OpenSUSE :: GDB Breakpoint Fails - Debug Info For Libc Not Available
Aug 4, 2010
When I debug a program with GDB and hit a break point I get:
Missing separate debuginfo for /lib64/ld-linux-x86-64.so.2
Try: zypper install -C "debuginfo(build-id)=2b25d8329d1aa713a2cc61b5cb701f6f89c78022"
Missing separate debuginfo for /lib64/libm.so.6
Try: zypper install -C "debuginfo(build-id)=f05646449077dfbf15d5f78251a3b25e0f955dd9"
Missing separate debuginfo for /lib64/libc.so.6
Try: zypper install -C "debuginfo(build-id)=9b674b2caef1982db55e436bca814060e1705b7f"
When I install the suggested packages the debuginfo does not match the object files. I get:
warning: the debug information found in "/usr/lib/debug//lib64/libc-2.11.2.so.debug" does not match "/lib64/libc.so.6" (CRC mismatch).
warning: the debug information found in "/usr/lib/debug/lib64/libc-2.11.2.so.debug" does not match "/lib64/libc.so.6" (CRC mismatch).
Missing separate debuginfo for /lib64/libc.so.6
Try: zypper install -C "debuginfo(build-id)=9b674b2caef1982db55e436bca814060e1705b7f"
The last package is not found. Is this missing from the repositories?
I am a hobbyist interested in development and have recently decided to migrate from Windows to Linux and after trying out a number of distributions decided to use openSUSE. After getting somewhat more comfortable with Linux and openSUSE I decided to try and get comfortable with development on this platform.
And hence my problem: After compiling the simplest of programs (hello world) and trying to step through it with gdb I get this output: Missing separate debuginfo for /lib64/ld-linux-x86-64.so.2 Try: zypper install -C "debuginfo(buildid)=2b25d8329d1aa713a2cc61b5cb701f6f89c78022" Missing separate debuginfo for /lib64/libc.so.6 Try: zypper install -C "debuginfo(buildid)=9b674b2caef1982db55e436bca814060e1705b7f"
As soon as I type the command start in gdb. Naturally I tried the recommended zypper downloads to no avail. Furthermore I sought out the glibc debuginfo in the software manager and found nothing which I have not already installed.
On my system are:-devel-32bit glibc glibc-32bit glibc-devel glibc glibc-info glibc-locale glibc-locale-32bit glibc-profile glibc-profile-32bit gdb gdbm gdbm-32bit
All of these are the latest version from the standard 11.3 repositories. I believe these are all the requirements I should need to not have gdb give me errors but sadly that is not the case. Lastly I am running on a 64bit version of openSUSE 11.3.
root@yes-desktop:/usr/src/eglibc-2.11.1/malloc# uname -a Linux yes-desktop 2.6.32-21-generic #32-Ubuntu SMP Fri Apr 16 08:10:02 UTC 2010 i686 GNU/Linux 2 gcc version root@yes-desktop:/usr/src/eglibc-2.11.1/malloc# gcc -v Using built-in specs.
[code]....
1 why cann't i debug the test? because i can't step into the "operator new" function.
I am developing an application whose executable is generated inside a certain folder hierarchy (say: /DevPath/MyProject/bin). My source code is located in a different branch of this hierarchy (say: /DevPath/MyProject/src). When my app crashes, its core files are stored in /DevPath/MyProject. I'm developing the app on a pc, but running it on a separate platform in which i can only execute it. Folder hierarchy is the same as above on both computers. Usually, when a new executable version is ready, we update both the executable and the source code on the target platform, transferring all the new /DevPath/MyProject folder on it. But sometimes it can really be a bother, so we update only the executable.
1)In the case we only update the executable, keeping an old source code version, and the app generates a core file, can i trust the backtrace produced by gdb in this case? I.e., does gdb need the latest source code files or it just needs the debugging information?
2) (More radical question ) Do i really need to keep the source code on the target platform for core dump analysis or i just need the executable?
