Quantcast
Viewing all 10494 articles
Browse latest View live

Re: Media object copy from command line; how?


André Beck wrote:


Ignoring the manpage for a second, I had a look into omniobjcopy -help and most of what I know from the GUI seems to be reflected there one way or another.


Hi André,

you have to be very careful with the information provided by omniobjcopy -help. The synopsis is not as accurate as it seems. I had this once where I had to replicate all objects included in a session from a StoreOnce Catalyst store to another. If you know the source objects the command will work ok.

Please use the Accept Solution button next to my post and assign a KUDO (thumbs up icon) if this works for you.

Regards,
Sebastian Koehler


Re: Move the Data Protector \ProgramData\OmniBack directory


tsambran wrote:

 I was wondering if there is a possibility of moving the C:\ProgramData\OmniBack directory to a new drive (i.e. E:) after the installation has been in use. I have noticed there are references to that directory in the registry and was hoping there was a documented procedure for this. (v9.08 is installed on a Windows 2012 server)


When I last investigated, the only supported (by rumour) option to do so was through reinstallation followed by a readdb of the previously dumped IDB, manual move of DCBFs and the Config tree. But that was for 6.11 (see this thread where I migrate to a new OS, 64bit platform and target drive letter in a single step). I assume something like that can be done with 9.08 as well, but it requires testing. I'm rather confident the IDB restore (readdb) and DCBF move will work out, but 8.x+ has new components compared to prior versions and just dropping a copy of Config into the right place might not suffice.

BTW, I prefer to install (both code and data) to O:\OmniBack which is a dedicated volume. Way more flexible in the future than splicing it into the system drive and no hassle with that pesky ProgramData tree, either.

HTH,
Andre.

Advanced Scheduler Randomly Stops Scheduling DP 9.06

Hi all,

We have a problem whereby the jobs we have using the advanced scheduler stop scheduling themselves for "no" reason, randomly, across multiple environments using seperate cell managers. We came in on Monday and found that on Friday lunchtime the advanced scheduler stopped running any jobs so missed out on a whole weekend of backups! Luckily we only use the advanced scheduler for SQL backups, not for files and other databases, so not all backups were missed, but enough to annoy us and users!

The only way we can work out if the scheduler stops is the jobs just stop getting scheduled. We don't see any error messages, or anything useful, just jobs stop scheduling.

We are going to raise a case with HP but thought we'd ask here also, anyone any ideas as to why this can happen and how we can avoid it? Given the advanced scheduler is completely integrated to the database with no files or anything to look at we can't find errors which could help us work out why it stops. Only way to get it going for us is to stop/start DP services, i.e. omnirc -stop & start, we are running LINUX by the way.

Thanks in advance for any help or advice.

Cheers,

Andy

Re: LT0-6 drive Block size modification in cell manager 9.08


dpx wrote:

OB2LIMITBLKSIZE=0 - i couldn't find in .omnirc file this parameter but i can add and try.


You absolutely have to. For historic reasons (more related to shortcomings of HPUX and Solaris than Linux) the Unix MA refuses to use block sizes >256KiB by default, and this exact tunable will override that nonsense.

BTW, you don't have to reformat any tapes, that will happen automatically. The first time you write at the new (larger) block size, appending to not-yet-full tapes written at the old BS will fail with a warning, the tape will be ejected and DP continues on the next expired/unprotected/blank medium from the pool as it always does. And given it has to relabel that tape anyway, it uses the opportunity to relabel it at the new block size. All you see is a warning that this happened, so you can keep track of the progress. The new tape is then marked with a better write order number as well, so it will be used for writing from now on instead of the inappendable ones. The latter will bubble up in protection until they expire, and finally will be relabeled to the new BS as well when their time comes to be overwritten. The entire migration happens without any need to manually interfer - at least it did in countless migrations I performed from 64KiB to 256KiB and later 256KiB to 512KiB, on real tape and VTLs alike.

