Tape backup speed - Erratic

Hi,
We have a 570 backing up a 3TB 4.7 Unicode database (save while active) to a 3592 tape drive. Some weeks the backup takes 12 hours. Other weeks the backup runs 24 hours or anywhere in between. We have not seen any appreciable difference in the system load at the time the backups are running. Has anyone else seen anything like this? It's really puzzling.
Thanks,
Craig

Could you show the exact command with all its parameters that you used?
Are you using internal or external disk subsystem? If external, do you dedicate entire ranks to i570? If not, you need to check if the shared ranks are being accessed with high IOs from other machine/LPAR while you do the save on i570.
Do you notice that the same set of tape cartridges are used when the save time exceeds 12 hours? This can indicate deteriorating tape media quality. To display tape error statictics:
- STRSST and provide correct service tool user/password
- Select 1 Start a service tool
- Select 1 Product activity log
- Select 4 Work with removable media lifetime statistics
- Specify proper "Removable media" (most likely = 2)
- Leave "Sort By" as 4 (All Error Rate)
You should then see a list of tape error rate with the Volume ID displayed.
Next time when the job running the save command runs more than 12 hours, please also try using WRKJOB command to look at its job log to see if there is any suspicious messages that can provide some more clues.
Satid S.
IBM Thailand

Similar Messages

Recomended TAPE backup for OS X Sever

Hi,
I would like to inquire regarding back-up device for OS X Server, My client requirement is a TAPE backup, do you suggest any brand of tape backup and software that works well with OSX server?
Thanks a lot,
val

You need to clarify your needs before you go much further.
There are many tape technologies, including things like DLT, LTO and AIT (amongst others). Each has their own pros and cons, including per-tape capacity, speed, etc.
Knowing the volume of data to backup and how frequently you want to backup is going to be critical in deciding what tape format to take.
For example if you have 1TB of data to backup you don't want to use a tape format that has 20GB per tape - you'll need 50 tapes for a full backup.
The volume of data and frequency also leads to speed. If you need to backup 1GB of data in an hour your options are dramatically different than if you need to backup 1TB of data in an hour. Some tape formats can keep up with the higher throughput, others can't.
Then, of course, there's budget. The larger, faster drives tend to cost more, but they're more capable. But you don't need to buy capabilities that you don't need.
So, you need to answer those questions in order to find the tape technology you want to use. Once you know the technology you can start looking at vendors.

Performance issue in tape backup

Friends,
Our tape backup is implemented with help of Veritas Netbackup,for one of our servers,tape backup is taking too much long time. I generated AWR report during the ongoing backup and noticed the below stats:-
Top 5 Timed Events
Event
Waits
Time(s)
Avg Wait(ms)
% Total Call Time
Wait Class
Backup: sbtwrite2
36,186
300,497
8,100
1,442.5
Administrative
CPU time
17,920
85.4
Backup: sbtbackup
38
1,892
48,000
9.0
Administrative
log file parallel write
112,632
1,378
12
6.6
System I/O
control file parallel write
57,983
1,075
19
5.1
System I/O
What will be the next troubleshooting steps ?
Any guidance will be highly appreciated.
Thanks in advance.

The tape drive is likely the constraint. It cannot write fast enough to handle 4 channels streaming datafiles to it.
I would guess that the datafiles are also going over the network to the media manager. You need to check the network throughput for that link as well.
Hemant K Chitale

How to include another partition in tape backup

Hi
I use a shell script written by my sysadmin to take backup of our oracle system on tape. now i moved some datafiles on another partition and hence want to include that partition also in tape backup.
for fs in $(cat /tmp/dailybkupfs | awk '{print $1}')
do
# For / replace the / with root for log filename
if [[ "$fs" = "/" ]]
then
fn="root"
else
# replace any / in fs name with _ for use in log filename
fn=$(echo $fs | sed 's:/::g')
fi
if umount ${fs}
then
echo "${fs} unmounted successfully" >>/tmp/logs/summary
else
echo "${fs} unmount failed" >>/tmp/logs/summary
fi
if ufsdump 0uf prddb:/dev/rmt/0n ${fs} >${log}.${fn}.${dt} 2>&1
then
echo "Full Backup of ${fs} completed successfully" >> /tmp/logs/summary
else
echo "Full Backup of ${fs} failed" >> /tmp/logs/summary
fi
if mount ${fs}
then
echo "${fs} mounted successfully" >>/tmp/logs/summary
else
echo "${fs} mount failed" >>/tmp/logs/summary
fi
done
echo "Full Backup of all DB file systems Complete" >> /tmp/logs/summary
Also the dailybkupfs file has the following contents->
/u01
/u02
/u03
/u04
/u05
/u06
Now to include /u07 partition for backup, I just need to include the /u07 partition in file dailybkupfs.right?
Thanks
Anmol

