During a fresh deployment, the default dashboard zip file is malformatted
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Prometheus libvirt exporter charm |
New
|
Undecided
|
Unassigned |
Bug Description
When adding the prometheus-
tracer: set flag libvirt-
tracer: -- dequeue handler reactive/
tracer: -- dequeue handler reactive/
2021-06-28 15:53:30 INFO juju-log juju-info:483: start_snap() libvirt-
2021-06-28 15:53:30 DEBUG juju-log juju-info:483: Installing dashboards from resource
2021-06-28 15:53:31 ERROR juju-log juju-info:483: BadZipFile: File is not a zip file
Investigating the files themselves, one can see:
ubuntu@
gzip: grafana-
and
ubuntu@
gzip: grafana-
Commands to deploy the charm:
juju deploy prometheus-
juju relate prometheus-
juju relate prometheus-
juju relate prometheus-
juju relate prometheus-
Charm info:
prometheus:
charm: cs:prometheus2-19
series: bionic
os: ubuntu
charm-origin: jujucharms
charm-name: prometheus2
charm-rev: 19
charm-version: ae30894
can-upgrade-to: cs:prometheus2-22
exposed: false
application
current: active
message: Ready
since: 28 Jun 2021 15:59:41Z
relations:
grafana-
- grafana
juju-info:
- canonical-livepatch
- filebeat
- landscape-client
- logrotate
- ntp
- telegraf-prometheus
nrpe-
- nrpe-host
target:
- prometheus-
- prometheus-
- prometheus-
- telegraf
- telegraf-maas-infra
- telegraf-prometheus
$ juju config keystone openstack-origin
cloud:bionic-ussuri
As a workaround, reattaching the dashboard solved this issue:
1) clone the libvirt.json file on the cloud in question
2) zip grafana-
3) juju attach-resource prometheus-