1285033382 Q * cehteh Ping timeout: 480 seconds 1285033394 J * cehteh ~ct@pipapo.org 1285033502 Q * fLoo 1285033694 J * Janno Janno@server3.raumopol.de 1285035588 J * eja ~user@c-219.redwoods.edu 1285035645 M * eja anyone have experience with linux-vserver and iscsi targets? 1285035923 M * kupo Bertl: ? 1285035928 M * Bertl eja: i.e. the guest as initiator? 1285035930 M * kupo problem with what? 1285035951 M * eja guest as target 1285035980 M * Bertl kupo: < kupo> does anyone have any experience running a tor exit node on vserver? 1285036003 M * kupo yes I'm choosing vps providers and they only offer vserver's 1285036010 M * kupo and I want to run a tor exit node 1285036020 M * kupo I'm wondering if there's any special considerations 1285036032 M * Bertl what I mean is, did you try and encounter any issues? 1285036053 M * kupo I've not purchased the vps yet, so no 1285036067 M * kupo but this information would help inform my choice of purchase 1285036080 M * Bertl eja: everything userspace should be fine 1285036101 M * Bertl kupo: I don't see why tor shouldn't work in a guest 1285036698 M * kupo ok 1285036707 M * kupo know anyone who's done it? 1285036717 M * Bertl nope 1285036737 M * kupo ok 1285037158 Q * Piet Ping timeout: 480 seconds 1285037803 J * Piet ~Piet__@82VAABDY8.tor-irc.dnsbl.oftc.net 1285039146 Q * balbir_ Ping timeout: 480 seconds 1285040245 Q * niki Quit: Ex-Chat 1285045373 M * Bertl off to bed now ... have a good one everyone! 1285045378 N * Bertl Bertl_zZ 1285047071 J * ntrs ~ntrs@77.28.24.180 1285048536 Q * arekm Quit: leaving 1285048894 J * arekm arekm@carme.pld-linux.org 1285048987 J * ncopa ~ncopa@180.40.189.109.customer.cdi.no 1285049352 J * sharkjaw ~gab@90.149.128.29 1285050002 Q * ntrs Read error: Connection reset by peer 1285050014 J * ntrs ~ntrs@77.29.196.167 1285050238 Q * derjohn_mob Ping timeout: 480 seconds 1285051441 Q * urbee Ping timeout: 480 seconds 1285051742 J * derjohn_mob ~aj@213.238.45.2 1285051990 J * ghislain ~AQUEOS@adsl2.aqueos.com 1285052224 J * alpha_one_x86 ~kvirc@131.30.17.95.dynamic.jazztel.es 1285052372 M * alpha_one_x86 setattr crash on my systeme 1285052671 J * petzsch ~markus@dslb-092-075-201-186.pools.arcor-ip.net 1285052872 Q * Hollow Quit: Lost terminal 1285052884 J * Hollow ~bene@shiva.xnull.de 1285052945 M * alpha_one_x86 http://pastebin.com/By2wQSM1 1285053205 M * alpha_one_x86 Environement variable see have influance: http://pastebin.com/z16nSWZT 1285053547 M * alpha_one_x86 I'm into 2.6.32 with corresponding vserver + util-vserver-0.30.216_pre2910 1285054175 J * thierryp ~thierry@zankai.inria.fr 1285055072 J * kir ~kir@swsoft-msk-nat.sw.ru 1285055176 J * omar ~omar@host27-68-static.24-87-b.business.telecomitalia.it 1285055268 P * kir 1285056127 J * petzsch1 ~markus@dslb-092-075-201-186.pools.arcor-ip.net 1285056232 J * michal ~michal@www.rsbac.org 1285056237 M * michal hey :) 1285056371 J * barismetin ~barismeti@zanzibar.inria.fr 1285056453 Q * petzsch Ping timeout: 480 seconds 1285056672 M * daniel_hozac alpha_one_x86: does your dietlibc work? 1285056677 J * imcsk8 ~ichavero@evdomip-75-146.iusacell.net 1285056718 M * alpha_one_x86 how check it? 1285056731 M * daniel_hozac diet -Os gcc -o helloworld helloworld.c 1285057014 Q * wibble Remote host closed the connection 1285057064 M * alpha_one_x86 this command work perfectly 1285057126 Q * ncopa Quit: Ex-Chat 1285057185 J * ncopa ~ncopa@180.40.189.109.customer.cdi.no 1285057390 M * daniel_hozac executing helloworld as well? 1285057521 M * harry i need 2.6.32.22 patches! :) 1285057560 M * daniel_hozac the previous 2.6.32 patch doesn't apply? 