1318293255 J * hparker ~hparker@2001:470:1f0f:32c:beae:c5ff:fe01:b647 1318297250 M * Bertl off to bed now ... have a good one everyone! 1318297255 N * Bertl Bertl_zZ 1318297698 Q * neofutur Quit: leaving 1318297702 J * neofutur ~neofutur@xena.ww7.be 1318303294 J * manana ~mayday090@178.162.120.195 1318308030 J * sannes1 ~ace@cm-84.209.106.118.getinternet.no 1318310893 Q * quasisane Ping timeout: 480 seconds 1318311137 J * derjohn_foo ~aj@213.238.45.2 1318311174 Q * hparker Quit: Quit 1318311651 J * thierryp ~thierry@home.parmentelat.net 1318311909 Q * thierryp Remote host closed the connection 1318313029 J * ncopa ~ncopa@3.203.202.84.customer.cdi.no 1318314412 J * quasisane ~sanep@c-76-24-80-97.hsd1.nh.comcast.net 1318315266 J * ghislain ~AQUEOS@adsl2.aqueos.com 1318316235 Q * LuckyLuk1 Quit: bbl 1318316559 J * fisted_ ~fisted@xdsl-87-78-213-197.netcologne.de 1318316688 Q * fisted Ping timeout: 480 seconds 1318316775 J * thierryp ~thierry@zankai.inria.fr 1318324562 N * Bertl_zZ Bertl 1318324566 M * Bertl morning folks! 1318325119 M * ghislain morning bertl :) 1318325166 M * Bertl arekm: any news on the memory/swap issue? 1318326001 Q * thierryp Remote host closed the connection 1318326043 J * thierryp ~thierry@zankai.inria.fr 1318326430 M * arekm Bertl: no but I'm back after few days of "vacation" and I'm going to test with that printk line 1318326448 M * Bertl ok, no problem 1318328307 P * kir Leaving. 1318330857 M * Bertl nap attack .. bbl 1318330867 N * Bertl Bertl_zZ 1318333295 J * kir ~kir@swsoft-msk-nat.sw.ru 1318333405 J * kir1 ~kir@swsoft-msk-nat.sw.ru 1318333594 Q * kir1 1318334635 Q * Aiken Remote host closed the connection 1318336589 J * Aiken ~Aiken@2001:44b8:2168:1000:21f:d0ff:fed6:d63f 1318336613 P * kir Leaving. 1318341206 N * Bertl_zZ Bertl 1318341210 M * Bertl back now .. 1318342352 Q * FireEgl Remote host closed the connection 1318343344 J * FireEgl FireEgl@2001:470:e056:1:a899:4c75:59df:2029 1318343833 Q * thierryp Remote host closed the connection 1318344427 J * BenG ~bengreen@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com 1318344521 Q * BenG 1318345758 Q * ncopa Quit: Leaving 1318347862 J * Crumbz ~Crumbz@host-78-146-188-170.as13285.net 1318347997 M * Crumbz Hey guys, there is a problem with the package : linux-headers-vserver for kernel 2.6.31 on the ubuntu repositories, it says "broken package", i have emailed the maintainer (christoph lukas) but there is no response. Unfortunately i can't install a new system until i have the kernel headers. Is there a workaround? 1318348360 Q * derjohn_foo Ping timeout: 480 seconds 1318349074 M * Crumbz anyone there? 1318349193 J * bonbons ~bonbons@2001:960:7ab:0:e805:fe4c:4609:73 1318349543 M * Guy- Crumbz: it would be my suggestion to forget the ubuntu packages and roll your own, using make-kpkg 1318349577 M * Guy- Crumbz: download your favourite kernel source, patch with vserver, configure, then build linux-image and linux-headers deb packages using make-kpkg (from the kernel-package package) 1318350669 M * Bertl Crumbz: first, what do you need the kernel headers for? 1318350705 M * Bertl second, why not use the mainline kernel headers (which should do nicely for 99% of all userspace needs) 1318350726 M * Bertl and finally, why use the obviously broken ubuntu package at all 1318350813 M * Bertl note: you can easily build your own kernel and tools, even as .deb 1318350958 M * Bertl Guy-: where downloading the kernel source isn't quite trivial atm, because kernel.org is still broken 1318351144 M * Bertl (i.e. the main releases are up, but the .x releases are still missing ... not speaking of rc and similar 1318353014 M * daniel_hozac mirrors still exist :) 1318353212 M * Bertl maybe we should link to them temporarily? 