Closed Riddle4045 closed 1 year ago
@Riddle4045 We've just released 17.0.9 - are you able to give that a go and report back? Also which version of the SQL driver are you using?
Can you share a reproducible test?
@Riddle4045 We've just released 17.0.9 - are you able to give that a go and report back? Also which version of the SQL driver are you using?
@karianna Sorry for the delay in response, I was travelling - I will try 17.0.9 today and get back to you. I was on the latest version 12.4, Also tried 11.2. This was not the only call flow that crashed the k8s containers, we also have a few customers reporting the crashes in a separate call flow not involving SQLserver driver.
@brunoborges we deploy a query engine (Trino) on K8s and query data from SQL server & ADLS gen2 via the query engine, unfortunately the deployment is a sophisticated and I will not be able to share simple repro here.
@Riddle4045 We've just released 17.0.9 - are you able to give that a go and report back? Also which version of the SQL driver are you using?
@karianna Sorry for the delay in response, I was travelling - I will try 17.0.9 today and get back to you. I was on the latest version 12.4, Also tried 11.2. This was not the only call flow that crashed the k8s containers, we also have a few customers reporting the crashes in a separate call flow not involving SQLserver driver.
@karianna I was not able to repro the crashed on 17.0.9, I have a consistent repro only on 17.0.8
Thanks for checking! We do only officially support the very latest point release, so I'll close this for now but if it re-appears in 17.0.9 let us know!
Describe the bug
we are running into frequent jvm crashes while running msopenjdk 17.0.8, we don't observe these crashes when we downgrade to 17.0.7
Desktop (please complete the following information):
OS & version [e.g. Windows 10, 1822]:
MSFT Build of OpenJDK Version:
To Reproduce: we are running mariner on Kubernetes pods, when calling into our service running on these pods we encounter these issues.
Screenshots, Logs etc hotspot error logs beginning pasted below, I can share the entire log file if needed. k8s restarts the failing container core-dumps are not written by the time container restarts, will be some work to get the core dumps, but it's passible if required.