1285057605 M * harry don't know :) 1285057611 M * harry probably will, yeah :) 1285057628 M * harry but actually, i just want the "latest" for 2.6.32 1285057657 M * harry it's still at 2..29.6 1285057669 M * daniel_hozac port it then. 1285057673 M * harry while there is a 2.blabla, yadda yadda.32 1285057776 M * harry it's better done by someone who writes the codebase 1285057786 M * harry like... Bertl !!! :) 1285058008 Q * petzsch1 Quit: Leaving. 1285058312 Q * imcsk8 Quit: This computer has gone to sleep 1285058432 J * BenG ~bengreen@cpc6-aztw22-2-0-cust100.aztw.cable.virginmedia.com 1285059305 M * ntrs daniel_hozac, I am using the latest patch and utils and I am having trouble stopping a guest. Can't even kill the processes inside the guest with vkill 1285059367 J * Romster ~romster@202.168.100.149.dynamic.rev.eftel.com 1285059370 M * daniel_hozac latest meaning what? 1285059375 M * daniel_hozac and what does dmesg say? 1285059378 M * daniel_hozac and ps? 1285059402 M * ntrs there is nothing out of the ordinary in dmesg 1285059408 M * ntrs # vserver guest exec ps auwwx 1285059408 M * ntrs USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND 1285059408 M * ntrs root 1 0.0 0.1 10408 736 ? Ss Sep12 0:34 init [3] 1285059408 M * ntrs root 1231 0.0 0.1 3708 996 ? R+ 08:53 0:00 ps auwwx 1285059408 M * ntrs root 9876 0.0 0.0 1484 288 ? D 08:24 0:00 reboot -d -f -i 1285059409 M * ntrs # vkill --xid 163 9876 1285059410 M * ntrs # vserver guest exec ps auwwx 1285059412 M * ntrs USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND 1285059414 M * ntrs root 1 0.0 0.1 10408 736 ? Ss Sep12 0:34 init [3] 1285059416 M * ntrs root 1231 0.0 0.1 3708 996 ? R+ 08:53 0:00 ps auwwx 1285059418 M * ntrs root 9876 0.0 0.0 1484 288 ? D 08:24 0:00 reboot -d -f -i 1285059425 M * ntrs This happened when stopping the guest 1285059437 M * ntrs now it cannot be restarted because it is not stopped yet. 1285059441 M * daniel_hozac well, it's stuck in kernel space. 1285059455 M * daniel_hozac what kernel and utils, exactly? 1285059494 M * ntrs 2.6.35.4-vs2.3.0.36.32 1285059502 M * ntrs 0.30.216-pre2913 1285059543 M * daniel_hozac how about a vps faux on the host. 1285059698 M * ntrs lots of output 1285059704 M * ntrs What about it? 1285059723 M * daniel_hozac pastebin is usually a good idea. 1285059740 M * alpha_one_x86 daniel_hozac: helloworld executed perfectly 1285059865 M * ntrs vps faux shows the same processes on the stuck guest 1285059933 M * daniel_hozac yes, the guest isn't interesting. 1285059955 M * alpha_one_x86 any idea for my crash? 1285060002 M * daniel_hozac what arch? 1285060010 M * alpha_one_x86 x86_64 1285060015 M * daniel_hozac which dietlibc? 1285060047 M * alpha_one_x86 0.30-r2 1285060053 M * ntrs x86_64 1285060069 M * ntrs daniel_hozac, were your last two questions for me or for alpha_one_x86 1285060077 Q * sharkjaw Quit: Leaving 1285060086 M * daniel_hozac for alpha_one_x86 1285060108 M * daniel_hozac alpha_one_x86: IIRC you need 0.31 for recent gcc's. 1285060173 M * alpha_one_x86 I have same version on my desktop and all work 1285060205 M * daniel_hozac that looks like Gentoo though, so you probably have different CFLAGS causing the problems to trigger... 1285060256 M * ntrs daniel_hozac, also, starting a guest that was stopped hangs with this process in D+ state: 1285060271 M * ntrs '/usr/lib64/util-vserver/secure-mount -a --chroot --fstab /etc/vservers/guest/fstab --rootfs no 1285060300 M * ntrs is it just waiting until it processes the disk utilization checking? 1285060315 M * daniel_hozac did you recently retag one of your guests? 