allthingssecurity commented on PR #30311:
URL: https://github.com/apache/camel/pull/30311#issuecomment-6972899966

   On the start time: I like it better than the gap, so I will switch to it in 
a follow-up. Two details I checked:
   
   - JQL `created` only takes minutes (`yyyy-MM-dd HH:mm`, 
NewIssuesConsumer.java:49), and the `>` acts like `>=` for that minute (:51). 
So the query with the start time also returns the issues created earlier in the 
start minute. The follow-up keeps the start time as an instant and filters on 
the full `getCreationDate()` of each issue, the way `getNewIssues` already 
filters against `latestIssue` (:164), so only issues created after the start 
are routed.
   - The JQL needs the start time in the user's time zone, which is one more 
reason to read the time zone before the lookup in `doStart`, as you suggest in 
your second point. I will do both in the same follow-up.
   
   What is left is the clock skew between the Camel host and Jira: a host clock 
that is ahead can miss issues created in that many seconds after the start, a 
clock that is behind can route issues created that many seconds before it. With 
NTP that is a few seconds at most, and it only applies while the startup lookup 
has not succeeded. I will note that in the class Javadoc and the description.
   
   _Claude Code on behalf of allthingssecurity_
   


-- 
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