isnt that a new function which should be declared in some header?
Is it only ppp that causes issues? I think the most failsafe way would be to say that once we have an interface with an IP, we are "not-offline". You think you could extend your 0.6 patch accordingly?
Further, we need a similar approach for NM 0.7. The other heuristic we could use would be to add a new "non-offline" state (MAYBE_OFFLINE), which is used instead of online if there is any unmanaged device. Combining this with the IP-is-set heuristic above, could make offline state more accurate even.
Feel free to ping me in #ubuntu-mozillateam or -devel to get all this sorted out and landed in intrepid (and hardy for 0.6.6).
thanks for your patch. I would like to consider this. Just one comment for now:
+nm_device_ other_update_ ppp (NMData *data,
+ gboolean is_connected)
isnt that a new function which should be declared in some header?
Is it only ppp that causes issues? I think the most failsafe way would be to say that once we have an interface with an IP, we are "not-offline". You think you could extend your 0.6 patch accordingly?
Further, we need a similar approach for NM 0.7. The other heuristic we could use would be to add a new "non-offline" state (MAYBE_OFFLINE), which is used instead of online if there is any unmanaged device. Combining this with the IP-is-set heuristic above, could make offline state more accurate even.
Feel free to ping me in #ubuntu-mozillateam or -devel to get all this sorted out and landed in intrepid (and hardy for 0.6.6).
Thanks a lot.