Bug 1829064

+++ This bug was initially created as a clone of Bug #1826781 +++

+++ This bug was initially created as a clone of Bug #1809498 +++

We are likely still awaiting proper images from ART since this is trying to boot an ES5 node.  ES6 will be used in 4.5.  What images does the EO say it should be using?  What images does the CLO say it should be using?

--- Additional comment from Anping Li on 2020-03-10 03:04:39 UTC ---

Yes, there isn't 4.5 channel in operator-registry redhat-operatrors-art. Waiting art.

--- Additional comment from Jeff Cantrill on 2020-04-14 20:08:20 UTC ---

Please retest now there are available images for 4.5. One suspicion I have is if you upgrade CLO before EO that records will be written to an index with a '-write' suffix and thus the new indexmanagement will be broken

--- Additional comment from errata-xmlrpc on 2020-04-22 18:28:05 UTC ---

This bug has been added to advisory RHBA-2020:51809 by OpenShift Release Team Bot (ocp-build/buildvm.openshift.eng.bos.redhat.com)

--- Additional comment from errata-xmlrpc on 2020-04-22 18:28:14 UTC ---

Bug report changed to ON_QA status by Errata System.
A QE request has been submitted for advisory RHBA-2020:51809-02
https://errata.devel.redhat.com/advisory/51809

--- Additional comment from Qiaoling Tang on 2020-04-23 01:25:38 UTC ---

I tested upgrade from 4.4 to 4.5 again, still hit the issue in https://bugzilla.redhat.com/show_bug.cgi?id=1809498#c0. 

For the issue in https://bugzilla.redhat.com/show_bug.cgi?id=1809498#c4, it didn't appear. I checked the logging 4.4, the env var `POD_IP` was added to the elasticsearch when deploying logging 4.4.

I also tested upgrade from 4.3 to 4.5, the issue in https://bugzilla.redhat.com/show_bug.cgi?id=1809498#c4 didn't appear, but I hit the same issue in https://bugzilla.redhat.com/show_bug.cgi?id=1809498#c0. The env var `POD_IP` wasn't added to the elasticsearch when deploying logging 4.3.


It seems the fix PR https://github.com/openshift/elasticsearch-operator/pull/316 only fixes the issue in https://bugzilla.redhat.com/show_bug.cgi?id=1809498#c4. @IgorKarpukhin, is my understanding correct?


4.5 logging images are from 4.5.0-0.ci-2020-04-22-212726.

--- Additional comment from IgorKarpukhin on 2020-04-23 11:23:58 UTC ---

Yes, that's correct. This fix isn't backported to 4.4 yet. I'm working on it.

--- Additional comment from IgorKarpukhin on 2020-04-23 11:54:32 UTC ---

Qiaoling Tang could you please post the logs from catalog operator during the upgrade from 4.3 to 4.4?

Ex: oc logs -f catalog-operator-7648d9cff4-rqx9j -n openshift-operator-lifecycle-manager

--- Additional comment from Qiaoling Tang on 2020-04-24 00:36:27 UTC ---

Catalog operator logs: http://pastebin.test.redhat.com/858655

Cluster version: 4.4.0-0.nightly-2020-04-23-192521

--- Additional comment from Qiaoling Tang on 2020-04-24 02:16:15 UTC ---

@IgorKarpukhin For the issue during upgrade from 4.3 to 4.4, we can track it in https://bugzilla.redhat.com/show_bug.cgi?id=1824006. 

Per our code merge process, we should fix issue in 4.5 firstly, then the fix for 4.4 could be merged. Could we set this bug as a blocker for 1824006 to complete our process? I think the issue in https://bugzilla.redhat.com/show_bug.cgi?id=1809498#c0 has nothing to do with https://bugzilla.redhat.com/show_bug.cgi?id=1824006. 

And Per comment 3 and comment 4, I think I can move this bug to VERIFIED, WDYT?

--- Additional comment from Jeff Cantrill on 2020-04-28 18:06:54 UTC ---

Moving to ON_QA.  Qiaoling if you feel it is reasonable to move this bug to Verified please do so.,