T2 IRC Log: 2010-02-18

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 Thu Feb 18 00:00:55 2010
00:41 -!- TobiX [tobias@zoidberg.org] has quit [Ping timeout: 264 seconds]
00:42 -!- TobiX [tobias@zoidberg.org] has joined #t2
01:24 < CIA-32> aldas * r35589 /trunk/package/develop/boost/boost.desc: * updated boost (1_41_0 -> 1_42_0)
01:54 < CIA-32> aldas * r35590 /trunk/package/network/bind/bind.desc: * updated bind (9.6.1-P3 -> 9.7.0)
09:11 < rxr> re
10:10 < CIA-32> rene * r35591 /trunk/package/archiver/lbzip2/lbzip2.desc: * updated lbzip2 (0.20 -> 0.21)
10:11 < CIA-32> rene * r35592 /trunk/package/kde/smb4k/smb4k.desc: * updated smb4k (0.10.4 -> 0.10.5)
10:11 < CIA-32> rene * r35593 /trunk/package/base/stress/stress.desc: * updated stress (1.0.2 -> 1.0.3)
10:24 < CIA-32> rene * r35594 /trunk/package/network/squid/squid.desc: * updated squid (3.0.STABLE23 -> 3.0.STABLE24)
10:52 < rxr> p 11 days, 23:35, 4 users, load average: 7.52, 3.84, 1.95
10:53 < rxr> ^- amd stress testing :-)
12:21 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has joined #t2
12:21 < arete_> quit
12:21 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has quit [Client Quit]
12:37 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has joined #t2
12:41 < arete_> quit
12:41 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has quit [Client Quit]
14:48 < CIA-32> rene * r35595 /trunk/scripts/ (Build-Pkg Commit): * fixed Crtl-C vs. Ctrl-C typos in the scripts
14:55 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has joined #t2
14:56 < arete_> hi, i have an error an build dbus in cross compile mode
14:56 -!- Baldzius [~aldas@188.141.89.135] has joined #t2
14:56 < rxr> hi arete_
14:56 < Baldzius> moin
14:56 < rxr> arete_: the one that scrolled by on the T2 mailing list?
14:56 < rxr> moin moin Baldzius
14:56 < Baldzius> hey rxr
14:57 < Baldzius> rxr: I hink boost failure might have something todo with my VM
14:57 < Baldzius> as I don't see any real error
14:57 < Baldzius> builds fine here on laptop
14:57 < rxr> Baldzius: I got a mostly LLVM built T2 system: http://rene.rebe.name/2010-02-16/llvm-clang-support-in-t2-sde-linux/
14:57 < Baldzius> dmesg doesn't show anything
14:57 < Baldzius> do you have any ideas? doesn't seem to be oom issue
14:58 < rxr> would need to take a look into the full log
14:58 < rxr> I just try on my other 64bit ref build
14:59 < arete_> i have send an email to t2 maillist, but no response
14:59 < Baldzius> please , as it builds on my laptop (32 bits)
14:59 < rxr> yeah - overall busy as usual , ...
14:59 < rxr> Baldzius: on my other VM (the xeon one) 32bit build anyway
14:59 < rxr> Baldzius: just the 64bit build was not yet at boost thru rebuilding, manually triggered it now for a test
15:00 < rxr> ok, on the other VM, the AMD one, the one you're on as well, built boost for x86-64
15:00 < rxr> -rw-r--r-- 1 root root 1080634 Feb 17 16:58 build/reference-8.0-trunk-generic-x86-64-linux/var/adm/logs/5-boost.log
15:01 < Baldzius> that might be 1.41.1
15:01 < Baldzius> I've updated it yesterday to 1.42.2
15:01 < Baldzius> but it should be the same
15:01 < Baldzius> so as I guessed something has to do with my VM
15:01 < rxr> my VM is a copy of yours after I insalled it, ...
15:02 < Baldzius> hm
15:02 < rxr> I guess the error is in stage 5 or so anyway, so this is in a fresh bootstrap, anyway?
15:02 < Baldzius> but yours in on xeon right?
