initctl unable to communicate with init

[Chris Packham]

On Mon, May 3, 2010 at 12:48 PM, Kees Jongenburger
<kees.jongenburger at gmail.com> wrote:
> Hello Chris,
>
> On Mon, May 3, 2010 at 6:57 PM, Chris Packham <judge.packham at gmail.com> wrote:
>> Hi,
>>
>> We've been using upstart in our embedded products for some time now
>> (we're using upstart 0.5.1). I'm in the process of porting our
>> software to a new hardware platform and I'm running into a problem
>> thats got me stumped.
>>
>> Our first init script mounts the file systems and then uses "initctl
>> reload" to reload the upstart config from the on-flash upstart config.
>> The problem is that "initctl reload" just hangs waiting forever. As
>> far as I've been able to debug so far it seems to be in the dbus
>> message dispatch which should have sent the message to init. I think
>> its waiting for a response it never sees, I also think init never got
>> the message.
>
> Are you using a initramfs? who mounts the required file system for
> upstart to work?

It is an initramfs (squashfs) mounted by the kernel/bootloader.

> try replacing upstart by a shell script and once you have a shell
> exec /sbin/init with extra debugging.

I'm already running with #define DEBUG in init/main.c and init=/bin/sh
so I can start init after getting a shell. I might just add -DDEBUG to
configure and see what else I get.

> Double check your kernel configuration to have networking and sockets enabled.
>

grep -e 'CONFIG_\(NET\|INET\|UNIX\)=' .config
CONFIG_NET=y
CONFIG_UNIX=y
CONFIG_INET=y

>> I've run the same code on one of our existing platforms and that
>> starts up normally so I'm thinking its a toolchain/kenrel setup issue
>> but the only differences I can see in that area are the things that
>> _are_ different.
>
> Your hardware might also be different causing a different events to happen
> in a different order (for example a tmpfs to be mounted over the place
> where upstart has it's unix socket waiting).
>

Hmm hadn't thought of that. One of the first things the script does is
mount tmpfs, procfs and sysfs. I might take that out and do it
manually as an experiment.