log☇︎
400 entries in 0.592s
assbot: Bug #134400 “glibc do not really static compile my project.” : Bugs : glibc package : Ubuntu ... ( http://bit.ly/1H4hmSN )
asciilifeform: https://bugs.launchpad.net/ubuntu/+source/glibc/+bug/134400 << as i understand, drepper 'did it'
asciilifeform: a #pragma in glibc ?
mod6: either have I, but glibc is full of trickery.
trinque: I have no strong opinion regarding uclibc vs glibc, as I haven't used the former at all before this
asciilifeform: glibc belongs in woodchipper regardless
trinque: older glibc?
asciilifeform: and/or glibc
mircea_popescu: not just from a "must be in there for we need all the dns-carried pores imported via glibc etc", but also in the much lower level "who's in charge of the it!!1" thing
trinque: it diddles glibc internal preprocessor flags, so on
mircea_popescu: what, like you stopped using glibc ?
ben_vulpes: justJanne: have you been following the glibc travails?
trinque: mircea_popescu: does seem that we keep encountering the rot of glibc
trinque: ben_vulpes: the actual dieharder code uses glibc internals in a way that used to work, now does not due to as yet undiscovered source of rust, with vague indications that compiling with std=c99 has implications for glibc
trinque: asciilifeform: turns out dieharder uses internal glibc preprocessor directives which cause it to explode when built as c99
trinque: ben_vulpes: probably faster than waiting for a glibc fix :^)
trinque: https://bugs.gentoo.org/show_bug.cgi?id=549754 << specific incantation that breaks the glibc headers
trinque: this guy's making me write a test program to fix an include in glibc
trinque: tragedy of the commons; everyone fucks glibc and no one pays her medical bills
mircea_popescu: trinque we were looking for someone to fix glibc earlier too
trinque: and I will, if only as an exercise in whether it's actually possible to get something fixed in glibc
trinque: guy sends me to go bother whoever can fix glibc
trinque: seems this is actually glibc fuckery re: dieharder
trinque: mircea_popescu: seems it might've worked against an older glibc
trinque: blowing up inside glibc
mircea_popescu: or was, at some point. before glibc
trinque: http://log.bitcoin-assets.com/?date=12-05-2015#1128895 << aha. glibc is dead to me. << gentoo's hardened/linux/uclibc/amd64 ? ☝︎
ascii_field: http://log.bitcoin-assets.com/?date=12-05-2015#1128895 << aha. glibc is dead to me. ☝︎
davout: from what i grasped, the issue is that glibc is pulling some random bits in, dynamically
asciilifeform: the uniquely idiotic error message, 'not found', refers to glibc and its turds (libnss etc)
mircea_popescu: i dunno, by now linux is windowized enough. they can release their shit for systemd/ubuntu/glibc-with-modules/etc and pretty much have it right.
mod6: ahh, so this is for embedded systems -- to replace a bulky/asinine glibc?
mircea_popescu: asciilifeform heh, so far nobody even compiles glibc. go away with your exotic temptations, fair lady.
mircea_popescu: ben_vulpes it ends up pulled through glibc which is the source of this poison
assbot: The GNU C Library: glibc iconv Implementation ... ( http://bit.ly/1EM1zFX )
mod6: <+ascii_field> http://www.gnu.org/software/libc/manual/html_node/glibc-iconv-Implementation.html << wtf is this thing doing in bitcoind << is this another automagically linked in pos to glibc?
assbot: The GNU C Library: glibc iconv Implementation ... ( http://bit.ly/1JMxYk6 )
ascii_field: http://www.gnu.org/software/libc/manual/html_node/glibc-iconv-Implementation.html << wtf is this thing doing in bitcoind
mircea_popescu: but it can't be started from the glibc i don't think. that's the middle.
mod6: <+jurov> the libnss was done as binary plugin to glibc << so there is no possible way to just build glibc by hand and not include libnss? or there are basically so many things that use libnss that even if you did, stuff wouldn't work anyway?
mod6: <+mircea_popescu> ~whether one even uses libnss or not~! << so even if we didn't even call "whatsMyIP()" or w/e it is, this would still be a dingleberry attached to glibc.
