default bootstrap timeout is too low for MAAS environments
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
juju-core |
Fix Released
|
High
|
Jorge Niedbalski |
Bug Description
I was bootstrapping a MAAS environment and hit 10 minutes timeout:
Waiting for address
Attempting to connect to host.domain:22
ERROR bootstrap failed: waited for 10m0s without being able to connect: ssh: connect to host host.domain port 22: Connection refused
Stopping instance...
Bootstrap failed, destroying environment
ERROR waited for 10m0s without being able to connect: ssh: connect to host host.domain port 22: Connection refused
I can work around it using "bootstrap-timeout" configuration option but this default seems too low for MAAS environments. Many physical servers take few minutes to POST and after that they need to deploy OS.
It would be good to have longer timeout used for MAAS environments, 20-30 minutes will likely be enough for most deployments. Another solution could be to prompt user whether to wait a bit longer for bootstrap or give up.
Changed in juju-core: | |
status: | New → Triaged |
importance: | Undecided → High |
tags: | added: bootstrap config maas-provider |
Changed in juju-core: | |
status: | Triaged → Fix Committed |
assignee: | nobody → Jorge Niedbalski (niedbalski) |
Changed in juju-core: | |
assignee: | nobody → Jorge Niedbalski (niedbalski) |
milestone: | none → 1.19.4 |
status: | Confirmed → Fix Committed |
Changed in juju-core: | |
status: | Fix Committed → Fix Released |
We know juju needs to set "bootstrap-timeout: 1800" or more bootstrap. Either the config "juju init" does the right thing, of the provider has a different timeout set.
From the docs.
Environments that need more time to provision an instance can configure 3 options the environments.yaml. MAAS environments often need to set bootstrap- timeout to 1800.
bootstrap-timeout (default: 600s) retry-delay (default: 5s) addresses- delay (default: 10s)
bootstrap-
bootstrap-