Date: Sat, 23 Sep 2000 03:12:02 +0200 (CEST) From: zrq501j@tninet.se To: Salvo Bartolotta <bartequi@inwind.it> Cc: freebsd-questions@FreeBSD.ORG Subject: Re: FreeBSD-problems... Message-ID: <Pine.BSF.4.21.0009230310360.56841-100000@localhost.tninet.se> In-Reply-To: <20000922.18493100@bartequi.ottodomain.org>
next in thread | previous in thread | raw e-mail | index | archive | help
I am trying this now... I will read the UPDATING-file to in the future =) Thankyou!! =) // Johan Andersson <zrq501j@tninet.se> > >>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<< > > On 9/22/00, 2:25:33 PM, zrq501j@tninet.se wrote regarding > FreeBSD-problems...: > > > > Hello!! > > > I did a mistake some days ago: I used cvsup to download the source of > > 5.0-CURRENT, but the kernel did'nt compile (the rest of the system > did, > > but not the kernel) so I tried to get 4.x-stable again by cvsup, but > > something was wrong: > > The system dont compile and the kernel dont want to compile ether. > > > I did like this first: > > cvsup'd down the 5.0-current-source > > make buildworld <-- it worked > > > > Dear Johan Andersson, > > If you read /usr/src/UPDATING (under -STABLE as well as under > -CURRENT), you will learn what the correct (=recommended, supported) > updating procedure is (in each case). > > N.B. Both -CURRENT and -STABLE follow analogous updating scheme; > -CURRENT, however, requires additional **special** considerations > (e.g. hints); which considerations are made (wait for it) in > /usr/src/UPDATING. > > > > > make installworld <-- it worked to > > config MY_KERNEL <-- did work, but a lot of strange warnings... > > make buildkernel KERNEL=MY_KERNEL <-- did not work > > > > You have not followed the prescribed method (N.B. here you dealing > with **CURRENT** sources), so no wonder something did not work. > > In fact, if you have a look at the /usr/src/UPDATING relative to > -CURRENT, you'll find the warnings (and the answers) yourself. > > By the way, since SMP code is being worked on under -CURRENT (ie > -CURRENT **may** potentially be found in a wild unpredictable > [quantum] state :-), a static tag was created a couple of weeks ago. > If you really wish to update to a safe -CURRENT, you should use it in > your cvsupfile; alternatively, you can specify a date in the cvsup > file, as per the instructions in cvsup (1). Needless to say, you > should browse the -current archives before **even** thinking of > attempting such an operation. > > > > > So... then I cvsup'ed down 4.1-stable again. > > make buildworld <-- dont work > > make installworld <-- dont work ether > > > > Why try to install a world that has not even been built (or has been > partially built) ? > > > > > config MY_KERNEL <-- Dont work... > > > Again, please read /usr/src/UPDATING. > > > > make buildkernel <-- becouse the config didn't work this dont work > ether. > > > <snip> > > > Summing up (If I correctly interpret all you have done): > > you built and somehow installed a -CURRENT world; upon which world you > tried to install part of a -STABLE world; your kernel, however, is > still the old (ie "stable") kernel. > > In the language of Q. M., I would say that your system is described by > a linear superposition of the (quantum) states -CURRENT and -STABLE, > with *unknown* coefficients :-) > > I have little idea of the present "state" of your system; I am not > quite sure what the most painless recovery strategy (if any) is at > this stage. I am not quite sure even that such a downgrading operation > can be successfully performed, either. > > > ******************************************************************* > ************************** Disclaimer ***************************** > ******************************************************************* > > Mutatis mutandis, the following suggestion, described in Ray > Kohler's own words on July 6, 2000, ***might*** work: > > <blockquote> > > >> Is building 4.0-stable on 5-current supported? If so, whats the > proper > >> procedure supposed to be, other than cvsup the sources and make > buildworld > >> (as a first step)? > >> > >> If its not supposed to work, or is just plain broken for a while, ill > try > >> a sysinstall "upgrade" to 4-s bins. buildworld is bombing for me in > >> cc1plus. > > > >Go to /usr/src/gnu/usr.bin/binutils and do make all install clean. > >You need to downgrade binutils this way before you can downgrade > >gcc. > > > >-- > >Ray Kohler <rjk191@psu.edu> > > Thanks, that worked. I did a buildworld and made a kernel after that > and installed both and everything seems to be working nicely now. > > </blockquote> > > ******************************************************************* > ********************** End of Disclaimer ************************** > ******************************************************************* > > If you succeed in making buildworld, you will ... read > /usr/src/UPDATING, and then try to go on. > > I hope I am not missing any more gotchas myself <pant, pant> > > Needless to say, as extrema ratio (~ last resort), you can back up all > your valuable files (config, data) and reinstall. > > Never forget to read /usr/src/UPDATING again :-)) > > Best regards & best of luck, > Salvo > > > > > All programmers are optimists. Perhaps this modern sorcery especially attracts those who believe in happy endings and fairy godmothers. Perhaps the hundreds of nitty frustrations drive away all but those who habitually focus on the end goal. Perhaps it is merely that computers are young, programmers are younger, and the young are always optimists. But however the selection process works, the result is indisputable: "This time it will surely run," or "I just found the last bug." -- Frederick Brooks, "The Mythical Man Month" To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0009230310360.56841-100000>