1285060326 M * daniel_hozac or tag /dev? 1285060327 M * ntrs any of them? 1285060342 M * alpha_one_x86 I have force my CFLAGS to -O2 -pipe 1285060343 M * ntrs not just the ones causing the problem? 1285060365 M * daniel_hozac the ones with the problem. 1285060417 Q * BenG Quit: I Leave 1285060441 M * ntrs no, they had the same tags as always 1285060596 M * daniel_hozac what does strace say about secure-mount? 1285060888 M * alpha_one_x86 make: *** [bin-x86_64/elftrunc] Segmentation fault with dev-libs/dietlibc-0.33_pre20100626 1285061532 M * harry daniel_hozac: do you sitll use dietlibc 0.30 ? 1285061537 M * harry or a later one? 1285061555 M * harry (i saw in your repo's the 0.30 one only, that's why i ask ;)) 1285061697 M * alpha_one_x86 here only diet version 0.30 cna be installed 1285061732 M * harry #define here 1285062205 J * petzsch ~markus@dslb-092-075-201-186.pools.arcor-ip.net 1285062277 M * daniel_hozac yes, i still use a patched 0.30. 1285063052 Q * petzsch Read error: Connection reset by peer 1285064634 M * mnemoc hi, has anyone tried to port 2.6.32.21 patch to 2.6.32.22 yet? there are massive changes in the scheduler code 1285064673 M * mnemoc they seem to have backported the whole scheduler from 2.6.35... 1285065793 M * Wonka btw, 2.6.35.5 is out, 428b01232e08df89751ed5d1683953ebcc6a9035 being one of the more interesting changes 1285066130 M * mnemoc i hope they finally solved the scheduler stalling problems 1285066174 Q * barismetin Remote host closed the connection 1285067248 J * jordi ~jordi@115.Red-213-96-69.staticIP.rima-tde.net 1285067283 M * jordi hi 1285067317 M * jordi I just upgraded a legacy server that was running Debian etch to lenny 1285067345 M * jordi suddenly a bunch of mysql servers in containers don't work. 1285067389 M * jordi it seems they should never have worked actually, as the config has a bind-address for 127.0.0.1 only and some failing apps can't contact the remote mysql server. 1285067448 M * jordi so I'm puzzled, what changed in lenny vserver that "broke" this? 1285067510 M * jordi oh, some of these failing mysqls are old too (etch) 1285068370 M * ghislain i think you upgraded the vserver from a version to another and that your tools/settings are not compatible no ? 1285068381 M * ghislain like a vserver 2.1 to 2.3 1285068445 M * ghislain you upgraded the host or the guest ? 1285068734 Q * ntrs Ping timeout: 480 seconds 1285069370 M * jordi ghislain: I upgraded the hosts 1285069384 M * jordi ie, no vserver guest changes 1285069561 N * Bertl_zZ Bertl 1285069566 M * Bertl morning folks! 1285069595 M * Bertl jordi: what was the old kernel version and what is the new one? 1285069620 M * jordi Bertl: 2.6.18 vs 2.6.26 1285069665 M * Bertl http://linux-vserver.org/Installation_on_Debian scroll down to the 'Issues' section 1285069703 M * alpha_one_x86 I can't upgrade dietlibc, setattr crash under chroot bug not into the chroot 1285069740 M * Bertl jordi: in short, I definitely cannot recommend the 2.6.26 debian kernel for anything 1285069792 M * Bertl jordi: also make sure to get recent util-vserver, otherwise your guest isolation will be incomplete 1285069859 M * Bertl alpha_one_x86: dietlibc is not something used once stuff is compiled, it is more like gcc, i.e. used to _build_ stuff 1285069902 M * Bertl i.e. you can upgrade and build util-vserver, then downgrade back to your version (not sure that helps, as your problem is in the different build options) 1285069911 M * alpha_one_x86 what do? because "setattr" command crash by segfault 1285069926 M * alpha_one_x86 I have try all version of util-vserver 1285069945 M * Bertl what achitecture do you build on? 1285069952 M * Bertl *architecture 1285069952 M * alpha_one_x86 x86_64 1285069965 M * Bertl and you are building for that arch as target? 