Hi,
HP Data Protector 'EXEC_INTEGUTIL' Arbitrary Command Execution is solved in HP DP 9.06 for sure.
Regards
tech88kur
↧
Re: HP Data Protector 'EXEC_INTEGUTIL' Arbitrary Command Execution
↧
Re: Dataprotector and StoreOnce VSA integration
Hi Grimaldy,
I try to answer your questions. Not sure what you mean with the image backup issue, but maybe you can open a new post for that.
What would SO VSA allow me to accomplish concerning DP?, right now we are going to use DP as a File Library, we are sending the Backups to a LUN.
The StoreOnce VSA runs the same code as a physical StoreOnce appliance. It is a highly optimized backup device that offers superior deduplication for efficient data storage. With the VSA you can device on which hardware the StoreOnce code runs. You can turn local or SAN storage into a dedupe backup storage.
If the SO VA its used as a Dedup 'device' what is the purpose of the 1TB HDD assigned to this VM?
You can scale the StoreOnce VSA using 1TB VMDKs. Each of the VMDK pieces are bound together into a file system where the deduplication engine runs. Depending on the VSA license it can scale in 1TB increments up to 50TB total.
What should I configure on the SO VA,a Catalyst store, a NAS, a VTL or what?, bearing in mind that DataProtector has been configured as File Library.
StoreOnce Catalyst is the best option. While a File Library on a CIFS/NFS share would also dedupe (on the VSA) it is integrated deeply into the product. Data Protector allows you to use different kind of backup devices simultaneously.
Hope this helps.
Regards,
Sebastian Koehler
↧
↧
Re: Dataprotector and StoreOnce VSA integration
Already wrote about the disk image issue with DP, here
Really appreciated your help, lets the tests begin.
↧
Re: Full backup works but not the incrementals
Thank you Sebastian.
This has worked before. I checked the logs and the incrementals used to work. The logs show that they stopped about one month ago. Something obviously has changed but no one around seems to know. (I just recently inherited this task by the way, so I'm a bit of a newbie). :-)
Here are more details:
The backup spec has three clients: engtaps, sat-taps and tel-taps. All three are linux CentOS boxes. The engtaps and sat-taps servers only have local mount points but tel-taps is my culprit with an NFS mount point to a volume on a NetAPP NAS appliance.
Every single "local" mount point are completed quickly with the exception of the NFS one. I unfortunately cannot exclude any trees from that mount point.
I'm afraid that increasing the timeout would simply have that task take longer but for diligence sake, where would I change this timeout?
I will look into upgrading to a newer version.
Again, thank you.
Guy
↧
Re: Cell manager - TAPE and VTL configuration on physical server
Hi sebastian,
Thank you for sharing valuable thoughts. Your explanation has cleared my doubts.
Thanks indeed,
Cheers
Dpx
Thank you for sharing valuable thoughts. Your explanation has cleared my doubts.
Thanks indeed,
Cheers
Dpx
↧
↧
Re: StoreOnce CoFC and Federated Catalyst
Hello Sebastien,
Many thanks for you answer. I just probably understand the problem, in fact I zoned only the node 1 to the server, this allowed me to see the catalyst over node 1 &2, then I zoned node 3 to the server, in my mind I was thinking that y should see like for node 1, node 3 & 4, but the node 4 was always offline
If I remove node 4 this is working, so I believe that I have to zone also node 4.
I will zone node 4 on Monday and give you an update.
Best regards,
Roland
↧
GUI very slow in Data Protector 9.06
Hi all, I have a HP Data Protector 9.06 system installed on a Windows Failover Cluster Server 2012.
I am experiencing many delays in the response of the GUI.
Continuously indicates the following in the bar below:
Working ...
or
Contacting the Cell Manager ....
Sometimes it seems to work a little faster but a few second delay returns
Do you know that you can be?
regards
↧
Re: GUI very slow in Data Protector 9.06
Hi mainland,
Please add the following omnirc parameter both all cluster nodes. Does this work?
# Reduce DNS timeouts from 30 to 2 seconds OB2_DNSTIMEOUT=1
Regards,
Sebastian Koehler
↧
Re: StoreOnce CoFC and Federated Catalyst
I can't explain why you're able to see stores from node 2 if it is not zoned. Is service set 2 in failover state (and active on node 1)?
Regards,
Sebastian Koehler
↧
↧
Re: Rawdisk image backup & Restore
Hi Grimaldy,
before we dig too deep we should go one step back. If I understand correctly you try to prepare for disaster. Hardware or software failure that requires operating system and data recovery. What you're looking for is Data Protector EADR.
Enhanced Automated Disaster Recovery requires that you install the AutoDR component along with the Disk Agent on a client. Now perform a file system backup of critical parts (Operating system disk and /CONFIGURATION object) and you're ready to go. With EADR you can create an image (specific to this client) that can be used for complete operating system and data recovery. You could also restore to dissimilar hardware configurations (for Windows). Linux also works pretty well and is used by most of my customers.