I ran GDB on a program and am receiving following errors, Code: anisha@linux-uitj:~/junk> g++ -g jk.cpp anisha@linux-uitj:~/junk> gdb a.out GNU gdb (GDB) SUSE (6.8.91.20090930-2.4) Copyright (C) 2009 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later [URL] This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "x86_64-suse-linux". For bug reporting instructions, please see: [URL] ... Reading symbols from /home/anisha/junk/a.out...done. (gdb) b readline Breakpoint 1 at 0x400b90: file jk.cpp, line 19. (gdb) r Starting program: /home/anisha/junk/a.out Missing separate debuginfo for /lib64/ld-linux-x86-64.so.2 Try: zypper install -C "debuginfo(build-id)=591af1afa33f255704fb6a60859b93d00e205302" Missing separate debuginfo for /usr/lib64/libstdc++.so.6 Try: zypper install -C "debuginfo(build-id)=62220ad5c8941afb5d332c0c47d32f8beec8ac50" Missing separate debuginfo for /lib64/libm.so.6 Try: zypper install -C "debuginfo(build-id)=57fc1891d8d9f419fb8c7fc06a8285563b53a47e" Missing separate debuginfo for /lib64/libgcc_s.so.1 Try: zypper install -C "debuginfo(build-id)=0206e11fa8ca0db0633073adcbf1349a7871e1dc" Missing separate debuginfo for /lib64/libc.so.6 Try: zypper install -C "debuginfo(build-id)=c5a3dfd66bf61fcdec9bc22153b2fbd0d6697960" can't open input file (null) Program exited with code 01. (gdb)
Googled this and found the same problem mentioned in the yum site: [root@plouton ~]# yum update kernel Loaded plugins: refresh-packagekit Setting up Update Process Resolving Dependencies --> Running transaction check ---> Package kernel.x86_64 0:2.6.33.6-147.fc13 set to be installed --> Finished Dependency Resolution --> Running transaction check ---> Package kernel.x86_64 0:2.6.33.4-95.fc13 set to be erased --> Finished Dependency Resolution Dependencies Resolved
I'm totally new to this linux thing and I have a very strange problem. I downloaded the .iso and burned it to CD as per instructions. I had intended to install alongside existing WinXP Pro Sp3. Everything seemed to be going fine until it the installer got to the "User Info" Screen where it asks for my name, the computer's name, password, etc. At that point nothing I did would allow the "forward" button to work. I backed up a step or two and went forward again to no avail. I ended up hitting the hard reset button to get out of it. I never saw any error messages even though I scrolled through the install log at the bottom of the install window. Now my HDD is 60 Gb smaller (The part intended for Ubuntu) under Win XP and I have no idea what to do about it.
My system:
Asus P5NE-SLI Mobo Intel Core2 Duo 8400 3.0 Ghz 2 Gb ram at 800Mhz Nvidia 9800GT GPU 512Mb
Win XP seems to be working normally except for the loss of HDD space.
How do I get debug symbols for glibc? I'm using oprofile when profiling applications. Unfortunately, the default glibc version that comes with SUSE 11.1 has the debug symbols stripped-out. That means that all libc functions get aggregated into one big blob under "glibc.so" and you can't tell whether you spend most of the time in memory allocation, syslog or lock related routines.
So, I've tried a myriad of different ways of getting the symbols. From trying to build glibc myself to hunting for dozens of different rpms on this site and on the Internet. So far - no positive results.
I followed the instructions at OpenSUSE: Bugreport KDE - openSUSE to install debug symbols for okular which seems to crash quite often for me. However gdb and the kde crash application are unable to find them. The kde app says its not a good crash report and I should follow the guide.. which has already been done.
I installed kdegraphics4-debuginfo kdebase4-runtime-debuginfo kdebase4-workspace-debuginfo kdelibs4-debuginfo libqt4-debuginfo
Is there a step I'm missing to get gdb to load the symbols or something?
I would like to build a sticky subject about repairing openSUSE boot.This article,once complete (not by me alone) should be made sticky (to have it in front of the forum)- these problems are frequent and disturbing.I know the great work done by please-try-again. But very often we don't need any script (and running a script from rescue mode is difficult) - that said I vote to have the pta scripts included in the rescue system !So I would like to focus on command line fixes.
Recently, I upgraded 11.1 to 11.2. Everything seemed to go fine. However, for some reason the system boots with the debug kernel - 2.6.31.5-0.1-debug. According to Yast every kernel under the sun seems to be loaded on the system. Grub, however, gives two choices - one for the regular and one for the failsafe system. They both boot to the debug kernel.
How do I get this thing to load into the regular, as opposed to the debug, kernel?
so i cannot install anything because update-info-dir file is missing from /var/lib/dpkg/info/ .. I've searched for the last day and a half for a way to fix this, but nothing. can't even update dpkg because of this. so how do I bypass or fix this so I can install stuff (this is a fresh install of ubuntu 10.04 lts Lucid Lynx).
Now I've got debug, default, desktop, ec2, trace, vanilla, xen installed im my system. I usually only boot with desktop, and I was wondering if it is OK to remove other stuff except debug, default, and desktop.
