[
https://issues.apache.org/jira/browse/CAMEL-25943?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18132239#comment-18132239
]
Claus Ibsen commented on CAMEL-25943:
-------------------------------------
Fixed by https://github.com/apache/camel/pull/28469, merged to main as
e48face394f190f43df2bba2a4190c1badaeee93 (Camel 4.24.0).
_Claude Code on behalf of davsclaus_
> camel-jira - a consumer with maxResults 0 or less repeats the same search
> forever on Jira Server / Data Center
> --------------------------------------------------------------------------------------------------------------
>
> Key: CAMEL-25943
> URL: https://issues.apache.org/jira/browse/CAMEL-25943
> Project: Camel
> Issue Type: Bug
> Components: camel-jira
> Reporter: shashank
> Assignee: shashank
> Priority: Major
> Fix For: 4.24.0
>
>
> {{AbstractJiraConsumer.getIssues}} says that {{maxResults}} is ignored when
> it is 0 or less, but it first does {{if (maxResults < maxPerQuery)
> maxPerQuery = maxResults;}} (main 9f839f4bcc9d), so the page size of the
> search becomes 0 (or negative). On Jira Server / Data Center the loop then
> asks for a page of 0 issues at {{startAt=0}}, gets no issue and the total,
> and moves on with {{start += 0}}. None of the exit conditions holds while the
> search finds an issue ({{maxPerQuery >= total}}, {{start >= total}},
> {{maxResults > 0 && ...}}), so the consumer sends the same search request in
> a tight loop and never returns:
> * {{watchUpdates}} and {{newComments}} read the issues in {{doStart}}: the
> route never finishes starting, and the CamelContext startup hangs;
> * {{newIssues}} hangs in its first poll when the JQL found no issue at
> startup (no latest issue yet, so it searches with {{getIssues(jql)}});
> * a negative {{maxResults}} moves the start index backwards.
> The Jira server is called as fast as the consumer thread can go. Claus Ibsen
> pointed at the line in the review of #28198 (CAMEL-25806), which made the
> newIssues consumer page through all new issues and already guards its own
> page size against {{maxResults <= 0}}; the shared {{getIssues}} was not
> changed there.
> *Fix*: lower the page size only for {{maxResults > 0}} ({{if (maxResults > 0
> && maxResults < maxPerQuery)}}). The Jira Cloud path gets the same page size
> of 50 then, instead of 0.
> *Test*: new {{WatchUpdatesConsumerMaxResultsZeroTest}}: a watchUpdates route
> with {{maxResults=0}} and 60 issues, with a search fake that returns at most
> {{maxResults}} issues from {{startAt}} and the total, as Jira Server does
> ({{maxResults=0}} returns only the total). It throws when the same
> {{startAt}} is asked for more than 10 times in a row. On main the route does
> not start: {{FailedToStartRouteException: ... The search is repeated at
> startAt 0}}. With the fix the change of an issue on the second page is
> routed. camel-jira unit tests with the fix: 63 run, 0 failures.
> *How it was found*: Claus Ibsen's review comment on #28198, confirmed by
> reading {{getIssues}} on main and by the JUnit test above.
> *Affected*: main, camel-4.23.x, camel-4.22.x, camel-4.18.x, camel-4.14.x and
> camel-4.8.x (the same line and loop; read, not run).
> _Filed with Claude Code on behalf of allthingssecurity._
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