15:02 < rxr> arete_: I can later help looing at such a cross buidl issue
15:02 < Baldzius> is*
15:02 < rxr> arete_: you can try with other libtool quirks, or just patch the paths "somehow"
15:03 < rxr> I test build on both
15:03 < rxr> the one that is the copy of your's resists on the same AMD
15:03 < rxr> as yours
15:03 < rxr> can you copy the log somewhere so I can scroll thru it ?
15:03 < Baldzius> ok
15:06 < Baldzius> any service where I could attach file rather copy/paste content?
15:07 < rxr> maybe I should just break into your VM :-)
15:07 < arete_> i have find this solution in openwrt -Wl,-rpath=/usr/lib/
15:07 < Baldzius> sure :)
15:08 < rxr> Baldzius: yeah - with kvm it is not that easy :-), with openvz it was just a chroot :-) vzenter, ...
15:08 < rxr> Baldzius: easiest with kvm would be to inspect your disk image, ...
15:08 < rxr> I guess
15:08 < rxr> anyway, don't want to practise that right now
15:08 < rxr> do you have no http server somewhere?
15:08 < rxr> otherwise just cut'n paste some sniplet where you can find something error-like into a pastebin
15:09 < rxr> arete_: have you tried in t2, does it work, too ?
15:11 < arete_> was is the env variable for LDFALGS?
15:12 < rxr> changing LDFLAGS in the .conf should work
15:12 < rxr> t2 also hs this nice wrappers
15:12 < rxr> I'd usually recomment
15:12 < rxr> however, I noticed recently that libtool does not see them and thus somethimes does stupid, wrong things then
15:12 < rxr> so changing LDFLAGS here should be ok
15:13 < Baldzius> rxr: http://188.141.89.135/ERROR-LOG
15:14 < arete_> thanks rxr,bye
15:14 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has quit [Quit: ircII EPIC4-2.6 -- Are we there yet?]
15:15 < Baldzius> my new cisco cable modem's web interface looks like from past century
15:15 < rxr> heh
15:15 < rxr> Baldzius: no idea, indeed strange
15:15 < rxr> maybe it's the jam exit value
15:15 < rxr> e.g. from "failed updating 2 targets"
15:15 < Baldzius> I tried with debug
15:15 < rxr> maybe it's timing related
15:16 < Baldzius> no luck
15:16 < rxr> strange thing
15:16 < Baldzius> I get no *real* error
15:16 < Baldzius> dmesg ir quite also
15:16 < Baldzius> is*
15:16 < rxr> I guess this boost-jam thing returns an return value != 0
15:16 < rxr> and breaks the build
15:17 < rxr> maybe due this "failed updatign 2 targets" or "skipped 6 targets" ...
15:18 < Baldzius> is it possible to tweak vm just temporary to see if it helps
15:18 < Baldzius> myself I am out of ideas
15:18 < rxr> my build does not have this lines
15:18 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has joined #t2
15:18 < Baldzius> not sure how many packages need boost in real
15:19 < Baldzius> I would ignore if it was some small stupid not needed package
15:19 < rxr> yeah - not many use it, but openoffice I guess
15:19 < Baldzius> not a problem if only that
15:20 < Baldzius> I will skip it for this moment, we'll se how it goes
15:20 < rxr> I guess the error really is where it stopped
15:20 < rxr> common.copy stage/lib/libboost_regex.a
15:20 < rxr> ...failed updating 2 targets...
15:20 < rxr> ...skipped 6 targets...
15:20 < rxr>
15:20 < rxr> on my build it continues there
15:20 < Baldzius> maybe it's optional for others
15:21 < rxr> mine goes on like this:
15:21 < rxr> common.copy stage/lib/libboost_regex.a
15:21 < rxr> gcc.compile.c++ bin.v2/libs/iostreams/build/gcc-4.4.3/release/link-static/threading-multi/file_descriptor.o
15:21 < rxr> gcc.compile.c++ bin.v2/libs/iostreams/build/gcc-4.4.3/release/link-static/threading-multi/mapped_file.o
15:21 < rxr> gcc.compile.c++ bin.v2/libs/iostreams/build/gcc-4.4.3/release/link-static/threading-multi/zlib.o
15:21 < rxr> gcc.compile.c++ bin.v2/libs/iostreams/build/gcc-4.4.3/release/link-static/threading-multi/gzip.o
15:21 < rxr> gcc.compile.c++ bin.v2/libs/iostreams/build/gcc-4.4.3/release/link-static/threading-multi/bzip2.o
15:21 < rxr> gcc.archive bin.v2/libs/iostreams/build/gcc-4.4.3/release/link-static/threading-multi/libboost_iostreams.a
15:21 < rxr> ...on 800th target...
