Web console should not try to display a null timestamp.
+++ This bug was initially created as a clone of Bug #1766747 +++
Description of problem:
In the dashboards page you see a lot of activity from 1AM. In my case it's an illogical time since this setup has just been installed.
Version-Release number of selected component (if applicable):
4.3.0-0.ci-2019-10-29-152254
How reproducible:
Unknown
Steps to Reproduce:
1. Look at the activity card in the dashboards page.
Actual results:
Illogical timestamps
Additional info:
Screenshot attached
--- Additional comment from Rastislav Wagner on 2019-11-04 08:56:59 UTC ---
I also saw this once and the issue was that the event itself has a wrong timestamp (1.00 AM 1 January 1970). Can you take a look at the event itself if this is the same case ? If yes, than its not a bug in UI.
--- Additional comment from Udi on 2019-11-06 14:44:46 UTC ---
The timestamp is <unknown>:
oc get events -n openshift-marketplace
<unknown> Normal Scheduled pod/redhat-operators-7779d777f8-np5ph Successfully assigned openshift-marketplace/redhat-operators-7779d777f8-np5ph to rhhi-node-master-2
<unknown> Normal Scheduled pod/redhat-operators-784b69c4b8-d8pcv Successfully assigned openshift-marketplace/redhat-operators-784b69c4b8-d8pcv to rhhi-node-master-2
It's still a problem that in the UI these events are always at the top of the list (they're the most recent), and the timestamps are confusing in the GUI if they're from a different date.
--- Additional comment from Samuel Padgett on 2019-11-12 14:22:09 UTC ---
I'm going to change the component to kube-apiserver to investigate since these events should have timestamps, but also clone this bug to have console handle missing values.
--- Additional comment from Samuel Padgett on 2019-11-14 13:47:18 UTC ---
Here's the YAML for one event with the missing timestamp:
- action: Binding
apiVersion: v1
eventTime: "2019-11-14T10:50:00.212773Z"
firstTimestamp: null
involvedObject:
apiVersion: v1
kind: Pod
name: curator-1573728600-7z2sj
namespace: openshift-logging
resourceVersion: "1179999"
uid: 339b7484-bc2c-4775-a419-7ca41dbf2681
kind: Event
lastTimestamp: null
message: Successfully assigned openshift-logging/curator-1573728600-7z2sj to ip-10-0-167-133.ap-south-1.compute.internal
metadata:
creationTimestamp: "2019-11-14T10:50:00Z"
name: curator-1573728600-7z2sj.15d701d2ec8e357d
namespace: openshift-logging
resourceVersion: "1180004"
selfLink: /api/v1/namespaces/openshift-logging/events/curator-1573728600-7z2sj.15d701d2ec8e357d
uid: 198674a0-9392-4d25-b34c-ae558880e977
reason: Scheduled
reportingComponent: default-scheduler
reportingInstance: default-scheduler-ip-10-0-160-4
source:
component: default-scheduler
type: Normal
--- Additional comment from Samuel Padgett on 2019-11-14 14:03:16 UTC ---
I'm going to clone this bug. We should fix the missing timestamps in the resource, but web console shouldn't try to display a null timestamp.
4.3.0-0.nightly-2019-11-17-224250
Tested on ocp4.3 with above payload. Create resource in project, check "Activity" part on dashboard, the timestamp is a meaningful time.
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.
https://access.redhat.com/errata/RHBA-2020:0062