Package: squirrelmail
Version: 1.4.9a-1
Severity: normal
Hi,
When /etc/squirrelmail/conf.pl is disconnected abruptly, it does not die
but spawns infinitely calls to clear via clear_screen().
I got 600 forks/s and 100% CPU for hours because the admin of one of
my vservers closed his ssh session
Package: kernel-package
Version: 10.055
Severity: wishlist
Tags: patch
Hi,
Very simple patch to add a --verbose option so that it will call the make
with the V=1 option to see the full commands.
I found it much more useful than --noexec because I cound not just simulate a
full compilation, some
Package: laptop-mode-tools
Version: 1.11-1
Severity: normal
Tags: patch
When setting VERBOSE_OUTPUT=1 it truncated /var/log/acpid and only
the last few lines are present.
This is due to the /dev/stdout redirections.
This is solved by changing them to /dev/stdout
Hyper-compressed patch :-)
Package: acpid
Version: 1.0.4-3
Severity: normal
Hello,
I found a rather strange bug in a client of acpid: laptop-mode-tools
When on verbose mode, this handler writes its output with /dev/stdout
The very weird consequence is that /var/log/acpid is emptied by this!
A short experience to
Package: spe
Version: 0.8.4.h-1
Severity: normal
Hello,
It is possible to type Notes in unicode in the Notes tab per sourcefile
or in the global Notes tab (lower part of the GUI) but when saving I get
the error:
Spe warning: could not save workspace options in /home/phil/.spe/swifi.sws
Package: dnsmasq
Version: 2.40-1
Severity: minor
--- Please enter the report below this line. ---
Hello,
I'm using --listen-address=1.2.3.4 --bind-interfaces
ns0:/# netstat -ltupn
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address Foreign Add State PID/Program
tcp
Package: openvpn
Version: 2.1~rc4-2
Severity: wishlist
2.1 rc7 is available upstream, please update.
Thanks!
-- System Information:
Debian Release: lenny/sid
APT prefers testing
APT policy: (990, 'testing'), (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.22-3-vserver-686 (SMP
Package: lkl
Version: 0.1.1-1
Severity: minor
Tags: patch
Hi,
Here's a patch to reorder the test on getuid/getgid
to allow option -h (showing help) without being root.
I don't want to run a program as root before knowing at least how to use it ;-)
Phil
-- System Information:
Debian Release:
Package: lkl
Version: 0.1.1-1
Followup-For: Bug #460230
Hi,
I encountered the same problem.
It looks like lkl was polling every microsecond
while every millisecond is largely enough.
So here is a patch, I also moved from usleep() to
the more POSIX compliant nanosleep().
Package: icewm
Version: 1.2.32-1
--- Please enter the report below this line. ---
cf
http://www.nabble.com/-Bugs-in-1.3.1--Not-all-tray-icons-are-shown-in-icewmtray-t4277712.html
This one-line patch solved the problem for me.
--- icewm-1.2.32/src/yxtray.cc.orig 2007-08-07 14:12:03.0
Hi,
On small setups like vservers we don't need the (rather large)
package locales so /etc/environment doesn't exist.
So yes please make sure pre/post install does a
test -e /etc/environment ||touch /etc/environment
Phil
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of
Package: libpam-modules
Version: 0.99.7.1-4
--- Please enter the report below this line. ---
Hello,
I got also the same kind of error with a missing /etc/default/locale
This occurs on small installs where package locales is absent.
Sep 14 04:25:07 private su[30273]: pam_env(su:session): Unable
Package: fetchyahoo
Version: 2.10.9
Severity: grave
Hello,
Since a few days fetchyahoo failed with can't find message
From upstream there is now:
v2.11.0 08/28/07
* fix can't find message error with new Yahoo internals
* replace html character entities in msg summaries
Please integrate it asap.
Package: icewm
Version: 1.2.32-1
Followup-For: Bug #439061
Just to avoid double work of the volunteers fixing bugs...
