Search Results: "dannf"

18 June 2025

Sergio Durigan Junior: GCC, glibc, stack unwinding and relocations A war story

I ve been meaning to write a post about this bug for a while, so here it is (before I forget the details!). First, I d like to thank a few people: I ll probably forget some details because it s been more than a week (and life at $DAYJOB moves fast), but we ll see.

The background story Wolfi OS takes security seriously, and one of the things we have is a package which sets the hardening compiler flags for C/C++ according to the best practices recommended by OpenSSF. At the time of this writing, these flags are (in GCC s spec file parlance):
*self_spec:
+ % !O:% !O1:% !O2:% !O3:% !O0:% !Os:% !0fast:% !0g:% !0z:-O2  -fhardened -Wno-error=hardened -Wno-hardened % !fdelete-null-pointer-checks:-fno-delete-null-pointer-checks  -fno-strict-overflow -fno-strict-aliasing % !fomit-frame-pointer:-fno-omit-frame-pointer  -mno-omit-leaf-frame-pointer
*link:
+ --as-needed -O1 --sort-common -z noexecstack -z relro -z now
The important part for our bug is the usage of -z now and -fno-strict-aliasing. As I was saying, these flags are set for almost every build, but sometimes things don t work as they should and we need to disable them. Unfortunately, one of these problematic cases has been glibc. There was an attempt to enable hardening while building glibc, but that introduced a strange breakage to several of our packages and had to be reverted. Things stayed pretty much the same until a few weeks ago, when I started working on one of my roadmap items: figure out why hardening glibc wasn t working, and get it to work as much as possible.

Reproducing the bug I started off by trying to reproduce the problem. It s important to mention this because I often see young engineers forgetting to check if the problem is even valid anymore. I don t blame them; the anxiety to get the bug fixed can be really blinding. Fortunately, I already had one simple test to trigger the failure. All I had to do was install the py3-matplotlib package and then invoke:
$ python3 -c 'import matplotlib'
This would result in an abortion with a coredump. I followed the steps above, and readily saw the problem manifesting again. OK, first step is done; I wasn t getting out easily from this one.

