`flyctl postgres create` fails

Questions / Help
1

I’m unable to deploy a new PostgreSQL cluster to LHR. I’ve tried twice and the Monitoring Deployment status just hangs with no output for a few minutes until it eventually aborts and reports:

Monitoring Deployment
Error 1 error occurred:
	* No deployment available to monitor

The logs look like this:

2021-06-19T13:51:30.634642822Z app[4003f5eb] lhr [info] proxy    | 2021-06-19T13:51:30.631Z	INFO	cmd/proxy.go:286	proxying to master address	{"address": "[fdaa:0:2109:a7b:a98:0:2617:2]:5433"}
2021-06-19T13:51:31.089803309Z app[c96449af] lhr [info] keeper   | 2021-06-19T13:51:31.088Z	INFO	cmd/keeper.go:1505	our db requested role is master
2021-06-19T13:51:31.091423138Z app[c96449af] lhr [info] keeper   | 2021-06-19T13:51:31.090Z	INFO	cmd/keeper.go:1543	already master
2021-06-19T13:51:31.106579280Z app[c96449af] lhr [info] keeper   | 2021-06-19T13:51:31.105Z	INFO	cmd/keeper.go:1676	postgres parameters not changed
2021-06-19T13:51:31.107883775Z app[c96449af] lhr [info] keeper   | 2021-06-19T13:51:31.106Z	INFO	cmd/keeper.go:1703	postgres hba entries not changed
2021-06-19T13:51:32.709654755Z app[4003f5eb] lhr [info] keeper   | 2021-06-19T13:51:32.708Z	INFO	cmd/keeper.go:1557	our db requested role is standby	{"followedDB": "fb7d11c8"}
2021-06-19T13:51:32.710752740Z app[4003f5eb] lhr [info] keeper   | 2021-06-19T13:51:32.709Z	INFO	cmd/keeper.go:1576	already standby
2021-06-19T13:51:32.726536549Z app[4003f5eb] lhr [info] keeper   | 2021-06-19T13:51:32.725Z	INFO	cmd/keeper.go:1676	postgres parameters not changed
2021-06-19T13:51:32.727830121Z app[4003f5eb] lhr [info] keeper   | 2021-06-19T13:51:32.726Z	INFO	cmd/keeper.go:1703	postgres hba entries not changed
2021-06-19T13:51:33.500734387Z app[c96449af] lhr [info] proxy    | 2021-06-19T13:51:33.497Z	INFO	cmd/proxy.go:268	master address	{"address": "[fdaa:0:2109:a7b:a98:0:2617:2]:5433"}
2021-06-19T13:51:33.717012723Z app[c96449af] lhr [info] proxy    | 2021-06-19T13:51:33.713Z	INFO	cmd/proxy.go:286	proxying to master address	{"address": "[fdaa:0:2109:a7b:a98:0:2617:2]:5433"}

That just repeats over and over.

I’m running flyctl v0.0.227.

2

Exact same issue in AMS, for the record.

3

I’m sorry you’ve been having troubles. The “No deployment available” error means flyctl gave up waiting for a deployment to start. That typically means a region is having hardware, network, or capacity issues, which shouldn’t normally happen. We’ll investigate.

Based on the logs, the pg cluster was created and ready to accept connections, it probably just took longer than expected. Those repeated lines are heartbeats between the two nodes. You can see the health of the cluster with flyctl status which will show if the cluster is unhealthy.

1 Like