gabrioux-2024-09-16_13:13:56-orch:cephadm-wip-guits-main-2024-09-13-1248-distro-default-smithi

"2024-09-16T13:35:11.847325+0000 mon.a (mon.0) 209 : cluster [WRN] Health check failed: 1/3 mons down, quorum a,c (MON_DOWN)" in cluster log

SELinux denials found on ubuntu@smithi063.front.sepia.ceph.com: ['type=AVC msg=audit(1726494165.677:10846): avc: denied { nlmsg_read } for pid=61066 comm="ss" scontext=system_u:system_r:container_t:s0:c64,c509 tcontext=system_u:system_r:container_t:s0:c64,c509 tclass=netlink_tcpdiag_socket permissive=1']

Command failed on smithi098 with status 5: 'sudo /home/ubuntu/cephtest/cephadm --image quay-quay-quay.apps.os.sepia.ceph.com/ceph-ci/ceph:8293d73f8690540e843a81caec373f9cc29cf705 shell -c /etc/ceph/ceph.conf -k /etc/ceph/ceph.client.admin.keyring --fsid 3240e16a-7431-11ef-bceb-c7b262605968 -- bash -c \'set -e\nset -x\nceph orch apply node-exporter\nceph orch apply grafana\nceph orch apply alertmanager\nceph orch apply prometheus\nsleep 240\nceph orch ls\nceph orch ps\nceph orch host ls\nMON_DAEMON=$(ceph orch ps --daemon-type mon -f json | jq -r \'"\'"\'last | .daemon_name\'"\'"\')\nGRAFANA_HOST=$(ceph orch ps --daemon-type grafana -f json | jq -e \'"\'"\'.[]\'"\'"\' | jq -r \'"\'"\'.hostname\'"\'"\')\nPROM_HOST=$(ceph orch ps --daemon-type prometheus -f json | jq -e \'"\'"\'.[]\'"\'"\' | jq -r \'"\'"\'.hostname\'"\'"\')\nALERTM_HOST=$(ceph orch ps --daemon-type alertmanager -f json | jq -e \'"\'"\'.[]\'"\'"\' | jq -r \'"\'"\'.hostname\'"\'"\')\nGRAFANA_IP=$(ceph orch host ls -f json | jq -r --arg GRAFANA_HOST "$GRAFANA_HOST" \'"\'"\'.[] | select(.hostname==$GRAFANA_HOST) | .addr\'"\'"\')\nPROM_IP=$(ceph orch host ls -f json | jq -r --arg PROM_HOST "$PROM_HOST" \'"\'"\'.[] | select(.hostname==$PROM_HOST) | .addr\'"\'"\')\nALERTM_IP=$(ceph orch host ls -f json | jq -r --arg ALERTM_HOST "$ALERTM_HOST" \'"\'"\'.[] | select(.hostname==$ALERTM_HOST) | .addr\'"\'"\')\n# check each host node-exporter metrics endpoint is responsive\nALL_HOST_IPS=$(ceph orch host ls -f json | jq -r \'"\'"\'.[] | .addr\'"\'"\')\nfor ip in $ALL_HOST_IPS; do\n curl -s http://${ip}:9100/metric\ndone\n# check grafana endpoints are responsive and database health is okay\ncurl -k -s https://${GRAFANA_IP}:3000/api/health\ncurl -k -s https://${GRAFANA_IP}:3000/api/health | jq -e \'"\'"\'.database == "ok"\'"\'"\'\n# stop mon daemon in order to trigger an alert\nceph orch daemon stop $MON_DAEMON\nsleep 120\n# check prometheus endpoints are responsive and mon down alert is firing\ncurl -s http://${PROM_IP}:9095/api/v1/status/config\ncurl -s http://${PROM_IP}:9095/api/v1/status/config | jq -e \'"\'"\'.status == "success"\'"\'"\'\ncurl -s http://${PROM_IP}:9095/api/v1/alerts\ncurl -s http://${PROM_IP}:9095/api/v1/alerts | jq -e \'"\'"\'.data | .alerts | .[] | select(.labels | .alertname == "CephMonDown") | .state == "firing"\'"\'"\'\n# check alertmanager endpoints are responsive and mon down alert is active\ncurl -s http://${ALERTM_IP}:9093/api/v1/status\ncurl -s http://${ALERTM_IP}:9093/api/v1/alerts\ncurl -s http://${ALERTM_IP}:9093/api/v1/alerts | jq -e \'"\'"\'.data | .[] | select(.labels | .alertname == "CephMonDown") | .status | .state == "active"\'"\'"\'\n\''

reached maximum tries (50) after waiting for 300 seconds

Command failed on smithi029 with status 1: 'sudo /home/ubuntu/cephtest/cephadm --image quay.ceph.io/ceph-ci/ceph:squid shell -c /etc/ceph/ceph.conf -k /etc/ceph/ceph.client.admin.keyring --fsid acb34024-7430-11ef-bceb-c7b262605968 -e sha1=8293d73f8690540e843a81caec373f9cc29cf705 -- bash -c \'ceph versions | jq -e \'"\'"\'.rgw | length == 1\'"\'"\'\''

