Device Delays Since 2.2.4

I’ve rebuilt this SA routine a few times from scratch. Deleted, reboot, rebuild SA routine and no go.
It’s two lights that worked before, work manually every time. One SA is. MWFS at a sunset offset the other is the same lights but TTS at a different offset

Thanks. I will investigate.

I know what the problem is, and we will get it fixed. FYI, it has to do with the Days of Week restriction. A work-around is to not have that restriction in the interim until a fix is available.

2 Likes

:+1:t2:

Appreciate your attention to this. I'll verify that the other instances of this issue that I've been seeing fall into this same situation... Looks like I'm back to zwave gremlins then.

I've performed both a global network repair, and individual device repairs as I have seen issues with them, since upgrading to 2.4.4. The only things of note is that 1) the repairs always complete without issue and 2) I seem to have a LOT more devices that have routes going straight point to point from the hub (not routing through other devices) according to the zwave details page. The switches that seem to be having issues now fall into that second category: they all have a path straight from hub to itself.

Still having issues and still unable to find a pattern. Still all related to automation (Rule Machine or Simple Automation Rules) with a contact being a trigger and a switch being turned on or off being the action. None of my motion related actions are having issues, which I also find odd if this is a mesh issue.

What is most troublesome is I can be actively tasking a switch from a dashboard without issue (can task a switch to turn on and off from a dashboard or from the devices page). As soon as an automation fires an action at the same switch, it hangs 8 - 10 seconds before it works. During this hang time, I can no longer activate said switch from a dashboard or the devices page.

The only consistency is the automation: if there is no automation interacting with the switch, it works fine. Some period after the automation action completes (if it ever does), the switch becomes responsive again to dashboard and device page interaction.

Since the devices in question work right up until the time of an automation action firing, I do not believe it to be a network mesh issue (in fact, most of these devices are reporting network paths are don't have routing nodes involves, they're supposedly communicating hub to device directly).

I'm at a complete loss. This was not the case before the 2.2.4 rollout; these same automations were working correctly and devices were not hanging like this. I have a Z-stick arriving today so I can start monitoring traffic.

Same thing again for me this morning.

I have a rule to turn my boiler (shelly1 switch) off at 658am and it missed this morning. I updated to 2.2.4.145 a couple days ago. I have never encountered this before. Of course I cant see the logs here from work and im sure the logs will be unavailable by the time i get home. I wish there was a way to dump logs somewhere before they are lost.

I need to check this, but sharing here to see if I'm on to something... I now believe certain contact sensors in my mesh are causing a hangup with z wave traffic... Might have something to do with secure inclusions and which level.

I'm realizing that certain contact event changes in my network seem to cause traffic to hang. The contract event changes seem to show up immediately on the hub... It's anything immediately occurring after that contract event that seems to hang... And the hanging period is a period of time, not just certain devices or number of events. The length of hang can be anywhere from a few to 20 seconds... Seems very inconsistent, but that might be dependent on the offending contact sensor.

I'm going to try excluding contract sensors and re-including using different security levels to see if I can confirm this out rule it out.

If you can confirm it, and it is a certain device or certain security level, I would contact support and/or tag bcopeland (Bryan) here so he might be able to track this down. If it is indeed a bug of some type.

The sad part for me is that the worst time is usually in the morning, so it's hard to do cause and effect.

Which brand contact sensors?

Spent three hours away from home in the middle of the day. Came back to a new personal best/worst: 33 seconds, per the log. After exercising it a little, it came back.
This is on .147.
Ecolink zwave+ contact sensors and Enbrighten switches, if they make any difference.

Finally getting time to troubleshoot more here.

I have several Ecolink zwave+ contant sensors as well, many of which are involved in automations seeing issues, but I also have Monoprice and Aeotec contact sensors in use. I've seen issues with automations hanging with all of them, but the more consistent issues have been with the Aeotec and Ecolink devices. Almost all of my switches at GE Jasco (both zwave and plus) / Enbrighten.

After updating to 2.2.4.148 this morning, followed by a proper shut down / wait 2 minutes / startup, I'm still seeing the same delay issues in a few places, but it's still inconsistent in others. I have removed all of the Aeotec contact sensors from my network, as I have other issues with them. Removing those had no impact on delays.

I continue to see the same issue: switch works perfectly fine until an event occurs related to a contact, then the same switch with lag / hang / be unresponsive to commands and then suddenly responds to the contact action. Some period of time later, the device returns to normal behavior (can be controlled via dashboard or device page). The period of time is random. As I have a house full of people and dogs, it is challenging to isolate activity to ensure only one contact is being activated at a time. I can say I have had ZERO issues with motion based devices and automations (almost all of my motion sensors are HomeSeer Multi-Sensors, with one Aeotec multisensor 6 and tri-sensor in the mix).

Next step is to try removing / re-adding the ecolink contact sensors...

I went down my path (not advised) of hard reset which involved removing and re-adding everything, plus the rules. So, everything was 'new', and it didn't do a damn bit of good.

What Aeotec devices did you have issues with? I am several Aeotec door sensor 6...the triangular shaped ones...that I had issues with when I updated to 2.2.4. I have since reverted back to 2.2.3

I have those same devices, but I believe I'm having a different / device specific issue with these. I detailed those issue in a different post earlier today:

https://community.hubitat.com/t/aeotec-door-window-sensor-6/56984

1 Like