Mark,
This sort of corruption is usually due to a crash causing HDS not to close
down correctly. It's a bit strange if more than one file is affected though.
The cure is, as you suggest, just to delete the file(s) and start again.
Regards
Alan
> -----Original Message-----
> From: Mark Taylor [mailto:[log in to unmask]]
> Sent: 29 August 2003 10:59
> To: [log in to unmask]
> Subject: corrupted HDS container file
>
>
> Chaps,
>
> I'm looking at a user's CCDPACK problem. It seems that some
> of his ADAM parameter files ($ADAM_USER/calcor.sdf etc) are corrupted:
>
> % hdstrace calcor
>
> CALCOR <STRUC>
>
> ADAM_DYNDEF <DEFAULTS> {structure}
> {structure is empty}
>
> KEEPIN <_LOGICAL> TRUE
> USEEXT <_LOGICAL> FALSE
> TYPE <_CHAR*132> 'NONE'
> OUT <_CHAR*132> 'ro55_dsub.sdf'
> CAL <_CHAR*132> 'ro41.sdf'
> EXPOSE <_CHAR*132> '1'
> PRESERVE <_LOGICAL> TRUE
> SETSAT <_LOGICAL> FALSE
> TITLE <_CHAR*132> 'Output from CALCOR'
> !! Locator refers to an object 'IN' which no longer exists
> (possible
> ! programming error or corrupted HDS container file).
> ! DAT_NAME: Error enquiring the name of an HDS object.
> ! Unable to get information about current object
> ! Application exit status DAT__INCHK, Integrity check
> ! calcor
>
> Not sure yet whether deleting this and starting over will fix
> the problem, but in any case: does anybody know what is
> likely to have caused this corruption? Can misuse of PAR
> cause this? or misuse of HDS (not that CCDPACK is poking
> around in these files directly .. as far as I can remember)?
> Or is the most likely explanation that the file has been
> corrupted by something other than an ADAM application?
>
> I may have once had an idea about the answers to these
> questions but a year or so of intensive java seems to have
> displaced my knowledge of ADAM internals somewhat.
>
> thanks for any help,
>
> Mark
>
> --
> Mark Taylor Starlink Programmer Physics, Bristol University, UK
> [log in to unmask] 0117 928 8776 http://www.star.bris.ac.uk/~mbt/
>
|