T2 IRC Log: 2010-01-20

This is the log as captured by an IRC bot in the channel. The statements are those of the individual people and might not neccessarily reflect the policy and legal rules as set forth by the T2 SDE Project.

« prev | next »

--- Log opened Wed Jan 20 00:00:22 2010
01:14 -!- digitaloktay_ [n=pardus@dslb-084-056-212-012.pools.arcor-ip.net] has quit [Client Quit]
02:33 -!- koan [n=koan@unaffiliated/koan] has quit [Read error: 60 (Operation timed out)]
02:34 -!- koan [n=koan@unaffiliated/koan] has joined #t2
07:07 -!- B-2 [i=stealth@sourcemage/guru/stealth] has joined #t2
07:08 -!- Stealth [i=stealth@sourcemage/guru/stealth] has quit [Read error: 92 (Protocol not available)]
07:08 -!- B-2 is now known as Stealth
08:31 < mjungwirth> moin
08:34 < mjungwirth> someone here ?
08:36 < mjungwirth> i build a fresh system. util-linux needs cvs on the build-system! is this right?
08:44 < mjungwirth> samba can't build. error in patching file with fix_smb_mtab.patch
08:54 < mjungwirth> this patch-file isn't need longer. the samba group has patch the mtab.c.
09:03 < mjungwirth> for the mc-mp package the mc-mp-intr_flag.patch is no longer needed.
09:14 < mjungwirth> hmm i cant build php 5.3.1
09:16 < mjungwirth> why dues php search the libxml2 on my host? /usr/lib/libxml2.so? the t2-build abort this. can you help ?
09:18 < rxr> re
09:19 < rxr> maybe it uses a fixed path? or some old style -config script to detect libxml presense
09:19 < mjungwirth> i remove libxml2 development packet from my host. the php build abort by making configure. (xml2-config not found. Please check you libxml2 installation)
09:19 < rxr> mjungwirth: I can try to give it a try to help
09:20 < mjungwirth> rxr have you reed the message from 08:35
09:20 < rxr> yep
09:20 < mjungwirth> you can remove to patch-files
09:20 < rxr> will update / delete the patches accordingly in a sec
09:20 < mjungwirth> perfect
09:42 < rxr> this samba patch ? fix_mtab_stat.patch
09:43 < rxr> the mc-mp patch is only needed when built withou one of it's supporting terminal libraries
09:43 < rxr> curses or some slang whater stuff it was
09:57 < CIA-28> rene * r35360 /trunk/package/base/lsof/lsof.desc: * updated lsof (4.82 -> 4.83)
09:57 < CIA-28> rene * r35359 /trunk/package/security/john/john.desc: * updated john (1.7.4 -> 1.7.4.2)
09:57 < CIA-28> rene * r35361 /trunk/package/base/pcre/pcre.desc: * updated pcre (8.00 -> 8.01)
10:12 < mjungwirth> i have add some --with parameters to the configure script. Now the php package can build without any errors but the installer process gives me an error (fl_wrapper.so write outside basedir (/os): /usr/lib/php.depdblock) have you any idea to fix this ?
10:12 < rxr> maxbe it needs some more configure switches or makefile options
10:12 < rxr> or some missing DESTDIR variables in the Makefile (or simillar typos)
10:12 < mjungwirth> ok i search
11:26 < mjungwirth> what is the correct variable for $root/usr/lib path for the .conf file
11:27 < rxr> $root$libdir
11:28 < rxr> but if there was $root/usr/lib that is not too bad
11:28 < rxr> at least it should not install into your host system, that is what $root is for
11:28 < rxr> just the hardcoded "lib" is not good for mulitilib builds, e.g. lib64 and other variants
11:29 < mjungwirth> ok thanks
11:29 < mjungwirth> yea
11:30 < mjungwirth> php build succesfully
11:30 < mjungwirth> i send you the php.conf file
11:30 < rxr> good to hear
11:32 < mjungwirth> mail is out
11:37 < mjungwirth> one question
11:38 < rxr> yeah=?
11:38 < mjungwirth> the e2fsprogs need pkgconfig for the build
11:38 < mjungwirth> but e2fsprogs build bevor pkgconfig is build
11:38 < mjungwirth> how can i handle this ?
11:38 < rxr> chage the currently static package Priority tag
11:39 < rxr> but I guess it needs it in the host to run the tool?
11:39 < rxr> than the solution would be rather to enable it for the stage 0, toolchain
11:39 < rxr> , if it is not yet
11:40 < CIA-28> rene * r35362 /trunk/package/gnome2/gmime/gmime.desc: * updated gmime (2.4.12 -> 2.4.13)
11:41 < mjungwirth> i test it
11:41 < CIA-28> rene * r35363 /trunk/package/develop/php/php.conf:
11:41 < CIA-28> Mario Jungwirth :
11:41 < CIA-28> * more guidance for php cross build
11:42 < mjungwirth> ok e2fsprogs runs
11:49 -!- digitaloktay [n=pardus@dslb-084-056-233-080.pools.arcor-ip.net] has joined #t2
11:50 < mjungwirth> is there a variable existing for the include directory ?
11:50 < rxr> sure, $inclduedir
11:50 < rxr> includedir
11:50 < mjungwirth> nice thanks
11:59 < mjungwirth> i send you a new .conf file for nfs-utils. have fun..
12:01 < mjungwirth> how can i find a binary file in all packages ?
12:02 < mjungwirth> not the build package! all packages for the complete t2-project
12:02 < rxr> for the T2 $*dir variables no extra slash (/) should be needed
12:03 < rxr> the variable should start already with a slash, double slashes can lead to subtile errors in some exotic packages / combinations
12:03 < rxr> (and lock ugly in the log and running binary output at run-time, man/info pages etc. otherwise)
12:04 < rxr> I removed that, test-build and commit
12:07 < CIA-28> rene * r35364 /trunk/package/filesystem/nfs-utils/nfs-utils.conf:
12:07 < CIA-28> Mario Jungwirth :
12:07 < CIA-28> * fix nfs-utils includedir for the new tirpc depencency
13:16 < mjungwirth> ok thanks sir :)
13:21 -!- digitaloktay [n=pardus@dslb-084-056-233-080.pools.arcor-ip.net] has quit ["Bin mo weg longer............ /exit hilft bei Kopfschmerzen oder hängt ein paar Poster von Merkel auf :)))"]
13:21 -!- digitaloktay [n=pardus@dslb-084-056-233-080.pools.arcor-ip.net] has joined #t2
13:28 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: mqueiros, koan, aloof
13:30 -!- Netsplit over, joins: koan, aloof, mqueiros
13:35 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: mqueiros, @ChanServ, koan, LMJ, CIA-28, aloof, TobiX, digitaloktay, Stealth
13:42 -!- Netsplit over, joins: Stealth
13:42 -!- Netsplit over, joins: koan, aloof, mqueiros
13:43 -!- Netsplit over, joins: @ChanServ, digitaloktay, TobiX, CIA-28
13:44 -!- LMJ [n=serwou@laf31-4-82-236-42-164.fbx.proxad.net] has joined #t2
13:49 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: mqueiros, koan, aloof
13:49 -!- Netsplit over, joins: koan, aloof, mqueiros
13:52 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: Stealth
13:53 -!- Netsplit over, joins: Stealth
14:11 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: mqueiros, koan, aloof
14:14 -!- Netsplit over, joins: koan, aloof, mqueiros
14:16 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: Stealth
14:17 -!- Netsplit over, joins: Stealth
14:29 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: mqueiros, @ChanServ, koan, LMJ, CIA-28, aloof, TobiX, Stealth, digitaloktay
14:35 -!- Netsplit over, joins: @ChanServ, CIA-28, TobiX, digitaloktay
14:35 -!- Netsplit over, joins: koan, aloof, mqueiros
14:39 -!- Stealth [i=stealth@sourcemage/guru/stealth] has joined #t2
14:40 -!- mpp [n=mpp@2a01:48:251:107:203:dff:fe13:6c9e] has joined #t2
14:41 -!- LMJ [n=serwou@laf31-4-82-236-42-164.fbx.proxad.net] has joined #t2
14:43 < mpp> moinsen
14:43 < rxr> moin moin
14:47 < rxr> we currently get 80% more out thru our inet pipe than the nominal provider bitrate
--- Log opened Wed Jan 20 14:57:32 2010
14:57 < rxr> ok - that was it
14:58 < rxr> stupid linux kernel oopsery
15:00 < mjungwirth> rxt
15:00 < mjungwirth> r
15:01 < rxr> I have to spent some more time on my own microkernel design and replace our router here ASAP :-)
15:01 < mjungwirth> can you delete the hot-fix.patch from the php package. it is not longer needed
15:02 < mjungwirth> the PHP_SUBST(.... is now in the php source includes!