mod6: so libnss is dynamically compiled and built/linked to glibc, and can not be avoided? ☟︎
jurov: thus truly statical compilation of glibc is impossbile
mod6: ok maybe that's the part I was missing - how libnss is somehow tied to glibc.
jurov: the libnss was done as binary plugin to glibc
jurov: mod6 i can explain, too. to support different configurations for DNS/users/whatever resolving without glibc recompilation and without interprocess communication
mod6: asciilifeform: I'm just trying to put together the monthy address; In one to three sentences cna you help me summarize what is going on with glibc/libnss? ☟︎
trinque: with the normal glibc, not any of the alternatives
asciilifeform: in other news, #glibc is one of the quietest channels one could imagine short of an entirely dead one.
mircea_popescu: let's make glibc no longer compile statically and github not work. that'll make the foss so much better.
assbot: Understanding glibc malloc | sploitF-U-N ... ( http://bit.ly/1Ct1SAJ )
mats: https://sploitfun.wordpress.com/2015/02/10/understanding-glibc-malloc/
nubbins`: " The problem comes, I think, mainly from statically linking other GLIBC libraries, notably "libpthread"" <<< we're using that one
mircea_popescu: less than glibc. weighs about the same as uc
ascii_field: mircea_popescu: i began reading the glibc source last night
mircea_popescu: you will be helped by the glibc team but you.absolutely.must.know.what.you're.doing.
ascii_field: and other things. It's supposed to make application programs independent of the separately configured actual network environment of the machine. A nice idea, but changes to GLIBC can lead to problems loading it. And you can't statically link "libnss", since it is configured for each machine individually. The problem comes, I think, mainly from statically linking other GLIBC libraries, notably "libpthread", "libm", and
ascii_field: '"I suppose the idea is that everything will be in the downloaded file, so nothing depends on the local libraries on the target system. Unfortunately with Linux, and I think anything else using GLIBC, this still isn't quite true. There's this "libnss" (name service switch, some people seem to call it network security system) which provides functions for accessing various databases for authentication, network information,
decimation: asciilifeform: part of the problem is that glibc has a 'colorful' history
asciilifeform: <decimation> I didn't see it. So " --enable-static-nss" is useful for glibc << as i understand, this results in random breakage (a binary which only runs with any degree of certainty on your machine)
asciilifeform: 'I do not know where to find the historic references, but yes, static linking is dead on GNU systems. (I believe it died during the transition from libc4/libc5 to libc6/glibc 2.x.) The feature was deemed useless in light of: Security vulnerabilities. Application which was statically linked doesn't even support upgrade of libc....'
decimation: I didn't see it. So " --enable-static-nss" is useful for glibc
decimation: http://log.bitcoin-assets.com/?date=05-04-2015#1089276 < I apologize for failing to convey my meaning well. I meant that redhat 'provides' this as a source package, and therefore one could examine exactly how they did it - not to swallow the binary without inspection. Source is here > http://ftp.redhat.com/redhat/linux/enterprise/6Server/en/os/SRPMS/glibc-2.12-1.149.el6_6.5.src.rpm ☝︎
mircea_popescu: what was bitcoind using that ended up pulling libnss via glibc ? gethostbyaddr() was it ?
mircea_popescu: http://log.bitcoin-assets.com/?date=05-04-2015#1089208 << apparently we were not the only ones to notice glibc got raped. ☝︎
decimation: which gives us glibc turdlets
decimation: the amount of bloat in glibc is quite shocking
decimation: 6) uClibc does not support NSS (/lib/libnss_*), which allows glibc to easily support various methods of authentication and DNS resolution. uClibc only supports flat password files and shadow password files for storing authentication information. If you need something more complex than this, you can compile and install pam.
decimation: ok from uclibc docs/Glibc_vs_uClibc_Differences.txt
decimation: well, the redhat glibc static build is pretty much just building glibc with -static
decimation: ascii_modem: did you see the earlier note about glibc compiled statically
decimation: it is apparently possible to create a static glibc
assbot: Logged on 02-04-2015 14:54:13; assbot: Logged on 02-04-2015 03:26:58; nubbins`: "glibc uses libnss to support a number of different providers for address resolution services. Unfortunately, you cannot statically link libnss, as exactly what providers it loads depends on the local system's configuration."
