Feature request: "Caching" job schedules (small sites)

[pkelly_sts]

It's a long shot I know but if you don't ask, you don't get!

I've just set up a couple of small remote sites (two hosts, a few VMs) where we've configured a local proxy/repository VM, replicate locally & manage from/send backup copies back to HQ.

It all works well but the weak point is the fact that if we lose the (point-to-point VPN) link back to HQ then EVERYTHING stops. It would be really nice if there was some method of caching "scheduled job commands" locally, say on a Proxy, so that even if the link back to HQ fails, which I know will obviously stop any off-site copies, at least it will do local backups/replicas as per default schedule until told otherwise.

I know it's a slightly odd request but I'm sure plenty of folks out there would appreciate such a facility and, as I said, if you don't ask...

[Shestakov]

Hello Paul,
Thanks for your request!
Have you considered installing Backup & Replication Console on the remote sites?
To make users' work uninterrupted the remote console maintains the session for 5 minutes if the connection is lost. If the connection is re-established within this period, you can continue working without re-logging to the console.

[pkelly_sts]

Thanks for the suggestion but I'm talking about day to day job execution reliability rather than connectivity to the console so 5 mins won't have much impact anyway.

What I'm thinking is if, for whatever reason, the firewall has a problem or the internet connection (back to B&R server) goes down and isn't resolved until next day some time. Without the connection the B&R server can't issue/orchestrate the jobs to that site. It'd be great if the site had an option to say "unless I hear otherwise, if I lose connection to HQ I'm going to keep running the *local* jobs as per the last-known schedule until I connect to HQ again & am told otherwise - if HQ comes back online, then I'll continue the jobs that send data off-site".

I appreciate it may not be simple, but I bet Veeam do plenty these days that they originally thought would be very difficult!

[Shestakov]

Paul,
taking the above-mentioned case into consideration, if you schedule a job it will work even without connection between the backup server and remote site with backup console.

[pkelly_sts]

Oh, sorry do you mean that if I install the console locally (say, on one of the local proxies) then that effectively creates the kind of schedule cache I'm talking about? If so, does that assume I need to have the console open/logged in permanently for it to work?

[Shestakov]

Yes, once the schedule is created connection between local console and VBR server is not needed unless you want to make some changes in the job or infrastructure settings.
Thanks!

[pkelly_sts]

Oh wow that changes things significantly! I shall have a play, thanks!

[dellock6]

I think guys you are talking about two different things...

Nikita, if I understood correctly, Paul is asking for veeam.agent.exe not to fail when is running a job, even if the connection to veeam.manager.exe is lost during the execution. Actually, to my knowledge this is not the case as the manager process is always executed at the VBR server, and not in the server where the console is installed, and this connection cannot be broken otherwise the job would fail.

[pkelly_sts]

Thanks Luca, I'm actually (cheekily!) asking for a little more than that, I'm asking/wishing that a scheduled job would even start without a connection back to veeam.manager.exe if it was already scheduled to do so in a kind of "Even if I lose connectivity to management, just carry on as I was scheduled to do anyway".

I was a little surprised when it appeared the functionality might already be there as I'd assume you guys would be shouting about it from the hill-tops!

[Shestakov]

Hmm, looks like Luca is correct. Sorry for misleading Paul.
In this case that is indeed a good feature request, thanks for that!
However you may be tolerant to disconnections if you have VBR servers on each site and Enterprise Manager in the middle.

[pkelly_sts]

Yes, I appreciate that but A) these sites are too small to warrant a local VBR server & B) I want to integrate tape copies as well which are difficult/impossible in cross-vbr server scenarios.

Thanks.

[Vitaliy S.]

You can always combine Veeam B&R server role with existing repository/proxy server, should not affect the overall job performance.

[dellock6]

Paul, where are the tape machines going to be located? At branches or at headquarter?

[tsightler]

Yes, I appreciate that but A) these sites are too small to warrant a local VBR server & B) I want to integrate tape copies as well which are difficult/impossible in cross-vbr server scenarios.

I'm assuming that, if the sites are too small for a VBR server, they almost certainly don't have vCenter, which is also pretty critical to the backup process. Sure, you could potentially add only the hosts, but that has other negative side effects if those hosts are managed by vCenter.