Hub has ground to a halt. Extremely slow

Just an update for anyone following this.
I've been in touch with support and have completed the following.
Support identified delays to a couple of pistons that were basically writing lux levels to a virtual device.
All my pistons were then paused by me as the hub isn't really doing that much.
Still running slowly.
Hub rebooted and then updated to 114.
Everything now running smoothly albeit with only 2 simple rules and mode changes by time running.
Support also requested that I did a backup and restore as the database looked excessively large.
This was completed.
Will now monitor and see what happens.
I really hope it is not webCoRE that is causing the issues as I find this easier to use than RM as for some of my pistons I would need Virtual devices configured and about 4 rules to do what I can do with 1 piston.

Looks like webcore to me. A rule in RM appears 2x-3x faster than an identical piston in WC. And I'm still on 1.1.1.119

I've been reading a lot and kind of come to the conclusion that webCoRE may be the cause. I've backed up, and removed webCoRE for now. This will hopefully go one way or the other. Support basically told me the same thing when I had issues the other night. Massive pauses on a few pistons.

1 Like

I come to the same conclusion. I have 200+ devices now and I have no delay issue like some of the folks here. I was a heavy user of Webcore in ST but manage to get most of my automations without Webcore in HE.
Another possible culprit could be Phillips Hue integration as well. I removed that maybe 5 firmwares back.

2 Likes

Whilst webcore may be the problem. Something changed between firmware versions that caused this, so we know what changed?

1 Like

I agree. Don't know what changed though.

I don't disagree here, although for stability's sake, I've removed all to see how I go. I do hate not being able to see the logs myself, and see whats causing issues. Trawling through logs looking for potential causes is my day job :joy:

Read something somewhere about "internal use only methods that were blacklisted in 1.1.3"

I don’t have WebCoRE and never have. Speed has been solid. Delays with the UI are for sure slower for some reason, but the devices I see a delay on are delayed for other reasons which I’m always able to duplicate when I isolate them from HE.

Don’t agree with the suspicion that Philips Hue is to blame in any way. That integration has been consistently one of the fastest and as I said, there are no slowdowns related to it being on my system.

Another point is that Z-Wave can experience and cause slow downs if there’s an issue with device. There was also recently a firmware to address issues with Z-Wave devices (for anyone late to this thread), hence the latest firmware update.

So I have webcore installed. Currently only one active piston to control my pc through eventghost. I haven't triggered it in sometime though. I plan to move this into RM when I have the time think it through.

My point is though I have not experienced the slowdowns others have reported. I also have the hue integration installed.

Just another data point if it helps anyone.

1 Like

Only reason I mentioned Phillips because it flooded my hub event logs before.
Hopefully they will figure out the root cause of the slowdown.

1 Like

Philips hue integration still floods my hub events. 10,000 a day. Never heard back from support about it. There’s a thread about it somewhere.

1 Like

Also have Hue with no issues. I have seen some slow down, but nothing severe. New update made everything better.

The Hue issue would present itself differently for different people. The issue seems to be that the Hue integration is initiating a SSDP discovery every minute and if you have a lot of devices that report SSDP then you can generate a lot of events for every call. This was the case on my thread. Luckily the Hub events logs are now getting properly cleaned up every night so it is no longer bogging down my hub.

2 Likes

I'm going to roll back to 1.1.2 and see what happens. Has anyone done this already and still seeing issues ? (I've updated to the latest version of webcore)..

UPDATE: All seems ok again..

Upgraded to the latest, and Hue lights triggered by motion, dead yet again. So back I go. This is most certainly an issue between the hub and Hue for me. I can see the motion being logged and then nothing triggered. No weCoRE installed, and just native apps.

Does your Hue bridge have a static IP address?

It does yes. Not sure if this is the same for all, but in HE it always shows offline in Devices, Yet I can see it syncing every second.

V1 bridge or v2 and is the firmware up to date if it’s v2?

V2, fully up to date :+1:

1 Like