This bug is also described in the upstream bug tracking tool:
http://sourceforge.net/tracker/index.php?func=detailaid=1771681group_id=31atid=100031
Phil
-- System Information:
Package: esniper
Version: 2.16.1-2+b1
Severity: minor
--- Please enter the report below this line. ---
Please package new upstream version, thanks
--- System information. ---
Architecture: i386
Kernel: Linux 2.6.21-1-686
Debian Release: lenny/sid
990 testing
Package: hping2
Version: 2.rc3-4
Severity: normal
Tags: patch
Hello,
When using hping2 on my system I get
[main] physical layer header size unknown
This is because my interfaces are named ETH0 and WLAN0 instead of eth0/wlan0.
I'm a fan of ifrename and this allows you to use any interface name
Package: net-tools
Version: 1.60-17
Followup-For: Bug #197282
Hi,
I'm amazed nobody replied to this bug for 3.5 years already!
I discovered myself the problem and I could get a behaviour closer
to the expectations by applying those changes:
--- inet.c.orig 2006-12-28 19:00:45.0 +0100
Bernd Eckenfels wrote:
I think no matter what flags are given, the remote address of a LISTENING
socket should be *:*.
(at least if ad == INADDR_ANY, however I guess there are no other cases)
Maybe, but in the fct we don't know in which case we are
(listening or not? local or remote addr?)
Package: linux-source-2.6.18
Severity: wishlist
Hello,
To use oprofile with kernel profiling enabled, we need the uncompressed version
of the kernel image, vmlinux.
When using the debian kernel image this is not possible, it is required to take
the kernel sources and to compile a new image with
To use oprofile with kernel profiling enabled, we need the uncompressed version
of the kernel image, vmlinux.
Use gunzip to get them (at least for architectures which simply
compresses it with gzip). Some architectures also uses uncompressed
files.
I'm using i386 and I tried many
Package: libnet-ssleay-perl
Version: 1.30-1
Severity: wishlist
Tags: patch
Hello,
(Patch submitted on behalf of Thierry Walrant)
Here is a small patch to add the possibility for a Perl script to
check the peer certificate fingerprints e.g. in a similar way
fetchmail does.
This adds the
Package: sudo
Version: 1.6.8p12-5
Severity: important
Hi,
I was relying entirely on sudo + sudo group to get access to the root account
on which the root password was disabled.
After upgrade to 1.6.8p12-5 nothing alerted me but the fact is that on
next login I could not get root access anymore.
Package: pwsafe
Version: 0.2.0-2+b1
Followup-For: Bug #407262
Hi,
Sorry but after a 100-day of total absence of reaction on this bug report,
I feel like saying me too!
Please either justify why you removed X11 support or provide an update
with X11 support reintegrated. Thanks.
Phil
-- System
Package: flickrfs
Version: 1.3.9.1-1
Severity: normal
Within Flickr, you can have several photos with the same title without any
problem.
Under flickrfs, each photo is represented as a file whose name is the Flickr
title.
The problem is immediate: only one of the photos with the same title will
Package: flickrfs
Version: 1.3.9.1-1
Severity: wishlist
Hello,
I see flickrfs as a great opportunity to have a backup of what's in Flickr.
In that context, I copy the /stream locally but there are still some
valuable infos not taken into account in the metadata:
- the geolocation
Apparently the
Subject: apt-proxy: malformed URI requests with http_proxy (extra /:
http://fqdn:80//dir)
Package: apt-proxy
Version: 1.9.32
Severity: normal
*** Please type your report below this line ***
When using apt-proxy behind a proxy, the external requests are
malformed.
