Backups happening at 5am rather than scheduled time

Not sure what's up, but it looks like on both my hubs, cloud backups are happening at scheduled (2:15am), but even though the local backup/cleanup is also scheduled for 2:15am, the backups are all datetime stamped at 5am (there is one at 4am, and then one for a hub update at 10am).

Is the cleanup taking that long? Seems odd it would align right with 5am though.

All mine have been 5 AM also for a while. All 5 of my hubs are exhibiting this behavior. I meant to post this issue but forgot so thank you for reminding me. I have adjusted the backup time a few times and it doesn’t change from 5AM.

My C3, C4, and C7 hubs are all still backing themselves up at ~2am like they always have. I wonder why some hubs have a 3 hour offset?

What time zone you in? (Random thought)

I am in EDT.

1 Like

I am also EDT.

1 Like

Yup. Got nothing then. I’m in PDT

@gopher.ny thoughts?

Is there a daily reboot happening at 5am?

Not for me. My backups across all 5 of my hubs has been happening at 5 am for a while. Only automation I have in place daily is downloading the backup which does occur at 5 AM. Would that cause this and replace the backup generated based on the setting?

2 Likes

Not for me either. Ahh. @ritchierich hit it on the head though. That is when my backup downloads.

2 Likes

Yeah, it's going to replace the backup made earlier on the same day.

3 Likes

Is there a HUB command that copies the existing backup?

This is the Windows 10 task I use for 10 days of backups
image

Not really, it requires either actual file name or latest as a query string.
I suppose you can parse HTML to get the file name, but I don't see enough practical value of doing that vs. just having it create a backup at the time of request.

2 Likes

Hi @gopher.ny,

Makes perfect sense. Would be nice if there was a command to just download latest rather than have it generate a new one. Not a huge deal though. I used to have it parse the html, but it kept breaking with new releases. Was super convoluted.

curl http://10.0.6.4/hub/backup | grep data-fileName | grep download | sed 's/<a class="download mdl-button mdl-js-button mdl-button--raised mdl-js-ripple-effect" href="#" data-fileName="//' | sed 's/">Download</a>//' | sed 's/ //g' | tail -1 | xargs -I @ curl -o @ http://10.0.6.4/hub/backupDB?fileName=@

But would consistently break.

1 Like

Download the Hubitat app