On Mon, Aug 5, 2013 at 1:12 PM, Andrei Gherzan <and...@gherzan.ro> wrote:
> > > > On Mon, Aug 5, 2013 at 2:51 AM, Rich Bayliss <richbayl...@gmail.com>wrote: > >> Sorry Paul - I am new to all this. I have checked and I am using Poky >> (1.4) >> >> Rich >> >> On 4 August 2013 23:35, Paul Eggleton <paul.eggle...@linux.intel.com> >> wrote: >> > On Sunday 04 August 2013 01:27:01 Rich Bayliss wrote: >> >> On 2 August 2013 13:47, Paul Eggleton <paul.eggle...@linux.intel.com> >> wrote: >> >> > On Tuesday 30 July 2013 11:43:43 Rich Bayliss wrote: >> >> >> I am trying to build for Raspberry Pi including "read-only-rootfs" >> in >> >> >> my image features. My aim is to have my SD Card read-only and at >> some >> >> >> point add a read-write overlay to certain directories. This should >> >> >> enable my system to boot fresh each time, and have some persistent >> >> >> storage for user files etc. >> >> >> >> >> >> However, after building my image I can SSH into the system and issue >> >> >> "touch test" to create a file in my home directory, then after a >> >> >> reboot it is still there. That isn't very read-only :) >> >> >> >> >> >> Am I missing something, or is this working incorrectly? >> >> > >> >> > It sounds like it's working incorrectly. Since meta-raspberrypi >> constructs >> >> > the SD card image using its own custom class I wonder if it has >> anything >> >> > to do with that. Andrei, do you know anything about this? >> >> > >> >> > BTW, are you using sysvinit or systemd in this image? >> >> >> >> I am using the standard sysvinit/systemd - I haven't specified either, >> >> so whichever is default. >> > >> > Except the default depends upon what DISTRO you are using. Which DISTRO >> are >> > you using - "poky" or something else? >> > >> > Cheers, >> > Paul >> > >> > -- >> > >> > Paul Eggleton >> > Intel Open Source Technology Centre >> > > Will check on this tomorrow. I've been on the roads these days. > Summary: I managed to find the root of the problem. We used to have an old fstab which i'm going to get rid of with a patch today - as the default one is working ok and fixes this issue. The problem was related to: tmpfs /run tmpfs mode=0755,nodev,nosuid,strictatime 0 0 We didn't have this specified in our fstab and would make udevd to keep the device busy => remount fail. Thank you for reporting this. And sorry for being late with investigation - managed to get some vacation :) -- *Andrei Gherzan* m: +40.744.478.414 | f: +40.31.816.28.12
_______________________________________________ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