15:02 < rxr> how's with the samba patch
15:02 < rxr> it still applies
15:02 < rxr> but it is still really superflous now ?
15:03 < mjungwirth> yes it is!
15:29 < CIA-28> rene * r35366 /trunk/package/network/samba/fix_mtab_stat.patch:
15:29 < CIA-28> Mario Jungwirth :
15:29 < CIA-28> * removed superfluous samba/fix_mtab_stat.patch
15:29 < CIA-28> rene * r35365 /trunk/package/develop/php/hot-fix.patch:
15:29 < CIA-28> Mario Jungwirth :
15:29 < CIA-28> * removed superfluous php/hot-fix.patch
16:30 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has joined #t2
16:30 < Baldzius> moin
16:31 < rxr> hey Baldzius
16:31 < Baldzius> hi rxr
16:40 < rxr> you are happy not to have an account on our box
16:40 < rxr> our line is so oversaturated today, ...
16:41 < rxr> pushing up to 90% more thru the pipe than our ISP nominally offer us ...
16:41 < rxr> didn't even know we can push that much thru :-)
16:41 < rxr> though the stupid linux kernel oopsed again under the load
16:41 < rxr> stupid typos in either the associated ethernet driver or ipv4 stack, ...
16:47 < Baldzius> which kernel version?
16:48 < rxr> 2.6.31.5-dist but this happened with previous ones already under high load as well :(
16:49 < Baldzius> not good
16:58 < CIA-28> aldas * r35367 /trunk/package/network/bind/bind.desc: * updated bind (9.6.1-P2 -> 9.6.1-P3)
16:58 < CIA-28> aldas * r35369 /trunk/package/mail/thunderbird/thunderbird.desc: * updated thunderbird (3.0 -> 3.0.1)
16:59 < CIA-28> aldas * r35368 /trunk/package/network/dhcp/dhcp.desc: * updated dhcp (4.1.1rc1 -> 4.1.1)
17:07 < Baldzius> rxr: btw would you mind to write mini howto about kvm maybe?
17:07 < Baldzius> I am not sure now, do I need qemu package to start using kvm
17:10 < rxr> well, in the past you had to use the kvm package
17:10 < rxr> because it comes with an modified / improved qemu
17:10 < rxr> but nowadays the basics are already merged in qemu
17:10 < rxr> so you could just use the qemu package and the in-kernel kvm
17:10 < rxr> but then you get only basic performance / not all features
17:11 < rxr> so for best results, enable the kvm package, so that during the kernel build (linux26) the newer kvm kernel mod is built with improved performance and features / correctnet
17:11 < rxr> correctness
17:11 < rxr> and then also use the qemu from the kvm package
17:11 < rxr> it has SMP support, live migration, etc. pp-
17:12 < rxr> (oh, and our kvm package comes with support to emulate a Mac, thus run Intel Mac OSX in the VM
17:12 < rxr> but that is currently broken, despite my extra care, at least for me tested just yesterday, though maybe that was just my bad luck, but probably one or the other Mac OS X virt patch needs some slap top behave again :-)
17:13 < Baldzius> ok, thanks that's pretty much information I needed
17:17 < Baldzius> too bad my work laptop don't have VT
17:18 < Baldzius> but I have a laptop at home with cpu that supports VT
17:19 < Baldzius> not sure what might be easier to administrate : kvm or xen
17:19 < rxr> btw. AMD VT is way more powerful
17:19 < rxr> can run real-mode code, has less overhead
17:20 < rxr> Intel VT wasts more cycles, here and there, and running real-mode code, such as the bootloader is fully emulated by an x86 in-kernel interpreter
17:20 < rxr> you can watch grub painting the menu on an Intel VT CPU, ...
17:20 < rxr> if you have the option, get an AMD
17:20 < rxr> btw. I would have implemented this real-mode quirk differently
17:21 < rxr> I would not trust an in-kernel emulator in any case, beside it's dog slow
17:21 < rxr> I would have let it fall back to the user-space, Qemu, emulate it
17:21 < rxr> more secure in any case, and also more performant, as Qemu would recompile with TCG
17:21 < rxr> that would be some 50 times or so faster, ...
