tdb/tdb.c: tdb_oob() significance (when not using MMAP)

[dhruva]

Hi,

On 10/18/06, tridge at samba.org <tridge at samba.org> wrote:
>  >  On VMS port, we are finding some cases where the call to stat returns
>  > a size smaller than the total lenght (file pointer offset + size of
>  > data to be written).
>
> I suspect that this indicates a real bug somewhere. Changing tdb_oob()
> to hide this bug is not the right approach - instead you need to find
> the bug :-)

I agree, will dig into our modifications.

> Does the tdb test suite pass?

Currently, we have not ported (even attempted to build) the test
suite. We will do it shortly.

> Also, does VMS have a pwrite() call? Using lseek/write is not nearly
> as good as using pwrite due to race conditions on inherited file
> descriptors.

pwrite() and pread() are available. However, I am not sure how it will
work on a clustered environment. I will give it a shot. Don't you
think fcntl() should handle probable race conditions.

 Thanks for clearing my doubt on tdb_oob(), glad I posted the query
without just blocking the portion of code and bursting elsewhere!

regards,
dk

-- 
dhruva (dk)
Contents reflect my personal views only!