On Fri, 2014-02-14 at 11:49 +0100, Stephan Sachse wrote: > > You didn't say if you had applied my experimental patch or not. I'm > > guessing not but I can't be sure.
> no, this was only the complete log of my "i lost my brain" mail. > > 2) Find the lxc-devsetup script (in lxc/config/init/systemd/lxc-devsetup > > in the source tree) and run that as root to see if we have better luck > > under devtmpfs. > output attached Maybe there's still some hope here, taking a closer look. Now... Was that run with the stock code that was still trying to do the mknod in setup_autodev or with the code you added attempting to do the bind mount? You still need the code doing a bind mount as you still can not do a mknod. Unfortunately, you didn't post your final patch but, from what I see in your initial patch, I see you have two error exits with almost identical messages, only one prints out a path and the other prints out the device name. What I see here is the device name: lxc-start 1392374433.579 ERROR lxc_conf - Operation not permitted - Error creating null That exit occurs after a failed mknod. That's telling me that either you're testing this with the original stock logic or there's a flaw in your patch and it's still trying to do the mknod, which we know will not work in a user namespace. > -- > Software is like sex, it's better when it's free! > > _______________________________________________ > lxc-devel mailing list > [email protected] > http://lists.linuxcontainers.org/listinfo/lxc-devel -- Michael H. Warfield (AI4NB) | (770) 978-7061 | [email protected] /\/\|=mhw=|\/\/ | (678) 463-0932 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0x674627FF | possible worlds. A pessimist is sure of it!
signature.asc
Description: This is a digitally signed message part
_______________________________________________ lxc-devel mailing list [email protected] http://lists.linuxcontainers.org/listinfo/lxc-devel