17:22 < rxr> but, well, in the Linux land so much is going insane these days, what should I say, ...
17:22 < rxr> then again that is just the Intel VT real-mode (read BIOS and boot loader until the OS is running) corner case, ...
17:23 < Baldzius> ic
17:24 < Baldzius> myself I always perfered AMD over Intel, but you know you don't have much options when it comes to laptops
17:24 < rxr> AMD Neo rocks
17:25 < Baldzius> X2?
17:25 < rxr> if Lenovo spins a Neo X2 bosting X100 I grab it
17:25 < Baldzius> I saw it too
17:25 < rxr> http://rene.rebe.name/2009-12-22/the-performance-stagnation-that-is-the-intel-atom/
17:25 < Baldzius> wrote down it's model :)
17:25 < rxr> might take to load, we're kind of slashdottet
17:25 < rxr> AMD Neo X12
17:25 < rxr> err
17:25 < rxr> X2
17:25 < rxr> thick fingers these days
17:26 < rxr> http://rene.rebe.name/2010-01-09/lenovo-x100e-nearly-the-perfect-netbook/
17:26 < rxr> the currently (starting to) ship X100 only has the older, singel core AMD silicon
17:27 < rxr> only a little faster than Atom
17:27 < rxr> X2 rocks more
17:27 < rxr> hope they find a more suitable battery 'till then
17:27 < Baldzius> :)
17:28 < rxr> and the site lods for ages?
17:28 < rxr> or still endurable ?
17:28 < rxr> given our current line saturation it should be pretty bad ...
17:30 < Baldzius> loaded
17:30 < Baldzius> took a while
17:30 < rxr> reasonably fast or slow like h*ll?
17:34 < Baldzius> it was in background
17:34 < Baldzius> so can't say
17:34 < Baldzius> http://shop.lenovo.com/SEUILibrary/controller/e/web/LenovoPortal/en_US/catalog.workflow:category.details?current-catalog-id=12F0696583E04D86B9B79B0FEC01C087¤t-category-id=AD7535A40F8845ABAEAC52D528B8C921&AID=10396483&PID=3434971&CJURL=http%3A%2F%2Fshop.lenovo.com%2FSEUILibrary%2Fcontroller%2Fe%2Fweb%2FLenovoPortal%2Fen_US%2Fcatalog.workflow%3Acategory.details%3Fcurrent-catalog-id%3D12F0696583E04D86B9B79B0FEC01C087%26current-category-id%3DA
17:34 < Baldzius> D7535A40F8845ABAEAC52D528B8C921
17:35 < Baldzius> you can customize like for Dell
17:36 < rxr> yeah - but not with an X2 CPU or did they already update that?
17:36 < rxr> some days ago when I checked they only has one CPU option
17:37 < rxr> but I saw some interview where some lenovo person said something of a dual core option coming then later in Q1, I hope that's true
17:37 < Baldzius> single only
17:37 < Baldzius> sec I will check one page
17:38 < Baldzius> yep, dual core, Athlon Neo X2 L335 or Turion X2 L625
17:39 < rxr> the X100 ?
17:39 < rxr> the link above is wrapped once, so I'm not sure
17:39 < rxr> the non-X100 Edge things had the X2 option
17:39 < rxr> but not the tiniest X100, some 2 days ago
17:39 < Baldzius> http://www.tcmagazine.com/comments.php?id=31778&catid=2&highlight=x100e
17:40 < Baldzius> Q1 availability
17:40 < rxr> Athlon Neo X2 L335 (Q1 availability),
17:40 < rxr> yeah, as I wrote :-)
17:40 < rxr> hope that's soon
17:40 < rxr> but I "just" got an Nokia Booklet, so maybe I'm not buying again too quickly ...
18:15 -!- digitaloktay [n=pardus@dslb-084-056-233-080.pools.arcor-ip.net] has quit ["Bin mo weg longer............ /exit hilft bei Kopfschmerzen oder hängt ein paar Poster von Merkel auf :)))"]
19:32 < rxr> http://www.spiegel.de/politik/deutschland/0,1518,672943,00.html
19:32 < rxr> via http://blog.fefe.de, ...
20:32 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has quit ["Leaving"]
--- Log closed Thu Jan 21 00:00:21 2010