1285069984 M * alpha_one_x86 on my desktop, all work, included dietlibc, but into chroot (or virtual machine) all failed 1285070020 M * Bertl what is your kernel? 1285070052 M * alpha_one_x86 2.6.32 1285070062 M * Bertl any grsec or selinux? 1285070094 M * alpha_one_x86 grsec under the virtual machine, vanilla kernel under chroot 1285070126 M * Bertl and on the host setattr works, but in the chroot it fails for you? 1285070151 Q * DreamerC Quit: leaving 1285070203 M * alpha_one_x86 I have install util-vserver on host and os into chroot, same os, same architecture, work on host, but not on guest 1285070222 M * alpha_one_x86 same when the guest is running into qemu virtual machine 1285070245 M * Bertl okay, let's get the nomenclature correct here .. chroot != guest 1285070253 M * alpha_one_x86 sorry 1285070284 M * Bertl because inside a 'guest' (Linux-VServer) you cannot execute Linux-VServer specific syscalls 1285070286 M * alpha_one_x86 both is gentoo, same use flags, execept that's my destkop have kde and debug enable 1285070335 M * Bertl qemu/kvm should behave the very same than your host system, so no differences to expect there, given that the OS is identical 1285070342 M * jordi Bertl: wow, this is pretty bad 1285070354 J * ntrs ~ntrs@77.29.196.167 1285070359 M * alpha_one_x86 gentoo desktop: kde4 + use flags debug -> chroot -> gentoo server hardened + vserer + grsec for the compiled kernel (when running into vm) 1285070399 M * Bertl you probably need to do a bunch of adjustments to run Linux-VServer stuff on hardened + grsec 1285070399 M * jordi Bertl: I've experienced no problems with my vserver setups running on lenny though 1285070402 M * jordi although the barriers problem seems bad enough 1285070449 M * Bertl alpha_one_x86: but let's get back to the chroot 1285070477 M * Bertl so, on the 'host' setattr works like a charm, in your chroot it segfaults, yes? 1285070524 M * alpha_one_x86 yes 1285070547 M * Bertl okay, do you have some disk space left? 1285070582 M * alpha_one_x86 yes 1285070619 M * Bertl good, I'd suggest to make a 'copy' of your host system into a directory, best do that with a low level tool like dump/restore 1285070653 M * Bertl for example, if your host system has / and /usr as mount points, and the filesystem is ext2/3, you'd do something like this: 1285070663 M * Bertl mkdir /test 1285070664 J * DreamerC ~DreamerC@122-116-181-118.HINET-IP.hinet.net 1285070667 M * Bertl cd /test 1285070676 M * Bertl dump 0f - / | restore rf - 1285070680 M * jordi Bertl: still, in the issues list I see nothing that affects networking 1285070680 M * Bertl mkdir usr 1285070685 M * Bertl cd usr 1285070689 M * jordi Bertl: is the squeeze kernel any beter? 1285070690 M * Bertl dump 0f - /usr | restore rf - 1285070691 M * jordi +t 1285070698 M * alpha_one_x86 yes I see 1285070719 M * Bertl jordi: I hope so 1285070742 M * jordi Bertl: it's marked as deprecated in the release notes though 1285070744 M * alpha_one_x86 my host system have custom CFLAGS I thing to: -march=native -g ... 