I am concerned that the config file resides in /tmp.
That filesystem normally gets cleared if the box reboots.
So as far as I can tell, if your box is rebooted, the backups stop working..

Persistent setting for rman in tape backup

Hi,
When your are going to take tape backup using RMAN utility, how can u control persistent settings with RMAN(show all). I meant how to set retention policy for datafile,controfile backup and archivelog as well.
And if i choose to take incremental backup how these things are changed?? Please help me on these issue as i have some confusion with this.
Regards,

Hi,
Can you please read following link: http://oit.scps.nyu.edu/~pallerm/11g/les_04_config_bu.pdf
Thank you

How to manually run a Yearly Tape backup of a Volume in DPM?

I can run a manual tape backup by right clicking on the volume and choosing Long term backup, but it only runs the monthly tape backup job.
How can I manually run a Yearly Tape backup job?
I am using DPM 2010.
There must be a way to create a job that uses Yearly backup settings.
Or an alternative is to change the expiry date of the Backup if that is possible.
Thanks,

Although you cannot pick the recovery point date in the past to make a new recovery point from, you can run the DPM Power shell script in this
post to select a recovery goal to make a current backup for. The recovery point date would be that of the latest disk based recovery point for any goal selected.
Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT]
This posting is provided "AS IS" with no warranties, and confers no rights.

Restore controlfile from tape backup without autobackup setting on

I am testing a scenraio of restoring control file if we loose all control files.
My RMAN settings for CONTROLFILE AUTOBACKUP is OFF. But, I have taken a backup of controlfile using 'BACKUP CURRENT CONTROL FILE' .
When I tried to restore control file. I am receiving following error:
RMAN> run {
2> allocate channel c101 type sbt_tape;
3> restore controlfile ;
4> }
allocated channel: c101
channel c101: sid=153 devtype=SBT_TAPE
channel c101: VERITAS NetBackup for Oracle - Release 6.0 (2006110304)
Starting restore at 24-MAY-2007 13:44:59
using channel ORA_DISK_1
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of restore command at 05/24/2007 13:44:59
RMAN-06563: control file or SPFILE must be restored using FROM AUTOBACKUP
My question is, How can I restore control file from tape backup without CONTROLFILE AUTOBACKUP setting ON.
It sounds like, There is no way we can restore control file from tape backup if you've not set the CONTROLFILE AUTOBACKUP is ON.
Please advice

When all controlfiles are lost and a recovery catalog is NOT used, RMAN has no information , where to find the controlfile backup. It's not enough to specify the channel, such a channel could hold many different backups from many different databases. You have tell , which database and which backupset. Modify your run block:
SET DBID <DBID of the database,for which you want to restore the controlfile>;
RESTORE CONTROLFILE FROM <name_of_backupiece_which_contains_the_controlfile backup>';
Werner

Restore ( duplicate) db from tape backup from one server to another server

I am looking for step by step process to restore (duplicate)db from tape backup from one server to another server
Thanks

Creating a Duplicate Database on a Remote Host will provide all the information.

Restore controlfile from leagato tape backup

I am using legato tool to perform rman tape backup.
I use the following script on legato to get the controlfile backup.
"backup current controlfile format '%d_%s_%p.ctrl'; "
If I lost the controlfile how do i restore controlfile from the tape. Assume that the DB in nomount stage.

Hi,
backing up autobackup controlfiles to legato is possible. All you have to do is to configure autobackup accordingly. By default autobackup controlfiles will get written to disk.
However i strongly recommend to use a rman backup catalog. This will save you a lot of time consuming trouble if you need to restore your database.
The rman catalog database itself does not need any license and is rather small (5 GB on disk, 512 MB memory).
Restoring the control file will be something like this:
set controlfile autobackup format for device type sbt to '%F';
restore controlfile from autobackup;
Ronny Egner
My Blog: http://blog.ronnyegner-consulting.de

Restore from Tape Backup

Hi All,
Oracle Ver. 10.2.0.4 on Sun-Solaris 9.
Need to do a restore and recover on the DR site using the tape backups from production site.
The path and file names are same in both sites.
In the DR site only the Oracle server software is installed. There are no oracle files/database.
Will a incomplete recovery solve the issue?
What will be the steps in RMAN.?

Hi,
If you know the location of the controlfile autobackup than
restore the spfile and controlfile first.
Than mount the controlfile and restore and recover the database to a PIT, sequence or SCN.
You can find a lot of examples of commands in this forum.
Regards,
Tycho

Disk/Tape backup?

Hi, All:
We have a new system installed Oracle 8.1.6. Now we are considering the method of Backup/Recovery.
In our system 2 disks and 1 tape device can be used for backup. And I think there are 2 ways to use them:
1). 1 disk : archive log;
1 disk : backup using by RMAN;
1 tape : backup's backup
2). 1 disk : archive log;
1 disk : the copy of the archive log;
1 tape : backup using by RMAN;
I don't know which is better? which is common?(The data size is about 20GB)
Can anybody give me some advices?
Thanks.
null

All you need to do is take the backup file from FRA to tape.
HTH--

Recovery window for tape backups

Hope someone can answer to my questions.
With RMAN you can maintain the number of tape backups that are needed for recovery purpose through the “delete obsolete recovery window” command. E.g. the command below ensures tape backups be kept for recovery within 31 days:
delete noprompt obsolete recovery window of 31 days device type 'SBT_TAPE';
Q:
1) If I use a disk area as pseudo tape area in my backups (through ORACLE.DISKSBT library parameter), I guess the above "delete noprompt obsolete recovery window" command would work on the pseudo tape area as well?
2) At my site the pseudo tape area is backed up to tape nightly through the system backup job, and the tape backups are kept for about 1 month by the tape system. To save space in the pseudo tape area on disk can I just set a small recovery window in the above DELETE command (say 1 day) , and relies on the actual tape system retention policy for oracle backups? How would this affect oracle’s ability to do recovery?
Thanks

1) I would never use a 'faked' tape as a permanent solution - only to test tape backup scripts - , as it slows down the backup/recovery process. From RMAN's point of view a pseudo tape works like a real tape, so all valid commands apply.
2) When you delete a backup, RMAN forgets it. RMAN is not aware of still available backups in the tape system, you have to recatalog the backups. That means overhead in manual maintenance and a source for possible errors.
Werner

Spfile restore from RMAN Tape backup to different server

Hi All,
First of all let me thank you to the forum from which I keep updating my knowledge on oracle.
I am having issues when trying to restore spfile from controlfile autobackup to different server. Problem and other details are below:
Target version : 10.2.0.2
Netbackup : 6.5.6
Taken the tape backup through rman on server ABC and trying to restore the spfile from controlfile autobackup to server XYZ. Tape backup has been taken today morning and trying restore now.
Both servers ABC and XYX have been added (as clients) to netbackup policy.
Below is the syntax I used for restoring spfile and getting error as mentioned.
RMAN> run {
2> set controlfile autobackup format for device type 'SBT_TAPE' to '%F';
3> allocate channel ch00 device type 'sbt_tape'
4> parms='ENV=(NB_ORA_SERV=Mediaserver,NB_ORA_CLIENT=XYZ)';
5> restore spfile to pfile '/oracle/product/10g/dbs/inittest.ora' from
6> autobackup maxseq 5 maxdays 3;
7> }
executing command: SET CONTROLFILE AUTOBACKUP FORMAT
allocated channel: ch00
channel ch00: sid=36 devtype=SBT_TAPE
channel ch00: Veritas NetBackup for Oracle - Release 6.5 (2010042405)
Starting restore at 03-AUG-2011
channel ch00: looking for autobackup on day: 20110803
channel ch00: looking for autobackup on day: 20110802
channel ch00: looking for autobackup on day: 20110801
channel ch00: no autobackup in 3 days found
released channel: ch00
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of restore command at 08/03/2011 14:03:26
RMAN-06172: no autobackup found or specified handle is not a valid copy or piece
Not sure where I am doing wrong. The same can be restored onto same host (I.E ABC) without any issues .
Request your help and guidance.
Thanks,
SSS