1318353259 M * Bertl as it seems getting kernel.org fully working is not a huge priority :) 1318353434 M * Bertl daniel_hozac: btw, which mirror is still on the old data? 1318353678 J * fback_ fback@red.fback.net 1318353679 Q * fback Remote host closed the connection 1318353761 M * daniel_hozac Bertl: http://ftp.sunet.se/pub/Linux/kernel.org/linux/kernel/v3.0/ 1318353763 M * daniel_hozac for instance. 1318353831 M * Bertl nice, tx! 1318355322 M * Crumbz Guy-, thanks, i will do that. I take it the vserver patch is on the vserver webpage? 1318355470 M * Crumbz Bertl, thanks, i need the headers to do some compiling. trying to get hold of the 2.6.31 kernel headers, from the Ubuntu repos, isn't easy. Would there be an issue with version umbers fi i had the 2.6.32 headers installed? Thanks. 1318355514 M * daniel_hozac 2.6.31 seems like an odd version to go for. 1318355632 M * Bertl indeed 31 is odd :) 1318355789 A * arekm going to test printk thing 1318356271 M * Crumbz 2.6.31 is the most recent vserver kernel image in the Ubtunu repos 1318356386 M * Bertl well, I wouldn't use that ... but YMMV 1318356450 M * Crumbz why not? 1318356537 M * Bertl well, first, it's quite old and probably wasn't maintained/updated to the recent fixes, second, it's not a very good choice kernel wise (has a lot of bugs/performance issues) and finally, it's completely unmaintained by now 1318356592 M * Bertl I'd use at least 2.6.32.28 + patch (if you feel like you don't want a recent kernel), but I'd go with 2.6.38.8 or 3.x 1318356606 M * Crumbz it's the only kernel available from the repositories. Where would i get a more recent version? 1318356629 M * Bertl http://ftp.sunet.se/pub/Linux/kernel.org/linux/kernel/v2.6/ 1318356662 M * Bertl and here are the patches: http://linux-vserver.org/Welcome_to_Linux-VServer.org 1318356694 M * pmjdebruijn Bertl: there's 2.6.32.45 now too :D 1318356729 M * pmjdebruijn or 43 1318356737 M * pmjdebruijn but it patches fine against newer versions I think 1318356770 M * pmjdebruijn Crumbz: we're actually using a homebuilt 2.6.32.45+vs with Ubuntu Lucid, though only few a few vserver yet 1318356783 M * pmjdebruijn most of our setup is still on lenny/2.6.27.x-vs :( 1318356809 M * Bertl sorry, I actually meant 2.6.32.43, just got the wrong line 1318356859 M * Crumbz Bertl, you are a good man, thanks! The links on the linux-vserver page point to broken kernel.org resources. + Where can i find good information on the kernel, like bugs etc. kernel.org best for that? or elsewhere? 1318356894 M * arekm Bertl: setting up test server, when swap is available but kernel cgroup swap limit disabled then: Swap: 1502072 -1240 1503312 1318356920 M * arekm Swap: 0 -740 740 1318356931 M * arekm and when swap is disabled. Ok, rebooting to kernel with printk 1318357002 M * Bertl Crumbz: in the first link I pasted 1318357195 M * Crumbz Bertl, for kernel news? + is it worth using the grsecurity patch? the extra security features would be nice, but i don't want a really unstable system. 1318357227 M * Bertl well, what security features are you going to use (from grsec)? 1318357269 M * Crumbz pax would be nice, i haven't looked over the features fully yet 1318357299 M * Bertl so then I'd suggest to look for a pax patch and not include the entire grsec 1318357365 M * Bertl anyway, the combined grsec patches are currently (mostly) unmaintained, so I wouldn't go for that unless you know what you are doing 1318357406 M * Crumbz which ones are maintained? 1318357477 M * Bertl hmm? 1318357485 M * arekm Bertl: ok got it 1318357486 M * arekm [ 78.845857] res=6873088/268435456 memsw=0/9223372036854775807 1318357486 M * arekm [ 79.798764] res=8134656/268435456 memsw=0/9223372036854775807 1318357504 M * arekm cgroup soft/hard limit was set to 256M 1318357511 M * Bertl okay, interesting ... let me check that with the code 1318357610 M * Bertl so the memsw_limit contains garbage in your case ... 