On 29/01/2015 16:49, Jeremy Harris wrote: > OK, thanks - I have it (I think). The previous bugfix was for the case
> of a header_remove in the transport but here we're in a router. It
> seems there is almost identical coding, so the fix is obvious. I
> should have checked before. Just to confirm, the problem that we (me and Gordon) reported earlier
was also from the router section, not transport -- i.e., it never hit
the transport as it could not pass through the router. Apologies if that
was not evident from the logs that I posted before. But glad that it is
out in the open, and we all know the direction to proceed.
>> Also, maybe I should mention I am
>> using FreeBSD 10.1 with Exim 4.85 -- though it should not matter as the
>> problem appears to be across the board.
> Yup, this is a basic application coding fault.
>
> Thanks for persisting with this. I should be able to get the
> development HEAD patched shortly, and the diff attached to bug 1553. Thanks and much appreciated.
