I am the IT manager for a small company with 3 sites, and am a gray-haired old network engineer. A few months ago, we setup Unimus at the smallest site, which I work out of. That Unimus instance communicates with the network infrastructure devices at the other two sites via Tailscale.
I saw occasional/intermittent backup failures, not enough to be a big concern. That local installation was primarily a trial run of Unimus, which I really like, and intend to continue to use, so I built a Unimus instance at the largest site. Surprisingly, the number of sporadic failures increased noticeably. The schedule ran once per day. In order to get a better look at these intermittent failures, I changed the default schedule to perform hourly backups.
Within a short time, a pattern became obvious... only my Cisco SMB switches were randomly failing to be backed up. I also have several Cisco Catalyst 9000-series enterprise switches, two ZyXEL switches, and four OPNsense firewalls in the mix, and I do not see recurring failures of backups of those devices. The Cisco SMB switches are Cat1200-series and CBS350 series switches of various models. IOS on some of the Cisco SMB switches is the latest, some are 1-2 years behind. I don't see any difference though in the incidences of backup failures. Also, the locations of the affected switches does not seem to matter - I am seeing failures with the SMB switches at all three sites. I also have no known ongoing issues with inter-site network communications.
With humble apologies, my team is short by two persons at the moment and I have very limited time to delve deep into this with an analysis of debug logs and Cisco switch logs. So I ask... are there any known issues with the Cisco SMB switch family device driver in Unimus 2.9.1
Backup issues with Cisco SMB switches
Currently, we are unaware of any persistent issues.
Do you have a single set of credentials in Unimus or several? If several are you using Credential Discovery or are you binding the appropriate credentials to each device?
When the job fails, what is the error given? I presume that manually running a discovery and backup works, but figure it's worth asking. (You can see the error if you hover over the red "failed job" dot for a device in the Devices screen or going through the most recent failed jobs menu on the Dashboard)
One diagnostic test I would be curious if you could run, enable Debug Logging (both for Unimus and Device Output) and then manually run a backup across your network. (If you get errors when you run a backup on a single device, then you don't need to do the whole network)
https://wiki.unimus.net/display/UNPUB/L ... +Subsystem
Once you have a failed job, disable the debugging and download the logs. They might be quite large. If you like you can shoot an email over to our support guys and they are more than happy to assist with figuring out what's happening.
Do you have a single set of credentials in Unimus or several? If several are you using Credential Discovery or are you binding the appropriate credentials to each device?
When the job fails, what is the error given? I presume that manually running a discovery and backup works, but figure it's worth asking. (You can see the error if you hover over the red "failed job" dot for a device in the Devices screen or going through the most recent failed jobs menu on the Dashboard)
One diagnostic test I would be curious if you could run, enable Debug Logging (both for Unimus and Device Output) and then manually run a backup across your network. (If you get errors when you run a backup on a single device, then you don't need to do the whole network)
https://wiki.unimus.net/display/UNPUB/L ... +Subsystem
Once you have a failed job, disable the debugging and download the logs. They might be quite large. If you like you can shoot an email over to our support guys and they are more than happy to assist with figuring out what's happening.