Initial debug The next step is to actually try to debug the failure. In an ideal world you get lucky and are able to spot what s wrong after just a few minutes. Or even better: you also can devise a patch to fix the bug and contribute it to upstream. I installed GDB, and then ran the py3-matplotlib command inside it. When the abortion happened, I issued a backtrace command inside GDB to see where exactly things had gone wrong. I got a stack trace similar to the following:
#0  0x00007c43afe9972c in __pthread_kill_implementation () from /lib/libc.so.6
#1  0x00007c43afe3d8be in raise () from /lib/libc.so.6
#2  0x00007c43afe2531f in abort () from /lib/libc.so.6
#3  0x00007c43af84f79d in uw_init_context_1[cold] () from /usr/lib/libgcc_s.so.1
#4  0x00007c43af86d4d8 in _Unwind_RaiseException () from /usr/lib/libgcc_s.so.1
#5  0x00007c43acac9014 in __cxxabiv1::__cxa_throw (obj=0x5b7d7f52fab0, tinfo=0x7c429b6fd218 <typeinfo for pybind11::attribute_error>, dest=0x7c429b5f7f70 <pybind11::reference_cast_error::~reference_cast_error() [clone .lto_priv.0]>)
    at ../../../../libstdc++-v3/libsupc++/eh_throw.cc:93
#6  0x00007c429b5ec3a7 in ft2font__getattr__(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) [clone .lto_priv.0] [clone .cold] () from /usr/lib/python3.13/site-packages/matplotlib/ft2font.cpython-313-x86_64-linux-gnu.so
#7  0x00007c429b62f086 in pybind11::cpp_function::initialize<pybind11::object (*&)(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >), pybind11::object, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, pybind11::name, pybind11::scope, pybind11::sibling>(pybind11::object (*&)(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >), pybind11::object (*)(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >), pybind11::name const&, pybind11::scope const&, pybind11::sibling const&):: lambda(pybind11::detail::function_call&)#1 ::_FUN(pybind11::detail::function_call&) [clone .lto_priv.0] ()
   from /usr/lib/python3.13/site-packages/matplotlib/ft2font.cpython-313-x86_64-linux-gnu.so
#8  0x00007c429b603886 in pybind11::cpp_function::dispatcher(_object*, _object*, _object*) () from /usr/lib/python3.13/site-packages/matplotlib/ft2font.cpython-313-x86_64-linux-gnu.so
...
Huh. Initially this didn t provide me with much information. There was something strange seeing the abort function being called right after _Unwind_RaiseException, but at the time I didn t pay much attention to it. OK, time to expand our horizons a little. Remember when I said that several of our packages would crash with a hardened glibc? I decided to look for another problematic package so that I could make it crash and get its stack trace. My thinking here is that maybe if I can compare both traces, something will come up. I happened to find an old discussion where Dann Frazier mentioned that Emacs was also crashing for him. He and I share the Emacs passion, and I totally agreed with him when he said that Emacs crashing is priority -1! (I m paraphrasing). I installed Emacs, ran it, and voil : the crash happened again. OK, that was good. When I ran Emacs inside GDB and asked for a backtrace, here s what I got:
#0  0x00007eede329972c in __pthread_kill_implementation () from /lib/libc.so.6
#1  0x00007eede323d8be in raise () from /lib/libc.so.6
#2  0x00007eede322531f in abort () from /lib/libc.so.6
#3  0x00007eede262879d in uw_init_context_1[cold] () from /usr/lib/libgcc_s.so.1
#4  0x00007eede2646e7c in _Unwind_Backtrace () from /usr/lib/libgcc_s.so.1
#5  0x00007eede3327b11 in backtrace () from /lib/libc.so.6
#6  0x000059535963a8a1 in emacs_backtrace ()
#7  0x000059535956499a in main ()
Ah, this backtrace is much simpler to follow. Nice. Hmmm. Now the crash is happening inside _Unwind_Backtrace. A pattern emerges! This must have something to do with stack unwinding (or so I thought keep reading to discover the whole truth). You see, the backtrace function (yes, it s a function) and C++ s exception handling mechanism use similar techniques to do their jobs, and it pretty much boils down to unwinding frames from the stack. I looked into Emacs source code, specifically the emacs_backtrace function, but could not find anything strange over there. This bug was probably not going to be an easy fix

The quest for a minimal reproducer Being able to easily reproduce the bug is awesome and really helps with debugging, but even better is being able to have a minimal reproducer for the problem. You see, py3-matplotlib is a huge package and pulls in a bunch of extra dependencies, so it s not easy to ask other people to just install this big package plus these other dependencies, and then run this command , especially if we have to file an upstream bug and talk to people who may not even run the distribution we re using. So I set up to try and come up with a smaller recipe to reproduce the issue, ideally something that s not tied to a specific package from the distribution. Having all the information gathered from the initial debug session, especially the Emacs backtrace, I thought that I could write a very simple program that just invoked the backtrace function from glibc in order to trigger the code path that leads to _Unwind_Backtrace. Here s what I wrote:
#include <execinfo.h>

int
main(int argc, char *argv[])
 
  void *a[4096];
  backtrace (a, 100);
  return 0;
 
After compiling it, I determined that yes, the problem did happen with this small program as well. There was only a small nuisance: the manifestation of the bug was not deterministic, so I had to execute the program a few times until it crashed. But that s much better than what I had before, and a small price to pay. Having a minimal reproducer pretty much allows us to switch our focus to what really matters. I wouldn t need to dive into Emacs or Python s source code anymore. At the time, I was sure this was a glibc bug. But then something else happened.

GCC 15 I had to stop my investigation efforts because something more important came up: it was time to upload GCC 15 to Wolfi. I spent a couple of weeks working on this (it involved rebuilding the whole archive, filing hundreds of FTBFS bugs, patching some programs, etc.), and by the end of it the transition went smooth. When the GCC 15 upload was finally done, I switched my focus back to the glibc hardening problem. The first thing I did was to yes, reproduce the bug again. It had been a few weeks since I had touched the package, after all. So I built a hardened glibc with the latest GCC and the bug did not happen anymore! Fortunately, the very first thing I thought was this must be GCC , so I rebuilt the hardened glibc with GCC 14, and the bug was there again. Huh, unexpected but very interesting.