15:21 < rxr> common.copy stage/lib/libboost_iostreams.a
15:21 < rxr> ...updated 800 targets...
15:21 < Baldzius> yeah it must be, it stops at same place without parallel
15:21 < rxr> my VM has 3 times as much RAM though
15:22 < rxr> yours 2G, mine 6G, ...
15:22 < rxr> maybe with this zillion targets this jam thing runs out of it :-)
15:22 < Baldzius> could you swap it for couple of min?
15:22 < rxr> though 2GB is not that tiny
15:22 < Baldzius> I would test
15:22 < Baldzius> yep , indeed
15:22 < rxr> hm, did not configure any kind of memory ballooning
15:23 < Baldzius> your MB maxed out at 8GB?
15:23 < rxr> guess you should get an out of memory oops in dmesg, though
15:23 < rxr> yes, officially
15:23 < Baldzius> ic
15:23 < rxr> though I think it's hard to get 8GB so-dimms anyway
15:23 < Baldzius> ok , no worries
15:23 < Baldzius> I will continue without boost then
15:24 < rxr> well - wait
15:24 < rxr> can you shutdown your vm ?
15:24 < Baldzius> yes
15:24 < Baldzius> now?
15:24 < rxr> shutdown -h 0
15:24 < rxr> just as usual
15:24 < rxr> it's kvm, it's like a real machine :-)
15:25 < Baldzius> done
15:25 * rxr hoping mine comes bach with half the system rebuild with LLVM :-)
15:25 < Baldzius> :)))
15:26 < Baldzius> it's possible to set it in config? llvm over gcc?
15:26 < Baldzius> somehow I coudln't find that
15:26 < rxr> yes, I addad the glue some days ago
15:26 < rxr> in the expert section C compiler
15:26 < rxr> but don't try yet
15:26 < Baldzius> hm
15:26 < rxr> wasted time
15:26 < rxr> won't bootstrap
15:26 < rxr> no cross build setup yet
15:26 < Baldzius> ok
15:27 < rxr> emerge , build into the system works for now, though
15:27 < rxr> for early testing
15:27 < Baldzius> compiles faster?
15:27 < rxr> before I ways days, or weeks getting it to canadian cross compile
15:27 < rxr> they say
15:27 < rxr> I think my build of clang has assertions on, so it's the same speed
15:27 < rxr> can try without assertions though
15:29 < rxr> yours should be up
15:29 < rxr> with 6G
15:30 < Baldzius> I think it showed 3.5
15:30 < rxr> now?
15:30 < rxr> or before ?
15:30 < Baldzius> before 2 now 3,5
15:30 < rxr> hm
15:30 < rxr> -smp 4 -m 6096
15:31 < rxr> you have a 64bit kernel, don't you
15:31 < rxr> oh wait
15:31 < Baldzius> yes shows 64 bits
15:32 < Baldzius> and boost stil fails: /
15:32 < Baldzius> crap
15:32 < rxr> can you paste your free output?
15:33 < Baldzius> total used free shared buffers cached
15:33 < Baldzius> Mem: 3607096 887204 2719892 0 93436 666072
15:33 < Baldzius> -/+ buffers/cache: 127696 3479400
15:33 < Baldzius> Swap: 0 0 0
15:33 < rxr> ok shutdown again
15:33 < rxr> mine vm was not fully down
15:33 < rxr> maybe kvm only grabed as much memory as was avail
15:33 < rxr> sstrange thing, expected it to error out, ...
15:33 < rxr> good thing is, my llvm VM still comes up
15:34 < rxr> first boot after sysvinit and co where rebuild :)
15:34 < rxr> pretty solid thing I'd say :-9
15:34 < Baldzius> done
15:36 < rxr> up again
15:36 < rxr> more ram, now ?
15:36 < Baldzius> same
15:36 < Baldzius> total used free shared buffers cached
15:36 < Baldzius> Mem: 3607096 81808 3525288 0 6492 37604
15:36 < Baldzius> -/+ buffers/cache: 37712 3569384
15:36 < Baldzius> Swap: 0 0 0
15:36 < rxr> what?
