Any way to retry backups after they fail..?

Unimus support forum
Post Reply
joairamd
Posts: 16
Joined: Thu Jun 30, 2022 4:49 pm

Mon Nov 28, 2022 5:42 pm

Hello,

I just wanted to know if there was an option in Unimus to retry the backups of devices if the backup fails? As an example, we have a Ruckus Switch that fails its scheduled backup due to a connection error, according to the logs in the latest failed job section, so I have to manually run a backup on it with this switch it always fails the scheduled backup but succeeds when manually doing it, sometimes it fails the manual backup but the second time I try to do it manually it always works. I was wondering if there was a way to retry the backups after they failed instead of waiting for the next backup schedule. Thank you and I appreciate the help.
Vik@Unimus
Posts: 177
Joined: Thu Aug 05, 2021 6:35 pm

Mon Nov 28, 2022 6:28 pm

First, let's mention that a scheduled job consists of both, a Discovery and a backup taken on all the affected devices under the particular schedule.

When a Discovery fails on a device, it should be because of the device not being available or some other issue and it would be difficult to know if a device would be just fine next time it attempt to rerun the Discovery as there could be a number of reasons why it fails.
Backups are similar to this, although there could be an argument that if the preceding Discovery succeeded, the following backup should succeed as well, but this is not a guarantee. This becomes even more complicated with the amount of devices which would fail during a discovery or backup retrieval.

This may change in the future, but at least for now, we prefer trying to figure out the cause why a Discovery or a backup fails. When there is an issue, it is the best to locate it and address it, so that a retry wouldn't be needed at all.

As for your particular Ruckus switch you mentioned as an example - let's say the hit rate is not the best. It might be a good idea to look at it closer. If you want, feel free to submit a support ticket and we can take a closer look at it.
joairamd
Posts: 16
Joined: Thu Jun 30, 2022 4:49 pm

Tue Nov 29, 2022 4:30 pm

Thank you for the reply, I apologize if I was being unclear but also thank you for the clarification. I brought up the question mainly because of that ruckus switch present in our environment wherein it would fail its scheduled backup but when I get into Unimus and manually run a discovery then backup it works. I'll start by changing the connect timeouts and observe from there and give an update on the situation via here or in a ticket.
Vik@Unimus
Posts: 177
Joined: Thu Aug 05, 2021 6:35 pm

Tue Nov 29, 2022 5:38 pm

This does sound to me like there could be a connection throttling present and considering the particular scenario, I reckon this might be one of the kind where there is some maximum amount of SSH connections established in some time frame (e.g. within 60 seconds).

In either of cases, it is good to check debug logs to get precise cause of any of the errors and if it is something an extended default timeout could aid, then try to do that.

In any case, let us know in case anything pops up. We can take a look at it as well.
Post Reply