A corrupt index

[Adam Sjøgren]

Adam writes:

>> However, it occurs to me you could try xapian-compact on the bad
>> database (if you didn't already).  That reads the entries in the table
>> sequentially for which the dividing keys don't matter, so it may give
>> you a good copy - I'd try that first.
>
> I hadn't tried that - I have now started a run on the copy of the index,
> to try it out.

Compacting went without any errors (run time around 10 hours):

    $ xapian-compact index index.compact-2011-11-18
    postlist: Reduced by 47% 64119424K (136063368K -> 71943944K)
    record: Reduced by 35% 635936K (1772776K -> 1136840K)
    termlist: Reduced by 41% 33945728K (81677456K -> 47731728K)
    position: Reduced by 38% 368390992K (945391144K -> 577000152K)
    spelling: doesn't exist
    synonym: doesn't exist
    $ 

However the following xapian-check failed (after 16 minutes):

    $ xapian-check /z/orne/nzdb/index.compact-2011-11-18
    record:
    baseB blocksize=8K items=10465475 lastblock=142104 revision=1 levels=2 root=601
    B-tree checked okay
    record table structure checked OK

    termlist:
    baseB blocksize=8K items=20930950 lastblock=5966465 revision=1 levels=3 root=343098
    B-tree checked okay
    xapian-check: DatabaseError: inflate failed (invalid distance too far back)
    $ 

Odd, or?


  Best regards,

    Adam

-- 
 "Hør nu lige her: Når Bill Gates personligt siger,         Adam Sjøgren
  at det er sådan, så må du altså tro på det !"        asjo at koldfront.dk