Updated from .39 to next release on a c-4 and it was showing cloud unavailable for a few days, so following some of the suggestions from @bobbyD in other similar thread I did a soft reset and restore from last backup. Now I'm stuck at initializing hub 20%. Any help much appreciated.
Does a reboot (gentle one from diagnostics tool) make any difference? If a single reboot doesn't help along, then no, and there's no point in trying it again.
Next step would be to roll back to a previous version, and if that still doesn't boot, do another soft reset. We're pretty much looking to go back in time to a configuration that worked. Once hub boots up, I can grab the logs and try to determine root cause.
Do not try full reset - that's sort of a nuclear option that will kill on-hub data backups and all radio pairings. The option has its uses, but we're nowhere near needing it.
2 Likes
Reboot from diags doesn't help.
I've tried rolling back to two prior backups and rolling back to prior release. At this point its off line, and can't get to diags screen. Light is blue.
- Correction in Opera I can see it's diags screen. I've rolled back to 2.2..3.148, and then to .143, then .141 and ended up at .139 the last available.
Update: Soft reset was not working, rebooting wasn't working, i.e either would leave me at 20%. I did get it to boot in safe mode but was getting network error... but I could see it. Can you get in and convince it to play nice from there?
I did do a soft reset from safe mode, but didn't get the restore from backup promt, just right back to the 20% screen.
At the moment it's Error 500 A server error has occurred.... and it's in safe mode.
It's not getting far enough to start support access (limited as it is) in normal mode, and safe mode doesn't provide any support access at all, so we're in a bit of a pinch.
So you're on the earliest version available on the hub right now... if you haven't tried doing a soft reset while on that latest version, please do so. Each version has its own starter database, and it's created right after soft reset is clicked, so chances are one will work.
If that doesn't help, it might be time for a full reset.
First, thanks for helping, I know officially this is outside normal support hours, so much appreciated.
Second, I'm in! Soft reset wasn't working, 'unknown error' it said. Did it in safemode, nothing but the same error... somehow I did get an older version to load, did a restore (so the soft reset did work even with the error) and have since gotten updated to current.
Still I wonder why cloud was unavailable for days... and this $hit$how is pretty stressful way of fixing that! Can/is it worth we dig into that?
1 Like
For future reference, I would first reboot the hub from Diagnostic Tool, without doing anything else. If that doesn't work, then I would check to see if hub's internal clock is in sync with the browser (Settings >> Hub Details). Those two solve a cloud connection issue in most cases. If you didn't do so already, please send us an email to support@hubitat.com and we will further investigate what may have been the cause of your problems, as your experience is not the norm.
2 Likes
I had this happen yesterday and discovered the same problem. I finally fixed it by doing the following:
- soft reset
- go to hub (instead of restoring right away)
- you will see there are no devices if you check
- go to settings, backup and restore, restore your backup
- do not reboot, shut down instead and pull the plug for a minute.
- plug it back in and say a prayer if you’re so inclined
@bobbyD. Is this a bug? I had this issue on both hubs and have done soft resets many times in the past without issue.
Most definitely. I got the logs from your hub and looking through them for clues.
2 Likes
@bobbyD - Yes, that's what I did, based on another similar thread. The clock was correct, the reboot did not help.
Related Question - Is there a difference between a reboot at the 'normal' screen vs port 8081 tool?
@Ken_Fraleigh - Curious, did you do the recent update(s) and then notice the cloud issue? That's what I'm guessing/blaming for the moment triggered it for me.
1 Like
I haven’t had any cloud issues, but I was and am still on .145
I was wondering if it was an issue with the radios. I initially had one hub getting stuck at 75% and the other at 20%. After a couple hours worth of rebooting, soft resets, restoring, and shut downs, the 1st one finally came back up. It took 4 more hours of this before I could get the second hub back up with the restored database. After a while it started getting stuck at 10%, but I was still able to access it from 8081 and soft reset it, and it would boot up. I tried at least a dozen times to restore it the normal way before deciding to try what I posted above, which worked on the first try.
I just had to do this, and unfortunately it did not work for me. I still am stuck at 75% or 20%, it depends on which backup I choose. I can however boot into safe-mode fine, but not sure exactly what I can do it there. Should I keep trying different backups and see if any work?
Update 1: Upon further inspection (Safe mode of previous backup) I found this which may be the cause?
Update 2: I managed to get into a state that is sort of functional enough to be able to enjoy Thanksgiving since in Safe Mode, none of my community integrations/drivers seem to be disabled nor the built-in stuff and even though the zigbee and z-wave radios are off by default turning them on seems to have worked. I hope I can figure this out this weekend. Any ideas why I can boot into safe mode and not normally? What exactly is being disabled in safe mode?
Update 3: This whole thing is just so weird, I sat down to watch a movie at around 7pm, I know it was working then because my zigbee plug detected the TV going on and turned the lights off. Then sometime after that but before the movie ended (around 9:30pm) it was borked since when I walked into the kitchen to get a glass of water the lights did not turn on.
Update 4: I also noticed that in safe mode, none of my rules that use a "delay" works anymore, including motion lighting. When I looked into safe mode it disables the scheduler, is there any way to manually enable that for the time being?
Update 5: So I tried reverting the version back to 2.2.4.143 from 145 and so far, so good, longer boot than normal, but it did boot. Updating to 147 now and going to see how it fairs. I would love to be able to help in any way possible, so if any logs are needed, or Support Access would be helpful I am game.
I see tons of socket related errors in the log, and it crowds out whatever else may be happening.
Please stick to 147 that I see is running on the hub right now. While these socket errors are unlikely to be related to the 20% issue, I'll have to get that figured out/fixed first. Or at least to find an app/driver to pin the blame on so that it can be disabled.
2 Likes
@gopher.ny Who U talkin to with this instruction? I for one am on .147
Adjusted the post above for clarity - thank you.
@gopher.ny So... based on that your comment wasn't to me, I saw an update and tried to update to .148.... Went fine on my C-7 but on the C-4 didn't go well, never finished applying, after an :8081 reboot now I'm back to cloud connection unavailable. 
Clearly something in the update process is jackin' the box! It seems to be back on .145. Suggestions?? Time is sync'd up and yet a reboot from :8081 give me a Error: Request failed with status code 401.
I believe that means your login has timed out. Log out and back in and then try to reboot.
2 Likes
As @Ken_Fraleigh indicated, this is an HTTP unauthorized error.
I’d recommend exiting your browser, or using a Private/Anonymous browser window.
1 Like
See if 2.2.4.148 is available in diagnostic tool (8081) under Restore previous version. If not, the update didn’t get downloaded properly. If it’s there, switch to it.
Thanks. That got past the reboot issue. Cloud is available and it's notifying me there's an update...
Just did the update via normal method, it now seems to be working... Strange that it keeps happening when updating this unit. Thanks to all.
1 Like
