Bug#377542: NMU ?

2007-03-21 Thread mechanix
Hi,

On Sun, Mar 18, 2007 at 11:41:56PM +0100, Vincent Fourmond wrote:
> 
>  Do you mind if I NMU the fix for that problem ? Now that I have the
> rights to do so ;-) It might close other segfault bugs as well. I
> leave you until Thusrday (March 22nd).

Not at all; by all means go ahead.


KR,

Filip


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Bug#384558: module-assistant build ieee80211-source fails

2006-08-26 Thread mechanix
tags 384558 + patch
thanks

On Thu, Aug 24, 2006 at 11:40:33PM -0400, Edward Allcutt wrote:
> 
> the error module-assistant reports is:
> debian/rules:6: /usr/share/dpatch/dpatch.make: No such file or directory
> make: *** No rule to make target `/usr/share/dpatch/dpatch.make'. Stop.

Another missing dependency is gcc-4.0 (explicitly), with just gcc and
thus gcc-4.1 the build fails, too.

> after installing dpatch building fails somewhat more verbosely,

Attached patch fixes this. I think that only even the first chunk is
needed, the dh_gencontrol options are probably not (and the DH_VERBOSE
environment variable is not at all, it was just a debugging bandaid).

Installing the resulting module package still fails because the included
modules are also in the linux-image package you have installed, which
should probably taken care of by using diversions.


Regards,

Filip

-- 
"Once upon a time a hacker named Xman wrote a library that used aio, and
 decided to use sigtimedwait() to pick up completion notifications. It
 worked well, and his I/O was blazing fast (since was using a copy of
 Linux that was patched to have good aio)."
-- opening of "A signal fairy tale" by Dan Kegel
--- modules/ieee80211/debian/rules
+++ modules/ieee80211.new/debian/rules
@@ -108,6 +108,7 @@
dh_clean -k
dh_installdirs
 
+   sed "s/_KVERS_/$(KVERS)/" debian/control.modules.in > debian/control
# Build the module
$(MAKE) modules KERNEL_DIR=$(KSRC) KVERS=$(KVERS)
 
@@ -136,7 +137,7 @@
#   > debian/ieee80211-headers-$(KVERS).$$FILE; \
#done
dh_installdeb
-   dh_gencontrol -- -v$(VERSION)
+   DH_VERBOSE=1 dh_gencontrol -s -i -- -v$(VERSION)
dh_md5sums
dh_builddeb --destdir=$(DEB_DESTDIR)
dh_clean -k


Bug#206804: l2tpd: connections fail when run as daemon

2006-09-08 Thread mechanix
tags + moreinfo unreproducible
thanks

On Sat, Aug 23, 2003 at 01:38:41AM +0100, Nuno Ferreira wrote:
> 
> When run as daemon connection fail and the following is logged:
> Aug 23 01:33:16 sylvester l2tpd[22245]: start_pppd: I'm running:
> Aug 23 01:33:16 sylvester l2tpd[22245]: "/usr/sbin/pppd"
> Aug 23 01:33:16 sylvester l2tpd[22245]: "passive"
> Aug 23 01:33:16 sylvester l2tpd[22245]: "-detach"
> Aug 23 01:33:16 sylvester l2tpd[22245]: "10.1.1.254:10.1.1.100"
> Aug 23 01:33:16 sylvester l2tpd[22245]: "auth"
> Aug 23 01:33:16 sylvester l2tpd[22245]: "require-chap"
> Aug 23 01:33:16 sylvester l2tpd[22245]: "file"
> Aug 23 01:33:16 sylvester l2tpd[22245]: "/etc/l2tp/l2tpd.pppd.options"
> Aug 23 01:33:16 sylvester l2tpd[22245]:
> Aug 23 01:33:17 sylvester pppd[22255]: no device specified and stdin is not a 
> tty
> Aug 23 01:33:17 sylvester l2tpd[22245]: child_handler : pppd died for call 1
> 
> if I run l2tpd with -D, I can login.

To me it looks like this has been fixed a long time ago:

Sep  6 07:16:07 pluto l2tpd[31121]: start_pppd: I'm running:
Sep  6 07:16:07 pluto l2tpd[31121]: "/usr/sbin/pppd"
<...>
Sep  6 07:16:07 pluto l2tpd[31121]: "/dev/ttyp0"

and connections work fine, without -D.

This is l2tpd version 0.70-pre20031121-2, been running it for a year and
a half or such.


Regards,

Filip


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Bug#425504: rdesktop: sound does not work ("select: invalid file descriptor")

2007-05-22 Thread mechanix
Package: rdesktop
Version: 1.5.0-1
Severity: normal
Tags: upstream

Sound redirection to the local machine is broken for me in rdesktop
1.5.0-1; the remote connection slows down to a crawl and hangs
eventually, until it fails with a spew of messages "select: invalid
file descriptor".