Diving into glibc and libgcc At this point, I was ready to start some serious debugging. And then I got a message on Signal. It was one of those moments where two minds think alike: Gabriel decided to check how I was doing, and I was thinking about him because this involved glibc, and Gabriel contributed to the project for many years. I explained what I was doing, and he promptly offered to help. Yes, there are more people who love low level debugging! We spent several hours going through disassembles of certain functions (because we didn t have any debug information in the beginning), trying to make sense of what we were seeing. There was some heavy GDB involved; unfortunately I completely lost the session s history because it was done inside a container running inside an ephemeral VM. But we learned a lot. For example:
  • It was hard to actually understand the full stack trace leading to uw_init_context_1[cold]. _Unwind_Backtrace obviously didn t call it (it called uw_init_context_1, but what was that [cold] doing?). We had to investigate the disassemble of uw_init_context_1 in order to determined where uw_init_context_1[cold] was being called.
  • The [cold] suffix is a GCC function attribute that can be used to tell the compiler that the function is unlikely to be reached. When I read that, my mind immediately jumped to this must be an assertion , so I went to the source code and found the spot.
  • We were able to determine that the return code of uw_frame_state_for was 5, which means _URC_END_OF_STACK. That s why the assertion was triggering.
After finding these facts without debug information, I decided to bite the bullet and recompiled GCC 14 with -O0 -g3, so that we could debug what uw_frame_state_for was doing. After banging our heads a bit more, we found that fde is NULL at this excerpt:
// ...
  fde = _Unwind_Find_FDE (context->ra + _Unwind_IsSignalFrame (context) - 1,
                          &context->bases);
  if (fde == NULL)
     
#ifdef MD_FALLBACK_FRAME_STATE_FOR
      /* Couldn't find frame unwind info for this function.  Try a
         target-specific fallback mechanism.  This will necessarily
         not provide a personality routine or LSDA.  */
      return MD_FALLBACK_FRAME_STATE_FOR (context, fs);
#else
      return _URC_END_OF_STACK;
#endif
     
// ...
We re debugging on amd64, which means that MD_FALLBACK_FRAME_STATE_FOR is defined and therefore is called. But that s not really important for our case here, because we had established before that _Unwind_Find_FDE would never return NULL when using a non-hardened glibc (or a glibc compiled with GCC 15). So we decided to look into what _Unwind_Find_FDE did. The function is complex because it deals with .eh_frame , but we were able to pinpoint the exact location where find_fde_tail (one of the functions called by _Unwind_Find_FDE) is returning NULL:
if (pc < table[0].initial_loc + data_base)
  return NULL;
We looked at the addresses of pc and table[0].initial_loc + data_base, and found that the former fell within libgcc s text section, which the latter fell within /lib/ld-linux-x86-64.so.2 text. At this point, we were already too tired to continue. I decided to keep looking at the problem later and see if I could get any further.

Bisecting GCC The next day, I woke up determined to find what changed in GCC 15 that caused the bug to disappear. Unless you know GCC s internals like they are your own home (which I definitely don t), the best way to do that is to git bisect the commits between GCC 14 and 15. I spent a few days running the bisect. It took me more time than I d have liked to find the right range of commits to pass git bisect (because of how branches and tags are done in GCC s repository), and I also had to write some helper scripts that:
  • Modified the gcc.yaml package definition to make it build with the commit being bisected.
  • Built glibc using the GCC that was just built.
  • Ran tests inside a docker container (with the recently built glibc installed) to determine whether the bug was present.
At the end, I had a commit to point to:
commit 99b1daae18c095d6c94d32efb77442838e11cbfb
Author: Richard Biener <rguenther@suse.de>
Date:   Fri May 3 14:04:41 2024 +0200
    tree-optimization/114589 - remove profile based sink heuristics
Makes sense, right?! No? Well, it didn t for me either. Even after reading what was changed in the code and the upstream bug fixed by the commit, I was still clueless as to why this change fixed the problem (I say fixed because it may very well be an unintended consequence of the change, and some other problem might have been introduced).