15:36 < arete_> quit
15:36 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has quit [Quit: ircII EPIC4-2.6 -- Are we there yet?]
15:37 < rxr> srange
15:37 < rxr> ok - then I probably had not more eiither
15:37 < rxr> either
15:37 < Baldzius> :)
15:37 < rxr> maybe kvm needs something for more ram, hugetlbfs or so
15:37 < rxr> ok - then it's less likely memory anyway, as then mine had only less in the VM container anyway, too :.-9
15:37 < rxr> good that we notice early :-9
15:37 < Baldzius> ok you can roll out same settings as before
15:37 < rxr> your vm has reiserfs as FS, too, right?
15:37 < Baldzius> I will skip boost
15:38 < Baldzius> ext4
15:38 < rxr> ah yes, /me too
15:38 < rxr> ok then it's also not the FS
15:39 < rxr> (usually I still use reiser3, for the performance in my scenarios of it)
15:39 < rxr> as I said when I tested the new box I made a copy of the test install, one for you one for me, the basics before I mangled mine with LLVM where the same
15:39 < Baldzius> so should I brind vm down now?
15:39 < rxr> but the reference test build is not affected of this as it's in the sandbox with the same gcc anyway
15:40 < rxr> yeah - take it down, I start with the config as before
15:40 < Baldzius> done
15:42 < rxr> should be up
15:42 < rxr> indeed, I also only got some 3.5GB of RAM
15:42 < rxr> good to know
15:43 < rxr> though the qemu process eastst the 6GB
15:43 < rxr> why the heck does the gust OS not see it
15:44 < Baldzius> strange indeed
15:45 < rxr> Memory: 3597892k/3670016k available (2574k kernel code, 388k absent, 71008k reserved, 1308k data, 420k init)
15:45 < rxr> hm - that is what the e820 map from kvm says
15:46 < rxr> ont he xeon VM my guest has 4GB, as much as I gave it
15:46 < rxr> maybe a new kvm bug
15:46 < rxr> will hunt another day
15:46 < rxr> should get back to work
16:04 < rxr> I think I'll try updating kvm later
16:04 < rxr> like tomorrow or the weekend
16:06 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has joined #t2
16:09 < Baldzius> cmake version compiles fine
16:14 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has quit [Quit: ircII EPIC4-2.6 -- Are we there yet?]
16:14 < rxr> please don't commit that
16:15 < rxr> maybe use locally for testing, this needs more review
16:15 < rxr> the cmake version broke a couple of my builds, ...
16:15 < rxr> and personally I do not find cmake particularly maintainable
16:15 < rxr> read: I won't like to hunt, fix bugs in cmakelist.txt fiels
16:15 < rxr> files even
16:15 < rxr> openoffice update would be interesting though
16:15 < rxr> particularly one that builds with gcc-4.4
16:20 < Baldzius> OOo build still goes here
16:20 < Baldzius> yeah i see cmake version gives some trouble
16:20 < Baldzius> one package compiled fine, the other just failed
16:21 < Baldzius> stupid boost
16:22 < Baldzius> OOo failed also
16:22 < Baldzius> feck
16:22 < rxr> you'll have fun
16:22 < rxr> as I have with my other work, and my LLVM/clang testing, and the upcoming KVM update etc. :-)
17:15 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has joined #t2
17:31 < CIA-32> rene * r35596 /trunk/package/emulators/kvm/kvm.desc: * updated kvm (0.12.1.2 -> 0.12.2)
17:37 -!- arete_ [~arete@93-44-145-163.ip97.fastwebnet.it] has quit [Quit: ircII EPIC4-2.6 -- Are we there yet?]
17:37 < Baldzius> seems that package format is giving troubles for OOo build
17:37 < Baldzius> will try different one
17:38 < Baldzius> and wait another 5 hours to see if it passes through
17:39 < Baldzius> my poor laptop
17:39 -!- T-One [~T-One@62.218.122.66] has quit [Quit: Verlassend]
17:43 < rxr> use your vm :-)
17:45 < Baldzius> ok, will try
17:46 < rxr> but if you are around for some more minutes (read 60 minutes) we could try an kvm update on the box :-9
17:46 < rxr> if I would have some more spare hardwrae I could live migrate it from one silion to another, but I haven't spare VM capacity at the momemnt ...
