Downtime this morning?

Anvil Q&A
1

Did anyone else experience a few minutes of Anvil downtime this morning (around 11am EST)? I was logged out of the Anvil backend and our app wasn’t reachable.

Thank you,

Jeff

2

I did.

2-3 minutes with apps working with some hiccups. They were slow or non responding, an app complained about the transaction non existing, another app showed this error

RuntimeUnavailableError: Server resources not available for 'python3-full'. Please contact support@anvil.works.

then I refreshed and everything restarted to work well.

3

Yeah, it’s been resolved now.

1 Like
4

Any word from Anvil HQ on the issue? We ran into the same errors that Stefano mentioned. We’re ramping up our usage of the tool so any hiccups cause friction–especially if I can’t explain them.

Jeff

1 Like
5

I emailed support and got a reply saying there was a short outage which had been fixed.

1 Like
6

and again just a minute ago for about a minute?
(at 01:50 UTC)

7

Hi all,

We did experience a few minutes’ degradation yesterday, starting around 14:25 UTC. 99th-centile latencies were unusually high but remained within limits. Unfortunately, once we had diagnosed the issue, the resolution required hard-restarting some components, and that would have interrupted ongoing sessions for a minute or two.

(For the record, we detected no anomalies overnight, and a squint at the metrics reveals nothing fishy-looking at 01:50 UTC.)

8

Thanks for the official response Meredydd! Long-term, it may be nice to have a dashboard and post-mortem for any degredations/outages.