1318357654 M * arekm I have memsw support compiled it but compile time setting is default to off (need some kernel boot option to enable it) 1318357675 M * Bertl but you do not enable it, yes? 1318357681 M * arekm yes 1318357732 M * Bertl okay, last time I disabled it with the boot option, maybe that's the difference 1318357827 M * Bertl would be great if you could verify that it works just fine when you enable it in the kernel, but disable it via the boot option 1318357848 M * arekm blahm, a lot of rebuilding ;> ok 1318357869 M * Bertl let me try first, it's not a lot of rebuilding here 1318357871 M * Guy- Bertl: ftp.kfki.hu also has a complete kernel.org mirror, I think (and may be closer to you topology-wise) 1318357935 M * Bertl thanks, but I have almost all kernel versions mirrored locally, it's just to have something to give folks looking for a kernel tar 1318357946 M * Guy- ah 1318358401 M * Bertl arekm: okay, I can reproduce the issue here with your config option 1318358744 M * arekm cool 1318359539 M * Bertl arekm: please try http://vserver.13thfloor.at/ExperimentalT/delta-memsw-fix01.diff and let me know if that fixes the issue for you 1318360079 M * arekm ok, wanted also to build with new patch but looks like there is nothing new after patch-3.0.4-vs2.3.1-pre10.1.diff 1318360092 M * arekm while there is ton of delta fixes - are these fixes worth applying? 1318360127 M * Bertl yeah, I haven't uploaded any new patch (yet), you might also want to include daniel_hozac's cow fix 1318360157 M * arekm so all these delta patches are fine? and where is cow fix? 1318360242 M * Bertl they should be fine, some are without feedback (like the memsw fix :) but presumed correct 1318360311 M * Bertl http://people.linux-vserver.org/~dhozac/p/k/delta-cow-fix25.diff 1318360400 M * Bertl daniel_hozac: btw, any pending patches on your side, except for that one? 1318360422 M * daniel_hozac no, unfortunately :) 1318360452 M * Bertl well, any issues you would love to have patches for then? 1318360467 M * daniel_hozac i'm not really aware of any issues. 1318360502 M * Bertl okay, same here, so I'll do a final release and start breaking that down for review then 1318360534 M * arekm does nfs02 obsolete nfs01 diff? 1318360545 M * Bertl IIRC, yes 1318360568 M * daniel_hozac hmm. 1318360586 M * daniel_hozac fix02 is the actual fix, fix01 is a fix for another potential issue, IIRC 1318360588 M * Bertl well, nfs01 is basically untested 1318360615 M * Bertl i.e. we do neither know if that actually happens, nor if the patch actually fixes it :) 1318360654 M * Bertl but of you feel like keeping it, I'm fine with that as well 1318360674 M * Bertl s/of/if/ 1318360863 M * daniel_hozac i think we can safely leave it out. 1318361251 M * Bertl so we have a vs2.3.1 then :) 1318361495 Q * fisted_ Read error: Operation timed out 1318361800 J * fisted ~fisted@xdsl-87-78-219-219.netcologne.de 1318361910 M * arekm Bertl: but why 3.0.4 :-/ 3.0.6 is latest one 1318362419 M * Bertl really? haven't checked ... 1318362551 M * arekm it was announced on lkml 1318362578 N * BobR_zZ BobR 1318362587 M * BobR PING 1318362873 M * Bertl hey BobR :) 1318362888 M * arekm Bertl: ftp://ftp.archlinux.org/other/linux/patch-3.0.6.gz as easiest place to get it (while the proper is to take diff from git tree) 1318363436 N * BobR BobR_zZ 1318364674 J * ryker ~jalberts@c-67-176-243-86.hsd1.in.comcast.net 1318365398 M * ryker can anyone help. i'm having this problem with 3 guests now. http://pastebin.com/SyLF9UtK 1318365409 M * ryker exec format error 1318365432 M * ryker it's centos 5.6 x86_64 host running centos 5.6 x86_64 guests 1318365442 M * ryker different hosts, same problem 1318365465 M * ryker the guests are on a seperate iscsi volume mounted at /vservers/guestname 1318365478 M * ryker fsck shows no problems 1318365481 M * ryker nothing in lost+found 1318365636 M * daniel_hozac i never saw any rpm results. 