Overflow stack bug
Hello,
I found this strange bug while coding, and wrote it in the
simplest way to be sure it was true. Try opening a new AS3 FLA
file, and write this code in the Actions Panel:
function someFunc(x1:String, x2:String, x3:Number=0):String
var my_result:String = "result";
return my_result;
someFunc("someVal1", "someVal2");
Calling someFunc will call a #1023 stack overflow error!!!
The only way not to have it is to remove the "x3" Parameter. That's
sooooooo strange (and annoying, 'cause before realizing this was a
bug i was getting crazy trying to find an error in my original
code, which was much more complex).
P.S. i'm using FlashPlayer debugging version 9.0.r47, i hope
it's his fault - or maybe there's another reason? Any idea?
Daniel
P.S. this is the output error code generated by that code.
I'm posting it because it's so different from other error outputs,
and by what it says it really seems that Flash is going crazy. When
you see "stackOverflow_01_fla", that's the name of the AS3 FLA i
created (stackOverflow_01.FLA) to debug this error.
verify stackOverflow_01_fla::MainTimeline/someFunc()
stack:
scope: [global Object$ flash.events::EventDispatcher$
flash.display::DisplayObject$ flash.display::InteractiveObject$
flash.display::DisplayObjectContainer$ flash.display::Sprite$
flash.display::MovieClip$ stackOverflow_01_fla::MainTimeline$]
locals: stackOverflow_01_fla::MainTimeline String? String?
Number *
0:getlocal0
stack: stackOverflow_01_fla::MainTimeline
scope: [global Object$ flash.events::EventDispatcher$
flash.display::DisplayObject$ flash.display::InteractiveObject$
flash.display::DisplayObjectContainer$ flash.display::Sprite$
flash.display::MovieClip$ stackOverflow_01_fla::MainTimeline$]
locals: stackOverflow_01_fla::MainTimeline String? String?
Number *
1:pushscope
stack:
scope: [global Object$ flash.events::EventDispatcher$
flash.display::DisplayObject$ flash.display::InteractiveObject$
flash.display::DisplayObjectContainer$ flash.display::Sprite$
flash.display::MovieClip$ stackOverflow_01_fla::MainTimeline$]
stackOverflow_01_fla::MainTimeline
locals: stackOverflow_01_fla::MainTimeline String? String?
Number *
2:pushnull
stack: null
scope: [global Object$ flash.events::EventDispatcher$
flash.display::DisplayObject$ flash.display::InteractiveObject$
flash.display::DisplayObjectContainer$ flash.display::Sprite$
flash.display::MovieClip$ stackOverflow_01_fla::MainTimeline$]
stackOverflow_01_fla::MainTimeline
locals: stackOverflow_01_fla::MainTimeline String? String?
Number *
3:coerce_s
stack: String?
scope: [global Object$ flash.events::EventDispatcher$
flash.display::DisplayObject$ flash.display::InteractiveObject$
flash.display::DisplayObjectContainer$ flash.display::Sprite$
flash.display::MovieClip$ stackOverflow_01_fla::MainTimeline$]
stackOverflow_01_fla::MainTimeline
locals: stackOverflow_01_fla::MainTimeline String? String?
Number *
4:setlocal 4
stack:
scope: [global Object$ flash.events::EventDispatcher$
flash.display::DisplayObject$ flash.display::InteractiveObject$
flash.display::DisplayObjectContainer$ flash.display::Sprite$
flash.display::MovieClip$ stackOverflow_01_fla::MainTimeline$]
stackOverflow_01_fla::MainTimeline
locals: stackOverflow_01_fla::MainTimeline String? String?
Number String?
6:pushstring "result"
stack: String
scope: [global Object$ flash.events::EventDispatcher$
flash.display::DisplayObject$ flash.display::InteractiveObject$
flash.display::DisplayObjectContainer$ flash.display::Sprite$
flash.display::MovieClip$ stackOverflow_01_fla::MainTimeline$]
stackOverflow_01_fla::MainTimeline
locals: stackOverflow_01_fla::MainTimeline String? String?
Number String?
8:dup
VerifyError: Error #1023: Stack overflow occurred.
at stackOverflow_01_fla::MainTimeline/someFunc()
at
stackOverflow_01_fla::MainTimeline/stackOverflow_01_fla::frame1()
Similar Messages
-
2960X 15.0(2)EX5 Stack Bug? Master Switch Ports link in Orange, no spanning Tree
Is anyone aware of a bug in version 15.0(2)EX5 for 2960X Switches that would cause a switch in the master role to stop linking in new ports in green (and passing traffic). I have 2 2960X-48FPD-L Switches in a stack and whichever switch I designate master will only link new connections in orange and not pass traffic. All ports linked in show up/up and can be seen in a show cdp neighbor but won't pass any other traffic.