Upstream takes over After obtaining the commit that possibly fixed the bug, while talking to Dann and explaining what I did, he suggested that I should file an upstream bug and check with them. Great idea, of course. I filed the following upstream bug: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120653 It s a bit long, very dense and complex, but ultimately upstream was able to find the real problem and have a patch accepted in just two days. Nothing like knowing the code base. The initial bug became: https://sourceware.org/bugzilla/show_bug.cgi?id=33088 In the end, the problem was indeed in how the linker defines __ehdr_start, which, according to the code (from elf/dl-support.c):
if (_dl_phdr == NULL)
   
    /* Starting from binutils-2.23, the linker will define the
       magic symbol __ehdr_start to point to our own ELF header
       if it is visible in a segment that also includes the phdrs.
       So we can set up _dl_phdr and _dl_phnum even without any
       information from auxv.  */


    extern const ElfW(Ehdr) __ehdr_start attribute_hidden;
    assert (__ehdr_start.e_phentsize == sizeof *GL(dl_phdr));
    _dl_phdr = (const void *) &__ehdr_start + __ehdr_start.e_phoff;
    _dl_phnum = __ehdr_start.e_phnum;
   
But the following definition is the problematic one (from elf/rtld.c):
extern const ElfW(Ehdr) __ehdr_start attribute_hidden;
This symbol (along with its counterpart, __ehdr_end) was being run-time relocated when it shouldn t be. The fix that was pushed added optimization barriers to prevent the compiler from doing the relocations. I don t claim to fully understand what was done here, and Jakub s analysis is a thing to behold, but in the end I was able to confirm that the patch fixed the bug. And in the end, it was indeed a glibc bug.

Conclusion This was an awesome bug to investigate. It s one of those that deserve a blog post, even though some of the final details of the fix flew over my head. I d like to start blogging more about these sort of bugs, because I ve encountered my fair share of them throughout my career. And it was great being able to do some debugging with another person, exchange ideas, learn things together, and ultimately share that deep satisfaction when we find why a crash is happening. I have at least one more bug in my TODO list to write about (another one with glibc, but this time I was able to get to the end of it and come up with a patch). Stay tunned. P.S.: After having published the post I realized that I forgot to explain why the -z now and -fno-strict-aliasing flags were important. -z now is the flag that I determined to be the root cause of the breakage. If I compiled glibc with every hardening flag except -z now, everything worked. So initially I thought that the problem had to do with how ld.so was resolving symbols at runtime. As it turns out, this ended up being more a symptom than the real cause of the bug. As for -fno-strict-aliasing, a Gentoo developer who commented on the GCC bug above mentioned that this OpenSSF bug had a good point against using this flag for hardening. I still have to do a deep dive on what was discussed in the issue, but this is certainly something to take into consideration. There s this very good write-up about strict aliasing in general if you re interested in understanding it better.

12 September 2010

Christian Perrier: Release team...

(update: yesterday's version was using http://wiki.debian.org/Teams/ReleaseTeam as reference, while http://www.debian.org/intro/organization is the one that's up-to-date. Those watching the current traffic in debian-release can probably realize the huge *thank you* deserved by the entire release team for the work they're doing. So, how about spamming the release team members with (private!) "thank you" messages when they unblock a package of yours? And (even more difficult) also one when they don't unblock your package...but spent time reviewing it and more time to explain you why they prefer not unblocking it... In any case, thank you, Neil Maulkin, Adam adsb, Dann dannf, Felipe faw, Jurij trave11er, Luk luk, Mehdi mehdi, Pierre MadCoder, Julien jcristau (doh, French Cabal!)...and Martin zobel (who's apparently forgotten on the page). Not to forget Adam adsb and Phil phil for managing stable releases....and the Wise Release Wizards (vorlon, aba, luk, HE). Hat off, guys (only guys there, yet another place for d-w to show up).

11 September 2010

Christian Perrier: Release team...

Those watching the current traffic in debian-release can probably realize the huge *thank you* deserved by the entire release team for the work they're doing. So, how about spamming the release team members with (private!) "thank you" messages when they unblock a package of yours? And (even more difficult) also one when they don't unblock your package...but spent time reviewing it and more time to explain you why they prefer not unblocking it... In any case, thank you, Neil Maulkin, Adam adsb, Dann dannf, Felipe faw, Jurij trave11er, Luk luk, Mehdi mehdi, Pierre MadCoder, Julien jcristau (doh, French Cabal!). Not to forget Adam adsb and Phil phil for managing stable releases....and the Wise Release Wizards (vorlon, aba, HE). Hat off....

16 September 2009

Dann Frazier: HP now supports etch on ProLiants

HP announced support for Debian 4.0 ('etch') on ProLiants today. See the HP/Debian page for details.

Dann Frazier: pysvn_load_dirs: a free implementation of svn_load_dirs

Users of subversion may have noticed that the svn_load_dirs script was removed from Debian due to a lack of a license from upstream. So far, attempts to get a DFSG-free licensed version have failed, so I've begun a new python implementation that is licensed under GPLv2. 0.1 is functional and uses the same syntax as the original, but is missing a few features that prevent it from being a drop-in replacement. I hope to remedy this in the coming weeks.

Dann Frazier: RPM Macros

I've been working with RPM macros quite a bit lately, and I've noticed some interesting properties. Though mostly obvious after the fact, they are surpisingly underdocumented. Macros don't seem to deal with floating point values. This works:
%define five 5
%if 0% ?five  <= 5
  %define string A
%else
  %define string B
%endif
While this causes a parse error:
%define five 5.3
%if 0% ?five  <= 5
  %define string A
%else
  %define string B
%endif
Quoting integers makes comparisons lexical. This will evaluate to false:
%define five 5
%if "0% ?five " == "5"
Perhaps the most frustrating one for me was that commented out macros are expanded. This has surprising results if your macro expands to multiple lines. Say my spec file includes the following:
%description
#%foo                                                                           
blah
And foo happens to expand to multiple lines:
%define foo one\
two\
three
My description will end up being:
Description :
two
three
blah

Dann Frazier: Controlling Power on a USB Hub

Andrew, I bought those wootoff lights as well, and have them connected to a hub on my mythtv system so I can activate them with a remote. I use the hub-ctrl.c utility from this page with this simple wrapper script that searches for the hub:
#!/bin/sh
bus=$(lsusb   grep TUSB2046   cut -d' ' -f2)
dev=$(lsusb   grep TUSB2046   cut -d' ' -f4   sed 's/:$//')
port=4
hubctrl=/home/dannf/hub-ctrl
if $hubctrl -b "$bus" -d "$dev" -v   grep "Port $ port :"   grep -q power; then
  toggle=0
else
  toggle=1
fi
$hub-ctrl -b "$bus" -d "$dev" -P "$port" -p "$toggle"
Note that not all hubs implement the port power feature - but luckily I had an unused one laying around that does. Unfortunately, one of my lights won't spin unless the physical power switch on the light is toggled - hopefully that's not true for yours.

Dann Frazier: svn-load now in unstable

svn-load, a DFSG-free replacement for svn_load_dirs, is now in unstable. John Wright has been working on adding support for doing preset pattern-based moves, which I hope will be ready in the next upload.

9 September 2009

Dann Frazier: Demuxing Personality Disorder (DPD)

I suffer from an in ability to successfully monitor multiple sources of information. For example, I rarely check on lists that I have procmailed out to their own folder. If its a list I need to stay on top of, I have to dump it to my primary inbox. For one list, I even dump messages to my inbox *and* keep a copy in a subfolder. That way I can keep on top of what's going on, but also keep a low-barrier-to-delete since I know I have an archival copy. I do think its important for me to check e-mail regularly, but there is some part of my brain that considers e-mail to be a time sink, and prevents me from going beyond what it considers the bare-minimum: Inbox messages. Another instance of this disorder struck me with RSS. There are web pages that I check every day, and not all of them have RSS feeds. I played with a few different RSS readers a while back, and decided that straw was my favorite. But I could not get myself in the habit of bringing up a second application. Later I started using Firefox, and I thought I'd have better luck with something like Sage. But even that was easy to avoid because it requires actually opening up the sage panel. I would either never check it or, in times of boredom, check it too often. Since then I've given up on RSS readers. These days I've stolen an idea from Alex Chiang and just keep a bookmark folder called "daily" and one called "monthly". Every morning I hit the "Open All in Tabs" item in the daily folder, and quickly ctrl-w through pages w/ no new content. My daily folder includes things like bug reports I'm monitoring for activity, gitweb views of files where I'm waiting for a fix, blogs, parcel tracking, comics, wiki watch lists, etc. Since a few of these pages are rather important, I always remember to do it and therefore force myself to browse the others as well. Most days I spend less than 10 minutes "wasting time" going through them. Its strange to both be aware of poor working habits, yet know from years of experience that I'll be more successful if I work around them rather than trying to retrain myself.