1318365699 M * ryker daniel_hozac: i couldn't get it to run. it kept givng me errors. i thought i pastebin'd it. let me find it. 1318365708 M * ryker hi, BTW. Thanks for helping. 1318365712 M * ryker :) 1318365733 M * ryker http://pastebin.com/teAFHu4W 1318365815 M * daniel_hozac that's with --root. 1318365820 M * daniel_hozac then i said to try with chroot instead. 1318365898 M * ryker sry, i can't find the paste for that. let me try that again and paste it. 1318365964 M * ryker I get this when I use chroot 1318365965 M * ryker chroot: cannot run command `rpm': Exec format error 1318365993 M * ryker The host was also rebooted, which didn't help. other guests on the same host are running ok 1318366002 M * daniel_hozac how about chroot /vservers//bin/true 1318366011 M * daniel_hozac with a space... 1318366084 M * ryker same thing. exec format error 1318366100 M * daniel_hozac and what does ldd /vservers//bin/true say? 1318366110 M * daniel_hozac and then file on all of those. 1318366258 M * ryker daniel_hozac: Here's the output of ldd and file on /bin/true for a working guest and a non-working guest. 1318366261 M * ryker http://pastebin.com/SFEuzEst 1318366281 M * ryker permissions seem correct 1318366302 M * daniel_hozac it's a junk file though. 1318366377 M * daniel_hozac try looking inside of it. 1318366387 M * ryker right. i don't understand what happend to all of the executables 1318366459 M * ryker the /bin/true on the bad guest looks totally different than the one on the good guest, though they are identical in size 1318366461 M * ryker :/ 1318366483 M * ryker so, it's a filesystem problem that fsck isn't detecting or fixing 1318366489 M * ryker apparently 1318366528 M * daniel_hozac or something just overwrote your files. 1318366659 M * ryker it was a filesystem crash. things like /etc/hosts, resolv.conf, etc look fine. this is weird. 1318366685 M * ryker really weird that it's happened multiple times now to different server and guests. 1318366786 Q * bonbons Quit: Leaving 1318366808 M * daniel_hozac sounds more like an exploit then. 1318366813 M * daniel_hozac unless you're using some broken kernel. 1318367092 M * ryker it's actually happened when our netapp server has had a controller failure and and switched over the other controller. The iscsi mount goes read only. Then when we stop the guest and mount it rw, we can't get it to start again. 1318367106 M * ryker I suspected file system corruption, but fsck shows all is fine 1318367125 M * ryker and I can read the text files fine. 1318367138 M * ryker I compared sizes of a couple binaries, but they were the same size. 1318367146 M * ryker so I assumed something else was going on. 1318367346 M * ryker daniel_hozac: thx for the help 1318367506 M * daniel_hozac you might want to use cmp instead ;) 1318367577 M * ryker yeah, I don't know why I didn't. or even diff, which I use all of the time 1318367668 Q * Crumbz Ping timeout: 480 seconds 1318369223 J * cuba33ci_ ~cuba33ci@111-240-169-234.dynamic.hinet.net 1318369224 Q * cuba33ci Read error: Connection reset by peer 1318369235 N * cuba33ci_ cuba33ci 1318369642 Q * Aiken Remote host closed the connection 1318369775 J * Aiken ~Aiken@2001:44b8:2168:1000:21f:d0ff:fed6:d63f 1318370886 Q * zbyniu Ping timeout: 480 seconds 1318371601 Q * ryker Quit: ryker 1318373720 J * zbyniu ~zbyniu@ip-62.181.188.13.static.crowley.pl 1318374114 Q * ghislain Quit: Leaving. 1318374172 M * Bertl arekm: but I presume, the 3.0.4 patch applies fine to more recent versions, yes? 1318374545 Q * Marillion Ping timeout: 480 seconds