Yeah I meant the offline bit not the cloud links.
Its possible the cloud dashboards failing is the first sign of a jumbo frames crash and no one has noticed before 
1 Like
I understand what you said.
A few (many) years ago I worked on a company selling Novell !!! Networks, so although I am not a specialist (obviously), I understand many of these concepts, and even now I can’t help getting into my company and home networks set up and configuration.
I will deactivate all UDM devices installed but not present.
I will also install Wire Shark on my laptop.
Unfortunately this week and probably next, I will not be able to leave my laptop at home because my office PC is locked for a couple of weeks.
My office is going through some remodeling/repairs, and instead of moving my PC to some other temporary office, I decided to simply “wrap it” to the desk until repairs are done, and for the moment I’m using my laptop and taking it from home to the office and back every day.
I will look at WS and maybe I can use a RPi I have somewhere without any use, for WS.
I will look deeper into my Unifi configuration at home. I have a UniFi smart switch (USW8PoE 150W) and also will review the specific configuration.
Thanks everyone for your support.
Its also possible cloud is down due to network issues between you and the cloud servers. You can try a traceroute to diagnose.
I´m going through Unifi configuration, and don´t see anything weird, and specifically on the USW8-PoE switch, I just can´t see anything related to jumbo frames.
Can anyone guide me to where should I look and what to look for?
Regarding tryint a traceroute to Cloud servers, can anyone provide an IP or Cloud address to trace?
Thanks
C:\windows\system32>ping cloud.hubitat.com
Pinging d194wzi4ri2kev.cloudfront.net [18.161.34.127] with 32 bytes of data:
Reply from 18.161.34.127: bytes=32 time=19ms TTL=246
Reply from 18.161.34.127: bytes=32 time=20ms TTL=246
Reply from 18.161.34.127: bytes=32 time=23ms TTL=246
Reply from 18.161.34.127: bytes=32 time=25ms TTL=246
Ping statistics for 18.161.34.127:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 19ms, Maximum = 25ms, Average = 21ms
also verify name resolution works in the settings/network setup/network test
ie
I was trying to trace the whole link, and not only cloud.hubitat.com 
Everything looks normal to me, but of course I rebooted the hub three hours ago, and everything is working as expected.
I will try to detect when something (if) fails, and then do all the testing again.
and
Any instructions on UniFi configuration?
BTW, I have all my devices up-to-date, except my Cloud Key Gen 1, stuck with ver 7.2.95.
Edit: Here is my inventory:
- UniFi USG Router
- UniFi Cloud Key Gen 1
- UniFi USW-8 150W Switch
- UniFi USP Plug (Connected to ISP Modem)
- UniFi USP Plug (Connected to HE Hub)
- UniFi AP IW HD (2)
- UniFi AP IW (3)
- UniFi AP ACLite (3)
- UniFi AP M (External)
My USG router is a little too old.
I can´t update the Cloud Key maybe because the microusb card bay is damaged, thus it cannot make a backup, so it won´t accept to be upgraded.
I´ve been procrastinating about instead of buying a Cloud Key Gen 2, buying a new set of Router-Switch-Cloud key, a Dream Machine with everything integrated, but that hasn´t happened yet.
Do you think that the problem could come from using Cloud Key Gen 1 with ver. 7.2.95?
Personally I use a Watchguard T35/T40 for my firewall and Unifi AP's... Has served me well. Watchguard is an industry leader in network security. Behind that I just uses cisco switches for the most part.
The UniFi network global settings:
An individual switch settings:
I doubt that is a likely issue, but I don't know. I'm running 7.3.83 on my UDM-Pro.
Everything is as you suggest, except for the options on your screen:
My screen shows as follows:
So I checked both IGMP Snooping options (they were unchecked)
Should I also check Multicast DNS options?
The switch is configured just like yours.
That may be pretty much the last release of the UniFi Network application that will run on an old UniFi Cloud Key gen1...
There you go! I missed this communication.
So, as I said, I´m stuck to 7.2.95
For the moment, there is nothing I can do about it, so I hope this is not related to my problem.
Multicast DNS is useful when you have devices on different VLANs and you want them to be able to communicate via the mDNS service. I doubt that it is the issues with your setup. If you do have multiple VLANs and Apple TVs or printers then you probably do want it enabled. There should be no harm in enabling it as part of your troubleshooting process.
THIS could be the same issue i had.. when my router did its weekly reboot. the hub did not recover correctly and dns was failing... reboot fixed it as well, and a test with a manual reboot did not cause the same issue..
/i believe there is some bug in the networking code of the hub and if it looses its connection at a critical poing or sometihng runs when it is momentarily disconnected.. dns does not recover and you get any cloud apps failing.. they have failed to answer my thread regarding this issue:
ie
Got the same problem. If my router reboots I lose cloud connection on my C8. C5 is fine. I have to reboot the C8 to get it back.
1 Like
I’ll try that, but it’s not dropping off the lan. Just no remote access including cloud dashboards not working.
Also seems strange that only the C8 has the problem. My C5 stays connected.
1 Like
I have some information I can share,
I rebooted the hub at 7:00 am, and all day long it worked as expected.
Around 8:00 pm I noticed there was a new version so I decided to upgrade. I also enabled a zigbee button switch I disabled yesterday.
At 10:30 pm I noticed that cloud dashboards didn’t work, but the hub was accessible via IP, it was not frozen, so I did all network test. Here are the results:
Everything looked ok, but weird things with cloud.hubitat.com
Was this useful?
It all looks normal. Are you on DHCP or Static IP
I can replicate my cloud issue. If I reboot my router the hub reconnects, using DHCP with reserved address in router. Local access via IP is fine. But no cloud access, dashboards, etc. Reboot hub and all is fine again. Tried this a couple times and same result every time.