Example config:
Package: openvpn
Version: 2.1~rc7-1
Severity: minor
When using the following verification method:
tls-remotevpn.mysite.be
it works fine on a certificate created with easy-rsa with Subject:
/C=BE/ST=Brussels/L=Brussels/O=MyVPN/CN=vpn.mysite.be/[EMAIL PROTECTED]
But on a CACert
Package: openvpn
Version: 2.1~rc7-1
Severity: normal
option --ns-cert-type server fails on CACert certificates
which have a broader scope than just server key usage:
- TLS Web Client Authentication
- TLS Web Server Authentication
- Netscape Server Gated Crypto
- Microsoft Server Gated Crypto
Package: gallery2
Severity: normal
After upgrade I got the following warning on the website wizard:
Contrôle d'intégrité des fichiers de Gallery
Attention
Fichiers modifiés (1)
lib/pear/HTMLSax3.php
Sorry it's in french, it's about sth like
Integrity check of Gallery files:
Warning
Modified
Package: initramfs-tools
Version: 0.92b
Severity: minor
Tags: patch
From /usr/share/initramfs-tools/hooks/cryptroot:
# Prepare the initramfs
copy_exec /sbin/cryptsetup /sbin
copy_exec /sbin/dmsetup /sbin
copy_exec /lib/cryptsetup/askpass /lib/cryptsetup
But without care of /lib/cryptsetup
Package: foo2zjs
Version: 20060124dfsg-1
Severity: normal
Hi,
From version 20060124dfsg-1_i386 the utility usb_printerid is again missing
while its manpage is present.
Related to this utility, another issue already mentioned in #279830 is that
it is very hard to guess how to enable firmware
Package: apt-index-watcher
Version: 0.3.9+b1
Followup-For: Bug #377709
Hi,
I got the very same problem, syslog complaints every 6 secs.
I could get rid of them by updating the vocabulary database of debtags
(after all, this is the file apt-index-watcher is complaining about)
The vocabulary was
Package: lkl
Version: 0.1.1-1
Severity: normal
Hello,
I just discovered lkl and tried it briefly.
$ sudo lkl -l -k /usr/share/lkl/keymaps/us_km
First remark (but it is said in another bugreport already), keymap
selection is less than intuitive, the README does not talk about having
to type the
Package: iceweasel
Version: 2.0.0.12-1
--- Please enter the report below this line. ---
Hello,
I generated a client certificate via cacert.org then I tried
to backup it (View certificates - Your certificates - Backup)
but it failed with the following message:
Failed to create the PKCS #12
Package: iceweasel
Version: 2.0.0.12-1
Severity: normal
--- Please enter the report below this line. ---
Hello,
I generated a client certificate via cacert.org then I tried
to backup it (View certificates - Your certificates - Backup)
but it failed with the following message:
Failed to create
Package: iceweasel
Version: 2.0.0.12-1
--- Please enter the report below this line. ---
Sorry I pushed the wrong button on reportbug-ng,
I wanted to create a new bugreport.
Please ignore!
Phil
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact
Package: iceweasel
Version: 2.0.0.12-1
--- Please enter the report below this line. ---
Let's continue that discussion at #469079
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Package: iceweasel
Version: 2.0.0.12-1
--- Please enter the report below this line. ---
Hi,
I downgraded to the only other available version:
libnss3-1d(= 3.12.0~1.9b1) | 3.12.0~1.9b1-1
And now it works, thanks for your cleverness!!
So I guess you guess what to fix now ;-)
Phil
The strange thing is that it works properly with both versions, here.
Can you try again with version 3.12.0~1.9b1-2 ?
Well yes indeed it works again with 3.12.0~1.9b1-2
I'm confused now.
For sure a day ago it failed with iceweasel, icedove and iceape
and after the downgrade (and updrade
What about other package upgrades ? (check /var/log/dpkg.log)
cat /var/log/dpkg.log.1 /var/log/dpkg.log|egrep (install|remove|upgrade)
of the week-end:
I did a apt-get build-dep icedove:
(as I wanted to try a recompilation)
2008-03-02 20:18:24 remove libcamel1.2-6 1.4.2.1-2 1.4.2.1-2
Was that the real problem?
Then should libnss3-1d conflict also with libnss3 and not only libnss3-0d?
There *shouldn't* be problems with libnss3, but just in case, can you
try reinstalling it ? BTW, when was libnss3-1d upgraded ?
Yes I just had the same idea and took the same
Package: xbase-clients
Version: 1:7.2.ds2-2
Severity: normal
--- Please enter the report below this line. ---
Hello,
When starting X with startx:
The client Xauthority file can be chosen via the $XAUTHORITY env var
but not the server one which is hardcoded as following:
Package: libbeidlibopensc2
Version: 2.6.0-3
Severity: normal
Tags: patch
Hello,
I just saw that my ~/.xsession-errors logfile was full of
Error: can't open /var/run/openct/status...
