[EMBOSS] Problems indexing with dbxfasta

[simon andrews (BI)]

I'm trying to use dbxfasta to index one of the WGS trace databases.
Unfortunately dbxfasta is falling over on me.  The session looks like
this:

$ dbxfasta
Database b+tree indexing for fasta file databases
Basename for index files: traces_oanatinus
Resource name: all
    simple : >ID
     idacc : >ID ACC
     gcgid : >db:ID
  gcgidacc : >db:ID ACC
      dbid : >db ID
      ncbi : | formats
ID line format [idacc]: simple
Database directory [.]:
Wildcard database filename [*.dat]: *.fasta
Release number [0.0]:
Index date [00/00/00]:
Processing file ./nisc-platypus-shotgun-1048960391.fasta
Processing file ./nisc-platypus-shotgun-1071756042.fasta
Processing file ./nisc-platypus-shotgun-1080815515.fasta
Processing file ./nisc-platypus-shotgun-1102160893.fasta
Processing file ./nisc-platypus-shotgun-1104879084.fasta
Processing file ./nisc-platypus-shotgun-1109000445.fasta
Processing file ./nisc-platypus-shotgun-1110804272.fasta
Processing file ./nisc-platypus-shotgun-1116844699.fasta
Processing file ./nisc-platypus-shotgun-1142973027.fasta
Processing file
./wugsc-ornithorhynchus_anatinus-cloneEnd-1113828608.fasta
Processing file
./wugsc-ornithorhynchus_anatinus-cloneEnd-1115655383.fasta
Processing file
./wugsc-ornithorhynchus_anatinus-cloneEnd-1119433885.fasta

   EMBOSS An error in ajindex.c at line 615:
Maximum retries (100) reached in btreeCacheFetch for page 14240710656

The same files have indexed OK with formatdb.  I havent' tried with
dbifasta as I'm trying to move everything over to the new dbx system
(and the rest of our databases have processed OK with dbx(fasta|flat)).

Anyone have any ideas about how to debug this?

Cheers

Simon.

-- 
Simon Andrews PhD
Bioinformatics Group
The Babraham Institute

simon.andrews at bbsrc.ac.uk
+44 (0) 1223 496463