"2024-09-16T13:52:16.966394+0000 mon.a (mon.0) 1352 : cluster [WRN] Health check failed: 1 stray daemon(s) not managed by cephadm (CEPHADM_STRAY_DAEMON)" in cluster log

Command failed on smithi079 with status 1: 'sudo /home/ubuntu/cephtest/cephadm --image quay-quay-quay.apps.os.sepia.ceph.com/ceph-ci/ceph:8293d73f8690540e843a81caec373f9cc29cf705 shell -c /etc/ceph/ceph.conf -k /etc/ceph/ceph.client.admin.keyring --fsid 3f7f3732-7431-11ef-bceb-c7b262605968 -- bash -c \'set -e\nset -x\nwhile true; do TOKEN=$(ceph rgw realm tokens | jq -r \'"\'"\'.[0].token\'"\'"\'); echo $TOKEN; if [ "$TOKEN" != "master zone has no endpoint" ]; then break; fi; sleep 5; done\nTOKENS=$(ceph rgw realm tokens)\necho $TOKENS | jq --exit-status \'"\'"\'.[0].realm == "myrealm1"\'"\'"\'\necho $TOKENS | jq --exit-status \'"\'"\'.[0].token\'"\'"\'\nTOKEN_JSON=$(ceph rgw realm tokens | jq -r \'"\'"\'.[0].token\'"\'"\' | base64 --decode)\necho $TOKEN_JSON | jq --exit-status \'"\'"\'.realm_name == "myrealm1"\'"\'"\'\necho $TOKEN_JSON | jq --exit-status \'"\'"\'.endpoint | test("http://.+:\\\\d+")\'"\'"\'\necho $TOKEN_JSON | jq --exit-status \'"\'"\'.realm_id | test("^[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}$")\'"\'"\'\necho $TOKEN_JSON | jq --exit-status \'"\'"\'.access_key\'"\'"\'\necho $TOKEN_JSON | jq --exit-status \'"\'"\'.secret\'"\'"\'\n\''

reached maximum tries (50) after waiting for 300 seconds

Command failed on smithi003 with status 1: 'sudo /home/ubuntu/cephtest/cephadm --image quay.ceph.io/ceph-ci/ceph:squid shell -c /etc/ceph/ceph.conf -k /etc/ceph/ceph.client.admin.keyring --fsid 101aa070-7432-11ef-bceb-c7b262605968 -e sha1=8293d73f8690540e843a81caec373f9cc29cf705 -- bash -c \'ceph versions | jq -e \'"\'"\'.rgw | length == 1\'"\'"\'\''

reached maximum tries (50) after waiting for 300 seconds

Command failed on smithi123 with status 1: 'sudo /home/ubuntu/cephtest/cephadm --image quay.ceph.io/ceph-ci/ceph:squid shell -c /etc/ceph/ceph.conf -k /etc/ceph/ceph.client.admin.keyring --fsid 023ad1fa-7432-11ef-bceb-c7b262605968 -e sha1=8293d73f8690540e843a81caec373f9cc29cf705 -- bash -c \'ceph versions | jq -e \'"\'"\'.rgw | length == 1\'"\'"\'\''

"2024-09-16T13:58:40.674134+0000 mon.a (mon.0) 789 : cluster [WRN] Health check failed: 2 stray daemon(s) not managed by cephadm (CEPHADM_STRAY_DAEMON)" in cluster log

Command failed on smithi162 with status 1: 'sudo /home/ubuntu/cephtest/cephadm --image quay.io/ceph/ceph:v18.2.1 shell -c /etc/ceph/ceph.conf -k /etc/ceph/ceph.client.admin.keyring --fsid 815f7116-7432-11ef-bceb-c7b262605968 -e sha1=8293d73f8690540e843a81caec373f9cc29cf705 -- bash -c \'ceph versions | jq -e \'"\'"\'.overall | length == 1\'"\'"\'\''

