-
Bug
-
Resolution: Done
-
Medium
-
Frankfurt Release
-
None
As a part of frankfurt deployment, we identifed an issue related to the mariadb-galera.
In the helm deployment trace also we observed that two times the mariadb-galera & cassandra charts are deploying, so 1st time its installing & 2nd time it's upgrading. We are attaching our environment traces in the helm_trace.txt file,which explain the issue step by step. Please have a look of that file once.
Trace:
Release "frankfurt-mariadb-galera" does not exist. Installing it now.
NAME: frankfurt-mariadb-galera
LAST DEPLOYED: Fri Jun 19 14:24:38 2020
NAMESPACE: onapSTATUS: DEPLOYED
Release "frankfurt-mariadb-galera" has been upgraded.
LAST DEPLOYED: Fri Jun 19 14:57:39 2020
NAMESPACE: onapSTATUS: DEPLOYED
Note : Similarly aaf & cassandra charts are also deploying twice. You can see the same in helm list revisions.
root@frankfurtrnd-nfs:/home/ubuntu# helm list
NAME REVISION UPDATED STATUS CHART APP VERSION NAMESPACE
frankfurt 1 Fri Jun 19 14:22:48 2020 DEPLOYED onap-6.0.0 El Alto onap
frankfurt-aaf 2 Fri Jun 19 14:49:23 2020 DEPLOYED aaf-6.0.0 onap
frankfurt-aai 1 Fri Jun 19 14:50:11 2020 DEPLOYED aai-6.0.0 onap
frankfurt-appc 1 Fri Jun 19 14:50:48 2020 DEPLOYED appc-6.0.0 onap
frankfurt-cassandra 2 Fri Jun 19 14:51:23 2020 DEPLOYED cassandra-6.0.0 onap
frankfurt-cds 1 Fri Jun 19 14:51:53 2020 DEPLOYED cds-6.0.0 onap
frankfurt-clamp 1 Fri Jun 19 14:52:26 2020 DEPLOYED clamp-6.0.0 onap
frankfurt-cli 1 Fri Jun 19 14:52:58 2020 DEPLOYED cli-6.0.0 onap
frankfurt-consul 1 Fri Jun 19 14:53:29 2020 DEPLOYED consul-6.0.0 onap
frankfurt-contrib 1 Fri Jun 19 14:54:00 2020 DEPLOYED contrib-6.0.0 onap
frankfurt-dcaegen2 1 Fri Jun 19 14:54:36 2020 DEPLOYED dcaegen2-6.0.0 onap
frankfurt-esr 1 Fri Jun 19 14:56:37 2020 DEPLOYED esr-6.0.0 onap
frankfurt-log 1 Fri Jun 19 14:57:08 2020 DEPLOYED log-6.0.0 onap
frankfurt-mariadb-galera 2 Fri Jun 19 14:57:39 2020 DEPLOYED mariadb-galera-6.0.0 onap frankfurt-modeling 1 Fri Jun 19 15:08:07 2020 DEPLOYED modeling-6.0.0 onap
In the process of helm deployment, mariadb-galera had got installed intially, but after sometime when a new job i.e(mariadb-galera-pre-upgrade job) got initialized the existing mariadb-galera pods had got terminated, again after sometime this pre-upgrade job cleared from the list and a new job (i.e mariadb-galera-post-upgrade job) had got created even that also cleared from list. We are attaching our environment traces in the Job_Pod_trace.txt file,which explain the issue step by step. Please have a look of that file once.
- duplicates
-
OOM-2442 Rolling upgrade of MariaDB-Galera doesn't work if there is no actual upgrade in charts
- Closed