[Solved] Controlling discovery jobs

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

Mon Oct 17, 2022 4:30 pm

Hello,

Just a question but is there a way to control or schedule when discoveries are run on devices? In my list of devices the discovery job sometimes fails and the connection becomes red and it doesn't go green until I force a discovery. Is there a way to control this so that a discovery is run and the device status stays green?
Vik@Unimus
Posts: 198
Joined: Thu Aug 05, 2021 6:35 pm

Mon Oct 17, 2022 5:59 pm

Hello,

My main question would be, what is your schedule like? When Unimus runs a scheduled backup, it first runs a discovery (on all devices using that schedule). If I may, I would like to link to our Wiki article for Unimus' Discovery


https://wiki.unimus.net/display/UNPUB/Discovery

If the discovery (or the backup, for that matter) on a device fails, it won't be run again during that particular schedule.

What is the time-frame after which you run a manual discovery? Is it shorter or longer than the gap between two scheduled runs? If it is shorter, you can edit your schedule to perform the scheduled job more often. If it is longer, this may indicate an on-going issue during the discovery or backup of that device. If that is the case, you may be wondering why it does not fail if you initiate a discovery. One of the possible cases might be that your device limits the number of successive connections to the device and that the discovery may get up to that limit and the following connection during the backup might be over it, hence failing. In the event that this turns out to be it, this can be remedied by extending one of our default timeouts (a delay between connections)


https://wiki.unimus.net/display/UNPUB/C ... t+timeouts

In either case, this is just me guessing. We'd need more context to determine the cause.
joairamd
Posts: 16
Joined: Thu Jun 30, 2022 4:49 pm

Thu Oct 20, 2022 3:12 pm

Hello,

Thank you for the reply, I have my switches backing up every day at around 2 AM. I see that in the Dashboard>Latest Failed job the devices that are failing the backup say that it's due to a CONNECTION_ERROR. Correct my if I'm wrong from my understanding of how the backup/discovery works. From how I'm understanding it, backups run a discovery first and once a connection on the device is established it takes the backup but when the discovery doesn't work a backup isn't taken. So in the scenario where a discovery failed and a backup isn't taken, Unimus will retry the discovery and backup during the next time a backup is scheduled? I appreciate the help and patience.
Vik@Unimus
Posts: 198
Joined: Thu Aug 05, 2021 6:35 pm

Thu Oct 20, 2022 3:54 pm

Yes, that is correct

https://wiki.unimus.net/display/UNPUB/D ... ledbackups

Also, as for the errors. If the possible reason why would the devices manifesting failing (an outage, etc) is unknown, then feel free to gather debug logs for us and we can take a peek at them to see what could be the cause of it and we can go from there.

To do so, go to Zones > Your Zone (the zone where you assigned the affected device(s)) > Debug mode and enable Debug mode and Device output logging.
Rerun a discovery on the device(s), and after it fails, download both log files (from the same Debug mode window). I would then recommend to submit a support ticket in our Portal and attaching them (I also recommend packing them into an archive in beforehand).
joairamd
Posts: 16
Joined: Thu Jun 30, 2022 4:49 pm

Mon Oct 24, 2022 3:36 pm

Thank you I will keep that in mind and submit another ticket if need be.
Post Reply