mgr/cephadm: handle HostConnectionError when checking for valid addr

[adk3798]

Otherwise, the error is not properly passed back up the chain and the user can get an error message like

TypeError: __init__() missing 2 required positional arguments: 'hostname' and 'addr'

when trying to add a host, despite the actual problem being

cephadm.ssh.HostConnectionError: Failed to connect to vm-01 (192.168.122.248). Permission denied

The tracker shows a bit more, but generally trying to add a host that doesn't have the proper pub-key set as an authorized key will get a misleasing error message. With this patch, the error message looks like

[ceph: root@vm-00 /]# ceph orch host add vm-01 192.168.122.29
Error EINVAL: Failed to connect to vm-01 (192.168.122.29). Permission denied
Log: Opening SSH connection to 192.168.122.29, port 22
[conn=1] Connected to SSH server at 192.168.122.29, port 22
[conn=1]   Local address: 192.168.122.156, port 49552
[conn=1]   Peer address: 192.168.122.29, port 22
[conn=1] Beginning auth for user root
[conn=1] Auth failed for user root
[conn=1] Connection failure: Permission denied
[conn=1] Aborting connection

which is much more useful

Fixes: https://tracker.ceph.com/issues/59081

Contribution Guidelines

• To sign and title your commits, please refer to Submitting Patches to Ceph.

• If you are submitting a fix for a stable branch (e.g. "pacific"), please refer to Submitting Patches to Ceph - Backports for the proper workflow.

Checklist

• Tracker (select at least one)
  • References tracker ticket
  • Very recent bug; references commit where it was introduced
  • New feature (ticket optional)
  • Doc update (no ticket needed)
  • Code cleanup (no ticket needed)
• Component impact
  • Affects Dashboard, opened tracker ticket
  • Affects Orchestrator, opened tracker ticket
  • No impact that needs to be tracked
• Documentation (select at least one)
  • Updates relevant documentation
  • No doc update is appropriate
• Tests (select at least one)
  • Includes unit test(s)
  • Includes integration test(s)
  • Includes bug reproducer
  • No tests

Show available Jenkins commands

• jenkins retest this please
• jenkins test classic perf
• jenkins test crimson perf
• jenkins test signed
• jenkins test make check
• jenkins test make check arm64
• jenkins test submodules
• jenkins test dashboard
• jenkins test dashboard cephadm
• jenkins test api
• jenkins test docs
• jenkins render docs
• jenkins test ceph-volume all
• jenkins test ceph-volume tox
• jenkins test windows

[phlogistonjohn]

Looks OK to me

[adk3798]

jenkins retest this please

[rkachach]

LGTM

[adk3798]

https://pulpito.ceph.com/adking-2023-03-15_01:34:51-orch:cephadm-wip-adk-testing-2023-03-14-1905-distro-default-smithi/

reruns of failed/dead jobs: https://pulpito.ceph.com/adking-2023-03-30_12:19:06-orch:cephadm-wip-adk-testing-2023-03-28-1716-distro-default-smithi/

After reruns, 6 failures:

• 2 failed staggered upgrade test, combination of https://tracker.ceph.com/issues/58535 plus the attempted fix not working
• 2 failed nfs-keepalive-only tests, newly added by a PR in the run, and not an issue in terms of merging any other PR
• 1 instance of https://tracker.ceph.com/issues/49287 (known issue)
• 1 failure in test_nfs task. The failure only happens in the test variation with the agent on (https://pulpito.ceph.com/adking-2023-03-29_11:57:55-orch:cephadm-wip-adk-testing-2023-03-28-1716-distro-default-smithi/7225304 passed) so I believe this is a timing thing related to the agent. Nothing to block merging on anything not specifically related to nfs.