Sometimes grafana-agent units (not necessarily all of them at once) get stuck in waiting status with message Waiting for TLS certificate.. This is related to the relation with grafana, and resolves if I bounce the relation between grafana and grafana-agent.
To Reproduce
This was observed in a test environment while testing something else; I don't have a neat reproducer yet. Opening the issue in case others have also observed it, and will fill in more info if I have time to investigate further. :)
Environment
Not complete environment, but this may help:
$ juju status grafana-agent
Model Controller Cloud/Region Version SLA Timestamp
teststack serverstack serverstack/serverstack 3.4.0 unsupported 14:28:18+10:30
SAAS Status Store URL
grafana active overcloud-microk8s admin/cos.grafana
loki active overcloud-microk8s admin/cos.loki
prometheus active overcloud-microk8s admin/cos.prometheus
App Version Status Scale Charm Channel Rev Exposed Message
cinder 20.3.1 active 1 cinder yoga/stable 664 no Unit is ready
cinder-mysql-router 8.0.36 active 0 mysql-router 8.0/stable 137 no Unit is ready
glance 24.2.1 active 1 glance yoga/stable 594 no Unit is ready
glance-mysql-router 8.0.36 active 0 mysql-router 8.0/stable 137 no Unit is ready
grafana-agent waiting 7 grafana-agent edge 77 no Waiting for TLS certificate.
mysql 8.0.36 active 3 mysql-innodb-cluster 8.0/stable 107 no Unit is ready: Mode: R/O, Cluster is ONLINE and can tolerate up to ONE failure.
nova-compute 25.2.1 active 1 nova-compute yoga/stable 722 no Unit is ready
openstack-exporter active 1 openstack-exporter 32 no
ovn-chassis 22.03.3 active 0 ovn-chassis 22.03/stable 222 no Unit is ready
Unit Workload Agent Machine Public address Ports Message
cinder/0* active idle 0 10.5.0.223 8776/tcp Unit is ready
cinder-mysql-router/0* active idle 10.5.0.223 Unit is ready
grafana-agent/16* waiting idle 10.5.0.223 Waiting for TLS certificate.
glance/0* active idle 1 10.5.1.75 9292/tcp Unit is ready
glance-mysql-router/0* active idle 10.5.1.75 Unit is ready
grafana-agent/17 active idle 10.5.1.75
mysql/0 active idle 3 10.5.0.8 Unit is ready: Mode: R/O, Cluster is ONLINE and can tolerate up to ONE failure.
grafana-agent/18 active idle 10.5.0.8
mysql/1* active idle 4 10.5.1.197 Unit is ready: Mode: R/W, Cluster is ONLINE and can tolerate up to ONE failure.
grafana-agent/20 active idle 10.5.1.197
mysql/2 active idle 5 10.5.2.227 Unit is ready: Mode: R/O, Cluster is ONLINE and can tolerate up to ONE failure.
grafana-agent/19 active idle 10.5.2.227
nova-compute/0* active idle 8 10.5.0.28 Unit is ready
grafana-agent/22 active idle 10.5.0.28
ovn-chassis/0* active idle 10.5.0.28 Unit is ready
openstack-exporter/1* active idle 18 10.5.3.70
grafana-agent/21 active idle 10.5.3.70
Machine State Address Inst id Base AZ Message
0 started 10.5.0.223 a2651ff9-a4e7-4b00-a303-01559d902f74 ubuntu@22.04 nova ACTIVE
1 started 10.5.1.75 4dc9a492-c20e-49a8-a779-48283e9d7e57 ubuntu@22.04 nova ACTIVE
3 started 10.5.0.8 5f641db1-3a58-4495-932c-6d11636c9f14 ubuntu@22.04 nova ACTIVE
4 started 10.5.1.197 591c673f-f469-4c79-945c-7f22a929c2e1 ubuntu@22.04 nova ACTIVE
5 started 10.5.2.227 900a92e2-b8dc-4159-bed7-cd3abc00167f ubuntu@22.04 nova ACTIVE
8 started 10.5.0.28 7e044527-5d11-44f0-bc19-bb597b54d4b5 ubuntu@22.04 nova ACTIVE
18 started 10.5.3.70 0fcc5a0f-9a14-45a5-839c-fa31fad19d75 ubuntu@22.04 nova ACTIVE
sed-i
commented
3 months ago
Bug Description
Sometimes grafana-agent units (not necessarily all of them at once) get stuck in
waitingstatus with messageWaiting for TLS certificate.. This is related to the relation with grafana, and resolves if I bounce the relation between grafana and grafana-agent.To Reproduce
This was observed in a test environment while testing something else; I don't have a neat reproducer yet. Opening the issue in case others have also observed it, and will fill in more info if I have time to investigate further. :)
Environment
Not complete environment, but this may help:
Relevant log output
Additional context
No response