vcb proxy

[fafa24]

Dear All,

I use vcb proxy with Backup Exec 2010 to backups my VMs. I use the FullVM switch. Since there is a time limitation in BE for running pre backup script I run each night two jobs. Each job saves three VMs. I use the latest VCB version, which is 1.5 Update 1

It used to run fine, but since a week I'm getting randomly pre script failure. I did some test with browse-start.bat to mount the VMs and it worked fine. For example yesterday both job ran fine, but today one job failed. I couldn't track it down to a certain VM.

I was thinking, there must be network connectivity issues. This is the only explanation I have for random problems. I was thinking the VM network adatper type might be a problem. We use E1000. I don't really know the difference and what's the recommendation.

Any hints are appreicated.

Thanks,

Edy

[DSTAVERT]

This isn't a direct answer to your issue but VCB is at the end of it's life. 4.1 is the last version of vSphere that it will be offered. Backup Exec 2010 is able to use the VMware storage API's and can take advantage of change block tracking. IMO it would be a far better approach to backing up your VMs.

[fafa24]

I agree, but the VMware option for BE is quite costly and I don't know, if it's worth the money.

Thanks,

[DSTAVERT]

I guess you need to be the judge of whether it is too costly.

Just guessing that it is a network issue isn't helpful in diagnosing a problem. Have you gone through your log files? What are the errors messages? Windows Event logs, Linux logs, Solaris logs VMware logs, BE logs etc. vCenter events? As much as you can dig through. You need to be looking for error messages, timeout errors to see if you can narrow it down.

[jamkey]

Disclaimer: I work for Symantec Backup Exec Engineering as the Beta program manager.

Now, with that out of the way, my first advice is to remove your network adapter that is the e1000 type and add a new one but change the type to VMXNET 2 (Enhanced). I was directed to do this to fix a problem by one of our lab gurus who said it seemed to particularly be a problem on x64 machines when trying to trasfer a lot of data (i.e. backups or large folder transfers). Check out this blog post for a discussion of dropped rx packets related to e1000 (it also links to the vmware article about the issue): http://www.yellow-bricks.com/2010/02/02/e1000-and-dropped-rx-packets/

As for the VMware (AVVI) agent in Backup Exec, there are quite a few advantages that might or might not be of value to you depending on your needs. The biggest (besides not having to reply on scripts that require manual monitoring) is probably the ability to get to the granular data for a restore in just one pass of the backup using our GRT (Granular Restore Technology) for either Files & Folders, Exchange or SQL. You can also pick a Transport mode priority list of SAN, NBD, NBDSSL, & Hotadd. Then, if you have the agent, the free management plug-in for vSphere is very handy to know what VMs are protected and when they were last backed up (& if successfully or not). Hope that helps.

[fafa24]

Thank you for your valuable response.

I also thought that I should change the VM network adapter, but I wasn't sure.

At the moment I'm testing the trial VMware option for Backup Exec. It works well. The granular option is a nice thing, although I haven't tested it yet. We use mainly application server in VMs, so granular option is not so important. In case of an emergency a FullVM restore is sufficient I guess, the granular option is nice to have.

It's a question of cost. We are a Backup Exce shop, maybe in the long run it's unavoidable to buy the vmware option.

Thanks,

Edy