> On Jul 31, 2016, at 9:38 AM, Brian <a...@cityscape.co.uk> wrote: > > A successful ping is an indication of association between interface and > AP, although there would be other ways of determining this. It says > little or nothing about the state of routing. Whatever is being pinged > needn't have a route to the internet. Which argues for users filling in > netmask and gateway for themselves and having something they can trust, > rather than relying on some iffy wicd method.
Yeah. Their gateway and netmask 'calculations' do leave a little to be desired, even though they were correct in my case. I spent some time with the source too, but I never saw the part where the gateway and netmask were determined. Without just getting those numbers from the user, they could be very wrong from their guessing; then they'd have built a bad routing table too. I wonder what they'd come up with on a 'real' network. I did see where one of the first things they do is clear the routing table. That would be another bad idea if there are more than the 2 entries. I'm on Wheezy/Jessie -- maybe these've been corrected in later versions. > Glad you solved your issue, incidentally. Note that if this so-called > verifying had been switched off it would never have occured. :) Tell me about it :-) If they'd been doing it everywhere it might have been easier to find. Or, of course, if I hadn't broken ping... Thanks again for your help, Brian. You led me on some very interesting trips through the wicd outputs. I hope I might be of some help to another bewildered wicd user in the future -- I learned a lot. -- Glenn English
