On Wed, Sep 14, 2005 at 04:52:45PM +0200, Tobias Walkowiak wrote: > On Wed, Sep 14, 2005 at 04:47:48PM +0300, Jari Ruusu wrote: > > Tobias Walkowiak wrote: > > > meanwhile i put it into /etc/init.d/bootmisc.sh and everything is > > > OK again. but that is not the solution as it should be > > > > Loop module is supposed to load automatically when needed. What > > version of module-init-tools are you using? > > 3.2-pre8-1 (the current package from unstable) I strongly suspect it is udev that prevents the autoload. When swapon is first run during boot (in S10checkroot.sh), only the /dev/loop/0 device node exists. Other loop devices than loop0 swapon can't open and hence the normal autoloading mechanism won't work. It probably worked with 2.6.11 because recent versions of udev in unstable disable themselves when they are running under a kernel < 2.6.12, and leave the old /dev in place which has the loop device nodes and allows the normal autoloading to work. Tobias, does it make a difference if you add loop to /etc/modules? It should be possible to explicitly load the loop module that way before the second swapon attempt happens in S35mountall.sh. cheers, Max BTW: You can ignore the error message from "Activating swap". Modules are only loaded after the first swapon, so this error is expected. It used to be silently ignore until Debian initscripts recently started using lsb functions in unstable. - Linux-crypto: cryptography in and on the Linux system Archive: http://mail.nl.linux.org/linux-crypto/
