Notice: /Stage[main]/Phabricator::Vcs/Base::Service_unit[ssh-phab]/Service[ssh-phab]/ensure: ensure changed 'stopped' to 'running'
From journalctl:
May 05 14:48:27 deployment-phab01 sshd[30653]: error: Bind to port 22 on 10.68.18.216 failed: Address already in use. May 05 14:48:27 deployment-phab01 sshd[30653]: fatal: Cannot bind any address.
Seems it listens on a secondary IP address and might rely on LVS. Anyway that does not work on labs??
The tasks was originally filled for some puppet failures:
Both deployment-phab01 and deployment-phab02 fail with:
Error: Could not retrieve catalog from remote server: Error 400 on SERVER: [{"environment"=>"www", "owner"=>"root", "group"=>"www-data", "phab_settings"=>{"mysql.user"=>"root", "mysql.pass"=>"labspass"}}, {"environment"=>"phd", "owner"=>"root", "group"=>"phd", "phab_settings"=>{"mysql.user"=>"root", "mysql.pass"=>"labspass"}}] is not a Hash. It looks to be a Array at /etc/puppet/modules/phabricator/manifests/init.pp:68 on node deployment-phab02.deployment-prep.eqiad.wmflabs
Reason is modules/role/manifests/phabricator/labs.pp pass an array instead of a hash of (name => config hash).
Used to fail with:
Error: /Stage[main]/Phabricator::Vcs/File[/etc/systemd/system/ssh-phab.service]: Could not evaluate: Could not retrieve information from environment production source(s) puppet:///modules/phabricator/sshd-phab.service