I am looking for explanations on how Opensuse utilizes the root directories. Like what decides if a program goes in /bin vs /sbin vs /usr/bin etc...In other words, a breakdown of the directory structure and how each part is utilized. I am trying to get more familiar with it so it will be easier for me to locate things and I will know where to look for X part of a program that does Y.
I've installed quantum GIS on opensuse 11.3 gnome, so far so good. I run the app and it opens fine. To get the app to use data on the HD you get a dialogue box with a browse button. It should just be a case of clicking browse and selecting the required data file. Unfortunately when I click on browse the app crashes. I've no idea how to get info about the crash in an attempt to solve the problem. Had a similar problem with gimp in suse 11 kde could only fix it by updating the os to 11.3.
When I say info, I get an empty document: nfo: (dir)Top, 1 linii --AllWhen I say info -f /usr/share/info/dir, I get the same.When I open /usr/share/info/dir in KWrite, I notice that there are tabulations in the Node: declaration.When I removed these tabulations, the directory suddenly displayed all right (info -f /tmp/dir).So I tried the same thing with the file extracted from the current package, and it contains tabulation characters, so I can see nothing there with info.
I am considering an OCZ RevoDrive SSD that takes the form of a PCI-Express x4 card. I have not been able to find compatibility or support information for Linux, though apparently it works OK under Windows 7. I want to make this a system / boot drive.
having repartitioned a disk yesterday but uwisely not obliterating a DOS OEM partition, came down to find the machine had woken up & booted the wrong disk (one I partitioned but without ne OS); with apparently dead (unwilling to seek main drive).
Trouble is the OEM diagnostics was old DOS and likes to **** on other disk partition tables and it seems to have got run, from what was the OEM Windows disk. Now I've fixed that, it seems to be screwed partition table issue.
No problem, it happened before and I recovered with gpart(8) & a spot of tweaking, but due to "redevelopment" I've now got my partition table backups on the disk that's lost. The old backup of them, got tidied. gpart's for once only finding the partitions I don't care about eg) swap & a Windows 7 NTFS backup partition.
Have some logfiles from 11.1 days, which may if it gets recorded somewhere by YaST or other in /var/log actually have the table output. Any one happen to know, where to look as a time saver?
I'd like to get these system partitions back, as I put in bugs on 12.1 M2, as well as an 11.4 and older 11.2 install for comparison. Wasn't planning on re-installing, when I get asked to try something later on.
how I'm using a computer then look up Smartnav by NaturalPoint for more info on guiding the cursor using head movements. Windows and Mac only. They have no plans for Linux, I already asked them. Linux now has a similar program which uses the webcam and its called Eviacam. It was developed in Catalonia and is free of charge. Its available along with the source code at Sourceforge.net. I have it working on Ubuntu 9.10 installed along with Windows 7, install it to Open Suse which I find to be a beautiful and slick distro. Ive played around with it in Vbox and I find it to be clean compared with Ubuntu which is quite buggy. this Quad and all disabled people to get onto OpenSUSE? There is very little in the way of these types of assistive technologies for Linux because of the costs.
I couldn't find any package containing a static version of the C library for 32-bit. Did I miss something or is it really not available, and if so is there a reason why?
Im using buildroot to generate a file system . in tat one package module-init tools need static version of libc.a . so may i know how to install tat in fedora 12 ?mine CPU is as given below
Linux Nakshatra 2.6.31.5-127.fc12.i686.PAE #1 SMP Sat Nov 7 21:25:57 EST 2009 i686 i686 i386 GNU/Linux
I'm on Ubuntu 11.04. I get the above error when running a utility to get code from a version control system (StarTeam). The error looks generic enough that I'm hoping its an easily resolvable system problem. Below is how I get it ...
Code:
$ /bin/sh -c stcmd label -x -nologo -stop -p username:password@mydomain.com:9999/NNA/NNA/Technology/nna/tools/selenium -nl selenium-utils-1.0 -b strings: '/lib/libc.so.6': No such file
I've been working with out servers for the past several years. We have a mix of OS. Fedora, Ubuntu, RHEL, etc. I've accidentally trashed the libc.so.6 link and now nothing but shell commands work.
I tried to echo "/lib/libc-2.10.2.so into the /etc/sestatus.conf file, but that didn't have any affect. Does anyone know where the search path is maintained besides PATH so I can try to point the apps to the /lib/libc-2.10.2.so file? I have a bunch of SVN repos on this server and right now, no one has access.
