i have reported it but it was basically ignored.. it would be good if someone could look at the engineering logs and possibly see if it is a hardware issue or software.
if it was hardware i assume it would lose all connectivity.. as i said it does not.. the hub continues to function and can reach cloud services such as amazon and honeywell etc..
all that stops working is the incoming cloud services..
it is reproduceable every time i reboot the router.. i have it set to reboot on a schedule once a week and that is how i first noticed it.
Ever tried a different Ethernet port on the router for the hub that loses access? Could you have a VLAN tag on the port with a firewall rule doing something?
In the past, when I have seen this type of download behavior it was related to Ethernet auto negotiation. Try making sure auto negotiation is set to auto. If this does not work, then make sure you set it to a fixed speed with full duplex, like 100-full, and make sure the cable modem is set to 100-full on the port as well. Auto-negotiate is problematic on some networks.
Like you scenario, I would see it downloaded fine up until a certain amount of data passed through then it would start to run very slow.
Loaded 2.3.9.155 just now and it loaded perfectly without any delay or errors. That is with Ethernet remaining at 100Mbps fixed. I wasn't clear if auto negotiated Ethernet was a good idea when someone claimed it could lockup the hub. That is a non starter for me as I am away from this home for a few months still.
Not sure why this worked fine this time but wanted to pass on since it didn't work well last update (until I rebooted and tried again that is). Maybe this issue is fixed or perhaps its an intermittent issue.
Update 6/25/2024
Loaded 2.3.9.156 this morning and it started with the familiar very slow crawl download. So stopped the upgrade. Then went into Ethernet setting and changed to auto negotiate (from fixed 100) and then reran update. Update completed the download quickly in about 30 seconds. So it does seem to be related to the the Ethernet speed as others have suggested.
I've not heard of that and would not expect it to be a common occurrence. I'd change my hub connection while away via VPN.
If you have a remote hub the best approach is a VPN on your router at the site and the hub on a Wi-Fi plug. That way you can force a reboot of the hub if you are unable to do so via the normal UI or Diagnostic tool. You can use Hubitat's Remote connect feature for most remote activities, but if the hub is truly locked up then that won't be an option.
This I've experienced it a few times over the years, I just toggle my current connection setting (auto to 100 or 100 to auto) and things are fine again.
That’s a good idea. I had several wifi cameras a few years ago that would often lockup. So I put a $6 electric timer on them and had then power off for 15 minutes every night at 3:30AM. That way if they locked up while I was away I’d just lose the camera for a few hours. Band aide for sure but it worked.
That's a pretty clever approach, as long as you have a device that doesn't remind a hard reboot, it works perfectly. Those timers are extremely reliable and inexpensive. Kudos.
Agreed I switched to that and it fixed my most recent download problem (see update earlier in this thread). That will be my standard setting going forward. Thanks.
I had this happen to me a week or so ago. Download was taking 30 seconds per 1% and would timeout. I decided to try a reboot and when the hub came back up there was a new version to download so I didn't know if it was the reboot or the new version. It has only happened once and I think it was on the same hub that locks up every couple of weeks. I just added a weekly reboot to that one.
Just did this for 2.3.9.174 and it worked. Probably does the same thing to the ethernet port as turning the unit off and on again. might have to swap every update.
I’ve recently migrated from a C7 to a C8 Pro and the new hub is exhibited this slow download and fail issue.
Tried switching Ethernet negotiation and rebooting but to no avail.
Plugged into the same switch/cable the C7 used to be on.
I haven’t tried moving it to another switch or using a different cable yet. Given this issue has been known for 6+ months was wondering if anyone was still experiencing it or knew of a fix coming in the pipeline?
I downloaded the fw via the diag tool, but as it had taken so long still I was concerned about the integrity of the data received and so chickened out on hitting restore. I'm assuming it normally does some checksum safety checks before installing, but wasn't sure if doing it via the diag restore skips some/all of those checks.
Now that the diag tool downloaded it, the normal update process I think will find that and then install it with its normal process.
Shut down, unplug/replug the ethernet cable or switching the negotiation setting I think has always resolved the issue for everyone else. So it seems like your problem goes deeper, probably at the network level in your LAN.
So I decided to rule out the Ethernet altogether. I’ve connected it to my WiFi and it’s still exactly the same!
So either my 1Gbps internet pipe accessed via an enterprise level LAN with 10Gbps backbone to the Router is very specifically throttling this download (it’s not!) or the hub has a problem or the server has a problem.
I figured removing the cable and using WiFi rules it out!
I reckon there is a fundamental flaw in the IP stack and poking the Ethernet negotiation just happens to reset it but actually in some environments something is triggering a higher layer of network congestion, hence why I can see it on WiFi too.
I've experienced it a few times over the years, I just toggle my current connection setting (auto to 100 or 100 to auto) and things are fine again.
