From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrey Borzenkov Date: Sun, 05 Feb 2006 11:46:26 +0000 Subject: Re: kernel 2.6.15, udev 084 Message-Id: <200602051446.27084.arvidjaar@mail.ru> List-Id: References: <200602051306.57238.arvidjaar@mail.ru> In-Reply-To: <200602051306.57238.arvidjaar@mail.ru> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To: linux-hotplug@vger.kernel.org -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Sunday 05 February 2006 14:01, Mark Carey wrote: > On 2/5/06, Andrey Borzenkov wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > > Hash: SHA1 > > > > On Sunday 05 February 2006 12:56, Mark Carey wrote: > > > All, > > > > > > In a quest to get pptp working for adsl (I hear kernel 2.6.15 has some > > > notable inclusions in this regard, ex pptpclient.sf.net) I have > > > compiled kernel 2.6.15.2 rebooted and note that udev version 057 no > > > longer works with this kernel. On reboot had no /dev/eth0 and > > > /dev/eth1. > > > > What are those devices and what do you need them for? Interfaces never > > had any device entry in /dev. I guess they have been created by accident > > before. > > As you rightly indicate these are interfaces and they have not been > created even on the working system, > > [root@angus:/dev]# ls eth* > ls: eth*: No such file or directory > > I have booted an Ubuntu Live CD and checked that the only entries in > /dev prior to mounting the tmpfs are console and null - IIRC these are > required at a minimum to get a shell running? > > Given that the devices {eth0, eth1} have never existed, in recent > history, why then does > > [root@angus:/dev]# ip link set eth0 up > SIOCGIFFLAGS: No such device > > fail with the new kernel/udev combination claiming SIOCGIFFLAGS and > work with the old? > is driver loaded in new kernel? Do interfaces still have the same names? > Also note that the ppp device has problems, > > /usr/sbin/pppd: pppd is unable to open the /dev/ppp device. > You need to create the /dev/ppp device node by > executing the following command as root: > mknod /dev/ppp c 108 0 > It is chicken and egg problem. /dev/ppp is created by loading ppp_generic (= at=20 least in 2.6.15.2) but someone has to load it ... having /dev/ppp in advanc= e=20 triggers module loading. Alternatively compile ppp_generic in kernel and no= t=20 as module. > Do I need to reenable hotplug, > > echo "/sbin/udevsend" > /proc/sys/kernel/hotplug > If kernel is properly setup, it should not be necessary. Do you have=20 CONFIG_KOBJECT_UEVENT=3Dy? > after processing queued events? Or should I no longer be using > hotplug with the new udev? > It should be possible to do everything with udev that hotplug did before. - -andrey -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) iD8DBQFD5eWTR6LMutpd94wRAmYXAKCN+ci/6Opd1Np5ORRIsScll03PgwCfZHgC k6HpkkeQ549vRuUHrNDBSVw=BDK7 -----END PGP SIGNATURE----- ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=103432&bid#0486&dat=121642 _______________________________________________ Linux-hotplug-devel mailing list http://linux-hotplug.sourceforge.net Linux-hotplug-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel