I have an issue with the force free tape in DPM2010.  It wont work!

I am using the script from Msoft blog http://blogs.technet.com/b/dpm/archive/2007/09/04/cli-script-force-mark-tape-as-free.aspx .

There are no visual errors, when I run the script I enter the server name, Tape Drive name and Location - I am even copying the info from DPM to ensure there are no anomalies. ....after a few seconds the cursor just starts blinking at me again... if anyone has any suggestions please let me know, I can provide more info if required - but hopefully I am using the right script, and hopefully I am inserting the right information - unless I dont know what my backup server is called.... :)

All of my backups to tapes are failing and I cannot resolve the issue.

Error: The back up to tape job failed for the following reason: (ID 3311)

View Detailed Errors:

Application: DPM 2012 R2 RU10
Server: Windows 2012 R2

We are running DPM 2010 with an Overland tabe library with 1 drive in it. Today when we checked in to DPM it showed the whole library as being offline and we can not connect to either the drive or slots. However, a new drive apeared as stand alone, but without any slots. This drive is the one inside the library but there havent been any recent changes that should make it appear seperate.

We tried rebooting the DPM server. We turned of the library and rescanned it in DPM, that made the stand alone drive disapear but when we turned it back on the actaul Overland library still said it was offline and when we rescanned that it re-added the stand alone drive. The only alert besides tape backups failing just says the number of drives detected is incorrect and needs to be refreshed.

Let me know if there is any more infromation needed, and thanks in advance for any help.

Hi guys,

Hoping you can help. My boss asked me to find out which tapes can be taken out of DPM - to drum up some kind of report (either via script or other) which shows tapes with an expired retention period? I didn't make much sense to be honest.

I think the scenario is - he is moving a bunch of servers across to a new DPM server and part of this I think is to re-catalogue a bunch of tapes which can be taken from the old DPM system. I think he wants to retain the old data on tapes but wants tapes he can re-use for the new server...

I'm not too clued up on DPM so that's probably why it doesn't make much sense. IF it helps, he made reference to the Veritas back-up system and how if a tape was marked 'blue' it could be overwritten.

If you can make sense of any of the above it would help me - I will likely ask my boss again what he meant but thought I'd some guys here too.

Thanks - Travis

DPM Protection Group shows  that a specific folder on a drive being backed up is being backed up. No alerts that the protection group is failing. (all green) However the specific folder does not show up when trying to restore it. All files and folders are available for restore above it. Any clues to, one, why this may be happening, and two, how to get the files and this folder to be backed up? (not listed in exception rules)

Thanks in advance.

Hi. We are examining options regarding the need for greater tape storage capacity. Our existing backup strategy is barely getting by with the tapes we have now, and we expect the data needing backed up to increase.  In one of our discussions, my manager asked me if we could attach more than 1 tape library to the DPM server and have it use both. We currently have a Dell Powervault TL4000 library connected via iSCSI.

I have not been able to find any reference in the DPM documentation that would answer this question.  Does anyone know if we can have 2 separate drives + libraries configured on the server, and have both be used by DPM as needed?

Thanks!

AndyM

Hello,

I'm managing a two nodes 2012 R2 file server cluster that contains a 16To CSV. I'm using DPM 2012 R2 to backup this entire shared volume directly to LTO 4 tapes, the job last about 55h.

Since SCCM 2012 client has been installed(I don't manage it), the tape jobs are failing ramdomly after several hours with the error:

Type: Tape backup
Status: Failed
Description: The DPM service was unable to communicate with the protection agent on serverX.xxxx.xxx . (ID 52 Details: The semaphore timeout period has expired (0x80070079))
 More information
End time: 19/07/2014 03:11:06
Start time: 18/07/2014 22:00:00
Time elapsed: 05:11:05
Data transferred: 768 289,56 MB
Cluster node serverX.xxxx.xxx
Source details: G:\
Protection group members: 1
 Details
Protection group: File Server Tape Protection
Library: Quantum PX500 Series Medium Changer
Tape Label (Barcode): File Server Tape Protection-00000230 (000043L4)

If I uninstall SCCM 2012 client, no more issue, backups succeed. I've asked our SCCM team, no specific task has been scheduled or deployed in SCCM.

I can't see anything abnormal in logs.

Any idea?

I have Tape Library Sharing configured and recently upgraded from DPM 2007 SP1 to DPM 2010. I have 4 tapes that have been inserted back into the library, and all DPM servers are reporting that these tapes "Belong to a DPM server sharing this library." It seems that none of my DPM servers own the tapes. I tried using the Force-FreeTape.ps1 script but I get the same "tape belongs to another DPM server" error.

I've tried recataloging the tapes and identifying unknown tapes, but nothing works. I don't care about the data on the tapes - how can I take ownership so they can be used by DPM?