It happens whenever icedove/iceweasel are open (so when the
libbeidpkcs11.so is loaded)
I found a similar
Package: esniper
Version: 2.17.1-1
Severity: normal
Release 2.18.0 of esniper is now available at
http://esniper.sourceforge.net/
This is a major update. It updates esniper to handle eBay's new bid
history page, which is now encoded in UTF-8.
In addition, it rolls many minor bug fixes made since
This is working as designed: dnsmasq is listening for DNS queries on
1.2.3.4 port 53 , both UDP and TCP. (the first and third lines of the
netstat output)
The second line is the socket that dnsmasq uses to communicate with
upstream servers. You can bind that to a specific interface, but
Package: signing-party
Version: 0.4.13-1
Severity: important
Tags: patch
--- Please enter the report below this line. ---
When working with e.g. the keyserver onak 0.3.5-1 available in Debian,
keylookup fails importing any key because onak returns 40-nibble keyids
instead of the more usual
Package: kgpg
Version: 4:3.5.8-1
Severity: normal
--- Please enter the report below this line. ---
When working with e.g. the keyserver onak 0.3.5-1 available in Debian,
kgpg fails to parse the keyids of a search because onak returns
40-nibble keyids instead of the more usual 8-nibble keyids.
I don't have a onak server running, so no easy way to do any debugging, but
maybe we can walk it thru together?
Mmm ok you won, I didn't want to install the 243Mb of build dependencies ;-)
subkey is the full string and section key means all the right part
so we've to get another section,
Mm sorry it's probably safer to use just , instead of , created
in case of servers replying with sth else than one space between , and
created:
--- keyservers.cpp.orig 2008-02-21 22:10:47.0 +0100
+++ keyservers.cpp 2008-02-22 00:50:16.0 +0100
@@ -315,8 +315,8 @@
Package: onak
Version: 0.3.5-1
Severity: normal
Tags: patch
--- Please enter the report below this line. ---
Stupid bug when more than one photo-uid is present in a key.
The indx arg was correctly fetched from URL but not used...
Here is a patch:
Phil
--- lookup.c.orig 2008-02-22
my own script.
Here it is for info, feel free to integrate it to the package
if you want.
#!/bin/bash
# License: GPLv3
# Copyright Philippe Teuwen phil a teuwen o org 2008
if [[ $# -ne 5 ]]; then
cat EOF
Usage: $0 input output datestring organizer title
Or:$0 - output datestring
Hi Thijs,
I think this would be very suitable to include. I have one question though:
it's currently licenced GPLv3. There are already two different licences for
the scripts in signing-party: GPLv2+ and BSD. To keep things simple I would
very much prefer not to add a third one to that. So
Package: php-openid
Severity: wishlist
--- Please enter the report below this line. ---
Hello,
I'd like to see some support for proxies.
The reason is that a typical webserver is allowed
to receive external connections but not to
initiate directly connections to the outside world
so now for
Since my cardreader does, in fact, use the OpenCT driver rather than the
PC/SC one, I re-added this driver to the beid packages. Disabling the
driver afterwards would be rather silly.
You're right, meanwhile I saw the video of your speech at Fosdem ;-)
(Mm ok the video itself didn't
Package: cryptonit
Version: 0.9.7-2+b1
Severity: normal
Hello,
I just tried cryptonit with my Belgian eID (and libbeidpkcs11)
Device-Certificates present on connected devices:
I see my 4 certificates, apparently sorted alphabetically.
When double-clicking on one of them I get the properties
of
Package: libopensc2
Version: 0.11.4-3
Severity: wishlist
Tags: patch
Hello,
update_binary() was not foreseen by the middleware creators of BELPIC
so I added it, which allows to write some data in the
MF/ID/EF(Preferences) of the card.
