I just wish we could check the "Solved" box, sooner than later.
Great point now that I've seen here the backups are the cause. I changed it to a week and will see how that goes.
I think I'll do the same. It didn't occur to me that i can still do daily local backups while doing weekly cloud backups.
Of course, i would like to see if my new rule actually works.
@bravenel This (cloudBackup event) is awesome.
It appears I'm looking at the contents of "%value%" -- "success" is a good thing. Is the only other value "fail" (or is there a way to see what specifically failed)?
I didn't see anything about "cloudBackup" in the documentation (probably since it's brand new). I'm hoping to take advantage of this--but need a bit more info to make sure I do so properly.
Thanks!
Note: It's "FAILED", all caps, lol.
I think this is a good idea.
You could %text% instead of %value% to test why the cloud backup failed.
So, in my test above where I disconnected the internet cable from the router, I got "Failed steps: backup upload to the cloud.
Maybe, and I'm just thinking off the top of my head, you'd (I'd, actually) want to do a router/modem power cycle as well as a hub shutdown/power cycle, since the internet connection stopped working. Of course, it'd be only a one-shot deal, dependent on the cloud backup cycle, but still, maybe.
I know there are Z-wave, some kind of file problem and likely Zigbee errors in %text%, so I think it's worthwhile doing some testing, making notes, and thinking about what's possible.
Ok, @JasonJoel , I changed the cloud backup to every two weeks. Still have the nightly local backup.
So, seeing as I've never had to restore the hub in earnest, knock on wood how would a likely scenario go?
I'm thinking:
-Something is screwed up. Radio not obviously involved.
-Do local restore
-If that doesn't work, or radio might be involved, do cloud restore
-Assume success
-Then, restore from a local backup to get it more current than 2 weeks?
What do you think? Am I on the right track?
Sounds right, except either change "backup" to "restore" on #2 and #4 your list - or change do to use.
That said, I expect I would never need the radio backup unless switching hardware (aka hardware failure). The radio database/config just doesn't typically get corrupted. Could it happen? Sure. But it is exceedingly uncommon.
Has there been any progress in resolving this problem?
If I'm not planning adding any additional devices and did a cloud backup how long would that backup be available or viable for in conjunction with daily local backups?
I just received this message this morning (cloud backup failed -- zWave etc.)--
Turning/unplugging my hub for a few minutes now. But..., I seem to have had problems since the end of December with my zWave iBlinds falling off of the network and then having issues with getting them back on again. One blind will not allow me to exclude it from the network -- weird stuff happening.
Anyone with Z-Wave radio offline after backup, please consider joining the beta program:
If you join the beta, as Bcopeland suggests, you may find your cloud backup issue solved (my C7 cloud backup issue is resolved on the 2.3.8.xxx beta), and if you are still having issues on the beta additional information gathered there will help HE find a final fix. Win/win.
I am part of the beta group. I will give my response there. Thanks
thanks but the hub that is having the both zigbee and zwave offline after a backup is in a remote house and i feel too risky to put on a beta program since i am not there.. let me know if there is anything else i can chime in..
Does the new firmware 2.3.8.117 fix this problem? No obvious mention of it doing so in the release notes.
Yes, we believe that updating the SDK has solved this issue. The feedback we received during beta has been promising, so we hope this issue is resolved :
Adding to this thread. I upgraded to the C8 from C7 in December hoping this problem would go away but I have had to power cycle my hub three times over the last four weeks due to this issue. I do use cloud backup and I have 51 zwave devices connected that all stop working when this failure shows its ugly head. I saw something in the thread that suggested decreasing the frequency of the online backup. How frequently do I need to be making backups provided I don't make any changes? I did install 2.3.8.117 this morning but didn't see anything in the notes about this getting fixed.
Now that you're on 2.3.8.117, you should be sorted on the scheduled cloud backup issue.
If you want to test "now" simply change your backup schedule to 10m in the future and let it run, and see how things go. You should be OK.
Here's to hoping that's true. I am curious though, how does updating the SDK help with a bug in the software?
There was a couple of major changes both fixed different problems that had the same symptom.
-
SDK (Software component provided by SiLabs) was updated, and this had major changes to backup/restore on 800 series Z-Wave.
-
Send queues were paused while backup runs. Some backups were failing when a packet got sent during backup.