Bug#1057623: Unnecessary debugging statements
Package: python3-apt Version: 2.7.0 Hi, We are using python3-apt in a configuration management system, LCFG. Because there are some debugging messages in the code, every machine sends us these messages when they do updates: Stdout: got fd: 5 Stderr: custom fork found got pid: 1054321 got pid: 0 These messages are leftover debugging messages that don't add value any more. They should be removed. (I could filter them, but the correct approach seems to be to remove them.) I created a pull request on Salsa which fixes this, but I never got a reply. I sent a bump message after a reasonable length of time, and still didn't get a reply, so I'm trying here instead. If this is the wrong place to address this problem, or I have done something else wrongly, please tell me. https://salsa.debian.org/apt-team/python-apt/-/merge_requests/84 Thanks, Bruce OpenPGP_0x3D3BB36EBFD0445D.asc Description: OpenPGP public key OpenPGP_signature.asc Description: OpenPGP digital signature
Bug#458016: fails to pick up WEP Type on key request
Package: network-manager-kde
Version: 1:0.2-2
Severity: minor
Tags: patch
Hey kids,
I've just switched to network manager to make dealing with a poor wireless
signal a little less painful. When the association fails, network manager
assumes I have provided the wrong key and prompts me for a new one. I know I
have the correct key and just want it to try again but it fails to notice
that the WEP key I have entered is in HEX format, rather than PASSPHRASE.
The "Encryption" combo-box therefore defaults to passphrase and I have to
change it manually before continuing.
I have attached a patch which might fix this. It's _completely_untested_ as
network-manager-kde currently depends on an obsolete version of libnl.
I hope that's clear and reproducible (entering the wrong key in anything other
than "passphrase" should produce the same behaviour)! It would be great if
someone could confirm my patch does the job.
Thanks,
Bruce
--- System information. ---
Architecture: amd64
Kernel: Linux 2.6.24-rc6
Debian Release: lenny/sid
500 unstablewww.debian-multimedia.org
500 unstableftp.de.debian.org
500 stable www.debian-multimedia.org
1 experimentalftp.de.debian.org
--- Package information. ---
Depends (Version) | Installed
===-+-===
kdelibs4c2a (>= 4:3.5.7-1) | 4:3.5.8.dfsg.1-4
libc6 (>= 2.6.1-1) | 2.7-5
libdbus-1-3 (>= 1.1.1) | 1.1.2-1
libdbus-qt-1-1c2 (>= 0.62.git.20060814) | 0.62.git.20060814-2
libgcc1(>= 1:4.2.1) | 1:4.2.2-4
libglib2.0-0(>= 2.14.0) | 2.14.4-2
libhal1 (>= 0.5.9) | 0.5.10-5
libice6(>= 1:1.0.0) | 2:1.0.4-1
libnl1-pre6 | 1.0~pre6-6
libnm-util0 | 0.6.5-4
libpng12-0(>= 1.2.13-4) | 1.2.15~beta5-3
libqt3-mt (>= 3:3.3.7) | 3:3.3.7-9
libsm6 | 2:1.0.3-1+b1
libstdc++6 (>= 4.2.1) | 4.2.2-4
libx11-6| 2:1.0.3-7
libxext6| 1:1.0.3-2
zlib1g(>= 1:1.2.3.3.dfsg-1) | 1:1.2.3.3.dfsg-8
network-manager (>= 0.6.2) | 0.6.5-3
diff -Naur knetworkmanager.old/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-dialogfab.cpp knetworkmanager/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-dialogfab.cpp
--- knetworkmanager.old/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-dialogfab.cpp 2007-12-27 22:47:23.0 +
+++ knetworkmanager/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-dialogfab.cpp 2007-12-27 22:43:20.0 +
@@ -126,7 +126,7 @@
_encryptionMap [_id_wep_ascii] = new EncryptionWEP (WEP_ASCII);
//cryptoWidget->stackEncAdvanced->raiseWidget (cryptoWidget->advancedWEPWidget);
- comboEnc->setCurrentItem (_id_wep_passphrase);
+ comboEnc->setCurrentItem (encryption->getType ());
} else if ((capabilities & NM_802_11_CAP_PROTO_WPA) || (capabilities & NM_802_11_CAP_PROTO_WPA2)) {
QComboBox* comboProto = cryptoWidget->comboProtocol;
int protoId = 0;
diff -Naur knetworkmanager.old/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-encryption.cpp knetworkmanager/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-encryption.cpp
--- knetworkmanager.old/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-encryption.cpp 2007-12-27 22:46:47.0 +
+++ knetworkmanager/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-encryption.cpp 2007-12-27 22:42:31.0 +
@@ -343,6 +343,12 @@
}
void
+EncryptionWEP::getType( void )
+{
+ return _type;
+}
+
+void
EncryptionWEP::setDefaults ()
{
setMethod (WEP_OPEN_SYSTEM);
diff -Naur knetworkmanager.old/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-encryption.h knetworkmanager/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-encryption.h
--- knetworkmanager.old/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-encryption.h 2007-12-27 22:47:08.0 +
+++ knetworkmanager/knetworkmanager-0.2/knetworkmanager/src/knetworkmanager-encryption.h 2007-12-27 22:41:59.0 +
@@ -138,6 +138,7 @@
void setMethod (WEPMethod);
WEPMethod getMethod (void);
void setType( WEPType );
+ WEPType getType (void);
bool serialize (DBusMessage*, const QString &);
virtual bool deserialize( DBusMessageIter *, int we_cipher );
smime.p7s
Description: S/MIME cryptographic signature
Bug#434418: new version
On Thursday 09 August 2007 16:57:18 Joost Yervante Damad wrote: > In my opinion the problem still persists. All you did is change it from > awaking a lot to awaking a little less (2x a second). My preferred solution > would only wake up when needed. Are you absolutely positive that blocking > infinite makes the the application not stop on SIGTERM? > I really doubt it. I agree that the correct solution is to not have the program wake up on a timer at all. Of course blocking forever prevents the program terminating, a quick test confirms this. Have you tried it? There is no reason to think that without some special handling the program will ever leave the while loop, even if it receives a signal. All the signal handler does is to set a flag. After it returns, control will simply pass back to whichever part of the dbus library is waiting for messages. There are a number of solutions to this problem (longjmp, dbus message, ignoring cleanup, ...), but I don't have the time to investigate them at the moment. Sorry... Bruce smime.p7s Description: S/MIME cryptographic signature
Bug#436519: passkey agent patch problem
Package: kdebluetooth Version: 0.99+1.0beta2-7 Severity: minor On line 116 of debian/passkey-agent/passkey-agent.c: if(buffer[1]!='P' && buffer[1]!='I' && buffer[2]!='N') I believe the first two conditions of this if statement can't be satisfied at the same time! It also seems to me that the three conditions should be or'd, not anded. I don't think this actually causes a bug, but it's wrong nevertheless. The patch which is referenced in the README doesn't have this little faux pas: http://www.kmobiletools.org/node/228 Bruce --- System information. --- Architecture: amd64 Kernel: Linux 2.6.23-rc2 Debian Release: lenny/sid 500 unstablewww.debian-multimedia.org 500 unstableftp.de.debian.org 500 stable www.debian-multimedia.org 1 experimentalftp.de.debian.org --- Package information. --- Depends (Version) | Installed ==-+- kdelibs4c2a (>= 4:3.5.7-1) | 4:3.5.7.dfsg.1-3 libacl1 (>= 2.2.11-1) | 2.2.42-1 libart-2.0-2 (>= 2.3.18) | 2.3.19-3 libattr1 (>= 2.4.4-1) | 1:2.4.32-1.2 libaudio2 | 1.9-2+b1 libbluetooth2 (>= 3.0) | 3.13-1 libc6 (>= 2.5-5) | 2.6.1-1 libdbus-1-3 (>= 0.94) | 1.1.1-3 libfam0| 2.7.0-13 libfontconfig1 (>= 2.4.0) | 2.4.2-1.2 libfreetype6 (>= 2.2) | 2.3.5-1+b1 libgcc1(>= 1:4.2-20070516) | 1:4.2.1-2 libice6 (>= 1:1.0.0) | 2:1.0.3-3 libidn11 (>= 0.5.18) | 1.0-0 libjpeg62 | 6b-13 libopenobex1 | 1.3-3 libpng12-0 (>= 1.2.13-4) | 1.2.15~beta5-2 libqt3-mt (>= 3:3.3.7) | 3:3.3.7-6 libsm6 | 2:1.0.3-1+b1 libstdc++6 (>= 4.2-20070516) | 4.2.1-2 libx11-6 | 2:1.0.3-7 libxcursor1 (>> 1.1.2) | 1:1.1.8-2 libxext6 | 1:1.0.3-2 libxft2 (>> 2.1.1) | 2.1.12-2 libxi6 | 2:1.1.2-1 libxinerama1 | 1:1.0.2-1 libxrandr2(>= 2:1.2.0) | 2:1.2.1-1 libxrender1| 1:0.9.2-1 libxt6 | 1:1.0.5-3 qobex | 0.99+1.0beta2-7 zlib1g(>= 1:1.2.1) | 1:1.2.3.3.dfsg-5 bluez-utils| 3.13-1 smime.p7s Description: S/MIME cryptographic signature
Bug#434418: More testing needed
Joost, I found this bug report while looking into the same problem, I was about to write a similar patch :) . Did you test whether passkey-agent dies gracefully with your patch? I believe the patch effectively prevents passkey-agent from noticing SIGTERM or SIGINT unless it receives a message. I think that passkey-agent just won't die if you send it one of those signals. I don't know enough about dbus to be able to say, but you might just be able to have the program send itself a dbus message from within the signal handler which will wake it up. It might also be the case that the cleanup is irrelevant since the program is terminating anyway, and the signal handlers could just be removed. Bruce smime.p7s Description: S/MIME cryptographic signature
Bug#432682: crash: less SEGV when permission is denied
tags 432682 + patch
thanks
Sorry, that bug report didn't actually say anything! Here's the real
backtrace:
Program received signal SIGSEGV, Segmentation fault.
0x00403828 in ch_length () at ch.c:508
508 if (ch_flags & CH_HELPFILE)
(gdb) bt
#0 0x00403828 in ch_length () at ch.c:508
#1 0x004041cd in ch_seek (pos=0) at ch.c:419
#2 0x0040b7cd in jump_loc (pos=0, sline=0) at jump.c:199
#3 0x0040baf4 in repaint () at jump.c:98
#4 0x004105fb in error (fmt=0x415328 "%s", parg=0x7fffd65b9a10) at
output.c:529
#5 0x004092b6 in edit_ifile (ifile=0x6277d0) at edit.c:311
#6 0x0040946a in edit_istep (h=0x6277d0, n=-1, dir=1) at edit.c:527
#7 0x00407167 in commands () at command.c:1412
#8 0x00402158 in main (argc=, argv=) at main.c:286
My interpretation of the crash is that less is attempting to get information
about the file it's failing to display in order to decide where to place the
cursor. It decides that the display is "squished", meaning that the first
displayed line is not at the top of the screen, and wants to repaint the
screen. This repaint causes it to try to find out the final cursor location
from the current position in the (non-existant) file, and crash.
This repainting behaviour seemed to have no effect in any test case I could
come up with, so I removed it. The patch that does this is:
--- output.c.orig 2007-07-15 18:51:25.0 +0100
+++ output.c2007-07-15 19:07:09.0 +0100
@@ -525,8 +525,6 @@
if (any_display && is_tty)
{
- if (!oldbot)
- squish_check();
at_exit();
clear_bot();
at_enter(AT_STANDOUT);
Of course, less's assessment that the display is "squished" is wrong anyway
(it blanked the screen before displaying this short file) which is why I
believe it's safe to remove it. However, if you disagree with this resolution
you might like to investigate making the squished detection more clever.
Bruce
smime.p7s
Description: S/MIME cryptographic signature
Bug#432682: crash: less SEGV when permission is denied
Package: less
Version: 406-0
Severity: normal
I have managed to crash less! I have reduced the crash to this test case:
$ echo $LINES $COLUMNS
30 239
$ echo hello > test
$ echo goodbye > test2
$ chmod -r test2
$ less test test2
It seems that when less views a file which does not fill the entire window,
followed by pressing ":n" to go to the next file (for which I have no read
permission) it segfaults.
The following is the end of an ltrace immediately before the crash:
open("test2", 0, 041) = -1
__errno_location() = 0x2b851ec26aa8
strerror(13) = "Permission denied"
strlen("Permission denied") = 17
strlen("test2") = 3
calloc(23, 1) = 0x627930
snprintf("\001\200\255\373\377\177",
4297040, "\377\377\377\377\377\377\377\377"...) = 22
--- SIGSEGV (Segmentation fault) ---
+++ killed by SIGSEGV +++
The strace was uninteresting, as was the gdb backtrace.
Hope that's of some use,
Bruce
--- System information. ---
Architecture: amd64
Kernel: Linux 2.6.20
Debian Release: lenny/sid
500 unstablewww.debian-multimedia.org
500 unstableftp.de.debian.org
500 stable www.debian-multimedia.org
1 experimentalftp.de.debian.org
--- Package information. ---
Depends (Version) | Installed
==-+-===
libc6 (>= 2.5-5) | 2.6-2
libncurses5 (>= 5.4-5) | 5.6-3
debianutils (>= 1.8) | 2.22.1
smime.p7s
Description: S/MIME cryptographic signature
Bug#430007: cover manager window shortcuts are useless
Package: amarok Version: 1.4.5-5 Severity: minor When flicking through a list of potential album covers (right-click on an album in the context view and do "Fetch from amazon.co.uk") the kekyboard shortcuts are unavailable because the window does not have the focus. It doesn't appear in the list of task-switchable (i.e. alt-tab) windows either, so it's impossible to give it the focus without using the mouse. Now that you have given it the focus, you might as well just click on the buttons instead of moving your hand back to the keyboard, thus rendering the shortcuts useless. The shortcuts are on the Save, New search, Next cover and Cancel buttons. Please could this window get the focus when it appears? It blocks the rest of Amarok anyway... Cheers, Bruce --- System information. --- Architecture: amd64 Kernel: Linux 2.6.20 Debian Release: lenny/sid 500 unstablewww.debian-multimedia.org 500 unstableftp.de.debian.org 500 stable www.debian-multimedia.org 1 experimentalftp.de.debian.org --- Package information. --- Depends (Version) | Installed ===-+- amarok-engines | 1.4.5-5 OR amarok-engine | ruby| 1.8.2-1 unzip | 5.52-9 kdelibs4c2a (>= 4:3.5.7-1) | 4:3.5.7.dfsg.1-1 libart-2.0-2(>= 2.3.18) | 2.3.19-3 libaudio2 | 1.9-2 libc6(>= 2.5-5) | 2.5-11 libfontconfig1 (>= 2.4.0) | 2.4.2-1.2 libfreetype6 (>= 2.2) | 2.2.1-6 libgcc1 (>= 1:4.2-20070516) | 1:4.2-20070609-1 libgl1-mesa-glx | 6.5.2-5 OR libgl1 | libglib2.0-0(>= 2.12.9) | 2.12.12-1 libgpod1| 0.4.2-3 libice6(>= 1:1.0.0) | 1:1.0.3-2 libidn11(>= 0.5.18) | 0.6.5-1 libifp4 | 1.0.0.2-3 libjpeg62 | 6b-13 libkarma0 | 0.0.6-2 libmtp5 | 0.1.5-2 libmysqlclient15off (>= 5.0.27-1) | 5.0.41-2 libnjb5 | 2.2.5-4.1 libpng12-0(>= 1.2.13-4) | 1.2.15~beta5-2 libpq5 | 8.2.4-1 libqt3-mt (>= 3:3.3.7) | 3:3.3.7-5 libruby1.8 (>= 1.8.6) | 1.8.6-2 libsdl1.2debian (>= 1.2.10-1) | 1.2.11-9 libsm6 | 2:1.0.3-1 libsqlite3-0(>= 3.3.17) | 3.3.17-1 libstdc++6(>= 4.2-20070516) | 4.2-20070609-1 libtag1c2a (>= 1.4) | 1.4-8 libtagc0 (>= 1.4) | 1.4-8 libtunepimp5| 0.5.3-2 libusb-0.1-4 (>= 2:0.1.12) | 2:0.1.12-7 libvisual-0.4-0 (>= 0.4.0) | 0.4.0-1.1 libx11-6| 2:1.0.3-7 libxcursor1 (>> 1.1.2) | 1:1.1.8-2 libxext6| 1:1.0.3-2 libxft2 (>> 2.1.1) | 2.1.12-2 libxi6 | 1:1.0.1-4 libxinerama1| 1:1.0.2-1 libxrandr2 (>= 2:1.2.0) | 2:1.2.1-1 libxrender1 | 1:0.9.2-1 libxt6 | 1:1.0.5-3 zlib1g (>= 1:1.2.1) | 1:1.2.3-15 smime.p7s Description: S/MIME cryptographic signature
Bug#429864: Fwd: Re: Bug#429864: help system relies on yelp
On Wednesday 20 June 2007 20:53:29 you wrote: > www-browser is not necessarily appropriate as a help browser. You are > free to ignore the recommendation and the help text is easily accessible > under /usr/share/doc/sgt-puzzles and in manual pages. Perhaps I could > arrange to call www-browser as a fallback, though. I see your point. www-browser maybe isn't the best choice. To address this particular bug, perhaps the puzzles could try yelp, then khelpcenter, then www-browser? Then the package could recommend any of those 3, keeping decent integration with the desktop. The trouble with your workaround is that the puzzles then give a cryptic error message when you try to use the help from the menu: "There was an error launching the default action command associated with this location." It took me a while even to figure out that yelp was required to fix this! Bruce smime.p7s Description: S/MIME cryptographic signature
Bug#429864: help system relies on yelp
Package: sgt-puzzles Version: 7446-1 Severity: wishlist Hi, sgt-puzzles recommends yelp to view the help files. This is irritating on a KDE system because it involves installing 30MB of gnome packages just to view the HTML files which would be perfectly viewable in Konqueror or any other browser! Please could you change this recommendation to a dependency on www-browser? Thanks, Bruce --- System information. --- Architecture: amd64 Kernel: Linux 2.6.20 Debian Release: lenny/sid 500 unstablewww.debian-multimedia.org 500 unstableftp.de.debian.org 500 stable www.debian-multimedia.org 1 experimentalftp.de.debian.org --- Package information. --- Depends(Version) | Installed -+-== libatk1.0-0 (>= 1.13.2) | 1.18.0-2 libbonobo2-0 (>= 2.15.0) | 2.18.0-2 libc6 (>= 2.5) | 2.5-11 libcairo2 (>= 1.4.0) | 1.4.6-1.1 libfontconfig1(>= 2.4.0) | 2.4.2-1.2 libglib2.0-0 (>= 2.12.9) | 2.12.12-1 libgnome2-0 (>= 2.17.3) | 2.18.0-4 libgtk2.0-0 (>= 2.10.3) | 2.10.13-1 liborbit2 (>= 1:2.14.1) | 1:2.14.7-0.1 libpango1.0-0(>= 1.16.2) | 1.16.4-1 libpopt0 (>= 1.10) | 1.10-3 libx11-6 | 2:1.0.3-7 libxcursor1 (>> 1.1.2) | 1:1.1.8-2 libxext6 | 1:1.0.3-2 libxfixes3 (>= 1:4.0.1) | 1:4.0.3-2 libxi6 | 1:1.0.1-4 libxinerama1 | 1:1.0.2-1 libxrandr2 (>= 2:1.2.0) | 2:1.2.1-1 libxrender1 | 1:0.9.2-1 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#421292: middle click in composer body tries to paste from wrong clipboard
Package: kmail Severity: minor --- Please enter the report below this line. --- (For clarity in the following description, I consider the X selection to be what you get when you select and then middle click a piece of text. I consider the X clipboard to be what you get when you ctrl-c and ctrl-v an object) When I middle click in the body of a composer window I expect it to paste the current X selection (in this case, an email address). Instead, it asks me for the name of the new attachment it's about to create (in this case, a screenshot I obtained an hour ago when I missed F12 and hit PrintScreen by mistake!). It works correctly when the contents of the X clipboard are text (i.e. I get the contents of the X selection when middle clicking) I can work around this by pasting the text into the subject and then dragging and dropping it, but that only works for a single line of text. I can work around it fully by replacing the contents of my X clipboard with arbitrary text. Cheers, Bruce --- System information. --- Architecture: amd64 Kernel: Linux 2.6.20 Debian Release: lenny/sid 500 unstablewww.debian-multimedia.org 500 unstableftp.de.debian.org 500 stable www.debian-multimedia.org 1 experimentalftp.de.debian.org --- Package information. --- Depends (Version) | Installed ===-+- kdelibs4c2a (>= 4:3.5.6-1) | 4:3.5.6.r1.dfsg.1-4 libart-2.0-2(>= 2.3.18) | 2.3.19-3 libaudio2 | 1.8-4 libc6 (>= 2.5) | 2.5-4 libfontconfig1 (>= 2.4.0) | 2.4.2-1.2 libfreetype6 (>= 2.2) | 2.2.1-5 libgcc1 (>= 1:4.1.1-12) | 1:4.1.2-4 libice6(>= 1:1.0.0) | 1:1.0.3-2 libidn11(>= 0.5.18) | 0.6.5-1 libjpeg62 | 6b-13 libkcal2b (>= 4:3.5.6) | 4:3.5.6.dfsg.1-2 libkdepim1a(>= 4:3.5.6) | 4:3.5.6.dfsg.1-2 libkleopatra1 (>= 4:3.5.6) | 4:3.5.6.dfsg.1-2 libkmime2 (>= 4:3.5.6) | 4:3.5.6.dfsg.1-2 libkpimidentities1 (>= 4:3.5.6) | 4:3.5.6.dfsg.1-2 libksieve0 (>= 4:3.5.6) | 4:3.5.6.dfsg.1-2 libmimelib1c2a (>= 4:3.5.6) | 4:3.5.6.dfsg.1-2 libpng12-0(>= 1.2.13-4) | 1.2.15~beta5-1 libqt3-mt (>= 3:3.3.7) | 3:3.3.7-4+b1 libsm6 | 1:1.0.2-2 libstdc++6(>= 4.1.1-12) | 4.1.2-4 libx11-6| 2:1.0.3-7 libxcursor1 (>> 1.1.2) | 1:1.1.8-2 libxext6| 1:1.0.3-2 libxft2 (>> 2.1.1) | 2.1.12-2 libxi6 | 1:1.0.1-4 libxinerama1| 1:1.0.2-1 libxrandr2 (>= 2:1.2.0) | 2:1.2.1-1 libxrender1 | 1:0.9.2-1 libxt6 | 1:1.0.5-2 zlib1g (>= 1:1.2.1) | 1:1.2.3-13 perl| 5.8.8-7 kdepim-kio-plugins | 4:3.5.6.dfsg.1-2 kdebase-kio-plugins | 4:3.5.6.dfsg.2-2 smime.p7s Description: S/MIME cryptographic signature
