Monitoring SQL Database backups

[dealproc]

Is there any plugin available today in this project to monitor any Microsoft SQL Server backups? Same idea/question for Postgres as well.

Our current backup is a shell script (powershell) that gets executed every 6 hours, but I'd like to have something that we can constantly monitor at any given threshold during the day/week. Opserver sounds close to what I'm looking for, but am unsure at this time.

As this is for a startup project, and not a mega-million dollar enterprise, I'm looking at open source initiatives to save a few green-backs instead of spending mega bucks to solve my problem.

Any help/advice would be appreciated.

[NickCraver]

@dealproc On the databases tab for a server there is the last time a backup ran, and the last time a full backup ran (which would be the same thing if you had only fulls running). There isn't anything more detailed than that, though I'm definitely for adding a backups tab in there since they are insanely important.

What I don't know is what information people want to see, since everyone does backups a little differently. We do fulls and t-logs every 10-15 minutes here, others do fulls + diffs, some just fulls, etc.

So what would you want to see? From a job view of what's running, maybe with a warning section (configurable threshold) for things not backed up in the last<insert timeframe>? (this could be hard if it's never successfully run). More of an overview of all backups, maybe with the job name/status as a column?

Give me ideas here, it wouldn't take long to implement, it just needs to be useful - of course we can always improve once it's in.

[dealproc]

We do need to roll to fills every 6 hours + transaction logs every 15-30 minutes. That may be the best starting point. the overview statements you made sound like a great starting point for details.

Will setup the project in the next 48 hours and see exactly what does what.

[DForck42]

I know in sp_Blitz it captures whether a database has EVER had a successful full backup and haven't been backed up within the last week. That seems like a good starting place as any.

[NickCraver]

@thirster42 we monitor this on the databases tab of SQL, but we don't alert on it - alerting is another fun & big project that I'll certainly be considering this request in, it seems like a perfect fit that should be supported to me.

[ConstantineK]

For me, I think simple/full recovery models would drive different history requirements, due to the different requirements of RPO/RTO if you cant define some custom metrics, then the current default is sane, but I would love to have a method to click on a database that has been backed up at least once and see the last 10 backups, tlog or otherwise.

[NickCraver]

@ConstantineK I missed this - that's a fantastically simple idea for the database modal view to improve things. I'll make it happen.

[NickCraver]

@ConstantineK I'll polish it up a lot, but overhaul just got some initial last-100-backups love:

[dealproc]

👍

[GregWhiteDBA]

tracking backups across availability groups would be good