On Mon, 2019-07-29 at 18:11 +0100, Daniel P. Berrangé wrote: [...] > +virtsecretd.service: secret/virtsecretd.service.in $(top_builddir)/config.status > + $(AM_V_GEN)sed $(VIRTSECRETD_UNIT_VARS) < $< > $@-t && mv $@-t $@ > + > +virtsecret%.socket: remote/libvirt%.socket.in $(top_builddir)/config.status > + $(AM_V_GEN)sed $(VIRTSECRETD_UNIT_VARS) < $< > $@-t && mv $@-t $@ > + > +secret/virtsecretd.conf: remote/libvirtd.conf.in > + $(AM_V_GEN)sed \ > + -e '/[@]CUT_ENABLE_IP[@]/,/[@]END[@]/d' \ > + -e 's/[@]DAEMON_NAME[@]/virtsecretd/' \ > + < $^ > $@ Oh, you didn't replace $^ with $< here! > +secret/virtsecretd.aug: remote/libvirtd.aug.in > + $(AM_V_GEN)$(SED) \ > + -e '/[@]CUT_ENABLE_IP[@]/,/[@]END[@]/d' \ > + -e 's/[@]DAEMON_NAME[@]/virtsecretd/' \ > + -e 's/[@]DAEMON_NAME_UC[@]/Virtsecretd/' \ > + $< > $@ > + > +secret/test_virtsecretd.aug: remote/test_libvirtd.aug.in \ > + secret/virtsecretd.conf $(AUG_GENTEST) > + $(AM_V_GEN)$(AUG_GENTEST) secret/virtsecretd.conf \ > + $(srcdir)/remote/test_libvirtd.aug.in | \ > + $(SED) -e '/[@]CUT_ENABLE_IP[@]/,/[@]END[@]/d' \ > + -e 's/[@]DAEMON_NAME[@]/virtsecretd/' \ > + -e 's/[@]DAEMON_NAME_UC[@]/Virtsecretd/' \ > + > $@ || rm -f $@ All of the issues mentioned in the previous patch unsurprisingly show up here as well, and they should similarly be addressed. Just to reiterate, the secret/virtsecretd.aug rule is the one that should be used as a model. The comments above also apply to patches 22-32. -- Andrea Bolognani / Red Hat / Virtualization -- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list