Installation of on premise erroring out with Connection to Kafka failed #472
Assignees
Labels
Comments
|
Same error when updating Sentry from older 10.1.0-dev0 to latest but in case of update it seems to start correctly if i run |
|
Looks like reverting #465 will "fix" this. With kafka and zookeeper on 5.5 both containers kept failing for me. |
|
Had a similar issue in January: see #331 Removing some containers and Kafka volume helped me back then. But it does not resolve the issue this time. |
|
@maximal maybe you need to remove the zookeeper volumes and containers this time too? Apologies for the issues, I was expecting this upgrade to be painless but seems like that's not really the case. |
Removing
No worries! We’re using unstable branch here, testing it so I suppose we all realize the consequences :-) |
Eh, if only we were providing a "stable" version, you'd be right. This is around the corner though so stay tuned and thanks again! |
|
Solution proposed by @maximal works in my case (update from older 10.1.0-dev0) |
|
If anyone has still access to ZooKeeper logs, can you share them here? You can run |
|
I'm getting this error too. Here is the zookeeper logs: And kafka: |
|
Here's yours:
|
|
https://issues.apache.org/jira/browse/ZOOKEEPER-3056 this might fix the issue in zookeeper. but i don't know the correct way to pass this setting to the container in this setup. |
|
I think there's a fix upcoming. In the meantime, can someone try adding |
|
@BYK stops the loop (error appears once) and install continues |
|
@xf- so that means adding this value solves the issue? |
|
@BYK like i wrote. I get the same error but only once and the update continues. |
This solution did not work for me when upgrading from
This workaround by @BYK did not work either. PS: According to Update: At least I was able to revert to my previous version (with latest git checkout): |
|
@caugner that's quite peculiar that going back to an earlier Sentry image fixing this. I bet it's not Sentry getting reverted itself but instead Snuba and other services getting updated or restarted. I'd recommend running |
BYK
added a commit
that referenced
this issue
May 23, 2020
Fixes the "logs found but no snapshot" error when upgrading from a short-lived older version. Fixes #472. See https://issues.apache.org/jira/browse/ZOOKEEPER-3056.
BYK
added a commit
that referenced
this issue
May 24, 2020
Fixes the "logs found but no snapshot" error when upgrading from a short-lived older version. Fixes #472. See https://issues.apache.org/jira/browse/ZOOKEEPER-3056.
|
@BYK I've pulled 83160e8 and I'm still seeing the errors of this issue when I run |
|
@jansorg it is normal to see it a few times. If the installation finishes successfully, just ignore it. If your install doesn't succeed, please file a new issue and share your Kafka and Zookeeper logs. |
|
I was able to update from e54cbeb to df57809 now, but it could also be that my connectivity issues (#505 (comment)) played a role here. |
MaicolBen
pushed a commit
to hinthealth/onpremise
that referenced
this issue
Jul 20, 2020
Fixes the "logs found but no snapshot" error when upgrading from a short-lived older version. Fixes getsentry#472. See https://issues.apache.org/jira/browse/ZOOKEEPER-3056.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Cloning master version to ubuntu 18.04 and running the install script
2020-05-05 09:29:21,527 Connection to Kafka failed (attempt 59)
Traceback (most recent call last):
File "/usr/src/snuba/snuba/cli/bootstrap.py", line 56, in bootstrap
client.list_topics(timeout=1)
cimpl.KafkaException: KafkaError{code=_TRANSPORT,val=-195,str="Failed to get metadata: Local: Broker transport failure"}
Traceback (
ubuntu@ubuntu-s-6vcpu-16gb-sgp1-01:~/onpremise$ docker -v Docker version 19.03.8, build afacb8b7f0ubuntu@ubuntu-s-6vcpu-16gb-sgp1-01:~/onpremise$ docker-compose -v docker-compose version 1.25.5, build unknownThe text was updated successfully, but these errors were encountered: