2015-03-03 16:25 GMT+01:00 Thomas D. <[email protected]>: > Hi, > > Rainer wrote: > > OK, that clarifies. While it is nice to see a RFC5424 header on the local > > log socket, imuxsock does not support this. However, this is not new, it > > is so for quite some years (as a result of pre-journal systemd > integration, > > I think back in v5 days). So this should not be a new fail. I guess the > test > > was not run for some reason, maybe it was among a set I added. > > > > Of course, it would make sense to enhance imuxsock to support multiple > > formats. The question is if that really makes sense. In spit of the > recent > > movement towards journal and the trend to replace imuxsock by imjournal > on > > such systems, I think the days of this module are more or less over. > > > > At least as a short-term measure we should disable this test for Gentoo, > I > > just need to see how I can do that. Do you know how to check for Gento > > inside the testbench scripts? > > I really want to avoid something like that. If it is only an "invalid" test > I would patch the Makefile in our ebuild to disable this test (and document > the reason). But please don't disable it just for Gentoo in source. > > But I first want to understand what's failing and why. Would be nice if you > could answer some questions: > > 1) I am very sure that the same test passed in 8.7.0 when we added 8.7.0 to > our repository in February. Just tested v8.7.0 today and this test is (now) > failing, too ;) Any idea what could cause this? Could it be related to the > util-linux package which was bumped from v2.25.2 to v2.26 [1] recently on > Gentoo? >
Yes, that could be. In fact, any change to logger or the syslog() calls could cause the format change. > 2) If I understand you correctly you are wondering to see a RFC5424 header > at all. Is there something we should turn off? > Don't get me wrong: the 5424 header is nice to see, and years ago we (IETF) hoped that this would happen everywhere. However, it'd didn't materialize. So this probably is a good move (if the tools changed) but a bit surprising. > > 3) Gentoo is not systemd at all. The test system for example, has no > systemd > installed. We rely on imuxsock for chroots for example. Will the RFC5424 > header cause any trouble because it is not really support in rsyslog or is > it really just the test...? > It causes misformatting in all cases. I was under the assumption that this is not new and so it should have no further implications. If it is new, it could cause problems. The right cure would be to make rsyslog's imuxsock support it. If you say you didn't make a decision to move to systemd, it is probably worth at least checking what it requires for imuxsock to be upgraded. I just really don't like to make changes to a piece of code that is almost dead. Much better things to do ;) HTH Rainer > > > [1] http://www.spinics.net/lists/util-linux-ng/msg09816.html > > > -Thomas > > > _______________________________________________ > rsyslog mailing list > http://lists.adiscon.net/mailman/listinfo/rsyslog > http://www.rsyslog.com/professional-services/ > What's up with rsyslog? Follow https://twitter.com/rgerhards > NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad > of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you > DON'T LIKE THAT. > _______________________________________________ rsyslog mailing list http://lists.adiscon.net/mailman/listinfo/rsyslog http://www.rsyslog.com/professional-services/ What's up with rsyslog? Follow https://twitter.com/rgerhards NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE THAT.
