Ceph octopus incompatible with containerised chrony
Bug #1885689 reported by
Mark Goddard
This bug affects 2 people
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
kolla-ansible |
Fix Released
|
Medium
|
Radosław Piliszek | ||
Ussuri |
Won't Fix
|
Medium
|
Unassigned | ||
Victoria |
Won't Fix
|
Medium
|
Radosław Piliszek | ||
Wallaby |
Fix Released
|
Medium
|
Radosław Piliszek | ||
Xena |
Fix Released
|
Medium
|
Radosław Piliszek |
Bug Description
CentOS 8 appears to have chrony enabled and started by default. The chronyd service is disabled and packages uninstalled during kolla-ansible bootstrap-servers, if enable_chrony is true.
The problem with Kolla-containerised chrony is that Ceph Octopus node health checks don’t see it and this puts the node into a warning state. Also, Ceph Orchestrator does a check for time sync during pre-flight checks, and Ceph deployment normally runs before the containerised service is deployed by kolla-ansible.
Related kayobe bug: https:/
Changed in kolla-ansible: | |
importance: | Undecided → Medium |
Changed in kolla-ansible: | |
milestone: | 11.0.0 → none |
Changed in kolla-ansible: | |
status: | In Progress → Fix Released |
To post a comment you must log in.
I'm +1 on stopping containerizing chrony - NTP service is a critical part of the base host for its many functions, usually related to synchronization and security.