I don't know whether you can restore to a different server. The easier way should be
RMAN> restore spfile to pfile 'YOUR_PFILE_NAME' from autobackup;
then copy this to the new host.
Check this link
restore spfile to another host
Hope this helps

Cloning DB on a different server from RMAN Tape backup

RDBMS version           : 11.2.0.2 on Solaris 10
Media Management Layer : Netbackup v7
I have done only backups to Tape. I have never done restore,recovery from Tape.
We have a production DB named NEHPROD running on server qualmh214. Every Monday morning an RMAN Full backup for NEHPROD is taken to Tape.
Production server : qualmh214        DB Name:NEHPROD
run
ALLOCATE CHANNEL c1 TYPE 'SBT_TAPE';
SEND 'NB_ORA_CLIENT=qualmh214,NB_ORA_POLICY=qualmh214_rman,NB_ORA_SERV=qualmh238,NB_ORA_SCHED=Daily';
backup database   FORMAT 'NEHPROD_%d_%U_%t.rbk' tag='NEHPROD_full';
}I need to create a Test DB in a different server (qualmh199) by restoring directly from NEHPROD's Full backup in tape .
From qualmh199, how can I connect to the Tape Media server ?
Even If I do connect to the Tape media server, how can i get access to tape backups of another DB? In the RMAN run block you can't specify a DB name. Right?
Even if I get access to another DB's backup , How can i specify which Date's Full backup I need to use to restore ?

From qualmh199, how can I connect to the Tape Media server ?You have to allocate channels to tape & need to mention NB parameters as example below.
in RMAN script
ALLOCATE CHANNEL CH1 DEVICE TYPE SBT_TAPE PARMS 'SBT_LIBRARY=/usr/openv/netbackup/bin/libobk.so64';
SEND 'NB_ORA_SERV=<server.com>, NB_ORA_CLIENT=<target.server>';
Even If I do connect to the Tape media server, how can i get access to tape backups of another DB? In the RMAN run block you can't specify a DB name. Right?If the libraries are installed and you able to perform, you can go for either duplicate database or simple RMAN restore. You can mention other DB names also
Ex: duplicate target database to 'test';
Even if I get access to another DB's backup , How can i specify which Date's Full backup I need to use to restore ?You can mention TAG's if you are using RMAN restore.

