allthingssecurity opened a new pull request, #30249:
URL: https://github.com/apache/camel/pull/30249

   # Description
   
   [CAMEL-27037](https://issues.apache.org/jira/browse/CAMEL-27037)
   
   Found by the DST world in #30137 (CAMEL-26974), in two cases its default set 
leaves out: with `-Ddst.zookeeper.stopWhileSuspended=true` 7 of 5000 seeds 
fail, and with `-Ddst.zookeeper.stopDuringCreate=true` 100 of the first 1189 
seeds fail (the run stops at 100). Reproduced here with a plain unit test 
against a ZooKeeper server in the JVM, outside the DST harness.
   
   When the master consumer stops, or the route of the `MasterRoutePolicy` is 
removed, its `ZooKeeperGroup` is closed: it interrupts its thread and deletes 
its ephemeral znode (`doUpdate(null)`). The CuratorFramework is often shared 
and outlives the group (the client of the component, or the `curator` bean), so 
its session, and every ephemeral znode of that session, lives on. Two cases 
left the znode in the group:
   
   - the group is closed while the client is disconnected: `doUpdate(null)` 
deleted the znode only `if (isConnected())`, and then forgot its path. When the 
client reconnects to the same session (a connection loss shorter than the 
session timeout), the znode is still there;
   - the group is closed while its thread waits for the answer of a create: the 
interrupt ends the wait, but the server applies the create; the group never 
recorded the path and only logged "Ephemeral node could be created in the 
registry, but ZooKeeper group didn't record its id".
   
   The znode then stays in the group as long as the session lives, as a member 
of its own (the container id is new for each join, unless `runtime.id` is set). 
While it has the lowest sequence it is the master of the group, so no node runs 
the route, this node included after its route started again.
   
   ZooKeeper removes an ephemeral znode only when its session ends, so the 
client has to delete it. Curator's `DeleteBuilder.guaranteed()` is meant for 
this: a failed delete is recorded and Curator will "attempt to delete them in 
the background until successful" while the CuratorFramework is open. The znodes 
of a group already carry its uuid (`NodeState.uuid`), which `prunePartialState` 
uses to delete the duplicate of a create whose answer was lost.
   
   Change, in `ZooKeeperGroup.doUpdate(null)`:
   - when the client is disconnected, the znode is deleted with 
`delete().guaranteed().inBackground()`. The close does not wait for the 
connection (as before); Curator deletes the znode once the client is connected 
again, and if the session expired first, ZooKeeper removed it already. When 
connected, the foreground guaranteed delete stays as it was.
   - when a create did not answer and the client is connected, the group reads 
the children of its path and deletes those that carry its uuid. The reads come 
after the create in the order of the session, so they see the znode if the 
server applied the create. It asks the client itself whether it is connected, 
not only the view of the group: the close removes the listener of the group 
first, so the group may not have been told that the client is connected again, 
and the create may have been made just after the reconnect (the DST world found 
this case). While the client is disconnected it still only logs, as before. Not 
covered: the answer of a create lost with the connection, and the group closed 
before the reconnect; the znode then stays until the session ends, as on main.
   
   Tests, new `ZooKeeperGroupCloseTest` (a `ZooKeeperServer` in the JVM, as in 
`MasterRoutePolicyRouteLifecycleTest`, and a CuratorFramework with a 30 second 
session):
   - `groupClosedWhileDisconnectedDeletesItsZnodeOnceReconnected`: a group 
joins, the server stops, the group is closed while disconnected, the server 
starts again on the same data directory and the client reconnects to the same 
session (checked); the znode must be gone;
   - `groupClosedDuringACreateDeletesTheZnode`: the client applies the create 
and then waits for its answer until the thread is interrupted, as the ZooKeeper 
client does when the close interrupts it (a proxy of the CuratorFramework 
around the real one); the znode must be gone after the close.
   - `groupClosedDuringACreateBeforeItKnowsItIsConnectedAgainDeletesTheZnode`: 
the same, with the group told that the connection was lost and not that it is 
back (the client is connected).
   
   Without the change all three fail on main (the znode is still in the group 
20 seconds after the reconnect, and after the close); with it all three pass, 
and the camel-zookeeper-master unit tests pass (43 run, 0 failures). The ITs 
need a ZooKeeper container and were not run locally (CI runs them).
   
   With the change merged locally into the world of #30137, and two small 
changes to its fake (test only; the default set still passes on main with 
them): `-Ddst.zookeeper.stopDuringCreate=true` passes 5000 of 5000 seeds (100 
of the first 1189 fail without the change), 
`-Ddst.zookeeper.stopWhileSuspended=true` passes 5000 of 5000 (7 fail without 
it), and the default set passes 5000 of 5000 with and without it. The fake 
changes: it models a guaranteed delete in the background (main never makes 
one), and an interrupted retry sleep ends the retries of the client, as 
Curator's `SleepingRetry` does (without that, a create interrupted in its retry 
sleep was still made after the close, which a real client does not do; 12 of 
5000 seeds failed on that alone). They are in #30137 now (its commit "DST 
world: model the background guaranteed delete").
   
   No documentation or upgrade guide change.
   
   Backport: camel-4.18.x has the same `doUpdate`; the change applies there.
   
   # Target
   
   - [x] I checked that the commit is targeting the correct branch (Camel 4 
uses the `main` branch)
   
   # Tracking
   - [x] If this is a large change, bug fix, or code improvement, I checked 
there is a [JIRA issue](https://issues.apache.org/jira/browse/CAMEL) filed for 
the change (usually before you start working on it): 
[CAMEL-27037](https://issues.apache.org/jira/browse/CAMEL-27037).
   
   # Apache Camel coding standards and style
   
   - [x] I checked that each commit in the pull request has a meaningful 
subject line and body.
   - [ ] I have run `mvn clean install -DskipTests` locally from root folder 
and I have committed all auto-generated changes.
     (I built camel-zookeeper-master and the core modules with the formatter 
and import-sort plugins and ran the camel-zookeeper-master unit tests. I did 
not run the full root build.)
   
   # AI-assisted contributions
   
   - [x] If this PR includes AI-generated code, commits have proper 
co-authorship attribution (e.g., `Co-authored-by` trailers) and the PR 
description identifies the AI tool used.
     This PR was prepared with Claude Code (Claude Opus 5.5). The commits carry 
a `Co-Authored-By` trailer.
   
   _Claude Code on behalf of allthingssecurity_
   
   🤖 Generated with [Claude Code](https://claude.com/claude-code)
   


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to