Re: Wireless NICs not obeying 'SSID' setting.

From: Kevin Gerry <sfpoof_at_gmail.com>
Date: Tue, 10 Jul 2007 06:14:12 -0700
For sure, a mandatory option would be wonderful to be added to the wlan
driver. I just find it surprising that after being 'down' for 1-2 seconds,
it reassigns to another AP.

One of the big 'pains in the butt' is that if I was using that card to NAT
off of, I could easily be breaking the law in some US states.

On 7/9/07, Sam Leffler <sam_at_errno.com> wrote:
>
> Kevin Gerry wrote:
> > Well, this is all it gave when it switched:
> >
> > --
> > messages.0:Jul  9 17:21:22 storage kernel: ral0: sta_roam_check: currssi
> > 23 currate 108 roamrssi 14 roamrate 10
> > messages.0:Jul  9 17:21:37 storage kernel: ral0: link state changed to
> DOWN
> > messages.0:Jul  9 17:21:39 storage kernel: ral0: link state changed to
> UP
> > messages.0:Jul  9 17:21:54 storage kernel: ral0: sta_roam_check: currssi
> > 45 currate 108 roamrssi 14 roamrate 10
>
> Looks like you moved to another ap not because of roaming but because
> you were dropped from your original ap.  This can happen for various
> reasons but the fact that you did not go back to it means the station
> could not communicate w/ the ap so it chose another one.  There are
> other wlandebug knobs you can turn on to see exactly what happened (e.g.
> state+scan).  You can also look at wlanstats output and identify what
> happened (probably).
>
> I changed the behaviour of setting the ssid from being mandatory to
> "desired" for various reasons but it's likely I'll have to add back a
> way to make it mandatory.  But then you'll likely find yourself taking a
> long time to re-connect to your ap if the ral card isn't being reset
> properly or otherwise getting into a bad state (ral has known issues).
> A well-behaved card+driver should  locate the ap you want during the
> re-scan and/or it should roam quickly if it misses the ap but then picks
> it up on a bg scan.
>
>         Sam
>
Received on Tue Jul 10 2007 - 11:14:15 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:14 UTC