HTH,
Andre.

Re: DP701 slow VM restore


mula_e wrote:
We are experiencing simillar performance issues, having very fast backups and slow restores using StoreOnce over SAN.

What StoreOnce? With Catalyst to SO Appliances, rehydration performance is starting to get mostly reasonable with 24 spindles. SO Software on the same number of spindles is significantly slower to rehydrate (even though other factors like RAM are taken into consideration, my particular comparison box has 60GiB of it, while a typical SO Appliance has 72GiB, and CPUs are similarly beefy). SO rehydration may scale up throughput-wise if multiple objects are running in parallel (as single CPU-core throughput of the RMA is also a factor), but sadly that isnt the case when restoring a VM. Single-object rehydration can be as slow as 25MB/s. It's in the nature of the thing, sadly - rehydration is a fully random access pattern to the SO backend disks, at 4KiB blocks worst case (an SO store uses a lot of small files to stuff away individual blocks). So with less than 24 spindles, things will only deteriorate from here. I've seen SO Software on 6-disk RAIDs and they don't perform well on reading. So it all depends - what's your exact SO backend specs?

HTH,
Andre.

Re: Media object copy from command line; how?

I'd be happy if someone does read what I wrote, or ask what he doesn't understand before giving wrong or not helpful answers: I know the name of a medium (label) with protected objects on it. I wnat to copy all protected objects from that medium to another medium, recycling the objects on the first medium, keeping the protection on the target medium. This is what intercative "media object copy" can do. Please read again before answering: "media object copy"

I started an interactive session from the GUI that does exactly what I want, the I observed the process lists on the cell manager and the media agnet hosts: I could not see the command that DP uses to do the media object copy; specifically not the "omniobjcopy"

The media agent process (bma) uses options like "-sessiontype 16" and "-operation 21". No idea what that means.

 

My script produces debug output like the following:

[2] reading output of '/opt/omni/bin/omnirpt -report media_list -class File -protection 6666 -tab'
[2] matching medium Jukebox2_11 File_Jukebox2_SAP 130629.00 MB -111.2 21.8
[1] 1 of 1 media selected
[1] matching media list: Jukebox2_11
[1] checking medium labeled Jukebox2_11
[2] reading output of '/opt/omni/bin/omnimm -list_media Jukebox2_11 -detail'
[2] 8 of 1076 objects protected (0%)
[2] 21807249 of 147481644 kB protected (14%)
[1] reorganize Jukebox2_11 140 0.15 0.01
[1] 1 of 1 media to reorganize: Jukebox2_11

So all I know is that the medium named Jukebox2_11 is to be reorganized (copyied, then recycled). Of course "media copy" is not what I want!

LTO4 load balancing

Hi all,

we use two LTO4 libraries (16 slots, 1 drive).  The libraries are used as the second destination device in backup specification (he first one is virtual file tape library). But the main usage of these libraries is in copy jobs. We have two copy jobs (Copy - Object copy - Automated - Scheduled; the jobs aren't run concurrently). The both libraries are used as destination devices in the jobs. Is it possible to setup load balancing for the LTO4 libraries in the copy jobs? How to find out usage of the libraries (number of used LTO4 tapes per library)?

We use HP DP 9.07

Thanks a lot

Data Protector 7 Ein Store stoppt automatisch

Hallo zusammen,

Eingesetze Umgebung:

Winows Serve 2008 R2 Enterprise (2x Six Core Cpu, 64 GB RAM) mit eingebundenem Festplattenpool als D: (Grösse ca. 20 TB).

HP Data Protector A7 .00

Problem:

Wir hatten nun das zweite Mal innert rund drei Monaten, wo sich ein Store (von insgesamt 3 Stores) von selbst zwischen zwei geplanen Backups gestoppt hat. Ich konnte keinen genaueren Grund in den Log von Data Protector oder Windos Logs für das Stoppen feststellen.

