Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1788618

Summary: Overcloud Container Health Check Fails
Product: Red Hat OpenStack Reporter: David Rosenfeld <drosenfe>
Component: openstack-tripleo-commonAssignee: Adriano Petrich <apetrich>
Status: CLOSED ERRATA QA Contact: David Rosenfeld <drosenfe>
Severity: medium Docs Contact:
Priority: medium    
Version: 16.0 (Train)CC: aschultz, kecarter, mburns, michele, slinaber
Target Milestone: ---Keywords: Triaged
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2020-06-03 20:43:38 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On: 1779249    
Bug Blocks:    

Description David Rosenfeld 2020-01-07 15:54:19 UTC
Description of problem: Overcloud container health check test fails. In this test sudo systemctl list-units --failed command is used to determine the containers where health checks have failed. Script is seeing following containers with no healthcheck:

Check healtcheck services for podman containers on Overcloud 


podman containers without healthcheck on node: 192.168.24.37
ceph-osd-0
ceph-osd-12
ceph-osd-3
ceph-osd-6
ceph-osd-9

podman containers without healthcheck on node: 192.168.24.32
ceph-osd-11
ceph-osd-14
ceph-osd-2
ceph-osd-5
ceph-osd-7

podman containers without healthcheck on node: 192.168.24.24
ceph-osd-1
ceph-osd-10
ceph-osd-13
ceph-osd-4
ceph-osd-8

podman containers without healthcheck on node: 192.168.24.50
nova_virtlogd

podman containers without healthcheck on node: 192.168.24.16
ceph-mgr-controller-0
ceph-mon-controller-0
haproxy-bundle-podman-0
horizon
ovn-dbs-bundle-podman-0
redis-bundle-podman-0
swift_account_auditor
swift_account_reaper
swift_account_replicator
swift_container_auditor
swift_container_replicator
swift_container_updater
swift_object_auditor
swift_object_expirer
swift_object_replicator
swift_object_updater
swift_rsync

podman containers without healthcheck on node: 192.168.24.25
ceph-mgr-controller-1
ceph-mon-controller-1
haproxy-bundle-podman-1
ovn-dbs-bundle-podman-1
redis-bundle-podman-1
swift_account_auditor
swift_account_reaper
swift_account_replicator
swift_container_auditor
swift_container_replicator
swift_container_updater
swift_object_auditor
swift_object_expirer
swift_object_replicator
swift_object_updater
swift_rsync

podman containers without healthcheck on node: 192.168.24.7
ceph-mgr-controller-2
ceph-mon-controller-2
haproxy-bundle-podman-2
horizon
ovn-dbs-bundle-podman-2
redis-bundle-podman-2
swift_account_auditor
swift_account_reaper
swift_account_replicator
swift_container_auditor
swift_container_replicator
swift_container_updater
swift_object_auditor
swift_object_expirer
swift_object_replicator
swift_object_updater
swift_rsync

podman containers without healthcheck on node: 192.168.24.30
clustercheck
galera-bundle-podman-0

podman containers without healthcheck on node: 192.168.24.8
clustercheck
galera-bundle-podman-1

podman containers without healthcheck on node: 192.168.24.14
clustercheck
galera-bundle-podman-2

podman containers without healthcheck on node: 192.168.24.29
rabbitmq-bundle-podman-0

podman containers without healthcheck on node: 192.168.24.27
rabbitmq-bundle-podman-1

podman containers without healthcheck on node: 192.168.24.31
rabbitmq-bundle-podman-2



Version-Release number of selected component (if applicable): RHOS_TRUNK-16.0-RHEL-8-20200103.n.1


How reproducible: Every time DFG-df-deployment-16-virthost-3cont_1comp_3ceph_3db_2net_3msg-yes_UC_SSL-yes_OC_SSL-ceph-ipv6-geneve-RHELOSP-31893 job is executed.


Steps to Reproduce:
1. Execute DFG-df-deployment-16-virthost-3cont_1comp_3ceph_3db_2net_3msg-yes_UC_SSL-yes_OC_SSL-ceph-ipv6-geneve-RHELOSP-31893 in Jenkins
2.
3.

Actual results: overcloud container health check test case fails due to containers with no healtcheck.


Expected results: overcloud container health check test case passes


Additional info:

Comment 1 David Rosenfeld 2020-01-07 15:55:25 UTC
Logs to a failing test:

https://rhos-qe-jenkins.rhev-ci-vms.eng.rdu2.redhat.com/view/DFG/view/df/view/deployment/job/DFG-df-deployment-16-virthost-3cont_1comp_3ceph_3db_2net_3msg-yes_UC_SSL-yes_OC_SSL-ceph-ipv6-geneve-RHELOSP-31893/8/

Look at pytest_runner / pytest-repo.tripleo.container_sanity_v3.Overcloud_Testing.test_check_helthcheck_status_overcloud failure

Comment 2 Alex Schultz 2020-01-07 16:56:13 UTC
We've addressed the clustercheck, swift and horizon health checks upstream. I'm uncertain about the ceph/rabbit/galera health checks. I think these might need to be excluded from the test itself.

Comment 3 Michele Baldessari 2020-01-07 17:39:32 UTC
(In reply to Alex Schultz from comment #2)
> We've addressed the clustercheck, swift and horizon health checks upstream.
> I'm uncertain about the ceph/rabbit/galera health checks. I think these
> might need to be excluded from the test itself.

+1 any container that has '-bundle' in its name needs to be excluded since it is being helthchecked by pacemaker with its own mechanism

Comment 4 David Rosenfeld 2020-01-31 14:43:43 UTC
After podman overcloud healthcheck script was updated to not report status on containers with format: -bundle-, ceph-, and swift_ the overcloud healthcheck still fails.


The script reports:

Check healtcheck services for podman containers on Overcloud 

NODE: 192.168.24.32: 

● tripleo_horizon_healthcheck.service - horizon healthcheck
   Loaded: loaded (/etc/systemd/system/tripleo_horizon_healthcheck.service; disabled; vendor preset: disabled)
   Active: failed (Result: exit-code) since Fri 2020-01-31 13:33:01 UTC; 17s ago
  Process: 560167 ExecStart=/usr/bin/podman exec horizon /openstack/healthcheck (code=exited, status=127)
 Main PID: 560167 (code=exited, status=127)
Jan 31 13:33:00 controller-0 systemd[1]: Starting horizon healthcheck...
Jan 31 13:33:01 controller-0 podman[560167]: Error: exec failed: container_linux.go:346: starting container process caused "exec: \"/openstack/healthcheck\": stat /openstack/healthcheck: no such file or directory": OCI runtime command not found error
Jan 31 13:33:01 controller-0 systemd[1]: tripleo_horizon_healthcheck.service: Main process exited, code=exited, status=127/n/a
Jan 31 13:33:01 controller-0 systemd[1]: tripleo_horizon_healthcheck.service: Failed with result 'exit-code'.
Jan 31 13:33:01 controller-0 systemd[1]: Failed to start horizon healthcheck.

192.168.24.32 is controller-0. Other nodes pass(i.e.: compute-0, ceph-0, ceph-1, and ceph-2.)

This is why script fails for controller node:

[heat-admin@controller-0 ~]$ sudo systemctl list-units --failed --plain --no-legend
dhcp-interface   loaded failed failed DHCP interface ovs-system
tripleo_horizon_healthcheck.service loaded failed failed horizon healthcheck    <----- doesn't show failed on other nodes

Comment 6 Radomir Dopieralski 2020-11-16 13:57:12 UTC
*** Bug 1793881 has been marked as a duplicate of this bug. ***