17:47 < Baldzius> I have ~50 minutes
17:47 < rxr> oh
17:47 < rxr> ok, let's see if that is enough
17:48 < Baldzius> will go to uni for cisco classes
17:48 < rxr> lucky you
17:48 < Baldzius> do you want me to bring vm down for you?
17:48 < rxr> ok, then do now
17:48 < rxr> the 60m was more to coordinate that when the kvm update reached the bare metal OS, ...
17:49 < rxr> in any case it will be back online when the update is thru
17:49 < rxr> hope soon, still building
17:51 < Baldzius> let me know when done, I'll try to see how openoffice build goes
17:51 < Baldzius> at least in begining
17:54 < rxr> have you -h or -r'ed? your's still up
17:55 < Baldzius> downloading OOo :)
17:57 < rxr> ok then let it
17:57 < rxr> new kernel still building
17:58 < Baldzius> heh, I was lucky my connection freezed typing shutdown -h
17:59 < Baldzius> I experience those freezes from time to time
17:59 < Baldzius> not sure if they are from my side
17:59 < rxr> hm - maybe our line is just to filled up then
18:02 < rxr> ok - guess I could reboot anytime now
18:02 < rxr> kernel done
18:02 < Baldzius> give me a min
18:03 < rxr> ok
18:03 < rxr> let me know
18:03 < rxr> cool
18:03 < rxr> new kvm version fixed the memory bug it appears
18:03 < rxr> or the new seabios, whatever
18:03 < Baldzius> done
18:03 < rxr> now got 6G in my VM
18:04 < Baldzius> let's hope it fixes my boost
18:04 < rxr> doubt :-)
18:05 < rxr> do you think 2GB are ok? or should I permanently increase to let's say 3 ?
18:05 < Baldzius> I'd say 2GB is really enough
18:06 < rxr> should be up again
18:07 < rxr> Mem: 6119452 87928 6031524 0 8536 38336
18:07 < rxr> mine
18:10 < Baldzius> feels slower
18:10 < rxr> hm
18:11 < Baldzius> way slower
18:11 < rxr> ok - let me adjust my memory
18:11 < rxr> maybe overcommited
18:11 < rxr> better now ?
18:11 < rxr> only network i/o or cpu time?
18:12 < Baldzius> super
18:12 < Baldzius> now good
18:12 < rxr> you notice quickly, how did you notice ?
18:13 < Baldzius> svn st was taking ages
18:14 < rxr> stil good?
18:15 < Baldzius> looks good
18:15 < Baldzius> only network seems dodgy
18:15 < rxr> worse then before, the last days ?
18:23 < Baldzius> I'd say
18:23 < Baldzius> but it's fine
18:23 < Baldzius> no worries
18:23 < rxr> maybe I schould do some traffic shaping in favour of SSH packets :-)
18:24 < Baldzius> ok , have to go
18:24 < Baldzius> see you later
18:25 -!- Baldzius [~aldas@188.141.89.135] has quit [Quit: Leaving]
18:25 < rxr> cu, have fun
18:57 -!- aloof [~kjj@pool-74-107-128-126.ptldor.fios.verizon.net] has joined #t2
23:34 -!- LMJ [~serwou@laf31-4-82-236-42-164.fbx.proxad.net] has quit [Quit: LMJ]
23:51 -!- Smashtoad [~smashtoad@pool-173-66-233-96.washdc.fios.verizon.net] has joined #t2
23:52 < Smashtoad> anyone able to give me a hand installing t2 for the first time
23:52 < Smashtoad> I am getting a grub error
23:53 -!- cem [~cem@chello062178214022.10.15.vie.surfer.at] has joined #t2
23:53 < cem> disco
23:53 -!- digitaloktay [~pardus@dslb-084-056-236-096.pools.arcor-ip.net] has joined #t2
23:54 < Smashtoad> grub isnt mapping devices on install
23:55 -!- Smashtoad [~smashtoad@pool-173-66-233-96.washdc.fios.verizon.net] has left #t2 []
--- Log closed Fri Feb 19 00:00:55 2010