1285070759 M * alpha_one_x86 after this systeme, what I do? 1285070760 M * Bertl jordi: the network problems might be the result of single IP special casing 1285070787 M * jordi Bertl: can you elaborate on that? 1285070813 M * Bertl alpha_one_x86: when you copied the host filesystem into the /test dir, chroot into that and try to run setattr to see if that segfaults too 1285070849 M * alpha_one_x86 no segfault in this case 1285070852 M * Bertl jordi: between 2.6.18 (stable patch) and 2.6.23+ (development/testing patches) changed a lot 1285070877 M * Bertl alpha_one_x86: so that means that your chroot build is messed up, and it's not a problem with the chroot per se 1285070888 M * Bertl (the one you tested before) 1285070889 M * alpha_one_x86 same if I copy setattr into my system into chroot, chroot ., cd /root ; ./setattr 1285070903 M * Bertl yeah, kind of expected that 1285070926 M * Bertl i.e. your toolchain (gcc/ld/dietlibc) doesn't work as expected in your chroot/kvm 1285070946 J * BenG ~bengreen@cpc6-aztw22-2-0-cust100.aztw.cable.virginmedia.com 1285070949 M * alpha_one_x86 only the setattr binary on the systeme segfault, other application work (top, nano, ...) 1285070973 M * Bertl jordi: one change regarding IP isolation was the lback mapping and the special casing 1285070987 Q * BenG 1285070999 M * Bertl alpha_one_x86: so you have been lucky with top/nano .. do they even use dietlibc? 1285071036 M * alpha_one_x86 I have never use that's, and only util-vserver use that's 1285071070 M * Bertl so chances are good that your problem is a misconfigured/miscompiled dietlibc 1285071116 M * Bertl check your build options, grab a new version (source), rebuild it properly and configure util-vserver to use the same build options 1285071126 M * Bertl (and all your problems will go away) 1285071203 M * Bertl http://www.mail-archive.com/gentoo-hardened@lists.gentoo.org/msg02687.html 1285071203 M * alpha_one_x86 execpte debug use flags it's same options, I have try all version of util-vserver and only the version 0.30 of dietlibc can be compiled into the chroot 1285071262 M * alpha_one_x86 Wed, 21 Oct 2009 -> old, have change since 1285071274 M * Bertl I wouldn't count on it 1285071305 M * Bertl and why would only the 0.30 dietlibc work in your chroot? 1285071328 M * alpha_one_x86 because the other version segfault at the compilation 1285071407 M * alpha_one_x86 I'm trying recompil all my packet directly into qemu to isolate the environement variable of my desktop machine 1285071413 M * Bertl doesn't that give you a hint that your toolchain is broken? 1285071439 M * Bertl i.e. that dietlibc (>= 0.30) segfaults at compilation time 1285071454 M * alpha_one_x86 I have only difficult only on dietlibc and util-vserver, all other application compil and work perfectly 1285071456 M * Bertl (where it compiles fine for everybody else) 1285071508 M * Bertl alpha_one_x86: once again, util-vserver definitely pushes the limits, so sure, top might work fine, despite the fact that your toolchain is broken 1285071532 M * alpha_one_x86 how I can repair it? 1285071549 M * Bertl also, as you said yourself, nothing except util-vserver uses dietlibc on your system 1285071566 M * alpha_one_x86 I have try reinstall all the linux gentoo from the base 3x 1285071603 M * Bertl so, assuming that the dietlibc build (0.30) already failed (because of the toolchain) it is not really surprising that util-vserver built with that toolchain and the dietlibc fails too (in subtle ways) 1285071705 M * Bertl I'd start with building a recent dietlibc and try different gcc/binutils for that 1285071734 M * alpha_one_x86 I will try that then 1285071752 M * Bertl once you find a gcc/binutils combination which compiles dietlibc 0.32 fine, you should be fine with util-vserver too 1285071919 M * alpha_one_x86 gcc, I can only use the current version, because the other is bloqued by gentoo 1285071944 M * Bertl well, that is a gentoo problem I guess 1285072231 M * alpha_one_x86 or I can overwrite setattr elf binary 1285072240 M * alpha_one_x86 from working systeme 1285072292 M * Bertl sure, if that works for you, why not 1285072335 M * Bertl but it is rather likely that this isn't the only binary which got miscompiled 1285073240 M * harry Bertl: any chance on a backport of the features for 2.6.32 series? 