-
Notifications
You must be signed in to change notification settings - Fork 278
How should we configure to keep the schema after restarting the apicurio deployment? #3145
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
Can you try to remove custom topic & consumer group configuration properties to see if the error disappears? I have a hypothesis that there is a bug in the way we work with those configuration properties. Are there any interesting log messages? |
Thank you for your quick reply! I removed the consumer group configuration( I could reproduce this on my local laptop and the log was as follows. I checked the diff between the successful one and failure one and found an interesting log. Maybe
|
If we cannot specify the consumer group id, we have to use the default one, right? kafka-acls --bootstrap-server <bootstrap_servers> --add \
--allow-principal <principal> \
--allow-host "*" \
--operation read \
--group "apicurio-registry-" \
--resource-pattern-type prefixed |
Thanks for the helpful info, I'll see if I can fix the consumer group issue. The |
Thank you for your reply! I hope this information helps you solve your problem. For the time being, I will not specify the consumer group id, and l use the default consumer group ID. |
We have the same issue, when I tried change some other settings like client.id but nothings works as long as the |
For anyone looking at this in the future the way Kafkasql works is that for writes, a message is sent to the Kafka topic. There is then an internal H2 database that is used for reads. Fort this to work, each Registry replica has to be put on a different consumer group so all the replicas consume all the messages in the topic. If the replicas share the consumer group, then they will miss some part of the information resulting in potential schemas not being present. |
@carlesarnal thank you for your explanation! I could understand the reason for this. 🙇 |
@LittleWat That is for sure a fair ask, do you mind opening a new GH issue for it? Thanks |
Hello! I am relatively new to Apicuiro.
We set up the apicurio registry that uses kafka as its storage following the document.
The deployment file is like as follows:
The schema registration via the Apicurui UI works fine, however, after restarting the deployment (
$ kubectl rollout restart deployment -n apicurio apicurio
), the schema is lost.I would like to keep the schema after restarting the deployment. Is it possible?
If possible, how should I configure?
Thank you!
The text was updated successfully, but these errors were encountered: