You are viewing a plain text version of this content. The canonical link for it is here.
Posted to pr@cassandra.apache.org by "mykaul (via GitHub)" <gi...@apache.org> on 2026/10/01 06:40:39 UTC

[PR] Probe each distinct JDK once; raise NODE_WAIT_TIMEOUT_IN_SECS to 120 [cassandra-ccm]

mykaul opened a new pull request, #815:
URL: https://github.com/apache/cassandra-ccm/pull/815

   ## What
   
   Two small fixes found while running Cassandra integration tests on GitHub Actions.
   
   ### 1. `get_available_jdk_versions()` probes the same JDK repeatedly
   
   It runs `java -version` for every environment variable matching `JAVA([0-9]+)?_HOME`. Hosts commonly define several pointing at the same JDK (`JAVA_HOME` alongside `JAVA11_HOME`, `JAVA_HOME_11_X64`, ...), so the same binary is spawned again and again — each one a cold JVM launch. On a GitHub Actions runner with five matches this costs ~15s on the first node *before* it starts. Each distinct `JAVA*_HOME` is now resolved and probed once.
   
   ### 2. `NODE_WAIT_TIMEOUT_IN_SECS` 90 -> 120
   
   `wait_for_binary_interface()` waits for the `Starting listening for CQL clients` log message with a 90-second cap. A Cassandra 5 seed node has been observed taking 89-92s (106s under JDK 17), i.e. right at/over the cap, so a slow-but-healthy start can fail spuriously. 120s gives headroom.
   
   ## Testing
   
   - `python3 -m py_compile ccmlib/common.py ccmlib/node.py`
   - Confirmed the dedup keeps the same selected Java version while dropping the duplicate probes (5 env vars -> 2 distinct JDKs in a typical runner).
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org
For additional commands, e-mail: pr-help@cassandra.apache.org


Re: [PR] Probe each distinct JDK once; raise NODE_WAIT_TIMEOUT_IN_SECS to 120 [cassandra-ccm]

Posted by "mykaul (via GitHub)" <gi...@apache.org>.
mykaul closed pull request #815: Probe each distinct JDK once; raise NODE_WAIT_TIMEOUT_IN_SECS to 120
URL: https://github.com/apache/cassandra-ccm/pull/815


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org
For additional commands, e-mail: pr-help@cassandra.apache.org


Re: [PR] Probe each distinct JDK once; raise NODE_WAIT_TIMEOUT_IN_SECS to 120 [cassandra-ccm]

Posted by "mykaul (via GitHub)" <gi...@apache.org>.
mykaul commented on PR #815:
URL: https://github.com/apache/cassandra-ccm/pull/815#issuecomment-5926264571

   Closing — gocql's Cassandra lane is being reconciled against scylladb/scylla-ccm; this upstream was not the right target.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org
For additional commands, e-mail: pr-help@cassandra.apache.org