Continuing the discussion from GE/Jasco Z-Wave Plug In Outlets No Longer Logging On/Off Events:
This is related, I think. I have a number of Z-Wave wall switches from different vendors – and I have three GE/Jasco wall switches controlling my back patio lights. Their behavior became erratic this week, after functioning great for 6 months.
All of the switches respond to the rocker switch itself, but all are erratic operating with smartthings. One just does not respond at all… the other two respond, but after a VERY long lag…30 to 60 seconds…
…20 years of being a CTO started my spidey sense tingling: it’s pretty clear that something happened on the Smartthings server side… did anyone change anything?
I confirm that its happening here as well.
I would suggest that you try a Z-Wave Network Repair, and see if that shows anything weird, or possibly fixes the problem. And, turn off the hub for 15 minutes first.
Really? Just recently?
I’ve done the his power off @bravenel - no love.
Never heard of the z-wave network repair. It’s it a z-wave network monitor?
I have noticed the same behaviour in my GE Outdoor Module for the last few days.
1 Like
Good to know @Kooltaco and @ehoffman73 - recent behavior on both your parts?
Anyone at smartthings listening, or do I have to file a ticket?
Come to think of it, it started happening after the “Stability Enhancements/Changes to Sunrise/Sunset” a few days back.
You can do it in the IDE or in the mobile app. It is a utility that repairs the z-wave mesh network, so all the devices know how to talk to the hub. In the IDE it’s My Hubs, then View Utilities button. There you can start the z-wave network repair. It is a good idea to watch the log for the repair.
In the mobile app go left to your location, then touch the gear icon, and then scroll down to your hub. Inside the hub settings select Z-Wave Utilities, and then Repair Z-Wave Network.
It would be best to leave the hub disconnected from power for 15 minutes prior to doing this, and allow 30 seconds for the hub to boot after powering it up. Then do the repair. If you have any errors show up, repeat the repair.
1 Like
Yes, I’ve had problems over the last two weeks…the delays in particular affect specifically the GE/Jasco light switches, and I had some random issues with the GE/Jasco outlets (one of them specifically).
I have a strong zWave network, including a repeater to insure the further sections of the home stay strong.
Doing a ZWave repair, which you can kick off from the iOS app is really hit and miss if it does anything for you, but it doesn’t hurt. Its under the settings in Hubs, Z-Wave utilities from memory.
Z-wave repair (also called “healing” the network) is a utility that will force every node on your network to rebuild its routing table. This can rescue “orphan” nodes who are confused about who their nearest neighbors are and consequently have a difficult time getting or receiving messages.
For zigbee, healing the networks occurs automatically whenever the hub is powered off for about 15 minutes or more. It may take an hour or two before the routing tables rebuild completely, but you don’t have to do anything other than unplug the hub, wait at least 15 minutes, and then plug it back in.
For zwave, though, you have to use the Repair utility. This is available through the IDE. You will see a message in the log that the repair is complete, but that’s just at the hub, like zigbee it may take an hour or two before all the nodes have rebuilt their tables as well.
Most field techs will actually run the zwave repair three times in a row. This allows each hop of the network to be rebuilt.
So for maximum benefit:
- unplug the hub for at least 15 minutes. Then plug it back in.
- run the zwave repair utility. When you see the message in the logs that the repair is complete, wait another 15 minutes (don’t unplug the hub this time, though), then do a second zwave repair.
- again when you see the log message that the repair is complete, wait 15 minutes with the hub still plugged in, then do a third zwave repair.
The wait and repeat also improves the odds of catching all nodes, including ones that were asleep during the first pass.
So easy, but a little tedious. You should expect to see improved response rates within a few hours if node address tables were contributing to the problem.
2 Likes
Do file a ticket with support@smartthings.com
They don’t monitor these boards for individual issues.
Thanks @JDRoberts - I filed a ticket yesterday.
Also, thanks for the network repair tutorial for Z-Wave — I give that a try tonight.
1 Like
I’ve had the same exact problems with one of my GE/Jasco light switches. The multiple attempts to perform a z-wave repair did nothing to fix it. I’ll have to try what @JDRoberts mentioned. Although I’ve noticed my entire ST system has been so unreliable these days, so I don’t know what’s going on. Mis-fires, insane delays in actions, even my geofence triggered my “arriving home” mode although I was 3 miles away (instead of a block away). Good times.
I think the first couple of weeks in March were bad for a lot of people, but it does seem to have improved again since then. It certainly did at my house.
There’s another problem that can affect some people but not others, and that’s polling collisions. In a commercial install we usually aim for polling to be no more than 5% of network traffic. In a residential install it’s up to you if you want to pay the extra cost for what are essentially “hey, buddy, you awake?” messages, but it is something to be aware of. Some people in the community have posted activity logs showing polling in excess of 99% of their total network traffic! That can cause other messages to be delayed or even lost altogether.
You can view the activity logs with either the mobile app or the IDE. It’s good to review once a month or so or if you’re having issues, just to see what the traffic looks like.
If you do see a lot of poll or refresh messages, it’s worth tracking down when they’re coming from and if you really need them.
May not have anything to do with the problems you’re seeing, but it’s an easy one to check, so you might as well rule it out.
@JDRoberts it would be stunning revelation to see that polling is 99% of home network traffic – that would be an indication of a ST hub that’s just gone nuts, or a serious software glitch. I monitor my home network traffic pretty closely (as I said, I’m a CTO) for all sorts of reasons, and I don’t see any out of the ordinary hub activity.
Now Z-Wave operates on it’s own frequency (900MHz, I think) and has it’s own protocol, so I am not sure what that traffic is like, since I don’t monitor that band – so it’s possible the devices themselves are causing so much chatter that they are stepping on each other and misbehaving, but communication from the IDE or the mobile apps is happening through the hub, which pretty much contains that situation to ethernet traffic.
[quote=“uberrob, post:17, topic:13561, full:true”]
@JDRoberts it would be stunning revelation to see that polling is 99% of home network traffic – that would be an indication of a ST hub that’s just gone nuts, or a serious software glitch. [/quote]
Neither: it was people using custom smartapps, including Pollster, to poll all their devices every 5 or 10 seconds in an attempt to force the statuses in the mobile app to always match the true statuses.
In some cases people cut and paste code from other smartapps without really understanding each line, and end up with a pollster wrapper around their code to get it to run rather than subscribing to an event so it runs on demand.
If you’ve only ever worked on single device code, or in a star network like WiFi, the whole concept of a mesh network can be really foreign. The idea that many devices are asleep most of the time and that the hub neither knows nor cares about true status of individual devices drives people crazy. Not to mention the fact that messages can arrive out of sequence. But mesh is a very cost and energy efficient way to operate in most home automation set ups, which is why zwave and zigbee are so popular.
So it’s really easy for people to create massive amounts of polling traffic just by installing someone else’s custom code.
2 Likes
Same boat.
I had to exclude and reinclude the offending switches in the end…which became the “uninstall all smart apps on each switch and reinstall” game.
Not a fun Saturday. Today all my AEON minimotes won’t do anything.
Its just HA whack-a-mole.
I am looking to see if the fixes this week make any impact that was promised in the other stability thread. Would be great…but as a seasoned CTO and currently doing challenging big data things…I understand the challenges they face and have been patient (in between curses hurled at the sky)