If I unplug the Stacking cables both switches become masters and ports linked in green on the previous member switch stay green, but after it switches to master any new connections plugged in only link in orange.
If I switch priorities and reboot the problem switches to the new master switch and the problem goes away on the member switch.
Also, a switch in the master role does not show any spanning tree instances for ports in the orange link state.
Has anyone seen this issue and do you know of a solution?
JimA quick update for those with this same problem.
1. 15.2(3)E turned out to be very unstable causing my switch stack to randomly lockup/reboot one of the switches about once a week.
2. I downgraded back to 15.0(2)EX5 but found a workaround. It turns out the switch stack with the 15.0 versions does not like the switchport voice vlan command on any of the interfaces on the master switch. I simply removed the voice vlan configuration on the interfaces and all the switch ports linked in just fine. I would prefer to run the phones on a voice vlan, but it still works without, just the PC's and phones are on the same vlan.
Jim -
Hi all,
I'm using Bridge to sort medias on a server. I added keywords and and created stackes with files.
My medias are in a "server" who allow other users with Adobe Bridge to used them.
I saw a tricky bug with my stackes. After fiews days of use, my stacks are deleted.
I know that those informations are in the .BridgeSort file and this file stay in the right place.
My problem is that this file is changed without my interference.
Has someone an idea to solve this problem?
In advance, thanks you for your help!
ByeSupposing with stacks deleted you mean the files are still there but not any more in a stack?
Stacks can be very tricky, and for some reason they can disappear all or partly in one folder. I have this sometimes when being to fast with renaming files or some other keystrokes. it is very troublesome indeed.
Also Bridge is not officially supported for use over a network. It is very sad but sometimes I can do weeks without groups unstacking themselves and out of the blue the are gone.
For me they are for saving space in the content window and can be easily redone. If you rely heavily on stacks you better use a second safety like rating or maybe better a specialized keyword so you can find them back. -
I have Stacks configured as Date Added, Folder, Fan for my Downloads folder. When new files are downloaded, they do not show up at the bottom of the stack like they used to. Instead, I have to switch from Date Added (to Date Created, usually) then back to Date Added to get that new file to show up. Needless to say, this is very annoying.
Bugreport wrote:
This is a follow up to Bug ID# 5749476. After further investigation it has been determined that this is a known issue, which is currently being investigated by engineering. This issue has been filed in our bug database under the original Bug ID# 5721820.
lets wait… -
Hi!
I was using Leopard for a half of a day now and some bug already found. Here's one :
When you click on a Stack in the Dock(same thing with both stack views) and them, after it's open - clicking the same place(without moving your mouse) - it won't open. Same with closing.
I wanted to see what's inside Downloads, just a tiny peak, when I found that I can't close it.
Seems very disturbing to me.there are definitely problems in the deployment from eclipse.
try deleting the server and adding it again, that normally fixes most problems... -
i have a problem with focus stacking
i try it in PS CS6, no problem.
With the same pictures in CC2014 there is no good result.
what do i do wrong?
i do it with 3 picture, 5, 6 7 etc..what do i do wrong?
Nothing, the Auto-Blend Layers feature has been damaged.
You could install Photoshop CC and work with that or wait for a possible fix in some future update. -
[Issue] Track Stacks Bug When Hidden
This is pretty hard to explain, but I'll do my best!
When a track stack (folder) is hidden in the arrange page using the 'H' icon, upon unhiding, the arrangement of tracks is changed in the mixer. The Folder header track is placed at the end of the mixer, instead of to the left of the tracks within the stack (where it should be).
Hope that makes sense.
SteveIt happens in Tracks view. Here's what happens here: In this pic, I have placed my groups into a folder stack. Everything as expected.
In this pic, I have 'hidden' the groups in the arrange windows. However, on the mixer, the track order becomes scrambled and the group 'header' is placed on the far right.
I would've thought that stacked tracks sholud maintain their hierarchy whether hidden, or not!
Thanks -
After the recent upgrade, I can no longer load vertical scrolling articles into my droid content viewers. Even when I built a 1-article jpeg folio I recieve, "This folio contains one or more PDF stacks which are not yet supported in content viewer for Android."
Any ideas?
-SOK,
We figured out that if you copy an article from folio to folio, it creates the aforementioned "PDF stacks" bug. If you recreate all assets from scratch, and build a new article in a new folio, it downloads in the droid content viewer just fine.
NOTES:
480 x 800
folio default format: jpg
article format: jpg
*must use "add open indesign document"
indesign CS6 (v8)
plugin version 12.3.3.20120911_m_699825 (8.0.9.23)
devices: exhibit 4g and htc one s
Thanks fellas,
-Stien -
I want to make my Java regex class interesting, even exciting.
Not by the volume of the regex pattern string, but by its inherent malign logic, the regexp I want should crash, i.e. overflowing stack, causing memory shortage or causing anything horrible.
I have searched the forum but not found nifty example(s).
