[Yocto-builds] buildbot failure in The Yocto Autobuilder on nightly-qa-systemd

[Stanacar, StefanX]

On Thu, 2013-09-19 at 09:50 +0100, Burton, Ross wrote:
> On 19 September 2013 09:16, Stanacar, StefanX
> <stefanx.stanacar at intel.com> wrote:
> > I assume it's the systemd+connman combination + the right timing
> > when we start the tests and the network get's restarted by
> > connman-applet.
> 
> The applet shouldn't be restarting the network.  Is that just a hunch
> or do you have something to back it up?
> 

It's mostly a hunch. Maybe not the applet, but the connmand process
which is started in the background?
Here what's a I see:

[  OK  ] Started Xserver startup without a display manager.
[  OK  ] Started Restore Sound Card State.
[  OK  ] Started LSB: Infrared port support.
[  OK  ] Started LSB: Remove /etc/nologin at boot.
[  OK  ] Started Permit User Sessions.
         Starting Getty on tty1...
[  OK  ] Started Getty on tty1.
[  OK  ] Started LSB: simple distributed compiler client and server.
[  OK  ] Found device /dev/ttyS0.
         Starting Serial Getty on ttyS0...
[  OK  ] Started Serial Getty on ttyS0.
[  OK  ] Reached target Login Prompts.
Installing knfsd (copyright (C) 1996 okir at monad.swb.de).
svc: failed to register nfsdv2 RPC service (errno 111).
ip_tables: (C) 2000-2006 Netfilter Core Team
systemd-logind[136]: Watching system buttons on /dev/input/event0 (Power
Button)

Poky Next (Yocto Project Reference Distro) 1.4+snapshot-20130919
qemux86-64 ttyS0

qemux86-64 login: systemd-udevd[85]: Failed to apply ACL
on /dev/snd/seq: Operation not supported
systemd-udevd[92]: Failed to apply ACL on /dev/sequencer: Operation not
supported
systemd-udevd[84]: Failed to apply ACL on /dev/sequencer2: Operation not
supported
systemd-udevd[85]: Failed to apply ACL on /dev/snd/timer: Operation not
supported
nf_conntrack version 0.5.0 (1910 buckets, 7640 max)
IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
e1000: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready

A few seconds after I started runqemu ... , I've started a ping in
another terminal:

[stefans at firebird b2]$ ping 192.168.7.2
PING 192.168.7.2 (192.168.7.2) 56(84) bytes of data.
64 bytes from 192.168.7.2: icmp_seq=1 ttl=64 time=8.39 ms
64 bytes from 192.168.7.2: icmp_seq=2 ttl=64 time=0.908 ms
64 bytes from 192.168.7.2: icmp_seq=3 ttl=64 time=5.26 ms
64 bytes from 192.168.7.2: icmp_seq=4 ttl=64 time=0.412 ms
64 bytes from 192.168.7.2: icmp_seq=5 ttl=64 time=5.79 ms
64 bytes from 192.168.7.2: icmp_seq=6 ttl=64 time=0.557 ms
64 bytes from 192.168.7.2: icmp_seq=7 ttl=64 time=1.24 ms
64 bytes from 192.168.7.2: icmp_seq=23 ttl=64 time=3.18 ms
64 bytes from 192.168.7.2: icmp_seq=24 ttl=64 time=1.00 ms
64 bytes from 192.168.7.2: icmp_seq=25 ttl=64 time=0.280 ms
64 bytes from 192.168.7.2: icmp_seq=26 ttl=64 time=6.97 ms
64 bytes from 192.168.7.2: icmp_seq=27 ttl=64 time=7.31 ms

Network starts up early (the image was still booting/starting services).
Then aproximately at the time it reached login: ping stopped (seq 7 to
23). The thing is we start the tests right when it reaches login: (we
had a problem before with the ping failing on systemd images because we
were hiting at the time of:
IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready). Then we increased
the number of pings/time waiting for a reply.
Again, I have the same build as on the AB, but I'm not seeing the ssh
thing...

Cheers,
Stefan



> Ross