Hi Morgan,

finally I found the problem with the MariaDB-Galera upgrade (also Cassandra, 
AAF,...) steps, which caused the initially mentioned problems (which do not 
appear in our Daily deployments).

In the NFS server startup script ( 
https://git.onap.org/integration/tree/deployment/heat/onap-rke/nfs_vm_entrypoint.sh
 ) the deploy.sh of the helm plugin is patched.

This causes somehow to install AAF, MariaDB-Galera, Cassandra and DMaaP before 
installation of each helm module in alphabetic order, where the mentioned 
components are upgraded (which fails for MariaDB-Galera).

After commenting out the workaround checkout and the patches on the deploy.sh 
in the nfs_vm_entrypoint.sh, the problems disappeared:

# Clone OOM repo
cd ~
git clone --recurse-submodules -b __oom_gerrit_branch__ 
https://gerrit.onap.org/r/oom
cd oom
if [ ! -z "__oom_gerrit_refspec__" ] ; then
   git fetch https://gerrit.onap.org/r/oom __oom_gerrit_refspec__
   git checkout FETCH_HEAD
fi
*#git checkout -b workarounds #git log -1*
... # install helm deploy plugin
rsync -avt ~/oom/kubernetes/helm/plugins ~/.helm/
# temporary workaround to throttle the helm deploy to alleviate startup disk 
contention issues
*#if [ ! -z "__helm_deploy_delay__" ] ; then # sed -i "/\^enabled:/a\ echo 
sleep __helm_deploy_delay__\n sleep __helm_deploy_delay__" 
~/.helm/plugins/deploy/deploy.sh # sed -i 's/for subchart in \*/for subchart in 
aaf cassandra mariadb-galera dmaap */' ~/.helm/plugins/deploy/deploy.sh #fi*

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#21463): https://lists.onap.org/g/onap-discuss/message/21463
Mute This Topic: https://lists.onap.org/mt/74984907/21656
Group Owner: [email protected]
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to