1285073263 J * barismetin ~barismeti@zanzibar.inria.fr 1285073273 M * Bertl harry: backport of what? 1285074678 Q * ex Ping timeout: 480 seconds 1285074843 M * harry don't know 1285074859 M * harry there's a 2..29.6 for 2.6.32 series 1285074872 M * harry and 2..32 for 2.6.35 1285074896 M * harry i asume there are feature differences 1285074907 M * Bertl there are mostly implementation differences 1285074919 M * harry no feature thingy's then... 1285074920 M * harry super :) 1285074940 M * harry just wanted to make sure that it makes sense to keep building for 2.6.32 kernels 1285074960 M * harry (since those are the base of ubuntu LTS and probably rhel6 and suse X etc... 1285075609 M * Bertl yeah, well, hopefullt they all soon update to working kernels (2.6.35+) 1285075614 M * Bertl *fully 1285076142 Q * alpha_one_x86 Quit: KVIrc Equilibrium 4.1.0, revision: 4696, sources date: 20100519, built on: 2010-08-29 21:19:12 UTC http://www.kvirc.net/ 1285077333 Q * michal Ping timeout: 480 seconds 1285077400 M * harry what's wrong with 2.6.32 ? 1285077433 M * Bertl IMHO basically unuseable for desktop and server 1285077443 J * michal ~michal@www.rsbac.org 1285077451 M * harry why's that? 1285077462 M * harry it seems to work fine on my new hetzner server 1285077477 M * Bertl under moderate I/O load, the kernel shows 'hickups' which can take 10-20 seconds 1285077507 M * Bertl mkdir over NFS takes 5-10 minutes!! unless you issue a sync command on the server, etc .. 1285077513 A * jordi hasn't experienced that. 1285077562 M * Bertl I'm experiencing that on at least 3 different machines with similar load 1285077596 M * harry and that's all fixed in 2.6.35+ ? 1285077613 M * harry what strikes me the most is... it's still the kernel in ubuntu 2.6.32 1285077616 M * Bertl I hope so, haven't got around testing the latest kernel releases on those systems 1285077625 M * harry so if the problems are that serious, they should know this, right? 1285077648 M * Bertl not sure anybody cares 1285077667 M * harry ahm... hickups of 10-20 seconds WILL have people care 1285077679 M * harry certainly in a LTS release 1285077702 M * Bertl if they just use a web browser, they won't experience it, will they? 1285077727 M * Bertl I'm talking about systems running quite some stuff in parallel 1285077745 M * Bertl but note that this worked before, e.g. with 2.6.22 quite fine 1285077964 A * hijacker has no delays with 2.6.33.2-vs2.3.0.36.30.4 1285078061 A * harry compiles a kernel in 1min 20sec on my new machine 1285078076 M * harry i asume that does some i/o... no performance problems at all 1285078150 M * Bertl create an encrypted volume, build the kernel on that with atime enabled 1285078172 M * Bertl I'm pretty sure that you will run into issues with that 1285078187 M * harry hmm... i don't have any time left today 1285078197 M * harry but i will have to test that... that's for sure 1285078201 M * harry is it a known bug? 1285078262 M * Bertl I guess so 1285078277 J * dna ~dna@dslb-088-074-220-247.pools.arcor-ip.net 1285078430 M * harry hmm.... /me has some problems in that reasoning... 1285078439 M * harry if it's a known bug, why can't i find anyone having it? 1285078454 M * harry nor anything in patches, nor with ubuntu, redhat/fedora/suse/... 