Hi Wallen,
Actually, the files Daniel listed were modified on 8/11 and then again on 8/15. In the time between 8/11 to 8/15, I belive there could have been any number of problems, including corrupt indexes and poor multithreaded performance. However, I think after 8/15, the files should be in good working order. If you are not sure if you saw problems with pre-8/15 or post-8/15 version of the code, is it possible for you to try the latest CVS and see if the problem exists now? If it does, it will of course require urgent attention.
Thanks very much! Dmitry.
Daniel Naber wrote:
On Tuesday 07 September 2004 17:41, [EMAIL PROTECTED] wrote:
A note to developers, the code checked into lucene CVS ~Aug 15th, post
1.4.1, was causing frequent index corruptions. When I reverted back to
version 1.4 I no longer am getting the corruptions.
Here are some changes from around that day:
http://cvs.apache.org/viewcvs.cgi/jakarta-lucene/src/java/org/apache/lucene/index/SegmentMerger.java http://cvs.apache.org/viewcvs.cgi/jakarta-lucene/src/java/org/apache/lucene/index/SegmentReader.java http://cvs.apache.org/viewcvs.cgi/jakarta-lucene/src/java/org/apache/lucene/index/IndexWriter.java
Could you check which of those might have caused the problem? I guess there's not much the developers can do without the problem being reproducible.
regards Daniel
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