Enviroment =  DPM 2012 SP1 on 20012 R2, database SQL 2012

Tape Library = Dell PowerVault TL2000

Tape Drive = IBM ULT3580 -TD4

---------------------------------------------------------------

Issue: The Library automatically cleans the drive when dirty. DPM tape backups that are in progress fail when this occurs.  How do I tell DPM to automatically retry the tape backup after a certain period?

Logging back in manually to restart the failed tape backup is not an effective solution.  I'm not always available to do this and I lose hours of time that could have been spent writing data to tape.

Hello,

Hi, my question is simple. I have four protection groups and want to run them all at the same time, on the same disk. How do I get DPM 2012 to do this? It runs the first job then ask for a new disk... I'm sure there is some simple setting but I can't seem to find it. i searched in the forums prior to posting this but nothing has came up. I'm sure this is a simple question.

Thanks!

Jim V. 

Hello,

I have a situation where I am looking to offload multiple large databases that my organisation no longer uses onto external storage. So far I am looking into offloading onto tapes since it'd be more suitable for archiving purposes for long term storage than hard drives. I've had a look at options for Recovery and I've noticed that there are options that allows you to specify "Copy to Tape" for old recovery points. There doesn't seem to be an option that allows you to choose the tape however hence, I was wondering if there was a way for me to specify that the data I'm offloading be copied to the same tape?

DPM 2012 R2 Rollup 3 on Windows 2012 R2.
12GB RAM.
Pagefile 110GB.
29TB allocated (according to DPM protection group)
37TB used (according to DPM management summary).
42TB Storage total.
91 days protection VMs in Cluster + weekly copy to tape. On wire compression enabled.
5 days protection applications (SQL/Exchange 2010). On wire compression disabled.
Both PG: Sync just before Recovery point. Express full backup once daily.

Hi. I'm having a problem with memory consumption in Windows when DPM is backing up members.

It appears that express full backups consume RAM as each member is backed up but the memory is not released and so, as each member is synch'd, the RAM consumption climbs over the hours until it hits about 95%-97% and system runs very slowly. The job runs on interminably and the whole process effectively stalls (but not actually, just very slow) until I reboot.
I can't free that shared memory back up. Restarting SQL & DPM services makes no difference.
In Task Manager the Processes Memory Usage hits 97% when the issue is worst but the actual processes (DPM & SQL) only display 100-400MB of use. However, the RAMMAP utility shows that the bulk of RAM (maybe 10GB) is consumed under the heading Shareable Memory.

I guess it might be related to the Dynamic Volumes in disk manager or VHD(X)s being mounted or maybe even the network access during synchronisation.

The replication to tape causes no issue and memory usage is low during the copy to tape (which takes maybe 8 hours) and freed up when completed.

I have read about known issues regarding SMB2/3 and VHDXs but I appear to have the approriate KB patches installed already.

FYI: I also have other issues on this server.
The external storage controller is resetting itself periodically but intermittenly failing to recover and so disks disappear from disk manager and backups fail.
Also, after the PGs term of protection expires, old recovery points are not being flushed/cleared/deleted and so I have to do this manually with a custom powershell script (http://social.technet.microsoft.com/Forums/en-US/fa700e75-de4f-4d11-89a7-d80c07406941/dpm-2012-r2-returned-tapes-with-expired-data-not-freeing-up-not-showing-as-expired?forum=dpmtapebackuprecovery#e5bdca77-e758-4542-85cc-0d49281d39a2). Incidentally, running this powershell script - which queries DPM for recovery points older than 91 days and deletes them using DPM powershell commands - also appears to induce the 'hogging and failing to release memory' issue.

Background

DPM 2012 R2 4.2.1594.0

As usual, last week we had a set of tapes reported as being required onsite by DPM (as their useful offsite period has expired). They were shipped into us and we inserted the tapes. Usually the tapes are either no longer marked as Offsite Ready but are not marked expired.

DPM PruneShadowCopies has never appeared to run at all and I have been unable to troubleshoot this properly. Instead, I have three scripts that I customised (based on an original script that I found online but I can't recall where) and each script iterates through one of my three PGs, identifies RPs that are over the allotted period (e.g. 25 days for one group) and then force expire the RPs on tape.

I do this manually every week and it’s been working well for some years now.

Only recently did I notice that one of these scripts was actually expiring disk backups too and much earlier than their allotted limit of 25 days. Before correcting the script I decided to try running PruneShadowCopies again and it now seemed to work but for only one of the PGs (but this was the most significant PG that backs up all VMs). It seemed to expire the tape RPs and lave the correct number of disk based RPs in place .

I now run PruneShadowCopies (8/05/2017 pruneshadowcopiesDpm2010.ps1) first each week, then my two other script to expire the tape backups for the remaining two PGs.

Issue

In this case, the VM PG tapes still had the Offsite Ready green tick despite the expiration date on each RP in the tape being listed as Jul 12th - over a week earlier. I left them for a few days more. They still did not show expired after running pruneshadowcopies2010.

The VM PG that pruneshadowcopies runs against is backing up the bulk of VMs from Hyper-V to disk for 25 days and tape for 13 weeks. There is a single express backup @ 18:00 Everyday.

