joerghoh commented on code in PR #1012:
URL: https://github.com/apache/jackrabbit-oak/pull/1012#discussion_r1282861611
##########
oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/index/cursor/TraversingCursor.java:
##########
@@ -159,10 +159,15 @@ private void fetchNext() {
readCount++;
if (readCount % 1000 == 0) {
- FilterIterators.checkReadLimit(readCount, settings);
- String caller =
IndexUtils.getCaller(this.settings.getIgnoredClassNamesInCallTrace());
- LOG.warn("Traversed {} nodes with filter {} called by {};
consider creating an index or changing the query" ,
- readCount, filter, caller);
+ if (readCount == 20000) {
+ LOG.warn("Traversed {} nodes with filter {}; consider
creating an index or changing the query",
+ readCount, filter, new Exception("call
stack"));
Review Comment:
Well, but then we have a inconsistent reporting here, and we should
harmonize it. And then this magic number ``20000`` makes me wonder what is so
special about it (and why this exception is not reported at 10k or 80k)?
What information is needed for analysis? I think that the full stack trace
is always noise, but you just need more information than currently provided by
``IndexUtils.getCaller()``. What information is this? Mabye 5 more lines of the
stack below that line which is currently reported?
--
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]