I expect some real genius could give it/them on this thread.
Thanks in advance.The regexes in both of those bug reports demonstrate two common mistakes. The first, as you said, is that they employ alternation when they should be using something more efficient. In each of those cases, every alternative matched exactly one letter, which means they should have been using a simple character class instead. When you're searching for multi-character sequences, you should look for opportunities to minimize the scope of the alternation, but don't bend over backward; readability matters, too. However, if you can factor out a prefix, you probably should. For instance, if you were looking for "you", "your" or "yourself", the regex you(r(self)?)? would be much more efficient than you|your|yourself.
The other mistake shown in both reports is the misuse of quantifiers in combination with parentheses. In each case, there's a subexpression inside a set of parens that matches exactly one character, and a quantifier outside the parens. Never do that. Parenthesized groups, even the non-capturing variety, always increase the overhead cost of regex matching; to put parens around a single character probably increases the cost of matching each character by several times, and capturing groups are even more expensive. Either get rid of the parentheses or put the quantifier inside them.
I wouldn't advise regex newbies to be as specific as possible all the time because it's too easy to create a regex that misses some of the things you want to match. But they should definitely learn how be specific, and not rely on the old dot-star all the time. In fact, for the sake of learning, I recommend trying to write all your regexes first without using the dot metacharacter at all. But in the real world, it usually works well enough.
As an example, suppose you're trying to extract certain parts of a bunch of reasonably well-formed web pages. How would you extract the HEAD section of each one? The BODY section? Finally, how would you extract all the SCRIPT sections from each page? The HEAD section can be fairly large, but it usually makes up a small portion of the whole page. If you used (?is)<HEAD>.*</HEAD>, the dot-star would consume the whole page, then backtrack one character at a time through the whole BODY section--a ridiculous waste. It makes much more sense to use a reluctant quantifier: (?is)<HEAD>.*?</HEAD>. It's not the most efficient way to do it, but it's probably good enough.
The BODY section is just the opposite. This time, you want the dot-star to gobble everything up, because you know it will only have to backtrack a few characters from the end of the page, so (?is)<BODY\b.*</BODY> is exactly what you want.
SCRIPT elements are a whole different story. They can be completely empty or thousands of characters long, and there can be a lot of them. This time, it's worth the effort to disable backtracking completely, and the inherent inefficiency of reluctant quantifiers will be too expensive. This is when you inject the nitrous oxide: (?is)<SCRIPT\b(?:[^<]++|<(?!/SCRIPT>))*+</SCRIPT> -
SYSTEM HANGS WHILE RUNNING EXP COMMAND IN ORACLE 10G
WHEN WE ARE RUNNING THE EXP (EXPORT) COMMAND IN ORACLE 10G
TO TAKE THE BACKUP IT GETS HANGED FOR A LONG TIME AND THEN IN THE LOG FILE WE FIND THE FOLLOWING ERROR MESSAGE.
SOMETIMES AFTER RESTARTING THE SERVER WE ARE ABLE TO RUN THE EXP COMMAND.
Connected to: Oracle Database 10g Enterprise Edition Release 10.1.0.2.0 - Production
With the Partitioning, OLAP and Data Mining options
Export done in WE8MSWIN1252 character set and AL16UTF16 NCHAR character set
About to export specified users ...
. exporting pre-schema procedural objects and actions
. exporting foreign function library names for user MANHAR
. exporting PUBLIC type synonyms
. exporting private type synonyms
. exporting object type definitions for user MANHAR
About to export MANHAR's objects ...
. exporting database links
. exporting sequence numbers
. exporting cluster definitions
*********************************** AT THIS POINT THE SYSTEM HANGS FOR A LONG TIME.
EXP-00056: ORACLE error 24324 encountered
ORA-24324: service handle not initialized
EXP-00056: ORACLE error 24324 encountered
ORA-24324: service handle not initialized
EXP-00000: Export terminated unsuccessfully
THE EXP COMMAND THAT WE USE IS AS FOLLOWS:
EXP MANHAR/MANHAR FILE=F:\BACKUP\MANHAR.DMP LOG=F:\BACKUP\MANHAR.LOG BUFFER=1000000 GRANTS=Y CONSTRAINTS=Y
WE ARE RUNNING THE EXP COMMAND ON THE SERVER ITSELF.
THE SERVER OPERATING SYSTEM IS WINDOWS 2003 SERVER.
Kindly give us a solution at the earliest.You should ask Oracle support for help. There's only one entry for this error stack (Bug 3361288 - Export of XMLSCHEMA may fail with ORA-24324 in UTF8), but it should have been fixed for your DB version and , as far as I can see, you don't have a Unicode database.
