During the poweroff/restart of a nfs rootfs based qemu runtime, th network is shutdown by connman, which cause the poweroff/restart process hang with the output "nfs: server 192.168.7.1 not responding, still trying"
Lianhao, Couple things to clarify for the bug: 1. nfs rootfs, you mean unfs right since there's old nfs support in poky-qemu 2. poweroff/restart of qemu you mean select the "shutdown" icon in qemu?
(In reply to comment #1) > Lianhao, > > Couple things to clarify for the bug: > 1. nfs rootfs, you mean unfs right since there's old nfs support in poky-qemu yes, it's should be unfs. > 2. poweroff/restart of qemu you mean select the "shutdown" icon in qemu? I booted the qemu into run level 3 instead of 5, and typed poweroff/restart. The run level 5 Sato should have the same issue.
Currently when quit connmand, it will disconnect the network interfaces that are under its control. When shutdown or reboot, system will kill all the processes including connmand, which caused this bug. Just have a talk with the MeeGo Connman owner, he said this was a design flaw that connman may not cover currently. A workaround is to detect if system is using nfs. If so, connman should NOT take over the interface ownership.
fix by commit: http://git.pokylinux.org/cgit.cgi/poky/commit/?id=771fb55d6b13a92e1b30fbc95fdb1c34ddec7ebe
Verified in M4 RC2.