Bug#885463: Submitted to SourceForge also

[Nick Gorham]

Hi,

I have taken a look at this and like Steve, I think the solution is to 
find the race condition that causes the problem.


I have committed a change to 2.3.8pre that will check the cause of the 
failure to open the file and only open as "w" when its safe to do so (at 
least thats the plan).


I don't want to just make it a "a" as that may cause a problem when the 
code does want to delete and recreate the file when ini file entries are 
made.


If you have chance, give it a try and see if it helps your situation.

--

Nick Gorham

Bug#885463: Submitted to SourceForge also

[Ján Jockusch]

Hi,

I submitted the bug report plus patch to SourceForge also, because I see 
no change here at Debian.


https://sourceforge.net/p/unixodbc/bugs/46/

For almost two years, we have a confirmed problem with libodbc1 *and* 
the solution, and the patch is just not going into mainline.


I wonder if the category of the bug can at least be raised to "patch 
submitted" instead of "information required", because I feel I have 
provided all information already.


Please, help. I'm getting a bit desperate now.

The next thing I can think of is to open a github repository for our 
patch and deliver it from there.


I hope it doesn't have to come to that.

Cheers,

Ján