Why did I add this? Just because it's possible :-)
---
Yeah, that was bad. You did also see my screencast, right? :-)
Yes I first saw the screencast actually
If something could be done in libbeidlibopensc2, it's to make a silent
test about /var/run/openct/status (or cf ct_format_path() ) before
calling the openct lib
Package: cron
Version: 3.0pl1-116
Severity: minor
--- Please enter the report below this line. ---
After upgrade from cron 3.0pl1-114 to 3.0pl1-116 I had the following
problem:
My log reporting system began to send me a lot of error messages from cron.
Digging down it appears that the following
After upgrade from cron 3.0pl1-114 to 3.0pl1-116 I had the following
problem:
I can confirm the general issue, however this should have first appeared
in -110, not -114 (see #581612, where we accidentally sent mails on exit
!= 0, which we fixed in -112).
Could you check your logs and
What I'd like to do is just leave this bug report open for now so other
possibly affected users can see it, and fix after Squeeze. This change
has also been present in Ubuntu since 10.10 and nobody noticed so far,
so I guess the impact on users upgrading from Lenny - Squeeze should be
To follow on:
This is unlikely to succeed on most systems. I did at one point let it run
for multiple minutes to see if it's just a display issue and if it'd actually
do something sensible, but it does not seem so.
Well it does something: it fills all the empty space of the partition
where
Package: pwsafe
Version: 0.2.0-3
Severity: minor
Manpage specifies following syntaxes:
--list[=REGEX]
--add[=NAME]
--edit=REGEX
--delete=NAME
but:
$ pwsafe --list=foo
pwsafe: option '--list' doesn't allow an argument
$ pwsafe --add=foo
pwsafe: option '--add' doesn't allow an argument
$ pwsafe
Package: onak
Version: 0.3.7-1
Severity: minor
Tags: patch
Hello,
I discovered a few minor bugs when I was tracking down an internal
problem with gpgwww:
Sometimes the key IDs are not left padded when printed:
--- gpgwww.c.orig2009-06-02 17:15:08.0 +0200
+++ gpgwww.c
Package: webcalendar
Version: 1.2.0+dfsg-2
Hello,
I faced the same problem after upgrading (very painfully) from 1.0.15 to
1.2.0
I digged a bit into it:
Actually the downloadable week.php is empty and Apache logs report
thing like this when accessing week.php:
child pid 14963 exit signal
Package: squirrelmail
Version: 2:1.4.15-4
Severity: normal
Tags: security
X-Debbugs-CC: secure-testing-t...@lists.alioth.debian.org
ANNOUNCE: SquirrelMail 1.4.18 Released
May 12, 2009 by Paul Lesniewski
The SquirrelMail Team is pleased to announce the release of
SquirrelMail version 1.4.18.
Did you have a look at upstream? They're at 2.1.2 now and maybe they
already implemented such thing (I didn't see it but I didn't spend too
much time on looking for it, either).
I inspected the new sources but I don't see anything new related to
proxy handling.
Phil
--
To
Hi,
I was also previously impacted by bug #616372 which I've locally solved
by simply rebuilding version 0.7.44+dfsg-2 from sources (bzr).
BTW this also solved #609705
Now it got being upgraded to 0.7.50+dfsg-1 and... segfault too.
So... I tried again to rebuild, now version 0.7.50+dfsg-1, from
Hello,
Is this but entirely solved?
I could now build an ISO which displays the semi-graphical boot menu but it
only contains the other options sub-menu.
And when I check into syslinux-themes-debian-squeeze/syslinux-live templates I
see tentative of inclusion of live.cfg which is apparently
Package: virtualbox
Version: 4.1.6-dfsg-2
Severity: normal
Dear Maintainer,
Upgrade from virtualbox-4.0 to virtualbox package (with 4.1.6 dependency
atm) failed
apt-get install -t unstable virtualbox
...
Setting up virtualbox (4.1.6-dfsg-2) ...
insserv: script virtualbox: service
Package: evolution-mapi
Version: 3.2.1-3
Severity: normal
--- Please enter the report below this line. ---
Hello,
I tried to use evolution-mapi plugin but when launching Evolution I got
the following errors:
(evolution:16411): evolution-plugin-lib-WARNING **: can't load plugin
Package: scanmem
Version: 0.12-2
Severity: minor
--- Please enter the report below this line. ---
$ scanmem
0 help
[...]