Werner -
Bridge doesn't load files to layers in Photoshop
I recently uninstalled Photoshop CC 2014 and loaded Bridge CC 14.0 because of a known bug that ca
uses focus stacking to not work in Photoshop CC 2014. Now Bridge doesn't recognize Photoshop. Adobe help told me to update my Photoshop, but that would bring me to v. 14.2.1 which has the focus stacking bug. Any ideas?Yep if I zoom in like 1000% I can clearly see which layer is too big. It's like when I'm using the marquee tool to select the area, snapping to the grid of guides, some times it's selecting 1 row or column of pixels too many. I can then fix it but I basically have to redo it those sections (bring back in the original bg elements etc).
The margin of error has to be what it is. The buttons are not just floating in space, they are on a background that I want to be a part of the popup menu. So the background becomes part of the button on the popup. If it's not perfect it is noticable.
On some occassions the highlight is the culprit, it's too big or needs to be moved a tad to fit within the rectangle.
If there was a slice to layer option it seems it would solve everything.
thanks -
Differences between -Xss[size] in linux and windows
Simple thread test program which runs 3000 threads:
startup options on windows:
-server -Xms16m -Xmx16m -Xss7k
startup options on linux:
-server -Xms16m -Xmx16m -Xss97k (Why 90k bigger stack size??)
Linux is using newest kernel and NPTL threads. With 'regular' threads the linux version overflows stack unless -Xss2m is given...
I Think Sun needs to come up with clear specification of threading models, libraries etc... used in both environments. If you search these forums - one of the most common and baffling error is the sig11 on linux. I think the main cause behind it are the library 'mismatches'.
Answers to these questions are really needed:
What is the recommended threading library on linux with each VM version?
Against which library is the VM tested?
If anyone has experiences with NPTL threads on 1.4.2 VM please contribute. How many threads did you manage to create, how much memory that took? What distribution, thread library, start up parameters, etc... did you use.
I'm going to do a bit more experimenting on a real life application during the followin weeks. With simple test program we were able to create over 16000 threads, but if you do the math with the minimum stack size you can guess that the virtual memory usage was sky high!
P.S.
I'll post more exact platform specs when I'm back to work tomorrow..http://java.sun.com/docs/hotspot/VMOptions.html - see under -Xoss option.
BTW, I ran my test case again, and now i finally succeeded to prove that 'rule' (see that 'another' thread): Xss*nThreads + Xmx < Xmx_MAX. The conclusions are (at least under 32-bit Linux):
1) Xss does limit native stack size
2) the 'rule' seems to be independent of whether u r actually using the allocated stack space or not
Below a test case is attached. Under 32-bit Linux it crashes before it can create 30 threads with
Exception in thread "main" java.lang.OutOfMemoryError: unable to create new native thread
at java.lang.Thread.start(Native Method)
at testlab.StackTest.main(StackTest.java:46)exception, if start up parameters are
java -cp ../../classes -Djava.library.path=./ -Xms1870m -Xmx1870m -Xss1m testlab.StackTest 100 200 1000000The crash point can be moved/eliminated by modifying either Xss param (the smaller the size the later the crush would happen) or Xmn/Xmx (the smaller...., the later....). Exact Xmx_MAX number might depend on what threads library you are running. On my hardware it is as follows:
java -Xms1920M -Xmx1920M
Error occurred during initialization of VM
Could not reserve enough space for object heap
java -Xms1910M -Xmx1910M
Error occurred during initialization of VM
Could not reserve enough space for card marking array
java -Xms1900M -Xmx1900M
<RUNS OK>See how error messages differ. So, in the case above Xmx_MAX is around 1.9G.
Native code was compiled with gcc2.96, under 32-bit RH Linux, no optimizations.
Java code compiled with 1.4.2_02 compiler, run with JVM is 1.4.2_02
Java file:
package testlab;
* Author: volenin
* Date: Dec 4, 2003
* Time: 10:40:20 AM
* under GPL license
public class StackTest extends Thread {
static byte[] arr;
static {
System.loadLibrary("stacktest");
boolean isStarted = false;
int allocSize;
StackTest(int allocSize) {
this.allocSize = allocSize;
setDaemon(true);
// arr2 = new byte[allocSize];
public void run() {
isStarted = true;
allocate(allocSize);
synchronized (this) {
try { wait(); }
catch (InterruptedException Ie) {}
public native void allocate(int size);
public static void main(String[] args) throws Exception {
int nThreads = Integer.parseInt(args[0]);
int allocSize = Integer.parseInt(args[1]);
int initSize = Integer.parseInt(args[2]);
arr = new byte[initSize];
for (int i = 0; i < nThreads; i++) {
int n = i+1;
System.out.println("Creating thread #"+n);
StackTest test = new StackTest(allocSize);
System.out.println("Starting thread #"+n);
test.start();
System.out.println("Thread started #"+n);
synchronized (StackTest.class) {
if (!test.isStarted) StackTest.class.wait(100);
System.out.println("Thread running #"+n);
Native .h file:
#include <jni.h>
#ifndef _Included_testlab_StackTest
#define _Included_testlab_StackTest
#ifdef __cplusplus
extern "C" {
#endif
JNIEXPORT void JNICALL Java_testlab_StackTest_allocate(JNIEnv *, jobject, jint);
#ifdef __cplusplus
#endif
#endif
Native .cpp file:
#include "testlab_StackTest.h"
JNIEXPORT void JNICALL Java_testlab_StackTest_allocate(JNIEnv *env, jobject jobj, jint allocSize) {
int arr[allocSize];
int arrSize = sizeof(arr);
printf("array allocated: %d, %d\n", allocSize, arrSize);
getchar(); -
JavaScript blocked or failed to load
Hello,
Sometime ago I suddenly realized that JavaScript was not working in my Firefox. I get a very explicit error message from Stack Overflow:
"Stack Overflow requires external JavaScript from another domain, which is blocked or failed to load."
