Heiko Schlittermann wrote: > citing Frank's first mail: > > 2011-02-09 11:28:25 1Pn7Hd-00077B-Dv unable to set gid=104 or uid=104 > (euid=103): local delivery to list<[email protected]> > > The local delivery *should* start as euid=0, since should be setuid > root.
It MAY. It doesn't always have to be so (at least thru 4.6X). Mine are neither. > As far as I understood deliver.c, exim fork()s, exec()s and then > changes the UID. But there are some cases where this uid change does not > take place, .e.g if "deliver_drop_privileges" is set. .. or when and where the exim binary is NOT setuid root, delivery was not invoked by root, and the particular delivery transport doesn't need either because it calls for delivery under an EGID of which the exim daemon runner is a *member*... Or .. just maybe applicable here .. as above, but the running UID/EUID is NOT a member of the egid the tranport calls for to make this delivery (gid 104)? The Devil is in such details... Bill > > Probably Frank can put some logging into deliver.c, or at least > ask for logging the PID, it should show us the PID transition… > > -- ## List details at http://lists.exim.org/mailman/listinfo/exim-users ## Exim details at http://www.exim.org/ ## Please use the Wiki with this list - http://wiki.exim.org/
