Em 10-02-2014 21:18, Zbigniew escreveu: > 2014-02-11 0:09 GMT+01:00, Predrag Punosevac <punoseva...@gmail.com>: > >>> Yes, you're right; after adding "0x" at the beginning it works! Thanks. >>> But that's not the end of the problems, unfortunately... :( >>> If I create - according to "man rum" - an /etc/hostname.rum0 file >>> containing: >>> #v+ >>> nwid <nwid> >>> wpakey <wpakey> >>> dhcp >>> #v- >> What happens if you put the following and reboot >> >> dhcp nwid <nwid> wpakey <wpakey> > Sorry, guys: "human factor". When editing /etc/hostname.rum0 I tried > to use <Del> key, which instead of removing character, inserted two > new - one of them I left by mistake. > > Now it works (with simple setup, taken from man)! Thanks again. > > But my questions are still actual: > > 1. Is wpa_supplicant flawed? No it is not, but it is not well suited for the much simpler OpenBSD style of doing interface initialization. > 2. If it somehow can be forced to work: where should I execute it, to > make it act _before_ the system reads /etc/hostname.rum0 ? You could thinker with your /etc/rc to call wpa_supplicant before it calls /etc/netstart or you can edit your /etc/netstart script to call it before initializing your wireless interfaces. > > Now it's rather matter of curiosity, can I rely on wpa_supplicant at > all (in other situations), or it's not reliable.
Yes, it is reliable. But if you do not need EAP authentication methods, you won't need it and I should say that OpenBSD's wpa-psk implementation, is faster. Cheers, -- Giancarlo Razzolini GPG: 4096R/77B981BC
