Twice I have ran into this issue updating C7 on the 2.2.9 code has failed. Moving from 2.2.8 to 2.2.9 and now 2.2.9 build 128 to 129. C5 updates without issue and it is quick. C7 never completes. First time major build update. It failed I had to pull the power after 15 minutes of "updating". Revert to old Platform, soft reset, (corrupted DB) restore from cloud back up and then apply update again, this worked.
Updating from 2.2.9 build 128 to 129.
C7 and C5 was up for less than 24 hours. C5 No issue updating
C7 Fails on update again. Second attempt, rebooted through diagnostics. Hub console still didn't' come up. Connected through the diagnostics again. Reverted Platform build 128. Didn't see any issue this time with the DB. Proceed with update again. This time it applied correctly.
I'm noticing the last few Platform builds that memory consumption has gone up and degrades faster over time. I don’t know what areas of the memory are being affected (kernel, system, device, etc). C5 and in the C7 and the update process seems to be affecting C7 and not the C5. Note most of my devices and services are on c5 and all the S2 devices are C7.
Coming to my conclusion is I wonder if reboot c7 prior to the update. Would I have seen this issue? Is there a memory leaking issue that affects c7 and updating it?
As far as I can tell this is pretty consistent problem with the updating issue.
This may be a UI issue we just found. Give it a few minutes, and see if the hub comes up at its IP address. The UI problem causes the update progression updates to stop showing, giving the impression of failed update, when in fact it continues to progress.
Hub info shows the correct IP. I don't see any change. If I attempt to connect to the web console, I receive a broken page. I have left in that state for unusually long time well over 15 minutes +. No response. Diagnostic page showed it was update. However, I was unable to communicate with the hub other than Diagnostic.
It even showed up in the remote admin. Couldn't connect to console.
The first time I even saw the LED light go from blue and green and still no response.
Second time it was like the web page time out and got a broke webpage again IP was correct and Diags showed updated.
I couldn't get to the web interface. Page cannot be displayed. It didn't jump it was pingable and reserved in DHCP. Yes, I did try find my hub and it found it on the correct IP Address.
The web console was completely unresponsive.
Diagnostic wouldn't have come up on the device if the IP address had changed. You are access via IP on 8081 and the diagnostic page had the correct IP in it too.
When 2.2.8 > 2.2.9 yes reverted back to the last platform (2.2.8). it stated DB corrupted. Soft Reset and restored DB from a few hours ago. Rebooted 2.2.8 applied 2.2.9 right after that. no issue.
Second on from build 128 >129. I revert back to 2.2.9 128 after the unresponsiveness. Everything came back OK. (no DB Change). Applied 2.2.9 129 again. All working.
Only thing I see is the rebooting before applying is the ONLY change I can see.
I assuming that there isn't any logging of this process to understand what may have happened. There should be some sort diagnostic logging of updating. Even if it is a small text to download and review from the Diag page.
I appreciate your help and input. I'm just trying to make sense of this too.