DPM 2012 R2 crashes during tape backup for a specific server

Hi Community,
A customer of ours is experiencing an issue with DPM crashing after a period of time when backing up a large protection group to tape.
Customer is running DPM 2012 R2 URE 5
Upon looking in the DPM Logs, we see the following error, immediately prior to the system crashing.
MSDPMCurr.errlog
0968 29A8 04/08 12:41:48.218 28 MediaPool.cs(350) NORMAL Found PGSetId a914779a-8847-40e5-90fd-a1ca6cffee05 for MediaPoolId e10c1280-445f-4454-8edd-0cb3c1113c27
0968 29A8 04/08 12:41:48.221 28 MediaPool.cs(329) WARNING MediaPool 00000000-0000-0000-0000-000000000000 does not exist.
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING Caught unhandled exception : Microsoft.Internal.EnterpriseStorage.Dls.Utils.MediaPoolNotFoundException: Media pool 00000000-0000-0000-0000-000000000000 not found in the database.
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMDatabase.MediaPool.GetPGSetIdForMediaPool(DbContext sqlContext, Guid mediaPoolId)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMDatabase.MediaPool.IsMediaPoolCollocated(DbContext sqlContext, Guid mediaPoolId)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMDatabase.ArchiveMedia.IsMediaCollocated(DbContext sqlContext)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMDatabase.ArchiveMedia.CheckAndRaiseOffsiteReadyEvent(DbContext sqlContext, Boolean considerOfflineMedia)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMDatabase.Dataset.RaiseMediaOffsiteReadyEvent(DbContext sqlContext)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMDatabase.Dataset.SetLifeStatus(DbContext sqlContext, DatasetLifeStatus lifeStatusPassed)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMDatabase.TapeArchiveMedia.Reclaim(DbContext sqlContext)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMDatabase.MMCatalog.ReclaimMedia(ArchiveMedia media)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMInterface.ReserveResourcesForBackupHelper.GetRecycedUsableMedia(IList listOfMedia)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMInterface.ReserveResourcesForBackupHelper.GetRecyclableMedia(MediaPool mediaPool)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMInterface.ReserveResourcesForBackupHelper.GetMediaRequiredByArchiveTask(Guid expectedDriveId)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMInterface.ReserveResourcesForBackupHelper.AllocateMedia(Guid expectedDriveId)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMInterface.ReserveResourcesForBackupHelper.QueryResourcePreference()
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMDatabase.TaskElement.ReserveResources(Dictionary`2 libraryResourcesCache)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMDatabase.JobElement.ReserveResourcesToOneTask(Dictionary`2 libraryResourcesCache)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMDatabase.PriorityQueue.ServeTasksInQueue(Message msg, Boolean waitForResources)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.MMCommonBlocks.ReserveResourcesBlock.DoServeTasksInQueueAndPostCheckMessage(Object msgObject)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at Microsoft.Internal.EnterpriseStorage.Dls.EngineUICommon.DpmThreadPool.Function(Object state)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at System.Threading.ExecutionContext.RunInternal(ExecutionContext executionContext, ContextCallback callback, Object state, Boolean preserveSyncCtx)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at System.Threading.ExecutionContext.Run(ExecutionContext executionContext, ContextCallback callback, Object state, Boolean preserveSyncCtx)
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at System.Threading.QueueUserWorkItemCallback.System.Threading.IThreadPoolWorkItem.ExecuteWorkItem()
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING at System.Threading.ThreadPoolWorkQueue.Dispatch()
0968 29A8 04/08 12:41:48.236 68 DpmThreadPool.cs(121) WARNING *** Mojito error was: MediaPoolNotFound; 0; None
0968 29A8 04/08 12:41:48.242 68 DpmThreadPool.cs(149) WARNING Attempting to write error to the event log Media pool 00000000-0000-0000-0000-000000000000 not found in the database.
0968 29A8 04/08 12:41:48.246 02 EventManager.cs(143) NORMAL Logging event to NT log from DpmThreadPool.cs(163): FatalServiceError
0968 29A8 04/08 12:41:48.267 68 DpmThreadPool.cs(170) WARNING Attempting to invoke Watson
0968 29A8 04/08 12:41:48.280 09 everettexception.cpp(761) CRITICAL Exception Message = Media pool 00000000-0000-0000-0000-000000000000 not found in the database. of type Microsoft.Internal.EnterpriseStorage.Dls.Utils.MediaPoolNotFoundException, process will terminate after generating dump
The server is part of a protection group that is co-located, I am trying the backup not as part of a co-located set to see if this makes any difference.
Tape backups for other servers, which are part of the same co-location group, work fine and it seems to be just this one causing the issue.
I have seen a few other posts on the net referencing "WARNING MediaPool 00000000-0000-0000-0000-000000000000 does not exist." where people have similar issue with DPM crashing out but so far nobody seems to have any answers.
I am debating removing the protection group and trying to 'start from scratch' but wondered if anyone else had any thoughts ?
Thanks in advance,
Ed

Hi,
Open SQL management studio and connect to the instance used by DPM.
Run the following Query against the DPMDB and see if it returns any rows, if so please provide the output.
select MP.mediapoolid, MP.PGSETID from dbo.tbl_MM_MediaPool MP
Join dbo.tbl_IM_ProtectedGroup PG on PG.ProtectedGroupId = MP.MediaPoolId
where PG.MediapoolId is NULL
and MP.PGSetId is not NULL
Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT] This
posting is provided "AS IS" with no warranties, and confers no rights.

Tape backup speed - Erratic

Similar Messages

Maybe you are looking for