The restore of a raw disk backup requires that the disk is offline. You see that section locked error during restore. When the operating system is running you will never able to restore the operating system drive (C:). While you could also use a raw disk backup to prepare for EADR, this is not necessary and quite uncommon.
To guide you into the right direction you should check DisasterRecovery.pdf in the Docs folder.
Regards,
Sebastian Koehler
↧
Re: GUI very slow in Data Protector 9.06
Hi Sebastian. Is there to restart services data protector after adding the lines you mention?
Best regards
↧
Re: GUI very slow in Data Protector 9.06
This should work without service restart. Just open a new GUI and check if it works.
Regards,
Sebastian Koehler
↧
Re: GUI very slow in Data Protector 9.06
Sebastian Hello again, I added the lines you described and looks a little better.
But when I save a backup espacificación, still taking about 20 seconds to save or record the changes with the following message:
Contacting the Cell Manager
regards
↧
↧
(DP) Support Tip Internal error :DbaXXXX functions
During backup, we meet this error, we should check IDB and will find the problem media, export/import should solve the problem.
But in restore, this error should be related with the un-completed installation DP patches.
1. IDB check：
#omnidbcheck -extended
Check Level Mode Status
==========================================================
Database connection -connection OK
Schema consistency -schema_consistency OK
Datafiles consistency -verify_db_files OK
Database consistency -database_consistency OK
Media consistency -media_consistency OK
SIBF(readability) -sibf OK
DCBF(presence and size) -bf OK
OMNIDC(consistency) -dc OK
2. debug file ****DBSM****
[ 99] 2015-11-17 19:14:51 ("/svnstore/unix/panther/dp90_features_nightly/src/idb/src/facade/object_version_service.cpp":2627)
[ 99] <<=== (5) } /* virtual void idb::FindFilteredVersionsService::doProcess() */
[ 99]
[ 5] Dynamic exception type: std::runtime_error
std::exception::what: ERROR: column "copy_version_uuid" does not exist
LINE 1: ...T OBJVER_UUID, DP_NUMKEY, SESS_UUID, SESS_NUMKEY, COPY_VERSI...
[170] Returning from Service with internal error.
This is most likely caused by installing newer patches and then downgrading to older ones which is not supported.
Solution:
Confirm if upgrade has taken place and then downgrade. Upgrade again.
↧
(DP) Support Tip Backup pending on "COMPLETED Media Agent "HP:Ultrium xxxxxx"
Backup pending on "COMPLETED Media Agent "HP:Ultrium xxxxxx"
Solution:
1. In CM, stop DP services， check and kill all DP related process
2. edit file <DP dir>\config\server\options\global, and increase the value of parameter "BackupDeviceIdle=3600“
3. restart DP services.
Generally they would solve the pending problem.
Sometimes we need also change TCP/IP parameter as following:
4. change OS TCP/IP parameter "keepalive timeout", from 2 hours to 15mins.
↧
Re: GUI very slow in Data Protector 9.06
Hi mainland,
is the Windows firewall active on the cluster nodes? You can try to switch that off to see if it makes a difference. Is the GUI running locallay on the active cluster node or on a client system? You can try if this omnirc parameter makes a difference.
# Enables/Disables DNS Reverse Lookup for xSM OB2IPCREVERSELOOKUP=0
Regards,
Sebastian Koehler
↧
DataProtect split object setting
Hi
i use Data Protector A.06.20 and SAP Brtools backup my oracle database , when i use 4 Drives for backup loading balance but backup session can't use , one drive show running , the other status inactive/waiting , can tell me how to setting for loading balance work , Thanks a lot
↧
↧
Re: Full backup works but not the incrementals
with the backup of the NFS mount, is the volume writeable as well? The DP agent will attempt tree walks as well as try and modify archive bits for incrementals.
as its unix I would set the following in the backup spec "filesystem options"
make sure "display statistical information" is unticked
Tick "Backup POSIX hard links as files"
Tick "Do not preserve access time attributes"
set logging to log files or log dir....
see how that performs. This should stop the tree walk and base the incremental of file modification dates rather than writing file change information.
↧
Re: DataProtect split object setting
Hi YAKEN,
I guess each of your drives has a concurrency of 4 (or higher). This allows 4 data steams to be written in parallel to each device. You could either change the concurrency globally (Devices & Media) or in a specific backup spec. I would recommend doing it per backup spec. Go to Destination and change the Properties of all selected drives accordingly to your needs.
Regards,
Sebastian Koehler
↧
Re: DataProtect split object setting
Hi Sebastian:
yes , each of drives has a concurrency of 4, but can not work , the other spec can use , do you have any idea , thanks
↧