Dude, I thought I'd run it in the debugger to see if I could determine where the breakage was occurring... First I emerged it and validated that I too was getting the segmentation fault.
So I cd to /usr/portage/app-misc/colortail and issue "ebuild colortail-0.3.0-r3.ebuild unpack" to extract the contents to /var/portage/colortail-0.3.0-r3/work/colortail-0.3.0. Go to that directory and issue 'CFLAGS="-g" ./configure' followed by 'CFLAGS="-g" make'. Follow so far? Start up the colortail in gdb "gdb colortail", set the args as you have done in the script and then type 'run' - no break. Hmm, I thinks, so I decide to run it outside of the debugger... Run this version directly and, to my surprise, no crash. So it works for your simple test case; I'm assuming you have a more complex file that the crash occurred in and slimmed it down to this repeatable case. We'd have to try it on that case to see if it makes a difference. So for some reason compiling colortail with the -g flag will make it work through your example. You should be able to do the same thing to get it working in your environment. > -----Original Message----- > From: W. B. Maguire II [mailto:[EMAIL PROTECTED] > Sent: Friday, January 07, 2005 1:50 PM > To: Gentoo-User list > Subject: [gentoo-user] colortail: reproducable bug > > Hello all: > > I'm having some problems with 'colortail'. I tried to send all > this information to the package author, but it appears that > he/she no longer exists! ;-) > > Is anyone else using 'colortail'? It sure would be nice to not > have to reinvent the wheel... > > Anyway, since I went to the trouble to *fully* document---and > include driver scripts to demonstrate the problem; I hate to not > send it to *anyone*! ;-) So, since it is small (1140 bytes), I'm > just attaching the gzipped tarball (I'm hoping that this size is > small enough that no one minds!?). If anyone is interested, or > knows how to get this info to the package author/maintainer, > please have a look at the tarball (it is only four files and > un-tar-zips into its own directory "colortail_bug_example"). > > This *seems* like an app that would be commonly-desired; is there > an alternative that I didn't find? > > Sheesh! I can't believe how much time I spent to narrow-down the > problem to the simplest possible case, only to have no one to > look at it! >:-/ If anyone looks at it, enjoy my folly! > > Bruce. > > _________________________________________________________________ > > Original message: > _________________________________________________________________ > > [...deleted comments to the original author: "Joakim Andersson"...] > > $ uname -a > > Linux tron 2.4.26-gentoo-r9 #1 \ > Wed Sep 1 11:46:40 MDT 2004 \ > i686 Intel(R) Pentium(R) 4 CPU 1.70GHz \ > GenuineIntel GNU/Linux > > The version of colortail I'm using is: > > $ colortail --version > > Colortail version <0.3.0> > Copyright (C) 1999 Joakim Andersson <[EMAIL PROTECTED]> > Colortail comes with ABSOLUTELY NO WARRANTY; for details see the > COPYING file. > This is free software, and you are welcome to redistribute it > under certain conditions; see the COPYING file for details. > > I'm attaching a gzip'ed tarball of a few files that will allow > you to (hopefully) easily recreate what I'm seeing (free pointer > errors and segmentation fault errors): > > $ tar -tvzf colortail_bug_example.tar.gz > > drwxrwsr-x ... colortail_bug_example/ > -rwxrwxr-x ... colortail_bug_example/colortailbug_text_source.sh > -rwxrwxr-x ... colortail_bug_example/colortailbug_RUN_ME > -rw-rw-r-- ... colortail_bug_example/colortailbug_conf_file.txt > -rw-rw-r-- ... colortail_bug_example/colortailbug_output.txt > > You should be able to recreate the bug by just running the > colortailbug_RUN_ME shell script: > > $ tar -xvzf colortail_bug_example.tar.gz > $ cd colortail_bug_example > $ ./colortailbug_RUN_ME > > ==> colortailbug_output.txt <== > ==> colortailbug_output.txt <== > Simple line 1 > > Simple line 2 > > Simple line 3 > > ./colortailbug_RUN_ME: line 23: 3674 Segmentation fault \ > colortail -k colortailbug_conf_file.txt --follow > colortailbug_output.txt > > > Am I doing something wrong? I just don't see anything... > > Please let me know if you get this e-mail, and if you are able to > recreate the "free()"/"segmentation fault" errors with my > scripts. > > Thanks! > Bruce. > > p.s. Also, I wondered why I always got a double-spaced output > for any text which does not specifically match any of the > RegEx's in the config file? Second, it appears that each > output line has an "invisible" trailing character (the > newline?) which will cause any RegEx in the config file to > *not* match unless the RegEx ends with a "." or ".*" to > capture that character. I've tried using a set: "[\n\r]", > but that doesn't do it. What is this? Thanks again! > > -- > _________ [EMAIL PROTECTED] ________ |**=== > / |**=== > ( W. B. Maguire II, Ph.D. Tel: 303.772.1615 |===== > | Chief Executive Manager FAX: 303.651.6389 9|| > | Analytic Investments LLC _ | > | 700 Ken Pratt Blvd STE204 PMB166 ( ) The ASCII ribbon | > ( Longmont CO 80501-6455 X campaign against ) > \__________________________________ / \ _ HTML e-mail __/ -- gentoo-user@gentoo.org mailing list