Many other websites also do not work.
I currently am running Windows 7 and on typing "java -version", I get the following output:
java version "1.7.0_40"
Java(TM) SE Runtime Environment (build 1.7.0_40-b43)
Java HotSpot(TM) 64-Bit Server VM (build 24.0-b56, mixed mode)
I know Java is installed and working because I can compile and run .java files from NetBeans.
I am in a fix as I have spent many hours troubleshooting this problem but could not solve it. I am very hopeful for a solution here.
ThanksFirefox has a single global setting for JavaScript: either it's enabled or it isn't. If JavaScript works on any site, then that setting should be fine. If you're not sure, you can check here:
(1) In a new tab, type or paste '''about:config''' in the address bar and press Enter. Click the button promising to be careful.
(2) In the search box above the list, type or paste '''java''' and pause while the list is filtered
(3) If '''javascript.enabled''' is not set to its default value of true, double-click it to switch it back to true.
Firefox does not have a user interface for applying site-specific scripting restrictions, but this can be done using extensions or using "capability" preferences. I don't see the best known script blockers on your extension list (NoScript, YesScript, Controle De Scripts) so I suggest checking for the hidden preferences first.
Open your current Firefox settings (AKA Firefox profile) folder using
Help > Troubleshooting Information > "Show Folder" button
Scroll down to prefs.js and open it in a text editor (note: you can't save changes while Firefox is running, but you can exit Firefox if you find something you need to edit out)
Search for '''capab''' (the beginning of the word capability). Do you find any preferences that contain capability and javascript and noAccess? If so, that may explain why scripts cannot run on some sites. Feel free to post those lines here for analysis.
To test the extension possibility, could you try some of the problem sites in Firefox's Safe Mode? That's a standard diagnostic tool to bypass interference by extensions (and some custom settings). More info: [[Troubleshoot Firefox issues using Safe Mode]].
You can restart Firefox in Safe Mode using
Help > Restart with Add-ons Disabled
In the dialog, click "Start in Safe Mode" (''not'' Reset)
Any change? -
ArchLinux Newsletter for February 22
Greetings, and welcome to the ArchLinux weekly newsletter for the week ending February 22, 2004.
Front Page News:
NEW KERNELS ABOUND
Hi all,
First off, a huge thank-you to everyone that has donated over the last couple of days. We now have enough of a start to pay setup fees and the first couple months on a new server. You guys rock.
Now, an update regarding the new kernels... We're planning on using 2.6.3 as the default kernel for Widget's release, so I've renamed the 2.4 kernel packages for more consistency. We now have four kernel packages:
* kernel24
* kernel24-scsi
* kernel26
* kernel26-scsi
As you know, the kernel26 packages write their bzImages as /boot/vmlinuz26 so that they can coexist with a 2.4 kernel. I've left the kernel 2.4 bzImage as /boot/vmlinuz for now, to avoid breaking too many users' boxes. But keep in mind that I may rename them to /boot/vmlinuz24 sometime in the future. Watch out!
Another effect of the package renames is that the setup script had to be updated. This means that you will have to re-download the root.img disk image to do an ftp-based install. The old one will look for the 'kernel' package, which no longer exists.
http://archlinux.org/news.php#48
FUN WITH BANDWIDTH COSTS
As most of you know, the recent review on OSNews has brought a large influx of users to us. While we're overjoyed to have so many new users, the bandwidth costs associated with all these new --sysupgrade'ers is going to cost me a lot of money that I don't really have.
To curb the trend, I've had to cap transfer speeds on ftp.archlinux.org. They now sit at 20 K/s with a max of 30 users at a time.
So, long story short, I have a couple of favors to ask:
1. If you can volunteer to be a mirror, please contact me.
2. Please download this perl script, which will use the "netselect" program to re-sort the server lists in pacman.conf, lowest ping times first. You can run it like so:
# ./sortmirrors.pl </etc/pacman.conf >pacman.conf.new
Make sure pacman.conf.new looks okay, then move it into place:
# mv pacman.conf.new /etc/pacman.conf
3. If you can spare a few bucks for a bandwidth-hungry distribution, please donate. We're looking to get another server, but the donations just aren't there yet.
