In a small (7-unit) deployment a single ntp unit will stay stuck executing for 40+ minutes
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
NTP Charm |
New
|
Undecided
|
Unassigned |
Bug Description
I have a small kubernetes cluster on top of OpenStack which uses ntp. There are 7 total machines in the deployment. 6/7 of the ntp subordinates in this cluster have finished executing, however one unit has been stuck executing for 45+ minutes with no indication as to what is holding up completion of the setup:
2021-04-29-16:41:23 root DEBUG DEBUG:root:ntp/9 juju agent status is allocating since 2021-04-29 16:39:39+00:00
2021-04-29-16:41:29 root DEBUG DEBUG:root:ntp/9 workload status is maintenance since 2021-04-29 16:41:23+00:00
...
2021-04-29-17:26:46 root DEBUG DEBUG:root:ntp/9 juju agent status is executing since 2021-04-29 17:25:06+00:00
From the juju logs it looks to be running though ntp-peers-
The unit itself successfully sets its offset:
2021-04-29 16:42:13 DEBUG install 29 Apr 16:42:13 ntpdate[16833]: adjust time server 91.189.89.198 offset -0.000008 sec
2021-04-29 16:42:13 DEBUG install active
And the install hook completes. What is it still executing on?
crashdump