appsmith
appsmith copied to clipboard
appsmithorg
fix: Add mongo/redis to helm chart dependencies
Description
Add redis and mongo as individual pods to our helm charts.
Fixes 15929
PFA
Redis:
Mongo deployment pics:
Mongo: Demo of the deployment of how it looks.
https://user-images.githubusercontent.com/15846947/190350964-2ec81f73-9c46-4cb6-b476-a0b1908c8f93.mov
The latest updates on your projects. Learn more about Vercel for Git ↗︎
| Name | Status | Preview | Updated |
|---|---|---|---|
| appsmith | ✅ Ready (Inspect) | Visit Preview | Oct 14, 2022 at 10:00AM (UTC) |
![vercel[bot] avatar](https://avatars.githubusercontent.com/in/8329?v=4 "Published by a pub.dev verified publisher"){kind=small}
Unable to find test scripts. Please add necessary tests to the PR.
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
Unable to find test scripts. Please add necessary tests to the PR.
Deployment failed with the following error:
Resource is limited - try again in 1 minute (more than 100, code: "api-deployments-free-per-day").
Unable to find test scripts. Please add necessary tests to the PR.
Unable to find test scripts. Please add necessary tests to the PR.
Unable to find test scripts. Please add necessary tests to the PR.
Unable to find test scripts. Please add necessary tests to the PR.
@pratapaprasanna, thanks for opening this.
This new chart will need manual upgrading, right? For example, if someone is running the current Helm Chart (
v1.5.3), their MongoDB data would be in the Appsmith pod's PV. But with this chart, that data should be in the mongodb pods' PV. Which means moving from that Helm Chart to this one, will need some migration effort, right? Can we document the steps needed for someone already using current Helm Chart to migrate to this one please?Also, because of this reason, let's make this Helm Chart version
v2.0.0inChart.yaml. What say?cc @mohanarpit
Will this mean that we will deprecate v1.x for our Helm charts and urge users to follow the migration steps v2.x? If yes, will any critical fixes be made to both v1.x & v2.x Helm charts? Also, when users run the command helm upgrade today, will Helm pick up the v1.x chart or v2.x chart?
@mohanarpit, I'd rather we take a call on deprecating v1.x, after we know what it takes to upgrade to this Helm Charts. If it's a simple one or two steps, then lets deprecate. We are still hosting charts all the way from 1.3 I think, which is just becoming a baggage.
Yeah, critical fixes will need to be made in both, but, we haven't seen any such critical issues in our Helm Charts so far. 🤞
I don't think helm upgrade upgrades the version of the Helm charts. That command is to apply any changes in our values.yml configuration to our existing chart. @pratapaprasanna, please correct if this is not right.
Unable to find test scripts. Please add necessary tests to the PR.
@pratapaprasanna, I see that you've requested a review, but I don't see a comment with your thoughts on backwards compatibility and a migration path. Can you take a look at https://github.com/appsmithorg/appsmith/pull/16096#pullrequestreview-1094413731 and share your thoughts please?
This PR has not seen activitiy for a while. It will be closed in 7 days unless further activity is detected.
Unable to find test scripts. Please add necessary tests to the PR.
Deployment failed with the following error:
Resource is limited - try again in 13 minutes (more than 100, code: "api-deployments-free-per-day").
Deployment failed with the following error:
Resource is limited - try again in 19 minutes (more than 100, code: "api-deployments-free-per-day").
Deployment failed with the following error:
Resource is limited - try again in 16 minutes (more than 100, code: "api-deployments-free-per-day").
Unable to find test scripts. Please add necessary tests to the PR.
Hi,
im currently facing an issue with in an kubernetes environment which doesn't have access to the internet. The appsmith container is always in the INIT:0/2 state because the helm chart is using an alpine image and is installing the redis-cli through the command durring the initialization:
sh -c apk add redis; until redis-cli -h appsmith-redis-master.manufacturing-appsmith-dev.svc.cluster.local ping; do echo waiting for redis; sleep 2; done
that is why the init process is logging the following messages all the time:
sh: redis-cli: not found
waiting for redis
looks like the current helm chart is not usable in such an offline environment and should be improved.
