[pkg-wpa-devel] Bug#678327: / Bug#678077

[Stefan Lippers-Hollmann]

Hi

I won't have much time to attend to these two bugs before the end of 
next week, but I hope to find some time to actually debug this then.
Any help with these would be appreciated.

Bug#678327
On Wednesday 20 June 2012, Ben Armstrong wrote:
[…]
> I was excited to see in the changelog that 1.0 adds a 'blacklist BSSID'
> command, something I've long wanted for congested areas where I am
> accepting connections for any open network with any ESSID, but many
> are not free and therefore not suitable for my purposes.
> 
> Oddly enough, every time I use it, wpa_cli is answering FAIL, yet when
> I examine the blacklist with 'blacklist' the entry is shown. Also, if I
> bump up the log level to DEBUG, I see suitable messages in syslog, e.g.
> (actual BSSID obscured for privacy):
> 
> Jun 20 17:28:30 shade wpa_supplicant[30271]: CTRL_IFACE: BLACKLIST bssid='00:11:22:dd:ee:ff'
> Jun 20 17:28:30 shade wpa_supplicant[30271]: Added BSSID 00:11:22:dd:ee:ff into blacklist
> 
> Here is the wpa_cli input and output associated with these messages:
> 
> > blacklist 00:13:a3:0f:63:43
> FAIL
> > blacklist
> 00:13:a3:0f:63:43
> 
> Because of the FAIL response, I don't have the confidence the blacklist
> will actually do its job.

Beyond that, it doesn't actually have an immediate effect - an existing
connection doesn't appear to be disconnected in favour of a 
non-blacklisted cell. However once something else forces a disconnect,
the cell appears to be actually blacklisted. At the moment I'm not sure
about the intended behaviour in detail either.


Bug#678077
On Wednesday 20 June 2012, Ben Armstrong wrote:
> To clarify what I meant by "but only the first time", follow this example:
> 
> > 1
> > 2
> > 3
> > 4
> > <up-arrow>
> > 3
> > <up-arrow>
> > 2
> 
> That is, if you press two consecutive up-arrows, the line before last,
> and the line before that are retrieved from the history. The bug only
> occurs on the first of the up-arrows.

Confirmed, hopefully not too hard to fix - the internal readline code
seems to be rather straight forward.

Regards
	Stefan Lippers-Hollmann
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.alioth.debian.org/pipermail/pkg-wpa-devel/attachments/20120621/7aa41c11/attachment-0003.pgp>