Has anyone else seen this issue? This is the 2nd time we have seen the problem in a week.
We factory rest an MX because of it, but even that didn't fix the problem. Support had to "reset" something on their end to resolve. I can tell you it's a real pain to have to factory reset a 48 port switch in production because of some phantom problem like this. Not feeling particularly endearing.
Is it on the same device as before? I've only seen it a few times and with the number of devices I manage it's been a really small issue and a factory reset has always taken care of it for good.
No, it was switch and an MX. Two separate sites.
That's out of an install base of maybe 250 devices total - so still relatively infrequent. But we haven't seen the issue in the past 3 years, and then we see it twice in a week.
I recall that some of the very recent firmware notes made references to resolving an issue with devices not being able to retrieve their configs.
Here it is, in 14.35 for the MX:
"Resolved an issue that could result in devices failing to pull configuration updates from the Meraki Dashboard in rare circumstances"
Interesting, that would explain the MX as it was running 14.35. It doesn't explain the MS though, but maybe we have to just accept it and move on.
I can confirm I’ve resolved this with a couple of client MX devices by ensuring that at least the secondary DNS is google (126.96.36.199). Sometimes the local or ISP DNS does not quickly resolve the cloud so MX cannot connect to the cloud. When using 188.8.131.52, it can find the cloud faster. You might be able to turn that back off once the connection to the cloud is reached. Don’t let downstream devices properly connecting distract you into thinking all is good - there is a difference between administration dashboard & payload routing.