Backup Copy Job to HPE StoreOnce - CIFS or NFS Repository

[redZonk]

Hello all

I have a situation, where my customer would like to have 90 restore points with a backup copy jo. He uses a HPE StoreOnce as a backup target.
The StoreOnce 3640 he has, does only allow 14 restore point in one chain when running the backup against the Catalyst store. Since the copy job is doing forever forward incremental without any periodic full, which would re-start the backup chain, we can only configure the backup copy jobs with 14 restore points. Now we are looking for other options.
During a workshop we discuessed the following.
- 1 backup copy job with 14 restore point to the catalyst store of the StoreOnce.
- 1 backup copy job with 90 restore points running towards a CIFS or NFS repository from the same StoreOnce.

Has anybody experience with using a StoreOnce with NFS/CIFS as a Veeam backup repository?

I did some tests in our lab with a StoreOnce VSA and it seams to work to use the NAS implementation of the StoreOnce. But I sometimes get the following error with the mirroring copy job. The only why to fix that is an active full.

Code: Select all

28.05.2021 07:06:46 :: Processing srv01 Error: The request could not be performed because of an I/O device error.
Asynchronous request operation has failed. [readsize = 4096] [offset = 0]
Failed to open storage for read access. Storage: [srv01.vm-4853D2021-05-27T220229_E782.vib].
Failed to download disk 'srv01-flat.vmdk'.
Reconnectable protocol device was closed.
Failed to upload disk. Skipped arguments: [srv01-flat.vmdk];
Agent failed to process method {DataTransfer.SyncDisk}.

Question is, if anybody has already implemented this scenario with backup copy job going to a NFS or CIFS share on a StoreOnce with lots of restore points in real life?

Thanks and regards

Patrick

[foggy]

Hi Patrick, I wouldn't go with CIFS on StoreOnce. To overcome the chain length limit, you can enable GFS and the 'Read from source...' check box to perform periodic fulls, similar to the regular backup jobs.

[redZonk]

Hi Foggy, thanks a lot for your suggestion. Is this the setting you would recommend?

https://ibb.co/SrnLqs6

It tells me that I can't configure more than 7 incremental backup between full backups. Thats with the StoreOnce VSA.
It generates the same error message like if I don't tick "read the entire restore...."

Thanks and regards
Patrick

[foggy]

Seems you're seeing an issue similar to this one. Have you already installed the latest cumulative patch?

[redZonk]

Yes, it seems to be the same issue as in the other post. The latest cumulative patch is not installed yet. I will update tomorrow and provide a feedback. Thanks

[redZonk]

I installed the latest cumulative patch from the link foggy mentioned. Now I was able to configure 90 restore points in a backup copy job towards a catalyst repository. I ticked the option "read the entire restore point from source backup instead os synthesising it from increments". That did the trick. I created a new backup copy job to run some tests. I will provide a feedback in a few days, as soon I have a big number of restore points.

btw. is it possible to attach a picture / screenshot to this forum? Or is the only way have the picture uploaded on a public website and set a link here? Thanks

[Andreas Neufert]

Click on full editor and have a look at the bottom of the editor. See picture.

[redZonk]

Thanks Andreas! Its often very easy if you know how to do things

Indeed, after installing latest cumulative patch, I was able to run more then the 14 restore points on a StoreOnce catalyst store if I select "read entire restore point...."
With the StoreOnce VSA with 24GB RAM, I was able to get about 48 restore point. With a VSA with 48 GB RAM I got 78 restore point. So with a hardware box I could potentially extend that a bit.