Command failed on smithi002 with status 5: 'sudo /home/ubuntu/cephtest/cephadm --image quay-quay-quay.apps.os.sepia.ceph.com/ceph-ci/ceph:8293d73f8690540e843a81caec373f9cc29cf705 shell -c /etc/ceph/ceph.conf -k /etc/ceph/ceph.client.admin.keyring --fsid a2c39bca-7432-11ef-bceb-c7b262605968 -- bash -c \'set -e\nset -x\nceph orch apply node-exporter\nceph orch apply grafana\nceph orch apply alertmanager\nceph orch apply prometheus\nsleep 240\nceph orch ls\nceph orch ps\nceph orch host ls\nMON_DAEMON=$(ceph orch ps --daemon-type mon -f json | jq -r \'"\'"\'last | .daemon_name\'"\'"\')\nGRAFANA_HOST=$(ceph orch ps --daemon-type grafana -f json | jq -e \'"\'"\'.[]\'"\'"\' | jq -r \'"\'"\'.hostname\'"\'"\')\nPROM_HOST=$(ceph orch ps --daemon-type prometheus -f json | jq -e \'"\'"\'.[]\'"\'"\' | jq -r \'"\'"\'.hostname\'"\'"\')\nALERTM_HOST=$(ceph orch ps --daemon-type alertmanager -f json | jq -e \'"\'"\'.[]\'"\'"\' | jq -r \'"\'"\'.hostname\'"\'"\')\nGRAFANA_IP=$(ceph orch host ls -f json | jq -r --arg GRAFANA_HOST "$GRAFANA_HOST" \'"\'"\'.[] | select(.hostname==$GRAFANA_HOST) | .addr\'"\'"\')\nPROM_IP=$(ceph orch host ls -f json | jq -r --arg PROM_HOST "$PROM_HOST" \'"\'"\'.[] | select(.hostname==$PROM_HOST) | .addr\'"\'"\')\nALERTM_IP=$(ceph orch host ls -f json | jq -r --arg ALERTM_HOST "$ALERTM_HOST" \'"\'"\'.[] | select(.hostname==$ALERTM_HOST) | .addr\'"\'"\')\n# check each host node-exporter metrics endpoint is responsive\nALL_HOST_IPS=$(ceph orch host ls -f json | jq -r \'"\'"\'.[] | .addr\'"\'"\')\nfor ip in $ALL_HOST_IPS; do\n curl -s http://${ip}:9100/metric\ndone\n# check grafana endpoints are responsive and database health is okay\ncurl -k -s https://${GRAFANA_IP}:3000/api/health\ncurl -k -s https://${GRAFANA_IP}:3000/api/health | jq -e \'"\'"\'.database == "ok"\'"\'"\'\n# stop mon daemon in order to trigger an alert\nceph orch daemon stop $MON_DAEMON\nsleep 120\n# check prometheus endpoints are responsive and mon down alert is firing\ncurl -s http://${PROM_IP}:9095/api/v1/status/config\ncurl -s http://${PROM_IP}:9095/api/v1/status/config | jq -e \'"\'"\'.status == "success"\'"\'"\'\ncurl -s http://${PROM_IP}:9095/api/v1/alerts\ncurl -s http://${PROM_IP}:9095/api/v1/alerts | jq -e \'"\'"\'.data | .alerts | .[] | select(.labels | .alertname == "CephMonDown") | .state == "firing"\'"\'"\'\n# check alertmanager endpoints are responsive and mon down alert is active\ncurl -s http://${ALERTM_IP}:9093/api/v1/status\ncurl -s http://${ALERTM_IP}:9093/api/v1/alerts\ncurl -s http://${ALERTM_IP}:9093/api/v1/alerts | jq -e \'"\'"\'.data | .[] | select(.labels | .alertname == "CephMonDown") | .status | .state == "active"\'"\'"\'\n\''

Command failed on smithi100 with status 1: 'sudo /home/ubuntu/cephtest/cephadm --image quay.ceph.io/ceph-ci/ceph:squid shell -c /etc/ceph/ceph.conf -k /etc/ceph/ceph.client.admin.keyring --fsid cdcbf970-7432-11ef-bceb-c7b262605968 -e sha1=8293d73f8690540e843a81caec373f9cc29cf705 -- bash -c \'ceph versions | jq -e \'"\'"\'.mgr | length == 1\'"\'"\'\''

Command failed on smithi050 with status 1: 'sudo /home/ubuntu/cephtest/cephadm --image quay.ceph.io/ceph-ci/ceph:squid shell -c /etc/ceph/ceph.conf -k /etc/ceph/ceph.client.admin.keyring --fsid 416e6b7e-7433-11ef-bceb-c7b262605968 -e sha1=8293d73f8690540e843a81caec373f9cc29cf705 -- bash -c \'ceph versions | jq -e \'"\'"\'.rgw | length == 1\'"\'"\'\''