19 February 2009

Dann Frazier: HP ProLiant Servers Debian 5.0 "capable"

HP has posted a capabilities matrix for Debian 5.0 ('lenny'). Its linked off of the easier-to-remember url: hp.com/go/debian.

17 February 2009

Dann Frazier: Adding firmware blobs to install media

Here's a shell script I wrote to stuff non-free firmware bits into an existing lenny installer image. It currently can operate on either a initrd.gz file or an isolinux-based iso image. I've only tested it on a netboot initrd.gz and a netinst iso, but it should work on larger images as well. If you need support for some other media type or architecture, let me know.

17 January 2009

Andrew Pollock: [tech] USB power switch, part 2

Continuing with my mission to control the power to some dumb USB-powered lights... Dann Frazier confirmed my theory that a USB hub would indeed do the job. I'd already found the hub-ctrl.c program he mentioned, but couldn't get it to work with my built-in USB ports of my laptop. It seems it all depends on whether the hub will support per-port power switching or not. (lsusb -v will tell you). So off to the mighty institution that is Fry's Electronics I went. The first two attempts failed, as did a borrowed hub. It seems most (at least the cheap one) use a Genesys Logic chipset, which does not support per-port power control. Fortunately I chose products that weren't in blister packs, so I was able to return them to Fry's in as-new condition and try again. Now that I knew USB hubs would do the trick, I did some more targeted searching, and found this thread where someone had been messing around with hubs to do what sounded like what I wanted. (Incidentally, this email in the thread also provided a nice looking Python program. I'm going to look at refitting it to use the "real" Python USB module.) I emailed the poster to ask him what brand of hub he was using. The answer was the Linksys 4-port hub So I managed to track down one of them last night. Yes, it works, but the downside is the hub itself requires external power, which is a bit unfortunate. Not only do I need to use a hub to make these lights software controllable, I have to plug the hub into a power outlet. Bleh.

28 May 2008

Dann Frazier: Retrieving a NIC's permanent address

I needed to query a card to get its permanent mac address (the value programmed into the card, even if the admin has ifconfig'd it differently). This can be done with the linux kernel's ethtool API, but the ethtool command doesn't currently support it and google didn't know how either. I had to figure it out myself. Now google, you no longer have an excuse - I expect you to know next time I ask.

9 April 2008

Dann Frazier: etchnhalf kernels available for testing

2.6.24 kernels targeted for etch are available for testing in etch-proposed-updates. Some architectures are missing for this first upload, but fixes are pending for the next upload.

10 March 2008

Jan Wagner: bit nagios-plugin bugsquashing, stalling policyd-weight and my first perl module package

Last week I did again some work on nagios-plugins. After the announcement of Dann Frazier to upload NMU to fix a trivial bug, I thought it’s time again to give some extra care to the package. So I prepared 1.4.11-2 fixing the important bugs and uploaded it. I also commited some minor fixes to the svn, so these issues will get fixed by the next upload. Since the development of policyd-weigh stalled and unfortunately maybe get stuck, I was looking for an alternative, which maybe found with postfwd. It’s quite flexible but it also will take more time (and care!) to get a reliable configuration, which maybe effective as policyd-weight (still) is right now. While checking the dependencies for postfwd I noticed that Net::DNS::Async isn’t available in Debian (yet). So I decided to create a package starting with dh-make-perl, join the Debian Perl Group and let it review. Damyan Ivanov was so kind to review and upload it, Gregor Herrmann did also give some much useful hints. Thanks to both! And yes, I also found time to step forward with NM, since I was overloaded the last weeks with usual work and life. Thank to my AM to be so appreciative.

