Date: Thu, 20 Mar 2008 01:41:33 +0100 From: "barbara" <barbara.xxx1975@libero.it> To: "kris" <kris@FreeBSD.org> Cc: freebsd-bugs <freebsd-bugs@FreeBSD.org>, vwe <vwe@FreeBSD.org> Subject: Re: kern/121809: unable to umount Message-ID: <JY0799$D3D5C88B3D634555785DE8744BCF2085@libero.it>
next in thread | raw e-mail | index | archive | help
> barbara wrote: > > > Today, after a reboot, I did the test again and the problem was still= existing. > > So I've rebuilt world+kernel again after updating the sources. > > I still have to reboot to run make installworld but the problem seems= gone after that. Is this possible? Maybe I've missed an error during the= previous install(world)|(kernel) (?). > > I'm going to reboot on the next minutes and repeat the test again. > > If it will be all right, do you need other tests/info like booting th= e old kernel or update the sources at more or less the time they was upda= ted yesterday? If so, please, tell me. > > Hi Barbara, > > The problem is known and we have a fix in development. I'll follow up = > with you once it's ready for testing. > > Kris > > barbara wrote: > > > Today, after a reboot, I did the test again and the problem was still= existing. > > So I've rebuilt world+kernel again after updating the sources. > > I still have to reboot to run make installworld but the problem seems= gone after that. Is this possible? Maybe I've missed an error during the= previous install(world)|(kernel) (?). > > I'm going to reboot on the next minutes and repeat the test again. > > If it will be all right, do you need other tests/info like booting th= e old kernel or update the sources at more or less the time they was upda= ted yesterday? If so, please, tell me. > > Hi Barbara, > > The problem is known and we have a fix in development. I'll follow up = > with you once it's ready for testing. > > Kris > Ok Kris, let me know! In the meanwhile, I could add that, as I said, I was pretty sure that aft= er the new makeworld+kernel, the problem was fixed. But I'm still able to= reproduce it. This is the sequence in my test case: # mount /dev/ad4s1e /mnt/bsd6 # touch /mnt/bsd6/test # umount /mnt/bsd6 # mount /dev/ad4s1e /mnt/bsd6 # rm -f /mnt/bsd6/test # umount /mnt/bsd6 umount: unmount of /mnt/bsd6 failed: Device busy What tricked me into thinking that the problem was fixed, is the fact tha= t if I skip the 3rd and 4th steps, all seems working as expected. I've also collected (almost) all the informations asked by vwe, but from = what I'm arguing you have yet enough elements. Ask me if I'm wrong or if you need something else. Cheers
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?JY0799$D3D5C88B3D634555785DE8744BCF2085>