On Mon, Oct 20, 2008 at 6:07 AM, Harald Barth <[EMAIL PROTECTED]> wrote:
>
>> So I managed to reproduce this here, and after much instrumenting and
>> tracing, I think that the problem is that the "areq" structure is not
>> always properly initialized in afs_Conn() when we test, among other
>> things, for skipserver = 1.
>
> Thanks, I will test your patch as soon as possible.
>
> Is struct vrequest *areq ever initialized properly? I did not find yet
> where that should be done.

afs_FinalizeReq at the start of afs_Analyze, before anything else is
done, will certainly do it. But that's not always helpful, that late.
_______________________________________________
OpenAFS-devel mailing list
[email protected]
https://lists.openafs.org/mailman/listinfo/openafs-devel

Reply via email to