 6.11 to 7.0. The problem is that you are changing OS-versions at the same time
You may want to backup your 6.11 IDB onto tape, install DP 7.0 fresh on the new cell manager, create a temporary device to IMport that tape, and do the IDB restore
Once you have DP 7 installed, you can use the Client context to upgrade your clients to 7.0, if the Cell Manager hostname is not changing, thsi should be an easy task
One thing to remember, always plan as if the upgrade will fail. Make copies of the COnfiguration files
Program Files -> Omniback -> Config -> Server
and of the DP IDB
Program Files -> Omniback -> db40
so that, if things don't go well, you can go back to the previous version with a minimum of downtime
This is the problem:
WARNING server login failed: FATAL database "hpdpidb" does not exist
stop DP and restart it, and run
omnisv status
and make sure the 'hpdp-idb' serice is running
You will also probably need to check the MIgration logs
C:\ProgramData\OmniBack\log\omnimigrate*.log
Maybe, but I checked the Virtualization Support Matrix from January 2014 and it's not mentioned anything like that on page 3.
Thanks
Steve
Just downloaded the most recent support matrix version March 2014 and it's clearly mentioned on page 3.
about the offline restore. this usually happens if the cell manager is not accessable from the client on which you perform the disaster restore. i would suggest you to check those points:
- is the network adapter recognized by the os?
- was the DR Process able to set the IP on the network adapter?
- is dns configuration done? (can you ping the cell server with fqdn)
if those steps are sucessfull and the restore is still switching into offline mode, add an <ANY> <ANY> <ANY> user to the DP admin group. do this just for the restore. IMPORTANT, remove this user after the restore!!
ok thanks a lot.
I also got below error at the end of DR wizard:
Cannot Copy X:\Windows\Temp\inet.exe to X:\Windows\System32\OB2DR\bin\omniinet.exe Please verify that the destination location is writable, that there is enought space on the destination drive and the write target is not locked in any way. Do you want to continue copying Disaster Recovery files?
Hello mardicas,
as you wanted to know, if there are any problems, i have some :´(
Installed everything, xinetd runs, qnap (firmware 3.8.0), import fails.
I then checked manual by doing a telnet connection to port 5555. Here the result:
Escape character is '^]'. inet: error while loading shared libraries: librt.so.1: wrong ELF class: ELFCLASS32 Connection closed by foreign host.
# uname -a Linux eron 2.6.33.2 #1 SMP Wed Nov 14 12:07:54 CST 2012 x86_64 unknown
but there are is only 32bit librt....
# ldd omni/lbin/inet
linux-vdso.so.1 => (0x00007fff0a5a6000)
librt.so.1 => not found
libm.so.6 => not found
libdl.so.2 => not found
libBrandChg_64bit.so => /opt/omni/lib/libBrandChg_64bit.so (0x00007f985f365000)
libpthread.so.0 => /lib64/libpthread.so.0 (0x00007f985f148000)
libc.so.6 => /lib64/libc.so.6 (0x00007f985edc5000)
/lib64/ld-linux-x86-64.so.2 (0x00007f985f56a000)so o copied the missing libs to /opt/omni/lib
now ldd does not give any errors anymore.
Telnet to 5555: ok
Import: fails :´(
In debug.log i get messages:
04/02/14 09:01:21 INET.21550.0 ["/lib/cmn/unix.c $Rev: 40407 $ $Date:: 2013-12-03 16:46:54":160] A.08.00 b600 CmnGetUserInfo: getpwuid() failed, errcode=2 [euid=0] 04/02/14 09:01:21 INET.21550.0 ["/lib/cmn/unix.c $Rev: 40407 $ $Date:: 2013-12-03 16:46:54":176] A.08.00 b600 CmnGetUserInfo: getgrgid() failed, errcode=2 [egid=0]
I have no more ideas .....
warm regards,
stef
I think I have it figured out. It looks like the CM name was changed when the domain changed, but nothing else was changed. I've changed it back and it seems to be queuing correctly.
Hi, I'm still having this issue. Any ideas you guys have or should I go to HP support with this?