Thanks a lot everyone. We'll keep on making it if you keep on running it.
Editor's Note: The perl script now comes with the new version of pacman, and the download cap has been lifted.
http://archlinux.org/news.php#47
This Week on the Forum:
Rasat made a request that we try to clean up the forums to make it easier to read. One idea was to put [RMx] for remove, [MVx] for move or [FXx] for fixed / resolved, in our topic titles. Read about it here: http://bbs.archlinux.org/viewtopic.php?t=2773
Neocephas wrote a CUPS HOW-TO for Arch. It's still being worked on, so go help out over here: http://bbs.archlinux.org/viewtopic.php?t=2757
This Week on the Mailing List:
After the announcement by Judd about the bandwidth troubles, people raised some ideas for saving bandwidth. One was randomizing mirrors automatically via pacman. Another was to take the Full 0.5 ISO off of the main ftp site, because it's out of date anyway. Neither of these have been implemented yet, but discussion is still ongoing. Read the threads at:
http://www.archlinux.org/pipermail/arch … 00844.html
http://www.archlinux.org/pipermail/arch … 00853.html
The idea was brought up again about multiple flavors of kernels. DanG also asked if multiple 2.4 or 2.6 kernel packages could "live together". The answer was no. A 2.4 and 2.6 kernel can co-exist though. Judd stated that there were no plans for different flavors of kernels (like "failsafe") and instead people should be building their own. Read the thread at: http://www.archlinux.org/pipermail/arch … 00872.html
What's happening with the TUR mailing list:
We're currently looking to assimilate a couple more Trusted Users to run their own repository and help test packages from /incoming. The system now requires sponsorship. So if you have submitted a lot of packages to /incoming that have made their way into the repositories, the TUs will be willing to sponsor you. If you are interested, join and mail the TUR Mailing list: https://lists.berlios.de/mailman/listinfo/tur-users
This Week in Dev Land (by Jason Chu):
Manolis Tzanidakis (zen_guerilla) left the development team. He has decided to work on his own personal distro and won't have time for both. We wish him well.
Tobias Kieslich (neri) and Damir Perisa (dp) have joined the dev team. They've started adopting packages and are still learning how everything works. They are a welcome addition to the dev team.
We now have a new package server. All the regular traffic goes to the original server but ftp.archlinux.org is now on a different one. This one has a higher limit on bandwidth but if you use a mirror please continue doing so.
We are debating if/how to add f77 and gcj compilers to gcc. The best suggestion is to have seperate packages that depend on gcc but gcc is a scary package so it's still being researched.
The gcc question brought up another: should we implement multiple packages per PKGBUILD file? It was agreed that a single package per PKGBUILD is much simpler and more like the arch way. At the same time, we still have some issues to work through because of this simple method.
There are new disk images on ftp.archlinux.org. They're more modular and have support for keymaps, xfs/jfs support, and raid/lvm. The setup scripts don't support all of these new features but the advanced user will be able to use them. There will also be an expert install script to support these features.
Xfree86 licensing is still being researched. An interesting link was found: http://www.xfree86.org/legal/licenses.html
An 0.6-base-BETA1 was released. It's available on the archlinux ftp server. Feel free to test it.
An Archlinux goal/philosophy/social contract document was mentioned. Judd said he was thinking about this already and we will probably start one soon.
A number of packages were found to be missing urls and md5sums. Namcap will detect this and spit out an error. If anyone finds PKGBUILDs with these missing, they can submit a bug report for that package.
Package Highlights:
Updated init scripts were released this week. They provide a more cleaned up boot process.
reiserfsprogs 3.6.13 was released this week. This fixes a bug that can prevent a ReiserFS partition from booting. Read about it here: http://bbs.archlinux.org/viewtopic.php?t=2614
GCC, Coreutils, the kernel and some other core packages were updated this week. These types of upgrades can be risky sometimes, so be sure to submit any problems you have to the Bug Tracker.
Pacman 2.7.4 was released this week. Changes are:
- fixed a bug in conflict handling, where installing a conflicting package would fail even if the new package "provided" it
- if pacman sees a newer version of itself during an upgrade, it will ask to install itself first, then be re-run to upgrade the other packages.
- You can now use the --info option with --sync to display an uninstalled package's dependency info.
- Added a sane umask before db writes
- buffer overflow fix (bug #442)
FAQ of The Week:
Well, it's the same as last week and the week before. Unfortunately, we're still receiving the same questions. If you are having an issue, please review the past Arch Newsletters because your questions are probably answered.
