Soft Reset/Restore to a backup before the slowdown started makes no difference. Still barely running.
Before the outage was meant to minimize corruption from the power outages. Did you have a backup? It could be a device, mesh strength, some app or device driver, etc. These are the things in our control. Try one by one.
Just sent off an e-mail to support a few mins ago. I'm sure they're just going to say remove all custom apps & drivers. But I've had those apps and drivers since day 1 and the issue just started with these outages.
Interestingly, I did a full shutdown on the hub. When I hit the light switch at the top of the stairs the z-wave associated devices triggered instantly like they always had. Powered the hub back up and they were back to almost a minute lag. There's another thread where people are talking about stranded/ghost devices clogging up the mesh, I'm hoping this is my issue and there's a quick fix for it.
Bobby has never told me to do so and I have custom apps. Shouldn't automatically assume that't what they're going to tell you. I'm sure he'll take a look at things and advise accordingly.
3 Likes
I really don't like to be the bearer of bad news, but your wish came true 
Also check your lan connected devices, as some appear to no longer communicate with the hub. The following post is a good summary of your options for the night. I will reach out to you via email or PM, whichever works better for you.
The LAN devices all got new IP's with the new ISP, and I restored a backup from the day before so I haven't gotten in to change them all yet. I'll get that all straightened out again tonight.
I remember having to deal with a ghost device once on ST. It was as simple as creating a new virtual device, assigning it the same device ID, selecting a z-wave driver for it and then deleting the device. I'm hoping that process will be as simple here.
1 Like
No need for that. The nightly task will remove those automatically. If you could power off any devices that don't exist on the Devices page, or otherwise failed to pair or you no longer use, that would tell the radio that they are dead, so the hub will remove them tonight. Let the hub do its thing tonight and we can look at it again tomorrow. Things may be rocky for the next few days, but we will get you straight.
Since you have backups, you could always try a full reset of the hub, then restore your backup.
https://docs.hubitat.com/index.php?title=Soft_Reset
I don't know exactly how the restores work, but if the database of your backup is good, and the process of restoring it isn't jiving with the current corrupt database, a full reset SHOULD correct the database corruption allowing you to do a restore. BUT, you may have to pair all your devices again?
Maybe do this as a last ditch attempt and see if others have better suggestions.
Please be careful when using the R word. With "Full Reset" you cannot restore as it resets the radio. The link you shared is correct and it is for "Soft Reset" which accomplishes what you have described, with the exception that you don't have to pair devices again.
2 Likes
Interesting to know that and thank you for the clarification. I tried looking for some more detailed documentation, but there doesn't appear to be anything referencing the Full Reset specifically. I've seen Soft Reset used differently in the IT field. Usually just refers to rebooting a device, or something similar without clearing out the settings.
Actually, something else just occured to me. @MRobi said the ISP replaced the modem. Does that mean you have a new wireless access point as well? Maybe the new device is using different channels than were previously used.
The wireless settings could be screwing up your Zigbee/Zwave mesh and causing the delays. Try changing the wireless radio to a different channel to avoid interference.
1 Like
@cherbein03 I use ubiquiti gear for all my networking. The ISP's modem has the WiFi disabled and once I get my hands on a UDM-Pro I'll be taking their modem out completely. Good thought though!
@bobbyD has me firing on all cylinders again. Things are back to normal here. Had to re-write a bunch of rules I lost from the restore but that's certainly better than starting over.
With all this extra time on my hands these days I may even experiment with bringing my Schlage locks over from ST. 
1 Like
You’re into self-harm eh? 
Glad you’re back in business. Now get that equipment on a UPS! 
And tell any tech that walks in your door, if they unplug anything without asking first, they can unplug their modem and take it with them. 
1 Like
LOL I figure everything else is running smoothly and they're all that's left. Might as well experiment
I've looked at UPS' several times and then never pull the trigger once I see prices for what I'd need. My media server is a beast and then with everything else in the rack I'd need a 4,000w UPS at a minimum. And those run around $3k.
Backup up the critical pieces. I’m sure that’s much less than 4000 W.
The media server is what I'd consider the most critical piece in the rack. 10x the price of anything else and near 100Tb of data. That alone is 2800w.
Understood, so set it up to shutdown safely. I have a similar rack setup as you, but no where near 4000 watts (my equipment is new and energy efficient).
2800W? Wow! I have a 24TB server and it only draws, at most about 250W, under full CPU load, with all drives spinning.
Where is the 2.8KW going? Genuinely curious.
1 Like
2800w is just based on dual 1400w psu's. I don't have it completely full (yet) so I'm sure my avg load is lower. Just never measured it.
It's a commercial 4u server, dual 8core 3.8ghz CPU's, 128gb ram, 10Gbe nic, 48drive case. Sounds like I keep a jet plane buried in the corner of the basement lol
1 Like
So, those are most likely redundant power supplies if they’re both in the same server chassis. At work, we feed one power supply from UPS power, and the other fro non-UPS power. This way we’re protected from a power failure OR a UPS failure. So, if they are redundant, your power requirements just dropped in half! 
1 Like
