Hello, thought i'd post here before continuing fur...
# troubleshooting
n
Hello, thought i'd post here before continuing further debugging. We're upgrading out Pinot installation and it's been a while so we have to move to the newer bitnami zookeeper chart. I've done the upgrade in our sandbox env and everything seemed to go smoothly, no data loss and components recovered just fine. However, when proceeding to our dev env, post restart and recreation of pods, we seem to have lost important zookeeper data - specifically table and schema configs. I see snapshots in the
/bitnami/zookeeper/data
dir, so clearly some things were not lost, but curious if migrating charts requires some extra steps in the config. Also, couple of followup questions: • are there ways to recover the tables/schemas if zookeeper lost them on restart? • how can this happen in one env but not in another? the only difference i can think of is the addition of basic auth to the controller ui • ways to prevent this for the future? we want to update prod as well, but not sure how to prevent this scenario. Backing up tables/schemas manually is one thing, but is there other important data whose loss can prevent a healthy recovery of pinot?