Bugs Opened this Week:
k3b-0.11.4-1 does not work: broken permissions
autologin feature of kdm and gdm does not work
IP Forwarding in iptables pkg
missing dependency for kdegraphics: fribidi, lcms
Mozilla 1.6 package has broken permissions
makewhatis creates /usr/local/man/whatis via crond
kpilot missing from KDEPIM
Bugs Closed this Week:
Program additions for 0.6 installer
kernel upgrades should remind users to run lilo.
Add nano to base
Choose other server in ftp install
Option to use non us keymaps in installer
Add jfs and xfs support
incorrect url in download list
groff-1.19 now available from German mirror
Please add the graphics mode to the links package!
New Vim 6.2 supports GTK2 but in arch it's still GTK1
InitScripts outdated Copyright
util-linux: missing
PARPORT_1284 support in Kernel
N_HDLC support in Kernel
Joystick support in kernel
Please update mozilla-thunderbird
Reiserfsprogs 3.6.12 can't mount partitions read only
Links (graphics mode)
pacman -Syu behaviour
PATCH: Fix buffer overflow in Pacman
pacman: set umask before writing db entries
libgnomeui needs depends on gnome-common
Nautilus crashes immediately
gqmpeg fails to build
gdesklets missing dependencies
transcode fails to build
mp32ogg fails to build
mjpegtools fails to build
mplayer-plugin fails to build
transcode pkg needs to be updated.
missing dep in transcode pkg. again
avifile dies with failure to build codecs
flashplugin be updated to reflect firebird's new name change
small firefox changes
gtk2 deps screwed up between package and db
gtk2 won't build
New and Updated Packages:
k3b 0.11.5-1
barcode 0.98-1
glabels 1.93.3-1
avidemux 2.0.20-2
lcab 1.0b10-1
hd2u 0.9.0-1
maxima 5.9.0-1
kobodeluxe 0.4pre8-1
pth 2.0.0-1
kile 1.6.1-1
atitvout 0.4-1
kdevelop 3.0.1-2
kdevelop-devel 3.1.0.snap040221-1
fbi 1.30-1
pipemeter 0.9.1-1
zile 1.7b3-1
gqview 1.4.0-1
ringtonetools 2.19-1
dbp 0.5.7-1
dcraw 1.170-1
scite 1.59-1
device-mapper 1.00.07-2
coreutils 5.2.0-1
lzop 1.01-3
gtick 0.3.0-2
ogle 0.9.2-3
rar 3.3.0-1
ascii 3.6-1
whois 4.6.9-1
rasmol 2.7.2.1-1
stellarium 0.5.2-1
xfsprogs 2.5.6-1
xmame-sdl 0.78.1-1
lesstif 0.93.94-1
kaffeine 0.4.1-1
alsa-driver 1.0.2-2
xpdf 3.00-1
initscripts 0.6-6
cssed pre0.1.4-1
gdsl 1.0-2
glut 3.7-2
rsync 2.6.0-2
vim 6.2-2
links 2.1pre14-2
gvim 6.2-2
kernel24 2.4.25-1
kernel24-scsi 2.4.25-1
uml_utilities 20040114-1
words 1.0-2
user-mode-linux 2.4.24-1
kernel26-scsi 2.6.3-1
kernel26 2.6.3-1
reiserfsprogs 3.6.13-1
pacman 2.7.4-1
bzflag 1.10.4.20040125-1
galeon 1.3.13-1
toppler 1.0.6-1
gcc 3.3.3-1
wine 20040213-1
mozilla-firefox 0.8-2
less 382-1
iptables 1.2.9-4
flashplugin 6.0r79-8
cdrdao 1.1.8-1
bash-completion 20040214-1
zope 2.7.0-1
liquidwar 5.6.2-1
pcmcia-cs 3.2.7-1
fortune-mod-buffy 1-1
pango 1.2.5-4
gtk2 2.2.4-3
Packages in Staging:
To gain access to this repository and the TURs, visit http://tur.berlios.de
filelight 0.6.4-1
j 0.20.2-s1
inkscape 0.37-s1
abcm2ps 3.7.3-1
potracegui 0.3-s1
amule 1.2.4-1
conglomerate 0.7.12-s1
bogofilter 0.17.0-s1
fftw3 3.0.1-1
gp 0.26-1
fox 1.0.50-s1
ion-devel 20030814-1
libcapsinetwork 0.2.5-s1
labplot 1.2.2-1
hwd 1.6-2
libchipcard 0.9.1-1
libksba 0.4.7-1
liblrdf 0.3.1-1
libofx 0.6.4-1
gtk-smooth-engine 0.5.6-s1
cups-pdf 1.4.1-s1
mage 6.30.030607-1
mpck 0.04-1
pixieplus 0.5.4-1
potrace 1.2-1
prekin 6.30.030607-1
tla 1.1-2
lwm 1.2.0-1
tse3 0.2.7-1
windowlab 1.21-s1
universalkopete 0.1-1
xv 3.10a-1
smclone 0.55-1
kbarcode 1.6.1-s1
bs 2.6-1
mp3burn 0.3.1-1
shaaft 0.5.0-1
monopd 0.9.0-s1
giftui 0.3.1-1
gurlchecker 0.6.4-s1
mono 0.30.1-1
ttf-cheapskate 0.1-2
libxml++ 1.0.0-1
chromium 0.9.12-1
passepartout 0.4-1
gnome-cups-manager 0.17-1
libgnomecups 0.1.6-1
gweled 0.4-s1
logjam 4.2.4-2
xpenguins_themes 1.0-1
libmath++ 0.0.3-1
texmacs 1.0.3.3-s1
gtkatlantic 0.3.2-1
ms-sys 1.1.1-1
rox-system 1.9.0-s1
smb4k 0.3.1-1
ksambaplugin 0.5b2-1
materm 0.1-1
seahorse 0.7.3-2
parse-yapp 1.05-1
libavc1394 0.4.1-1
xml-dom 1.43-1
xml-regexp 0.03-1
xml-xql 0.68-1
prizm 0.2-s1
liferea 0.4.6b-1
panorama-tools 2.6bml12-s1
sylpheed-gtk2 0.9.8a-s1
amarok 0.8.3-s1
mlview 0.6.2-1
gossip 0.7.1-s1
loudmouth 0.15.1-s1
gtodo 0.13.5-s1
gnome-alsamixer 0.9.6-s1
scribus-devel 1.1.5-s1
source-highlight 1.8-s1
xtermset 0.5.1-s1
cdw 0.2.2-s1
gnotime 2.1.7-s1
xfce4-fsguard-plugin 0.2.0-s1
xfce4-xmms-plugin 0.1.1-s1
gksu 1.0.3-s1
pychecker 0.8.13-s1
gtkglext 1.0.5-s1
gtkglextmm 1.0.1-s1
pygtkglext 1.0.1-s1
spe 0.4.2c-s1
gmencoder 0.1.0-s1
scribus-docs 1.0.1-s1
ArchStats Update:
To participate, visit http://archstats.coding-zone.com
Number of registered systems: 135
Date first system was registered: 20031017 [22:24:30]
Most recent update occurred: 20040222 [16:30:46]
Longest recorded uptime: 166 days, 22 hours, 51 minutes, 56 seconds.
Average uptime: 8 days, 2 hours, 14 minutes, 33 seconds.
Lowest kernel version: 2.4.21
Highest kernel version: 2.6.3-mm2
Least packages installed on a system: 44
Average installed packages: 255
Most packages installed on a system: 563
That's it for this week, stay tuned for more exciting news.
-- Jason Chu and Ben MazerKeep up the good work!
-
Kernel_task using 16 Million Terabytes of memory - what's causing this?
According to Activity Monitor kernel_task is using 16,777,216 TB of virtual memory.
Obviously, there is some (overflow?) bug somewhere (is that number the maximum allowable memory on a 64-bit system?), and Virtual Memory doesn't really mean much in the real world, but is there a way to track down what is causing this, given that kernel_task going wrong can bring down the whole system.
This reappears after a restart - and in fact for the first time in memory my computer locked up a few days ago, so I'm wondering if there is a connection.
ThanksYou have a Mid 2010 15-inch MacBook Pro and your log is showing a problem with the NVIDIA graphic card.
These MacBooks have got a defective NVIDIA graphic card that may cause unexpected restarts, black screens, kernel panics... See > http://support.apple.com/kb/TS4088
Fortunately, Apple is repairing all of those MacBooks for free, so make a backup of your data and take your MacBook Pro to an Apple Store or reseller
Maybe you are looking for
-
PS CC - I have a B/W background which I flattened - Then, change to RGB and add an RGB color graphic - print out makes background a greenish color - Can't figure out how to adjust. Thanks, Dave
-
Shuffle with XP Home Edition vs XP Pro
I have a Dell 8100 Desktop running XP Home Edition SP2. I am getting -50 error and can't use the shuffle. Have a ticket open with Apple Support and they want me to try it on another PC and report back to them. I installed iTunes on my IBM T43 XP Prof
-
Hi every body , My requirment is : i have a tabstrip and the no.of tabs in the tabstrip is dynamic. So in each tab i am creating table ui element and binding the single context only in each tab .I am using the same node on each tab . Now my requirmen
-
Viewing old itunes library via external hard drive
'ello, I have an external hard drive that I had linked up to my old iMac, which has now died. I bought a new Macbook pro and want to plug in my old external into my wireless router and pick up my old itunes library. I have plugged in the external ha
-
I have a decent couple of routers at home.. but for some reason my phone is sticking with LTE. I'm glad to connect to LTE when I'm away from home and not using wifi..but if I have a good wifi connection.. I want to direct iPhone 5 to stick with wifi