25 January 2008

Dann Frazier: Etch and a half

For those unfamiliar with the project, etch-and-a-half has the goal of making the existing etch release work on more hardware, but with minimal risk to existing etch users. This will be done by *adding* new kernel packages to the existing release. Kernel selection isn't final yet - the kernel team would like to use 2.6.24 if it proves itself stable in time; 2.6.22 is the backup choice. etch-and-a-half isn't restricted to kernel updates by-definition; new X drivers, or proven-safe X driver updates are reasonable candidates as well (though at this point, no changes have been proposed). An oft-asked question is around updating desktop packages, e.g. new versions of GNOME/KDE. This is beyond the scope of etch-and-a-half because it introduces a great deal more change than just additional hardware support, its wouldn't be a transparent upgrade to existing etch users, and we cannot say with a high level of confidence that it wouldn't introduce regressions for existing users. That said - a localized change to a package that adds support for new hardware and clearly doesn't break existing hardware is a valid candidate for any stable release. Such requests should be submitted to the SRM team like any other stable update request. The current plan for etch and a half is to release as part of a point release. 4.0r4 is the planned point release, since 4.0r3 is planned to have a quick turnaround to fix issues with 4.0r2. This means that pre-release bits will be made available via proposed-updates - testers are greatly appreciated. There is no plan to add support for etch-and-a-half to the etch branch of d-i. Rather, support will likely be added in a beta of the lenny installer. A wiki page exists here, and a coordination list has been setup on teams.debian.net.

16 November 2007

Dann Frazier: On IRC proxies

Adeodato mentioned using ctrlproxy, so I thought I'd ramble a bit about my experience w/ IRC proxies in general. dircproxy was the first IRC proxy software I tried. It did the job, but the thing that annoyed me the most was that only one client could connect at a time. I'd get home and realize I was still connected at work, and then have to login and send a kill signal to xchat. When I found out about ctrlproxy, and that it supports multiple clients, I was very excited. It did the job, but occasionally it would hang and need to manually killed/restarted, and it had this weird problem of writing logs for one channel to the log file of another. Most recently (and for probably a year now) I've been using bip. I immediately hit an issue with one server - bip would constantly reconnect/disconnect. Upstream immediately went to work on the problem, determined it was a bug in the server itself, and (iirc) sent the server maintainers a patch. Now that's support! bip has been stable for me for nearly 2 years now. I'd also suggest comparing the bug pages for dircproxy, ctrlproxy, and bip. That's not always a good way to measure relative stability, but it resembles my experience in this instance.

6 August 2007

Martin F. Krafft: Sending email via IMAP (or not?)

I am playing with the thought of using IMAP to send email from my mobile clients. The way this works is that mail is actually placed into a Maildir locally and synchronised with the IMAP server using a tool such as offlineimap. The Courier IMAP server supports such an outbox feature, for other servers you can use inotify or a cron job (I am planning to publish a writeup about my new mail setup sometime soon, so no scripts linked here yet). Anyway, it's trivial to get it working, and Dann Frazier has written it all up. The question is more whether I want to, or not. I am soliciting feedback. The advantages of using an IMAP-synchronised outbox are: Any solution with advantages comes with disadvantages, and these are: Can you think of any other advantages or disadvantages? Do you use an IMAP-synchronised outbox and would like to share your experiences with me? Or do you have yet another solution? Write to me! NP: OSI: Free

18 July 2007

Dann Frazier: Installing Debian 4.0 ('etch') on a TVease Hannibal

Last night I decided to install etch on my pre-built MythTV system. The process is documented here

17 July 2007

Adrian von Bidder: HP and Debian

I recently learned that HP updated their Debian compatibility matrix to include Debian 4.0 a.k.a etch (which is obviously great, and I hope more major vendors would include this type of compatibility information.) Clicking around, I find it somewhat funny that if you go to the certification information for Red Hat or Novell SuSE (now there's a great name slowly dying...), you'll find a Debian logo, a Red Flag logo and a Mandriva logo. But no Red Hat or Novell/SuSE logos. Do they want their customers to try Red Flag, Mandriva or Debian instead of the Big Two(tm)? Because on the Debian page, there's no Red Hat or Novell/SuSE logo...

Next.