!= match all variables that have not changed since last scan
= match all variables that have not changed since last scan
First line is wrong, it should read:
Package: mediawiki
Version: 1:1.19.5-1
Severity: normal
Tags: security
X-Debbugs-CC: secure-testing-t...@lists.alioth.debian.org
Default allowed extensions for file upload are only:
$wgFileExtensions = array( 'png', 'gif', 'jpg', 'jpeg' );
Under Firefox Chrome it's indeed impossible to upload a
On 07/15/2013 01:00 PM, Henri Salo wrote:
On Mon, Jul 15, 2013 at 11:41:16AM +0200, Philippe Teuwen wrote:
Package: mediawiki
Version: 1:1.19.5-1
Severity: normal
Tags: security
X-Debbugs-CC: secure-testing-t...@lists.alioth.debian.org
Default allowed extensions for file upload are only
secur...@wikimedia.org
On Mon, Jul 15, 2013 at 1:27 PM, Philippe Teuwen p...@teuwen.org wrote:
On 07/15/2013 01:00 PM, Henri Salo wrote:
On Mon, Jul 15, 2013 at 11:41:16AM +0200, Philippe Teuwen wrote:
Package: mediawiki
Version: 1:1.19.5-1
Severity: normal
Tags: security
X-Debbugs-CC
On 07/16/2013 10:26 AM, Thorsten Glaser wrote:
On Mon, 15 Jul 2013, Philippe Teuwen wrote:
A quick inspection seems to indicate that the file extension is only
checked on the client side via javascript and IE does not do a proper job.
File extensions are a joke, really.
Note
Package: texlive-latex-extra
Version: 2013.20140408-1
Severity: normal
--- Please enter the report below this line. ---
cleveref v0.19 is still using functions of IEEEtrantool that have
disappeared at least since IEEEtrantools.sty 2012/12/27 version V1.3
E.g. it is using \if@IEEEissubequation
Package: logkeys
Version: 0.1.1a+git5ef6b0dcb9e3-2
Severity: normal
--- Please enter the report below this line. ---
Hi,
Logkeys can be used with a keymap (using option -m) OR using -u ==
--us-keymap
One could create a keymap and add the option in /etc/default/logkeys:
DAEMON_OPTS=--keymap
Package: wine
Version: 1.6.2-8
Severity: normal
Tags: patch
--- Please enter the report below this line. ---
Hello
When running alternatively wine32 and wine64, you get such error (or
vice versa):
wine: '/home/phil/.wine' is a 32-bit installation, it cannot support
64-bit applications.
One
Package: sed
Version: 4.2.2-4+b1
Severity: important
--- Please enter the report below this line. ---
Hello
sed version 4.2.1-10, correct result:
echo =abcABC[]\\_=|sed 's/[A-z]/*/g'
=**=
sed version 4.2.2-4+b1, error:
echo =abcABC[]\\_=|sed 's/[A-z]/*/g'
sed: -e expression #1, char
It appears that I was wrong, it's not related to the version of sed but
to the terminal configuration:
echo '=abcABC[]\\_='|LANG=en_US.utf8 sed 's/[A-z]/*/g'
sed: -e expression #1, char 11: Invalid range end
echo '=abcABC[]\\_='|LANG=C sed 's/[A-z]/*/g'
=***=
The order of chars in utf8
Package: quassel-client
Version: 1:0.10.0-2.4
--- Please enter the report below this line. ---
Here Quassel client doesn't crash but when trying to setup a sound
notification here is the behavior:
First time, I can browse select an audio file and I can play it with the
test button, no problem.
Package: python-pil
Version: 3.1.1-1
Severity: normal
Dear Maintainer,
Openjpeg (jpeg2000) support is missing in the binary packages of pillow.
I could fix it locally by adding some openjpeg headers:
apt-get source python-pil
sudo apt-get build-dep python-pil
sudo apt-get install
BTW after udev_monitor_receive_device() returns null,
strerror(errno) gives "Resource temporarily unavailable"
So on one side poll() says there is data ready to be read on
udev_monitor_fd and on the other side udev_monitor_receive_device() says
"Resource temporarily unavailable"
Hi Ludovic,
Version 1.8.16-1
Same problem.
See log.txt
Cheers
Phil
pcscdaemon.c:261:main() pcscd set to foreground with debug send to
stdout
0023 debuglog.c:310:DebugLogSetCategory() Debug options: APDU
[36m0003[0m [34mpcscdaemon.c:266:main() Force colored logs[0m
> Does the problem occurs when you connect the USB device VID: 0x2A70,
> PID: 0x9011?
> This is your phone?
Yes
> The first "USB Device add" is correct. You connect the USB device.
> But why a second "USB Device add" for the same device?
> And a third "USB Device add" 1 second after for again
On 03/27/2016 09:06 PM, Ludovic Rousseau wrote:
> +else
> +usbi_err(NULL, "udev_monitor_receive_device failed");
See att.
I've cut it as it's printing in the crazy loop as expected.
So it confirms what we already know:
fds[0].revents = 0
fds[1].revents = 1
Here are some backtraces
Normal CPU usage:
Thread 5 (Thread 0x7f6365c82700 (LWP 30482)):
#0 0x7f6366196b6d in poll () at ../sysdeps/unix/syscall-template.S:81
#1 0x7f6365c92b3c in ?? () from /lib/x86_64-linux-gnu/libusb-1.0.so.0
#2 0x7f6366462284 in start_thread
Mmm activating libudev debug didn't bring much I think.
I attached one log with LIBUSB_DEBUG=99 and one with recompiled pcscd
with udev debug support.
For both I just ran pcscd with the yubikey plugged in then plugged a USB
hub that triggered the CPU to 100%
libccid1.4.22-1
libusb-1.0-0
Hi Ludovic,
I'm getting the same problem with my Debian Testing.
I'll try to generate more logs related to libusb as explained in the bug
thread but here are some things I noticed:
I've a Yubikey Neo-n plugged constantly.
When the 100% problem arises, nothing special in pcscd logs, even with
On 03/07/2016 07:34 PM, Ludovic Rousseau wrote:
> printf("fds: %d %d\n", fds[0].revents, fds[1].revents);
fds: 0 1
always
I also printed udev_dev from udev_monitor_receive_device()
it's always null
So we get fds[1].revents but don't get anything from
udev_monitor_receive_device() so it's
_lock);
}
}
usbi_dbg("udev event thread exiting");
I added error msgs in the loop.
When 100% CPU, the poll() is non-blocking and the loop becomes a busy loop.
On 03/06/2016 11:49 PM, Philippe Teuwen wrote:
> #0 0x7f196a6cabdd in recvmsg () at ../sysdeps/uni
I forgot to say, removing TEMP_FAILURE_RETRY() didn't change anything.
> Interesting.
> Do you give arguments to libusb/examples/hotplugtest or just execute the
> command with no argument?
no argument
> The scl3711 is not supported by my CCID driver (at least I can't find
> the device VID: 0x04E6, PID: 0x5591 in my list)
> I guess you use the SCM driver with this
Package: chromium
Version: 52.0.2743.116-1
--- Please enter the report below this line. ---
There is a similar bugreport upstream:
https://bugs.chromium.org/p/chromium/issues/detail?id=623290
I've added a few info there.
--- System information. ---
Architecture: amd64
Kernel: Linux
Package: comix
Version: 4.0.4-3
Severity: normal
--- Please enter the report below this line. ---
Hi I got this error.
Traceback (most recent call last):
File "/usr/share/comix/src/thumbbar.py", line 126, in _load
_add_page_number(pixbuf, i)
File "/usr/share/comix/src/thumbbar.py", line
Package: gcc-arm-none-eabi
Version: 15:8-2019-q3-1
Severity: normal
Dear Maintainer,
Code containing PRIx64 macro after inclusion of does not
compile anymore,
the macro doesn't exist contrarily to the situation with
gcc-arm-none-eabi_7-2018-q2-6.
I tracked it down to the following difference
96 matches
Mail list logo
