Hey can someone help me understand why

datahub-upgrade

is needed to run datahub-gms? Some background: I am running datahub on kubernetes for ~2 years now, I don't use your provided helm charts. I simply have two pods - one for gms and one for frontend. ES and mysql are not on k8s. And for the longest time that's all I needed - both frontend and gms recovered after being restarted. But as of

v0.10.1

(I think) gms just won't start if I don't run the upgrade container:

2023-04-27 06:41:59,065 [R2 Nio Event Loop-1-2] WARN  c.l.r.t.h.c.c.ChannelPoolLifecycle:139 - Failed to create channel, remote=localhost/127.0.0.1:8080

I think it just gets stuck at this step:

2023-04-27 06:41:39,668 [main] INFO  c.l.metadata.boot.BootstrapManager:33 - Executing bootstrap step 1/13 with name WaitForSystemUpdateStep...

I understand the need to reindex ES indices when a certain upgrade requires it (which I've done adhoc when upgrading

0.9.6.1

->

0.10.1

), but what's the point of it outside of that? Is there any way to avoid to have to run the upgrade each time gms starts?

Hey there 👋 I'm The DataHub Community Support bot. I'm here to help make sure the community can best support you with your request. Let's double check a few things first: 1️⃣ There's a lot of good information on our docs site: www.datahubproject.io/docs, Have you searched there for a solution? Yes button 2️⃣ It's not uncommon that someone has run into your exact problem before in the community. Have you searched Slack for similar issues? Yes button

This job is intended as a central place to make sure the backend services are in a state that will not cause errors when upgrading. If you're managing things ad hoc and do not want to utilize it there is a flag to skip this check you can set to false on your gms deployment:

BOOTSTRAP_SYSTEM_UPDATE_WAIT_FOR_SYSTEM_UPDATE

Thanks. Am I right to assume that the upgrade job will do nothing with indices if I am not actually doing an upgrade but gms is just restarting?