Date: Tue, 19 Jun 2001 09:59:57 +0200 (CEST) From: Micke Josefsson <mj@isy.liu.se> To: Micke Josefsson <mj@isy.liu.se> Cc: freebsd-questions@FreeBSD.ORG, Greg Lehey <grog@FreeBSD.ORG> Subject: Re: Vinum trouble Message-ID: <XFMail.010619095957.mj@isy.liu.se> In-Reply-To: <XFMail.010613081148.mj@isy.liu.se>
next in thread | previous in thread | raw e-mail | index | archive | help
Just for the record this is what I did to make it work again. Apparently the /bin/sh and /sbin/vinum did not like the older kernel. With the 4.2-fixit CD I installed /kernel, /modules/vinum.ko, /bin/sh and /sbin/vinum. The machine booted up allright! Then before making world again (with 4_2_0_RELEASE) I copied the entire /bin and /sbin contents from CD to my root partition. I rebuilt OK and is yet again up and running! Thanks for the tip about the version mismatch of vinum/kernel. /Micke On 13-Jun-01 Micke Josefsson wrote: > Thanks for your quick reply. I'll have another go at it today... > > On 12-Jun-01 Greg Lehey wrote: >> On Tuesday, 12 June 2001 at 16:50:05 +0200, Micke Josefsson wrote: >>> After failing at "make world" with sources for 4_2_0_RELEASE trying to >>> upgrade >>> my 3.x-STABLE machine vinum seems to have lost its config: >>> >>> >>> >>> When other disks have been found clean is says: >>> >>> vinum kernel module not available >> >> Well, that's not exactly a lost config, is it? >> >>> So I kldload vinum.ko : >>> link_elf: symbol gd_proc not defined (I got a cvsw_dev(?) also at one time) >> >> Looks like you have a mismatch between your kernel and your modules. >> Given the situation you describe, that's not surprising. > > I will try to extract a kernel and vinum.ko from my 4.2 CD and try again. > >> >>> And still get: >>> >>> cant open dev/vinum/mirror: device not configured >>> >>> and it later prompts me to run fsck manually. >>> >>> What's up? Is there any way to rectify things? >> >> Yes. If you're able to build the module, do so and install it. >> Otherwise the simplest thing is probably just to reboot with your 3.x >> kernel and repeat your build. >> > > Unfortunately the 'make world' bailed out some time into the 'installworld' > part. Apparently the system got a signal 12 when installing 'sh' and later > also > when installing 'test'. (At that point I was able to replace the 'sh' with one > I > had from my backups and 'make installworld' again. To no avail...) I ended up > with a system which not entirely 4.2 (but probably near to it) and not 3.x > either > but a mix of the two. So I am unable to start a new make world since I cannot > access /usr, which is where vinum enters the scene. > > > >>> (If I am going to pull back my backup I will probably skip vinum >>> this time) >> >> This has nothing to do with Vinum. This is a broken upgrade. >> > > Of course you are right about that, but having vinum appears (for now) to be > the > only thing that stops me from accessing /usr and being able to cvsup > 3.x-STABLE > and do another make world. Could I do that then I would try a shorter jump > into > 4.x, say 4.1. > > This is no criticism of vinum. It has served well for over a year now. When > evaluating it I disconnected one of the drives and after half a second of > locomotion the other drive started pumping out data as if nothing had > happened! > It really seems to work as such, I just gives me trouble when trying to > upgrade, > which is not necessarily an easy task even without vinum. > >>> BTW. A normal upgrade (with binaries) does not recognize vinum >>> drives, right? I only saw da1s1e and da2s1e not the mirrored drive >>> they actually constitute. >> >> I'm not sure what this question means. > > When doing an upgrade with my CD with 4.2 on it. I am passed to the disklabel > editor and prompted to insert mount points for my drives and partitions. There > I > see the two drives that are mirrored but only as da2s1e and da2s2e (not as > /dev/vinum/rmirror). What shall I do? Is it ok to set the mount point to /usr > for > both of them (or just one of them since they are mirrored)? Will > vinum see to the rest? > >> >> Greg >> -- >> When replying to this message, please copy the original recipients. >> If you don't, I may ignore the reply. >> For more information, see http://www.lemis.com/questions.html >> See complete headers for address and phone numbers >> >> To Unsubscribe: send mail to majordomo@FreeBSD.org >> with "unsubscribe freebsd-questions" in the body of the message > > ---------------------------------- > Michael Josefsson, MSEE > mj@isy.liu.se > > This message was sent by XFMail > running on FreeBSD 4.3-STABLE > ---------------------------------- > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-questions" in the body of the message ---------------------------------- Michael Josefsson, MSEE mj@isy.liu.se This message was sent by XFMail running on FreeBSD 4.3-STABLE ---------------------------------- 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?XFMail.010619095957.mj>