Seems fine to me.   Although, someone who knows more shell than me
should probably look at it.

Why do we muck with the PATH variable in force_kmod_reload function?

Ethan

On Wed, Jun 29, 2011 at 15:55, Ben Pfaff <[email protected]> wrote:
> Otherwise its heartbeats start failing after the reload and the XenServer
> reboots after a minute or so.
>
> Thanks to Justin Pettit for figuring out that this was HA-related.
>
> Bug #5706.
> Reported-by: Henrik Amren <[email protected]>
> ---
>  xenserver/etc_init.d_openvswitch |   14 +++++++++++++-
>  1 files changed, 13 insertions(+), 1 deletions(-)
>
> diff --git a/xenserver/etc_init.d_openvswitch 
> b/xenserver/etc_init.d_openvswitch
> index 8103900..534b81d 100755
> --- a/xenserver/etc_init.d_openvswitch
> +++ b/xenserver/etc_init.d_openvswitch
> @@ -81,6 +81,18 @@ start () {
>     touch /var/lock/subsys/openvswitch
>  }
>
> +force_reload_kmod () {
> +    start force-reload-kmod
> +
> +    # Restart the high-availability daemon if it is running.  Otherwise
> +    # it loses its heartbeat and reboots the system after a few minutes.
> +    if pidof xhad >/dev/null && test -e /etc/xensource/xhad.conf; then
> +       PATH=$PATH:/opt/xensource/xha
> +       action "Stopping HA daemon" ha_stop_daemon
> +       action "Starting HA daemon" ha_start_daemon
> +    fi
> +}
> +
>  stop () {
>     $ovs_ctl stop
>     stop_daemon ovs-xapi-sync
> @@ -114,7 +126,7 @@ case $1 in
>         $ovs_ctl version
>         ;;
>     force-reload-kmod)
> -        start force-reload-kmod
> +        force_reload_kmod
>         ;;
>     help)
>         printf "openvswitch 
> [start|stop|restart|reload|force-reload|status|version]\n"
> --
> 1.7.4.4
>
> _______________________________________________
> dev mailing list
> [email protected]
> http://openvswitch.org/mailman/listinfo/dev
>
_______________________________________________
dev mailing list
[email protected]
http://openvswitch.org/mailman/listinfo/dev

Reply via email to