Closed vbotbuildovich closed 2 days ago
This one doesn't look like infra issue.. Originally that is what I thought, since it shows ssh issue and timeout of 20 seconds.
But I looked at this test code and the logic was not changed for almost two years.
It is related to upgrade, we wait for service to come up
Error message is:
ducktape.errors.TimeoutError: Redpanda service ip-172-31-6-14 failed to start within 20 sec
We could increase a timeout and see what happens, but since this code is not new, looks like there is a possible degradation with the product.. I will assign it to myslef and investigate more. Let's see if we can reproduce it
@rpdevmp wrote:
Error message is: ducktape.errors.TimeoutError: Redpanda service ip-172-31-6-14 failed to start within 20 sec
... but the error message from the stack in the top comment is:
ducktape.cluster.remoteaccount.RemoteCommandError: root@ip-172-31-11-46: Command 'curl -fsSL https://vectorized-public.s3.us-west-2.amazonaws.com/releases/redpanda/23.3.15/redpanda-23.3.15-amd64.tar.gz --retry 3 --retry-connrefused --retry-delay 2 --create-dir -o /opt/redpanda_installs/v23.3.15/redpanda.tar.gz && gunzip -c /opt/redpanda_installs/v23.3.15/redpanda.tar.gz | tar -xf - -C /opt/redpanda_installs/v23.3.15 && rm /opt/redpanda_installs/v23.3.15/redpanda.tar.gz' returned non-zero exit status 35.
This clearly looks like curl failing to connect to the s3 bucket (well it's hard to tell for sure because of the long pipeline but based on errors we see elsewhere the 35 is almost certain from curl
).
So I don't understand the comment about Redpanda failing to start. Where do you see that?
I suspect these are duplicates of https://github.com/redpanda-data/redpanda/issues/18607. We are using curl here rather than requests, but the endpoint is the same and curl error 35 is an SSL-related error just like we got in Python.
Automatically closing issue to match current state of CORE-2940
https://buildkite.com/redpanda/vtools/builds/13676
JIRA Link: CORE-2940