Date: Sun, 17 Nov 2002 22:48:19 -0400 (AST) From: "Marc G. Fournier" <scrappy@hub.org> To: freebsd-stable@freebsd.org Subject: Re: -STABLE was stable for long time (Re: FreeBSD: Server or Desktop OS?) Message-ID: <20021117224133.A23359-100000@hub.org> In-Reply-To: <20021118023431.60AA1286A5@mail.tmseck.homedns.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 18 Nov 2002, Thomas Seck wrote: > * Marc G. Fournier (scrappy@hub.org): > > > On Mon, 18 Nov 2002, Thomas Seck wrote: > > >> Basically, you want others to do the developing, testing and patching > >> and keep the complaining part for yourself. > > > > No, that is the opposite of my point ... my point is that I *want*, and am > > willing, to do the testing required, as well as do what I can to debug the > > crashes ... I spent several hours last weekend narrowing down a problem > > with -STABLE where it hangs on my server, to the point that a kernel from > > 2002.10.28 boots, but 2002.10.29 doesn't ... that involved at least a half > > dozen CVSup's, builds and reboots of the server ... I'm definitely not > > afraid to do the testing and work on the debugging ... but I don't have > > sufficient knowledge to *fix* the crash ... > > Ah, sorry for misinterpreting you. A'alright, its one of the limitations of email :) > > What I'm complaining about is that those that are MFCng to -STABLE seem to > > be leaving it as "it works on my hardware, so it must work" and drop'ng > > responsibility for their changes ... > > Well, this sounds somewhat familiar to me. But you must agree that the > situation at your site is hard to reproduce for any developer unless you > give him or her shell access. Actually, when Matt Dillon worked with me on one of the crashes, whenever it crashed, I'd compress the vmcore for him to ftp down to his server to work on, as well as the kernel.debug file ... I offered him shell access, but, in his opinion, it was easier for him to work locally where he already had his debug environment setup ... And I'm willing to make those vmcore's available to anyone that wishes to debug the problems ... but the only one that has ever step'd up to that plate *so far* has been Matt ... > > so what happens when its time to do a > > RELEASE? There have been soooooo many changes, how do you know which one > > caused the problem? > > Hm, I do not quite understand you here. Actually, it might not make complete sense ... but if 100 committers make 100 changes to the server, and it isn't until -RELEASE that anyone puts any *real* load onto it and it breaks ... which of those 100 changes might have caused that problem? If ppl like me take snapshots periodically of STABLE and pound the hell out of it in a real life scenario, then the bugs don't get a chance to pile up ... the overall system *would* stay stable ... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20021117224133.A23359-100000>