Right now, these tapes have no status next to them - neither “Expired” nor “Offsite Ready”. The date of each RP on the tape is July 12th and the backup date is April 12th. I cannot "Mark tape as free": "the selected tapes cannot be marked as free becuase they belong to protection groups."

I run “wilson-duetape.ps1”. This script lists the tapes that are offsite and the three tapes in the library that are Offsite Ready. It does not list anything to do with the tapes I'm having an issue with - the output ignores these tapes as if they're not in DPM at all.

I suspect that, if I run my old force script, it will force expire the data on the tapes successfully by rooting out the disk based backups that are responsible (but I don’t want to lose these disk based RPs when I don’t know which ones it will be).

Possibly Related

I have two Hyper-V VMs running Linux in this PG. Since a recent update, these Linux VMs no longer create Recovery Points when up and running. As such, they probably don't meet the 25 days limit specified. I have temporarily deleted their membership in the PG and their inactive disk based backups but still those tapes are not expired.

Question

I realise that DPM is probably doing exactly what it is supposed to and that I am missing something but that is because I don't know how to identify the issue. Can someone assist me in identifying what exactly on the disks or PG is preventing these tapes from expiring in DPM?

Hi all,

the environment is DPM2012 on win 2008r2. all the short term protection working perfectly and once the long term protection configured with tape library, the error comes saying "The operation failed because of a protection agent failure. (ID 998 Details: Access is denied (0x80070005)) ".

also note the tapes in the library are imported from a dpm 2010 library which was we had earlier.

any body gone through a problem like this?

Thanks

Hi,

I have a customer who has DPM 2012 R2 version 4.2.1589.0, we attached a Tape library of type: Hewlett Packard  MSL G3 Series Library (x64) and I can see its driver (version of the driver is 2013) in DPM device manager.

However when I try to inventory the tape in DPM I receive below errors:

Affected area:     (CYW256L6)
Occurred since:    2/18/2017 6:15:37 PM
Description:    The detailed inventory of tape failed for the following reason: (ID 3316)
    DPM could not reserve the drive resource because one of required drive resources is not online or it needs cleaning or servicing. (ID 24052)
    More information
Recommended action:    1) View and resolve all alerts for the library and its drive(s). Retry the operation after resolving the alerts.
2) Go to the Library tab in the Management pane and click on Rescan. Once the rescan operation has completed successfully, retry the operation.
    Run tape inventory job again...
Resolution:    To dismiss the alert, click below
    Inactivate

I also receive the below warning:

Affected area:    Hewlett Packard MSL G3 Series library  (x64 based)
Occurred since:    2/18/2017 6:15:36 PM
Description:    Tape Library Hewlett Packard MSL G3 Series library  (x64 based) is not functioning efficiently for the following reason:
 The drive information for Tape Library Hewlett Packard MSL G3 Series library  (x64 based) is incorrect and needs to be refreshed. (ID 3302)
    More information
Recommended action:    
    The number of drives reported for this library is different from the number of drives detected by DPM. If any drives are out of the library for servicing, ignore this alert; otherwise, remap the drives within the library manually. For information about remapping tape library drives, see the System Center 2012 R2 DPM Operations Guide.
Resolution:    To dismiss the alert, click below
    Inactivate

What could be possibly the issue?

We are using DPM 2010 on Windows 2008 R2 backing up various virtual machines at the machine level and physical machines at the volume level. We were trying to restore a VHD, this worked when the backup was on disk but once it was from tape we got "error id 1001".  We tried various restore locations and it always gave the same error.  Has anyone else came across this problem or know why this is happening.

Cary Wells

Hi,

after upgrading to DPM 2016 CU4 from DPM 2012 R2 we are experiencing issues with tape backups to AWS Storage Gateway. Tape backups for both virtual machines as well as SQL databases fail with the following error:

The DPM service terminated unexpectedly during completion of the job. The termination may have been caused by a system reboot (ID 910)

Obviously there is no reboot to DPM, Hyper-V host, SQL Server or anything during this time. We are failing constantly all scheduled tape backup jobs - some of the machine or smaller databases get to be successfully backed up, but it seems the backups fail after 15 minutes from the job start.

We are running DPM 2016 CU4, local instance of SQL 2012 SP4 and all of that on Windows 2012 R2. Before moving to DPM2016 we have been successfully running all backups using DPM 2012 R2 on the very same machine.

Please help me find the root cause, as my off-site backups are no longer completing :(

Thanks, Marcin

Marcin Dobija | MCSE:Security | MCITP:DBA,EA,EMA,SA | VCP5 | RHCSA

Hi

Because of the risky upgrade scenario from DPM 2007 to DPM 2010 we decided to set up a new DPM 2010 server and move one protected server after the other to the new DPM server.

The library will be moved to the new server as well.

Question: We will lose the whole tape catalog information, including which backup is on which tape. Is there a way to migrate or export that data to the new DPM server?

Thanks

Philippe