Recovery bugs, classified by me

[Vladislav Vaintroub]

Hello,
Here is my classification of falcon recovery bugs, those with F_RECOVERY. As
I see it, we have 4 real recovery bugs, near the end (or 6 if I count
duplicates).


Category : Persistency bug (not recovery)
http://bugs.mysql.com/bug.php?id=42903 
Bug #42903 Falcon: Inconsistent tablespace after recovery. 
Short description : Not a recovery bug. 
Difference in queries that supposed to return the same results. Classified
as index inconsistence. Search with simpler queries on the same tablespaces
normally does not reveal inconsistencies (but one counterexample produced
without recovery is attached).
Suspected cause: don't know.
Suggested fix: no clue.. Tried to reproduce, all data that was reporting as
missing from index, was there (found using simpler queries)


Category: Undebuggable one time glitch.
http://bugs.mysql.com/bug.php?id=39703 
Bug #39703 Assertion in Section.cpp failed during recovery after
falcon_limit test
Short description. seen in September. Crash dump is not usable for anyone
but Olav (Sparc, wrong endian).
Possible fix:
  Close as none-reproducible



Category: Won't fix 
http://bugs.mysql.com/bug.php?id=39130 
Bug #39130 Unbounded serial log growth with online ALTER
Short description: discussion on how gophers can lag behind
Suspected cause: gophers lag behind
Suggested fix: won't fix (works as designed)

http://bugs.mysql.com/bug.php?id=36993 
Bug#39139 Falcon reports Index SCHEDULE..PRIMARY_KEY in SYSTEM.SCHEDULE
damaged
Suspected cause:
 Kill -9 before system tables were completely created.
Suggested fix: won't fix (good workaround)
Workaround: delete all falcon spaces and serial logs.


Category: Real bugs
http://bugs.mysql.com/bug.php?id=44096 
Bug #44096 Exception: Recovery failed: corrupted serial log
Short description: after truncation, serial log cannot be recovered (Falcon
does not know where to start)

http://bugs.mysql.com/bug.php?id=41840 
Bug #41840 Unbounded tablespace growth during recovery
Was shortly inspected in Athens.
Short description: lost pages.
Suspected cause :
  Pageoverflow pages are lost. Recovery does not do anything with
SRLOverflowPages
Possible fix:
  Delete those pages in recovery (phase3).

http://bugs.mysql.com/bug.php?id=43106 
Bug #43106 Falcon recovery assertion (bdb->buffer->pageNumber == pageNumber)
in Cache.cpp 
Short description: 
 page of all zeros encountered while datapage is expected during redo of
blob delete.
 Hardly reproducible, usable tablespace attached.
Suspected cause:
 Blob datapage not logged (Ann, Kevin - that sounds familiar, right:)?)


http://bugs.mysql.com/bug.php?id=43502 
Bug #43502 Falcon assertion (idx->offset == 0 || (idx->offset >= indexEnd
... 
Sort description:  Datapage corruption and its duplicates. May or may not be
not recovery specific.
Duplicates : http://bugs.mysql.com/bug.php?id=44162 (Bug #44162 Falcon
recovery error DataPage::computeSpaceAvailable got a negative number)
             http://bugs.mysql.com/bug.php?id=43503 (Bug #43503 Falcon
attempting to access exorbitant page number during recovery)
Possible fix: Assign to Ann, she knows the datastructures best.