You can follow the memory to see if that is an issue.. here is my C-5. I am not having hub lockups (yet) and am only running the Zigbee service on this hub.
http://[YourHubIP]/hub/advanced/freeOSMemoryHistory
Note: I did this quick and dirty in LibreOffice Calc. You can see a slow increase in total JVM memory and a slightly larger decrease in free memory over the time span. It does look to be flattening out a little though. Also the blip at the beginning was likely a reboot.
I have no hub lockups on either of my C-5 hubs.
This one is zigbee and virtual devices only:
This one is zigbee & zwave:
They were both last rebooted on 2/5/2021. Both are running 2.2.5.131.
New user from the latest W**k Exodus. I've been digging in familiarizing myself with the system and really am amazed by it. Have had similar "lockups" recently. I can access port 8081 and reboot, but it seems the lockups are coming more frequently. I'm following this conversation for any updates. I'm running hardware Rev C-7 - HE Version 2.2.5.131 .
My lockup is only occurring every 12-14 days which is livable, but interesting that the second hub doesn't mimic the first in that it has never locked up. I all of my zigbee lights on the one that is locking up.
That is interesting.. I have a C-5 with the same setup (but no RM stuff). As I mentioned there does appear to be a decrease in available memory after only 4 days but things seems to be flattening out a bit.. (see my prior post). Will keep an eye out and report back.
edit: am on the latest firmware: 2.2.5.131
"+ 1" to random lockups/issues, usually noticed when automations stop firing and/or Z-Wave stops responding to commands. C7 HE running 2.2.5.131.
I was having issues on 2.2.4, it seems without a ton of diagnostics that the Z-Wave side just hangs up. A pretty basic setup. Running Webcore and Tonesto's Homebridge.
Mesh has been solid for years prior to migrating from ST to HE.
Memory trends from the last reboot just to share.
another lockup today -- was running from 2/14/21 @ 7:16am until ~9:42am today (2/16)
@gopher.ny I just rebooted after wife complained of not being able to turn lights on
I will try the backup & immediate restore that you mentioned by DM, but any insights??
+1 to this issue with my hubitat as well. Never had lockup issues but ever since updating to latest firmware it seems to be nightly. I have to unplug/power cycle to get it back up.
Platform Version:
hub-2.2.5.131
Hardware Version:
Rev C-5
Hub ID:
xxxxxxxxxx-xxxx-xxxxxxxxxxx
IP Address:
192.168.1.XX
Diagnostic Tool Version:
1.0.81
Please revert back to 2.2.4 if the hub keeps locking up on 2.2.5. There are changes in the upcoming 2.2.6 that will hopefully address the issue, but there's no point in having an unstable system in the meantime. If you no longer have 2.2.4 as one of the options to revert to, use endpoint http://your.hubs.ip.address/hub/advanced/download224 to force the download. The endpoint has no UI, so it will "hang" for a minute or two before returning a simple success/failure message.
thanks for the insight gopher.ny. I am a bit confused with this suggestion since in the beginning of the thread, I was running 2.2.4. I know most of the people that have joined in this thread are having issues much more than I am, since mine is happening once every 10-14 days, but I don't think I should go backward because it was happening on 2.2.4 also. I will wait and see what happens when the next release comes out. Hope that doesn't come across snide, I appreciate the work being done to stabilize the system, I just think maybe these people are seeing a separate issue than what I am seeing and what the start of the thread was about.
Can anyone confirm if 2.2.4 is more stable than recent updates? It seems to me ever since updating earlier this year there's been daily lockups of my C-5. Even a z-wave reset and soft-reset followed by a rebuild from scratch locks up the first night for me.
I'll roll back to 2.2.4 if so, but that'll take another weekend of re-adding everything.
Not for my C-7 or my C-5 - there's no issue with stability, and I've only needed to reboot for platform updates going back to over a year. I know that all of these are just anecdotal points, but if there was widespread instability associated with recent updates, we'd have a large number of community users indicating that.
Probably my device then, will see what support says I opened a case.
Recent updates have been very stable for me, and judging by forum posts, pretty good for others too. In fact 2.2.8 is running smoother and faster than ever in my opinion.
Oooh, I don't think I would do that, I don't think you are solving anything by doing so. And you arec creating more work for yourself.
A soft reset doesn't really clear every aspect of the hub, so technically I don't think you are "rebuild from scratch". So you might not really be clearing everything out of the system. However, a soft reset and restore of a backup can clean the database, which is often the cause of problems, and you don't lose everything. So it doesn't hurt to do a reset/restore, it can only help.
I would not roll back unless you absolutely have to. And don't reset the Zwave or Zigbee radios, I don't think that is doing any good and just causing more work. I also would leave it so support can take a look at the back end of the hub and see if they can determine what is going on.
What I would do is get it up and running, and watch the Hub Stats page and see if anything is flagged in red, OR if some app or device is consuming a large amount of memory. Things that are cloud based are always a prime suspect in my opinion. Judging by forum posts, it is often something that polls the cloud very often like a thermostat, weather, or a similar app. So keep an eye on those things.
Logs also would be a good thing to look at, see if there are errors/warnings, or if some device is incessantly spamming the log.
IMHO, there is a known memory leak and that memory leak will get worse the more complex and the more rules you create. It’s only a matter of how long it will take for your unit to deplete enough ram before it locks up. I’ve set up my hub to reboot daily now because of this memory leak.
This was the first thing I tried, didn't help. Soft reset without a restore lost all my devices, so at that point a z-wave reset wasn't causing any more work.
By who? Not arguing, just asking how you came to be aware of this.
I don't know if I would agree with that. I have many rules, some quite complex, and I know others do too and none of us are experiencing this.
I would suggest that you file a support ticket with all this information if you have not done so. It might be something specific to your hub or it might be a larger issue. But until staff can catch this so called memory leak, nothing can be done to fix it. The more reports the better so maybe a pattern will emerge.
Well that is why you have to restore the backup, it clears the database so the hub loses track of what is in there. The radio information and other things are still there, they just need the restored database to function.
A restore would have just made it all work without any (rhymes with mucking) around with the Zwave stuff. It is the difference between clicking the restore button and waiting 5 minutes, or hours of screwing around with inclusion/exclusion, naming, and all that crap let alone the rules and automations you lost. I guess if you want to go this way, you can, but it really is just more work than necessary. And frustrating on top if it all.
Download the Hubitat app