Someone reported the same problem upstream:
http://sourceforge.net/tracker/index.php?func=detail&aid=1695146&group_id=24366&atid=381347

I downloaded 1.4.1-1.1 from snapshot.debian.net and sound works fine
there.
The other person who originally reported it upstream talked about a
homegrown application; however for me at least any application using
sound causes it.

KR,

Filip

--- System information. ---
Architecture: amd64
Kernel:   Linux 2.6.20-1-amd64

Debian Release: 4.0
  900 testing ftp.belnet.be 
  899 stable  security.debian.org 
  899 stable  ftp.belnet.be 
   10 unstableftp.belnet.be 

--- Package information. ---
Depends (Version) | Installed
=-+-==
libc6(>= 2.3.5-1) | 2.5-8
libssl0.9.8 (>= 0.9.8c-1) | 0.9.8c-4
libx11-6  | 2:1.0.3-7


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Bug#249872: closed by Noèl Köthe <[EMAIL PROTECTED]> (Re: #249872 samba - module loading broken)

2007-06-27 Thread mechanix
> Am Freitag, den 22.06.2007, 23:36 +0200 schrieb Noèl Köthe:
> 
> email to reported bounced.
> 
> Closing...
> 
> > as requested 1,5 year ago we left this bug report open:
> > 
> > http://bugs.debian.org/249872
> > 
> > Any new information about it?
> > 
> > Or can we close it?

For the record: I have no idea on the current status, but won't be getting
to look into it in any near future if ever at all (again).
Better to close it, as has happened.

>   [EMAIL PROTECTED]
> SMTP error from remote mailer after RCPT TO:<[EMAIL PROTECTED]>:
> host mailproxy1.rack66.com [85.88.55.21]: 554 5.7.1 Service
> unavailable; Client host [70.103.162.29] blocked using bl.csma.biz;
> http://bl.csma.biz/cgi-bin/listing.cgi?ip=70.103.162.29

Aww. This is out my control :/


KR,

Filip



Bug#394052: ppp: mtu value in options ignored

2006-10-20 Thread mechanix
retitle 394052 ppp: report mtu value used
thanks

False report.
I played with mtu configuration option and it does effect the traffic;
it's not reflected in debug output though.

So, let's reformulate this: when pppd says it negotiated a certain client
mru value, which differs (being larger, it seems) from the actual mtu it
will use, it should say so and tell which mtu that is.

I still have the packets not reaching the client problem, but I'm not sure
at this point why they're not.

On Thu, Oct 19, 2006 at 07:56:41AM +0200, Filip Van Raemdonck wrote:
> 
> I'm trying to connect Windows systems as VPN roadwarriors to a gateway
> running ppp over l2tp over ipsec. The gateway is in Singapore, and has
> non-ideal internet connection (mainly, the MTU/MRU had to be lowered
> for some things to work properly).
> 
> For ppp, I set mtu and mru as well:
> 
> ...
> mtu 1200
> mru 1200
> ...
> 
> Checking the ppp log, the client asks for a mru of 1400, larger than what
> the ppp configuration has set as mtu; pppd happily acknowledges it despite
> its lower mtu setting. It looks like the mtu configuration is ignored and
> pppd starts sending packets larger than what the tcp-ip (and thus ipsec)
> layer of the gateway can send.
> 
> ...
> Oct 18 15:24:00 scotos pppd[16120]: rcvd [LCP ConfAck id=0x1  
> ]
> Oct 18 15:24:00 scotos pppd[16120]: rcvd [LCP ConfReq id=0x1  
> [local:95.21.d7.70.e2.45.4b.6a.ad.23.8b.3d.b6.46.05.06.00.00.00.02]>]
> Oct 18 15:24:00 scotos pppd[16120]: sent [LCP ConfAck id=0x1  
> [local:95.21.d7.70.e2.45.4b.6a.ad.23.8b.3d.b6.46.05.06.00.00.00.02]>]


Regards,

Filip


http://www.sysfs.be/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Bug#431564: python-feedparser: fails basic authentication

2007-07-12 Thread mechanix
On Tue, Jul 03, 2007 at 06:52:40PM +0200, Carlos Galisteo wrote:
> > This can be tested by running python interactively and entering the
> > commands from
> > http://feedparser.org/docs/http-authentication.html#example.auth.basic
> 
>  Seems to work for me:
>  >>> d = feedparser.parse('http://test:[EMAIL 
> PROTECTED]/docs/examples/basic_auth.xml')

That's not the example from the link above :)

I'm using (as the docs state) "the hard way" -- an authentication handler.
And the result is a 403 status, always.

Someone else reported the same issue upstream:
http://code.google.com/p/feedparser/issues/detail?id=40

I can confirm that the fix listed there works, reversing the two handlers
passed to urllib.


KR,

Filip


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]