Beim ersten Ausfall konnte der Store wieder gestartet werden als eine Dirty Datei unter dem betroffenen Store entfernt wurde.

Beim zweiten Ausfall reichte es aus, den Dienst C:\Program Files\OmniBack\bin\StoreOnceSoftware.exe neuzustarten und danach musste aber der betroffene Store auch nochmal gestartet werden (Keine Dirty Datei)

 

Seither (ca. 10 Tagen) laufen die Backups ohne Probleme durch.

Hat jemand eine Idee wieso ein Store von selbst stoppt?  Gibt es irgendeine Einstellung für das? Konfigurationsdateien?

 

Freundliche Grüsse

mro

===================Google Translate====================================

Hello everybody,

Location:

Winows Serve 2008 R2 Enterprise (2x Six Core Cpu, 64 GB RAM) with integrated hard disk pool as D: (size approx. 20 TB).

HP Data Protector A7 .00

Problem:

We had now the second time within approximately three months, where a store (of a total of 3 stores) stopped by itself between two planned backups. I could not find a more detailed reason in the log of Data Protector or Windos logs for stopping.

On the first failure, the store could be restarted as a dirty file was removed under the affected store.

On the second failure, it was sufficient to restart the C: \ Program Files \ OmniBack \ bin \ StoreOnceSoftware.exe service and then restart the affected store (no dirty file)

 

Since then (about 10 days) the backups run through without problems.

Does anyone have an idea why a store stops by itself? Is there any setting for that? Configuration files?

 

With kind regards

mro


Re: Data Protector 7 Ein Store stoppt automatisch

Hi,

First I want to mention that the s.dirty file has a function. It is there to indicate that the store was not cleanly stopped and it will trigger additional checks at the next startup and possibly repair existing corruptions. That said, if everything seems to be running fine for now, the only additional thing you could do is:

  • Stop the stores
  • Stop the StoreOnceSoftware service
  • Start the service with traces enabled (sc start StoreOnceSoftware --log_level=tracing)
  • Collect the generated logs after a while and send them to support for verification

I hope that helps.

Koen

Re: Hyper V backup failes with 'internal error'

Hello Friend,
What was the resolution of the case, I have the same problem

Re: Data Protector 7 Ein Store stoppt automatisch

Hi Koen;

Thanks for you impact.

I habe now activted the StoreOnceSoftware with Log-Level = tracing.

After a while I will send them to support for verification.

 

Have you or somebody another idea?

 

Kind regards

mro

Activation of Data Prootector

I have installed HP Data Protector 7.0, I would like to know if I can use a key of an upper version to activate ,For example the key of version 9.0

 

Unknow SSL protocol

I have finished installing HP data Protector 9.0 in GUI but I have this error message

'' unknown ssl protocol error in connection to port srv-nas: 7116

Please can you help me

 

Re: Activation of Data Prootector

Hi team,

Please help me , i want to know if i can use the generate key of Data Protector 9.0 on Data protector 7.0.

Re: Unknow SSL protocol

Hi team,

 

This is the message i got

''Unknown SSL protocol error in connection to srv_nas.gimac-afr.org:7116 "" and aslo the  see attched the certificate i got

 


Re: Activation of Data Prootector

The Licensing format changed at version 8 I believe,

So in short no you cannot license a version 7 install with a version 9 license (If I have read your post correctly)

You would need an active SAID and to regenerate your license via the portal. 

Re: Activation of Data Prootector

I want to know if licence version 9 can be installed on version 7

Re: Activation of Data Prootector

No you cannot license a version 7 install with a version 9 license, as I mentioned above.

Re: Activation of Data Prootector

Tape Drive is yellowed

Hello guys,

Today I found our dataprotector backup jobs failed last 2 days. After some digging, I found that one of the tape drive is yellowed. But cartridges are fine and had enough space to store backups. 

Have you ever faced with this kind of problem? 

Please see attached error screenshot

Thanks

Viewing all 10494 articles
Browse latest View live


<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>