asciilifeform: 'Using 'getaddrinfo' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking' << wtf are we even using getaddrinfo for.
assbot: Logged on 02-04-2015 03:26:58; nubbins`: "glibc uses libnss to support a number of different providers for address resolution services. Unfortunately, you cannot statically link libnss, as exactly what providers it loads depends on the local system's configuration."
assbot: Logged on 02-04-2015 01:18:00; decimation: http://stackoverflow.com/questions/2725255/create-statically-linked-binary-that-uses-getaddrinfo <glibc uses libnss to support a number of different providers for address resolution services. Unfortunately, you cannot statically link libnss, as exactly what providers it loads depends on the local system's configuration.
nubbins`: "glibc uses libnss to support a number of different providers for address resolution services. Unfortunately, you cannot statically link libnss, as exactly what providers it loads depends on the local system's configuration."
decimation: yes, the traditional glibc/dns client shit is quite turdly
decimation: http://stackoverflow.com/questions/2725255/create-statically-linked-binary-that-uses-getaddrinfo <glibc uses libnss to support a number of different providers for address resolution services. Unfortunately, you cannot statically link libnss, as exactly what providers it loads depends on the local system's configuration.
mircea_popescu: "/home/stas/bitcoin-v0_5_3_1/ourlibs/include/boost/asio/detail/impl/socket_ops.ipp:2900: warning: Using 'getaddrinfo' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking" ☟︎
mircea_popescu: re " warning: Using 'getaddrinfo' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking" ascii_field wtf is that shit !?
asciilifeform: just to underscore the sheer level of the braindamage i've uncovered - the thing won't build on any box with past 2+ yrs of glibc.
asciilifeform: http://pastebin.com/VkgE27Pd << undoes the shitgnomism, but wtf should i have to rebuild glibc and -everything- to use a perfectly ordinary proggy ?
asciilifeform: (and yes, as far as i can tell, the recent glibc pant-shitting was -also- drepper)
asciilifeform: glibc
assbot: Bug #4295: Issue between aspectator and new glibc at different architectures - C Instrumentation Framework - Open-Source Projects ... ( http://bit.ly/1EeRsYG )
asciilifeform: the_scourge: 'NixOS 14.12 “Caterpillar” has been released, the third stable release branch. It brings Linux 3.14, systemd 217, Glibc 2.20, KDE 4.14.1, and much more.'
punkman: "so 2013 glibc fixed a bug, 2014 google found+fixed the same bug https://code.google.com/p/chromium/issues/detail?id=364511 … and 2015 qualys found it again.."
mircea_popescu: decimation: re: glibc bug << was this an ulrich drepperism? << quite likely.
decimation: glibc bugs
BingoBoingo: 9 years ago the debian official random number was 6. Now glibc fuckery is uncovered. Imperial nudity advances.
decimation: re: glibc bug << was this an ulrich drepperism?
assbot: oss-sec: Re: GHOST gethostbyname() heap overflow in glibc (CVE-2015-0235) ... ( http://bit.ly/18qONRD )
assbot: oss-security - Qualys Security Advisory CVE-2015-0235 - GHOST: glibc gethostbyname buffer overflow ... ( http://bit.ly/1zr7C2Y )
assbot: sourceware.org Git - glibc.git/commitdiff ... ( http://bit.ly/1yLj1sr )
asciilifeform: https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=d5dd6189d506068ed11c8bfa1e1e9bffde04decd;hp=fef94eab0bd308d5059a2588c753bf9a4926845d << works
assbot: Bug 1183461 – CVE-2015-0235 glibc: __nss_hostname_digits_dots() heap-based buffer overflow ... ( http://bit.ly/1z6fvHR )
mircea_popescu: im sure you can have glibc ported over :D
mircea_popescu: a glibc forc
mircea_popescu: pankkake saywut ?! glibc ?