[Date Prev][Date Next]
Re: (ITS#7377) Poor libmdb error handling
I messed up last reply headers somehow. Trying again. I wrote:
> Also mdb could catch more user errors - including "cannot happen"
> mutex errors from mutex lock/unlock, since some threading user
> errors are be non-obvious from the doc.
That is, "cannot happen" when there is no user error.
> The fork() restriction needs to be documented, as does "don't
> open or close the same database twice in the same process".
Correction, "don't have the same database open twice at the same
time in the same process".