WFTO – new scavenger and more bug fixes
Our MySQL/Falcon team is back from vacation and we were very productive in fixing bugs. Last week Kevin pushed his new scavenger code and after a few bug fixes in the new scavenger code long outstanding memory leak issues were fixed.
Since my last report from 2009-01-09 we fixed around 8 Falcon related bugs. Compared to our last WFTO from 2009-01-09 we have 1 new test.
The development of our failed/passed tests ratio over time looks like:
- 1.28% – 3/234(this report)
- 1.29% – 3/233 (last report)
- 1.29% – 3/232 (report before last one)
News for this week:
- Kevin Lewis fixed
- Bug#36700 Running falcon_bug_28048 shows increasing memory usage and run time
- Bug#41521 falcon_bug_34174 fails with lock wait timeout, sometimes
- Bug#41831 Error: assertion (record->state != recLock) failed at line 1999 in file Table.cp
- Bug#41870 Unbounded tablespace growth when updating BLOB record
- Bug#42006 Running falcon_bug_22207-big on 32-bit Linux leads to crash
- Bug#42080 falcon_record_cache_memory_leak2 crashes in RecordVersion::scavengeSavepoint
- Olav Sandstaa fixed
- Bug#37281 Time stamps in Falcon on big-endian machine give wrong results
- Vladislav Vaintroub fixed
- Bug#35255 falcon_bug_22165.test fails with debug builds
The three failing test cases are:
- falcon_bug_30124-big
- falcon_bug_36294-big
- falcon_deadlock (a fix for this test failure is in review)
By Cody 2009-01-16 - 17:07
Thanks for adding the failing test cases.
Sounds like things are getting close to a big release.