Your message dated Tue, 6 Jan 2009 17:16:35 +0200 with message-id <[email protected]> and subject line Re: Bug#80360: seemingly reproducable dpkg segfault has caused the Debian Bug report #80360, regarding [SEGFAULT] seemingly reproducible dpkg segfault to be marked as done.
This means that you claim that the problem has been dealt with. If this is not the case it is now your responsibility to reopen the Bug report if necessary, and/or fix the problem forthwith. (NB: If you are a system administrator and have no idea what this message is talking about, this may indicate a serious mail system misconfiguration somewhere. Please contact [email protected] immediately.) -- 80360: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=80360 Debian Bug Tracking System Contact [email protected] with problems
--- Begin Message ---package: apt version: 0.3.19 On Fri, 22 Dec 2000, Sean 'Shaleh' Perry wrote: > Steps: > > grab base.tgz from potato > setup resolv.conf and sources.list for potato > chroot in > apt-get install dpkg-dev build-essential (due to a bug in apt, you have to > isntall dpkg-dev first) > apt-get install strace (you won't get the chance later) > apt-get install libncurses5-dev libreadline4-dev libssl09-dev libzvt-dev > ... > ... > segfault > > dpkg appears to die during the dpkg --unpack of the 30 some packages at once. > I could not figure out how to get strace to give me the entire command line so > I could try to run dpkg by hand. > > This dies for me every time I try it. Here's a little more info for the bug: If you add -o Debug::pkgDPkgPM=true, and then run the displayed dpkg commands by hand, it works fine, no segfault. It's only when run under apt that dpkg has problems. <doogie> #0 0x805c070 in strcpy () <doogie> #1 0xbfffea48 in ?? () <doogie> #2 0x80502cb in strcpy () <doogie> #3 0x8051c49 in strcpy () <doogie> #4 0x8050f68 in strcpy () <doogie> #5 0x804a752 in strcpy () <doogie> #6 0x4002da52 in __libc_start_main () from /lib/libc.so.6 <BenC> aha <BenC> argv is bad, I'm sure <BenC> I bet there's a string that apt isn't writing a NULL byte at the end of one of the argv params And the following patch to apt works around the problem. -- Index: apt-pkg/deb/dpkgpm.cc =================================================================== RCS file: /cvs/deity/apt/apt-pkg/deb/dpkgpm.cc,v retrieving revision 1.17.2.3 diff -u -r1.17.2.3 dpkgpm.cc --- apt-pkg/deb/dpkgpm.cc 2000/11/30 08:39:22 1.17.2.3 +++ apt-pkg/deb/dpkgpm.cc 2000/12/23 01:46:56 @@ -408,6 +408,8 @@ { if (I->File[0] != '/') return _error->Error("Internal Error, Pathname to install is not absolute '%s'",I->File.c_str()); + if (Size + strlen(I->File.c_str()) > 1024) + break; Args[n++] = I->File.c_str(); Size += strlen(Args[n-1]); } @@ -416,6 +418,8 @@ { for (;I != J && Size < 1024; I++) { + if (Size + strlen(I->File.c_str()) > 1024) + break; Args[n++] = I->Pkg.Name(); Size += strlen(Args[n-1]); } -- This may be a bug in libc6 2.1.3-13 that came with potato, but I don't feel like debugging that. ----BEGIN GEEK CODE BLOCK---- Version: 3.12 GCS d- s: a-- c+++ UL++++ P+ L++++ !E W+ M o+ K- W--- !O M- !V PS-- PE++ Y+ PGP++ t* 5++ X+ tv b+ D++ G e h*! !r z? -----END GEEK CODE BLOCK----- ----BEGIN PGP INFO---- Adam Heath <[email protected]> Finger Print | KeyID 67 01 42 93 CA 37 FB 1E 63 C9 80 1D 08 CF 84 0A | DE656B05 PGP AD46 C888 F587 F8A3 A6DA 3261 8A2C 7DC2 8BD4 A489 | 8BD4A489 GPG -----END PGP INFO-----
--- End Message ---
--- Begin Message ---Hi! On Fri, 2000-12-22 at 19:48:24 -0600, Adam Heath wrote: > package: apt > version: 0.3.19 > > On Fri, 22 Dec 2000, Sean 'Shaleh' Perry wrote: > > > Steps: > > > > grab base.tgz from potato > > setup resolv.conf and sources.list for potato > > chroot in > > apt-get install dpkg-dev build-essential (due to a bug in apt, you have to > > isntall dpkg-dev first) > > apt-get install strace (you won't get the chance later) > > apt-get install libncurses5-dev libreadline4-dev libssl09-dev libzvt-dev > > ... > > ... > > segfault > > > > dpkg appears to die during the dpkg --unpack of the 30 some packages at > > once. > > I could not figure out how to get strace to give me the entire command line > > so > > I could try to run dpkg by hand. > > > > This dies for me every time I try it. > > Here's a little more info for the bug: > > If you add -o Debug::pkgDPkgPM=true, and then run the displayed dpkg commands > by hand, it works fine, no segfault. It's only when run under apt that dpkg > has problems. > > <doogie> #0 0x805c070 in strcpy () > <doogie> #1 0xbfffea48 in ?? () > <doogie> #2 0x80502cb in strcpy () > <doogie> #3 0x8051c49 in strcpy () > <doogie> #4 0x8050f68 in strcpy () > <doogie> #5 0x804a752 in strcpy () > <doogie> #6 0x4002da52 in __libc_start_main () from /lib/libc.so.6 > <BenC> aha > <BenC> argv is bad, I'm sure > <BenC> I bet there's a string that apt isn't writing a NULL byte at the end > of one of the argv params > > And the following patch to apt works around the problem. > > -- > Index: apt-pkg/deb/dpkgpm.cc > =================================================================== > RCS file: /cvs/deity/apt/apt-pkg/deb/dpkgpm.cc,v > retrieving revision 1.17.2.3 > diff -u -r1.17.2.3 dpkgpm.cc > --- apt-pkg/deb/dpkgpm.cc 2000/11/30 08:39:22 1.17.2.3 > +++ apt-pkg/deb/dpkgpm.cc 2000/12/23 01:46:56 > @@ -408,6 +408,8 @@ > { > if (I->File[0] != '/') > return _error->Error("Internal Error, Pathname to install is not > absolute '%s'",I->File.c_str()); > + if (Size + strlen(I->File.c_str()) > 1024) > + break; > Args[n++] = I->File.c_str(); > Size += strlen(Args[n-1]); > } > @@ -416,6 +418,8 @@ > { > for (;I != J && Size < 1024; I++) > { > + if (Size + strlen(I->File.c_str()) > 1024) > + break; > Args[n++] = I->Pkg.Name(); > Size += strlen(Args[n-1]); > } > -- > > This may be a bug in libc6 2.1.3-13 that came with potato, but I don't feel > like debugging that. Given that this bug is around 8 years old, the backtrace is not that useful, that I cannot see anything related in recent versions of dpkg and that similar scenarios do not produce a segfault, I'm just going to assume this got fixed somewhere and close this bug report. regards, guillem
--- End Message ---
