Yeah I just can't face the manual migration to a C7, so have to stick with the C4 for now.
I might have been swayed IF the Hub Protection Service could backup / restore its own ZigBee network but that doesn't seem possible based on recent news, so there's not really much incentive for me to manually move to a C7.
At the moment a C4 plus external ZigBee stick is actually easier to recover from a hub disaster IMO as you can simply move the dongle to another C4 (fortunately I have a spare). As long as the dongle doesn't fail of course, but the Norteks are pretty solid from what I've seen.
How's everyone doing here? I'm at 10 days and just started seeing response times creeping (using /driver/list load times). This is my C4 in Yellow, and C5 in Green (C5 handles HEM-v1's via Z-wave and some LAN-focused items - Harmony Hub, AlexaTTS, Hubduino). Neither are using rule machine and do a majority of activities via Maker-API (within Node-Red).
3 weeks since my last reboot (and after some fine tuning of maker API calls) and free memory has again halved. However the big difference this time is that my hub UI response time has pretty much remained the same and I haven't actually experience much of a slowdown.
There must have been some pretty decent magic being cast in the last couple of updates!
FYI, I swapped my C4 for a C5 (literally moved z stick over and restore backup; thanks for the device @JasonJoel). Only things needing adjusting/reinstalling was cloud apps like Alexa and any of my scripts that call cloud URL’s (the urls are device specific). I also added 2 chatty devices and apps (HEM v2’s and influxdb app), but the C5 device is still very responsive. Memory is low now, but pages load immediately - on C4 they would take >5 secs at this memory level.
I’d you’re on the fence about C4 stability issues, I’d recommend trying migration to a C5 (using C4 stick). I blame the issues on CPU throttling due to heat (my C4 runs at 180) and 64bit JVM.
Are you able to add new devices? When I tried that I had issues but I did not fully troubleshoot just switched over to the internal chipset as I had a few hubs lying around. In fact my setup now is C-7 for Z-Wave, C-5 for Zigbee and C-4 for cloud only apps. All hubs talk to Homebridge and Node-RED as I consider them local "companion" servers.
Oh great. Now I have to dig up a device and try to add it. Thanks @erktrek
EDIT: I seem to be OK, unless S2 devices have issues. No way I'm unpairing a lock and re-pairing it now :). I also have another C5 device that I will probably move my HEMv2's to. Also have a C7, but it's unopened as I don't need any of those features, just couldn't turn down the spring deal. Like you, I plan on making my C4 device a LAN-only or Cloud-only hub, but putting in a location that gets better cooling.
Okay that is cool thanks for testing that!! I am not sure what I did wrong back then I wonder if it was trying to change Zigbee channels or something - there was some system config I could not change due to the external stick overriding the internal one.. but am happy with my current configuration so all good.
With the Hub Information driver, I was able to see the effects of memory decline on one of my C-5 hubs. It went from 5-6% cpu usage to a constant 22-25% and the hub’s temperature went up to 122 degrees constantly from 102. I hadn’t seen any slowdowns yet, but rebooting brought the cpu usage and temps back down to previous levels. @gopher.ny is this something that is being looked at?
BTW, this is one the latest firmware.
I'm on a C-4 and I'm seeing the memory decline as well. I've rebooted several times for one reason or another over the last month, not related to slowdowns(Illustrated by the peaks in the graph). I have no time lately and haven't even attempted to look at anything other than the graph, watching the memory. Both device and app stats look OK, no red numbers. That's about my extent on reading it. I finally hit an unreachable hub yesterday morning when I woke up. I was on .6.137 at the time and then upgraded to the latest. It was up for 9 days before it went down. When I get some time I'm going to start disabling community apps and whatever's associated with them to see if one or some may be an issue. I'm also going back to weekly reboots.
For me it was 17 days since my last hub update/reboot. The telltale sign was the cpu usage spiked from average 4-6% to 22-25% that I’m assuming was increased garbage collection/freeing up memory. If only the hub had another gig of memory….
I haven't looked at CPU usage but I forgot about the heat(related?). @brianwilson pointed out his findings and my temps peak daily at about 194F. Right now my hub is 167F with an ambient temp of 69. It is vertical and free standing. Same place it has always been. I have 2 temps sensors near it and the ambient temp runs from about 68 to 72 this time of year. But, I haven't had any real slowdowns. Just the crash. I use Hub Watchdog and the numbers are fine. The last readings before the crash the other day were
Virtual 04-27 01:29 - 0.172, Zigbee 04-27 00:30 - 0.363, Zwave 04-26 23:59 - 0.258
The only oddity there is Zigbee is usually faster than Zwave but those numbers are miniscule. That aside though, it would appear to be a memory issue. I was in the UI the day before and didn't notice that it was slow or sluggish.
My hub watchdog numbers had actually gotten faster. I mostly rebooted to see if it was a memory issue causing the CPU usage and temp increase. I figured that a reboot would fix it if it was. I think I would start to twitch if my hub temp was 194F.