1285078468 M * harry why are there no patches talking about it? 1285078487 A * harry has ubuntu 10.04 here, with a 2.6.32 kernel, fully encrypted homedir 1285078493 M * harry never seen problems with IO load 1285078565 M * biz_ well, cfq low latency mode was introduced in 2.6.32? http://kernelnewbies.org/Linux_2_6_32#head-e07f19bb79beab04cb9e68b942c9041b490b71db 1285078623 M * biz_ I'm running around 10 servers with NFS and some with luks+lvm+mdraid on 2.6.32 in production and also don't see these hickups 1285078692 M * harry biz_: that stuff is FIXED/improved in 2.6.32 1285078695 M * harry not worsened ;) 1285078713 M * harry if it would cause hickups, it wouldn't be in "the cool stuff" ;) 1285078722 M * Bertl harry: first, you found somebody (me) who experiences those issues :) 1285078734 M * Bertl with and without Linux-VServer that is 1285078738 M * biz_ Bertl: perhaps all of your servers share something similar hardware-wise whose driver is buggy? hrm... 1285078770 M * biz_ harry: but it's important to know that it defaults to _on_ 1285078775 M * Bertl no, definitely not, the systems I tested with have nothing in common 1285078797 M * harry Bertl: true :) 1285078811 A * harry has to go... massive servermove ! :) 1285078822 M * harry (aix stuff, power6 machines ;)) 1285078831 M * harry soon i'll play with power7!!! ;) 1285078846 M * biz_ :) 1285078904 M * daniel_hozac i see that too. 1285079047 M * Bertl for example, on my desktop, when playing a full HD recording (consumes about 5% cpu) from NFS, writing to a different NFS share (e.g. unpacking a kernel tree) will cause the playback to stall, despite the fact that the unpack is ioniced and the player has a raised priority 1285079123 M * Bertl if, for whatever reason, the NFS share runs full, then the desktop will go crazy and you cannot even logon anymore 1285079149 M * Bertl (well, that is an NFS issue, unless you use sync/slow NFS) 1285079236 M * Bertl note that this is a quad core2 cpu with 4GB memory and GBit ethernet, so no real limitation on the resources there 1285079312 M * biz_ sounds b0rked indeed... :/ 1285079376 M * biz_ I'm using nfs4 in soft mode with lots of sequential readers and writers accessing the share from different machines, however I did not encounter it (other services on the server and clients are always pretty responsive) 1285079410 M * daniel_hozac i'm not using NFS at all, just LVM on dm-crypt. 1285079460 M * Bertl yeah, it might be related to the LVM/dm-crypt stuff as well 1285079483 M * Bertl (which definitely is a common (software) element on my systems) 1285079532 M * biz_ and which IO scheduler do you use? I'm on CFQ (to support ionice) with low_latency mode disabled 1285079552 M * Bertl CFQ on the desktop, deadline on the server machine 1285079643 M * daniel_hozac yeah, CFQ here as well. 1285079680 M * biz_ the strange this is really, like harry, that there's not much noise about that on LKML as far as I follow it. I'd expect this to be a real show-stopper for any LTS kernel 1285079682 M * Bertl I did setup a test screnario a few weeks ago which basically emphasized the issues 1285079716 M * Bertl i.e. take 5 physical disks, encrypt each of them with dm-crypt (any cipher will do) 1285079728 M * Bertl and combine them to a raid 5/6 setup 1285079752 M * Bertl put a journaled filesystem ext3/jfs/xfs ontop of that 1285079774 M * Bertl and do some burst read/writes side by side with 'normal' kernel compiling 1285079822 M * Bertl I would be surprised if any 2.6.32 distro kernel can handle that nicely 1285079901 M * biz_ I don't really care for LTS on the desktop... so I guess it's not worth the effort and just skip to 2.6.35+ if it's fixed there 1285080404 J * balbir_ ~balbir@122.172.35.114 1285080487 J * petzsch ~markus@dslb-094-218-178-167.pools.arcor-ip.net 1285080715 M * Bertl off for a nap now ... bbl 1285080721 N * Bertl Bertl_zZ 1285081047 Q * thierryp Ping timeout: 480 seconds 1285081433 Q * ncopa Quit: Ex-Chat 1285082050 Q * balbir_ Read error: Connection reset by peer 1285082752 Q * barismetin Remote host closed the connection 1285082842 J * balbir_ ~balbir@122.172.162.206 1285082984 J * barismetin ~barismeti@zanzibar.inria.fr 1285083696 Q * petzsch Read error: Connection reset by peer 1285084408 J * thierryp ~thierry@home.parmentelat.net 1285084494 J * petzsch ~markus@dslb-094-218-178-167.pools.arcor-ip.net 1285084924 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d 1285085214 Q * petzsch Quit: Leaving. 1285085735 Q * barismetin Remote host closed the connection 1285086167 Q * omar Quit: Sto andando via 1285086457 J * petzsch ~markus@dslb-094-218-178-167.pools.arcor-ip.net 1285086497 Q * petzsch 1285086880 P * Snow-Man 1285087650 Q * dna Read error: Connection reset by peer 1285087664 J * dna ~dna@dslb-088-074-220-247.pools.arcor-ip.net 1285088223 J * ex ex@valis.net.pl 1285088982 Q * hel Remote host closed the connection 1285089390 J * hel ~hel@porthos.lennackers.de 1285091578 Q * derjohn_mob Ping timeout: 480 seconds 1285093196 Q * ntrs Read error: Connection reset by peer 1285093208 J * ntrs ~ntrs@77.28.16.52 1285093453 J * petzsch ~markus@dslb-094-218-178-167.pools.arcor-ip.net 1285094627 J * remember ~Marek@ip-95-223-36-117.unitymediagroup.de 1285094650 Q * remember 1285095164 J * rememberme ~rememberm@ip-95-223-36-117.unitymediagroup.de 1285095406 J * dna_ ~dna@dslb-088-074-220-247.pools.arcor-ip.net 1285095823 Q * dna Ping timeout: 480 seconds 1285095938 Q * petzsch Quit: Leaving. 1285097466 N * Bertl_zZ Bertl 1285097473 M * Bertl back now ... 1285097493 Q * Piet Remote host closed the connection 1285097556 J * Piet ~Piet__@82VAABENN.tor-irc.dnsbl.oftc.net 1285099071 J * petzsch ~markus@dslb-094-218-178-167.pools.arcor-ip.net 1285099297 Q * nkukard Ping timeout: 480 seconds 1285099354 Q * thierryp Remote host closed the connection 1285099365 Q * petzsch Read error: Connection reset by peer 1285100940 J * derjohn_mob ~aj@d062056.adsl.hansenet.de 1285101321 J * hijacker_ ~hijacker@87-126-142-51.btc-net.bg 1285101678 J * nkukard ~nkukard@41-133-236-167.dsl.mweb.co.za 1285101869 Q * bonbons Quit: Leaving 1285102467 Q * dna_ Quit: Verlassend 1285103163 J * petzsch ~markus@dslb-094-218-178-167.pools.arcor-ip.net 1285103370 Q * hijacker_ Quit: Leaving 1285103422 J * alpha_one_x86 ~kvirc@131.30.17.95.dynamic.jazztel.es 1285103480 M * alpha_one_x86 The patch work with 2.6.32.22? 1285103938 M * Bertl we'll see, I'll update shortly 1285104026 M * rememberme Bertl, the tun device howto is not realy understandable 1285104075 M * Bertl go ahead, improve it 1285104104 M * alpha_one_x86 and with grsec patch please too, thanks 1285104132 M * Bertl that is harry's part, but I'm sure he'll update when he gets around 1285105672 Q * alpha_one_x86 Quit: KVIrc Equilibrium 4.1.0, revision: 4696, sources date: 20100519, built on: 2010-08-29 21:19:12 UTC http://www.kvirc.net/ 1285106048 Q * petzsch Read error: Connection reset by peer 1285106620 Q * Piet Remote host closed the connection 1285106880 J * Piet ~Piet__@82VAABERB.tor-irc.dnsbl.oftc.net 1285107604 Q * ntrs Ping timeout: 480 seconds 1285108537 Q * DoberMann Ping timeout: 480 seconds 1285108658 Q * Piet Ping timeout: 480 seconds 1285109106 Q * Wonka Ping timeout: 480 seconds 1285109274 J * Piet ~Piet__@82VAABESE.tor-irc.dnsbl.oftc.net 1285110481 Q * ghislain Quit: Leaving. 1285110811 J * DoberMann ~james@2a01:e35:8b44:84c0:230:18ff:fea3:7188