Description of problem: Attempting to delete a host in the web ui while the capsule that it's registered to is down fails. Satellite tries to contact the capsule to delete the SSH known host key, fails to contact the capsule because it's down and host is not deleted. Version-Release number of selected component (if applicable): satellite-6.8.1-1.el7sat.noarch How reproducible: Always Steps to Reproduce: 1. Register a host to a capsule 2. Run a remote command against the host 3. Shutdown the capsule 4. Attempt to delete the host Actual results: Via web ui, error message "Failed to delete <hostname>: []" is displayed. Via hammer, "Host deleted." is returned with a 0 exit code but the host is not deleted. Expected results: Host is deleted. The capsule may have been removed/destroyed and may no longer be available. This should not block deletion of hosts. Additional info:
2020-12-14T16:13:13 [W|app|] Remove SSH known hosts for host.example.com task failed with the following error: ERF12-6886 [ProxyAPI::ProxyException]: Unable to remove host from known hosts ([SocketError]: Failed to open TCP connection to capsule.example.com:9090 (getaddrinfo: Name or service not known)) for Capsule https://capsule.example.com:9090/ssh 2020-12-14T16:13:13 [W|app|] Rolling back due to a problem: [#<Orchestration::Task:0x0000000011f995f0 @name="Remove SSH known hosts for host.example.com", @id="ssh_remove_known_hosts_interface_10.33.8.73_2", @status="failed", @priority=200, @action=[#<Nic::Bridge id: 15, mac: "52:54:00:14:4d:08", ip: "10.33.8.73", type: "Nic::Bridge", name: "host.example.com", host_id: 8, subnet_id: 1, domain_id: 1, attrs: {"bridge"=>true}, created_at: "2020-12-14 14:51:07", updated_at: "2020-12-14 15:03:18", provider: nil, username: nil, password: nil, virtual: true, link: true, identifier: "br0", tag: "", attached_to: "", managed: true, mode: "balance-rr", attached_devices: "", bond_options: "", primary: true, provision: true, compute_attributes: {}, execution: true, ip6: "", subnet6_id: nil>, :drop_from_known_hosts, 2], @created=1607962393.0890071, @timestamp=2020-12-14 16:13:13 UTC>] 2020-12-14T16:13:13 [I|bac|] Task {label: Actions::Katello::Host::Destroy, id: e81809da-1c33-4a0c-84ae-ab20536a62e9, execution_plan_id: a8f006df-a5ac-420d-90d3-b49336a39e42} state changed: stopped result: success 2020-12-14T16:13:13 [I|bac|] Task {label: Actions::Katello::Host::Destroy, id: e81809da-1c33-4a0c-84ae-ab20536a62e9, execution_plan_id: a8f006df-a5ac-420d-90d3-b49336a39e42} state changed: stopped result: success 2020-12-14T16:13:13 [I|bac|] Task {label: Actions::BulkAction, id: 22c644c1-e0d7-49c2-88f0-514ca9140cf1, execution_plan_id: 9d74e080-ea14-4eb0-8c8d-ca6e30a39790} state changed: stopped result: success 2020-12-14T16:13:13 [I|bac|] Task {label: Actions::BulkAction, id: 22c644c1-e0d7-49c2-88f0-514ca9140cf1, execution_plan_id: 9d74e080-ea14-4eb0-8c8d-ca6e30a39790} state changed: stopped result: success
*** Bug 1942366 has been marked as a duplicate of this bug. ***
On another thought, REX has diverged quite a bit since 6.9 went out and a cherry-pick would be better. The patches should still apply cleanly, if not, feel free to ping me.
Verified on Satellite 6.9.5 sn 2, host can be successfully removed even if the rex capsule it was registered to is down
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Satellite 6.9.5 Async Bug Fix Update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2021:3387