Power Outage restore switches to proper state

[integlikewhoa]

Yesterday I had a whole house power outage and don't have a UPS for the HS. So on boot back up all my virtual switches and light switches went to a previous state of a few days prior rather then what they were before the power outage.

Can someone tell me where the backups are stored and why they are not kept more current?

My virtual home and away, and most of my lights are non polling and until I go threw and flip them everything reads wrong and wont update.

Also I haven't checked but can I base an event on a reboot or restart so it sends me an alert letting me know HS3 just started back up again?

Also the biggest tell of when the devices were restored to (few days back) was all the last trip times for doors and lights were several days prior vs. I know I used them that morning.

I'm running HS3 on my own server CPU and data Storage is not an issue.

[cabldawg]

Yesterday I had a whole house power outage and don't have a UPS for the HS. So on boot back up all my virtual switches and light switches went to a previous state of a few days prior rather then what they were before the power outage.

Can someone tell me where the backups are stored and why they are not kept more current?

My virtual home and away, and most of my lights are non polling and until I go threw and flip them everything reads wrong and wont update.

Also I haven't checked but can I base an event on a reboot or restart so it sends me an alert letting me know HS3 just started back up again?

Also the biggest tell of when the devices were restored to (few days back) was all the last trip times for doors and lights were several days prior vs. I know I used them that morning.

I'm running HS3 on my own server CPU and data Storage is not an issue.

I'm not certain about that but for others that may be able to answer you which OS are you running on your Server? Some things are OS dependent. Are you running a real server with ipmi such as Dell idrac or HP ilo? You can manage power events through that but for any server I would highly recommend a UPS to allow for graceful shutdown so your system doesn't get hosed.

Sent from my SM-G935V using Tapatalk

[integlikewhoa]

I'm not certain about that but for others that may be able to answer you which OS are you running on your Server? Some things are OS dependent. Are you running a real server with ipmi such as Dell idrac or HP ilo? You can manage power events through that but for any server I would highly recommend a UPS to allow for graceful shutdown so your system doesn't get hosed.

Sent from my SM-G935V using Tapatalk

I run windows 10 LTSB on the server.

I'm not sure how windows or anything outside of HS3 would manage how HS3 is storing or doing its device status updates.

I do nightly zip backup of the HS3 folder using BLBackup and can use that if there was a major issue. This really is not enough for me to run a full restore, manual polling and flipping some virtual switches fixed the problem, but I would like it happen automatically on restore vs. me doing it.

I know a UPS is ideal for a graceful shutdown. I haven't wanted to invest yet in a rack mounted UPS that I need for POE switch and other things I run.

[sparkman]

You could run an event similar to this:


It can be run more frequently with little impact.

Cheers
Al

[integlikewhoa]

It can be run more frequently with little impact.

Thanks I have set this up and have a follow up question. Obviously I don't want to busy the system but what do you consider little impact? Is running this every 30 min or 1 hour to much?

[sparkman]

Thanks I have set this up and have a follow up question. Obviously I don't want to busy the system but what do you consider little impact? Is running this every 30 min or 1 hour to much?

Not from my understanding, but keep an eye on system resources to confirm. It'll likely depend on how many devices/events you have and what type of system it is. I haven't noticed any degradation on my system when that event runs.

Cheers
Al

[naellis]

This is an issue I am also still trying to find a solution to. In my case I have many battery powered zwave central heating radiator valves (StellaZ's) and if ever I have a power outage, when HS3 comes back I find some HS device values of these are wrong. It also happens on other zwave devices such as Fibaro lighting modules. Currently I have to manually poll or toggle incorrect HS devices just to get everything back in sync. A real pain. I raised this issue many many months ago trying to find out why HS restarts like this but I haven't had a good explanation yet. How is it possible for some HS devices to change value for no reason and others remain as they where before the outage ? Really would like to get to the bottom of this one.