1266367151 J * ViRUS_ ~mp@p579B5C6E.dip.t-dialin.net 1266367585 Q * ViRUS Ping timeout: 480 seconds 1266368923 Q * dowdle Remote host closed the connection 1266369560 Q * Piet Ping timeout: 480 seconds 1266369814 J * chatter ~chatter@124.81.85.222 1266370977 Q * taenzerme Quit: Leaving. 1266372401 Q * fzylogic Quit: fzylogic 1266372441 J * taenzerme ~Adium@static-87-79-237-223.netcologne.de 1266374398 Q * taenzerme Quit: Leaving. 1266376644 Q * ViRUS_ Quit: If there is Artificial Intelligence, then there's bound to be some artificial stupidity. (Thomas Edison) 1266376861 Q * chatter 1266379227 J * SauLus_ ~SauLus@c192030.adsl.hansenet.de 1266379576 Q * SauLus Read error: Connection reset by peer 1266379576 N * SauLus_ SauLus 1266379968 J * jrklein ~jrklein@2001:0:53aa:64c:0:6336:b4d8:6cf 1266382370 Q * balbir Ping timeout: 480 seconds 1266387887 N * Bertl_zZ Bertl 1266387891 M * Bertl morning folks! 1266389536 J * balbir ~balbir@122.248.161.59 1266390220 J * ncopa ~ncopa@245.39.189.109.customer.cdi.no 1266390729 J * dna ~dna@170-198-103-86.dynamic.dsl.tng.de 1266391460 J * sharkjaw ~gab@90.149.121.45 1266391935 J * Piet ~Piet__@04ZAAAGRF.tor-irc.dnsbl.oftc.net 1266394173 M * incd Good morning 1266394250 M * incd How do I get syslogd working inside container? I get "Error opening file for reading; filename='/proc/kmsg', error='Operation not permitted (1)'" with Gentoo container. 1266394271 M * Bertl usually you remove the klog setup, then syslog works fine 1266394282 M * incd ok thanks :) 1266394293 M * Bertl alternatively you can activate the syslog virtualization which will provide an empty klog 1266394362 M * Bertl btw, you reported (a month ago?) some issues where vserver-stat crashed the host? 1266394394 M * incd Yeh 1266394403 M * Bertl how was that resolved? 1266394449 Q * balbir Ping timeout: 480 seconds 1266394507 M * incd Finally got containers running perfectly by using different kernel, 2.6.22.19-vs2.2.0.7 1266394541 M * incd with dietlibc-0.31 1266394550 M * Bertl okay, what architecture you're on? 1266394576 M * incd sparc64 (UltraSparc T1) 1266394656 M * incd My network problems were caused by wrong routing (datacenter's fault, when they gave more IPs) :p 1266394658 M * Bertl okay, did/can you try with a recent kernel (e.g. 2.6.31/32) (from kernel org)? 1266394695 M * incd Tried linux-2.6.31-vserver-2.3.0.36.23 and linux-2.6.32-vserver-2.3.0.36.28 1266394719 M * Bertl what I mean is, try with a vaniller kernel and patch 1266394727 M * Bertl *vanilla even 1266394809 M * incd Tried by getting kernel from kernel.org and then patching it - with those kernels it hung up when running vserver-stat 1266394833 M * Bertl really? that's interesting ... 1266394847 M * incd Even ILOM borked 1266394888 M * incd But it just rebooted, took 20min it to reboot 1266394901 M * Bertl but vanilla without Linux-VServer patch works fine? 1266394902 M * incd With this kernel everything's been just fine. 1266394906 M * incd Yea 1266394936 M * incd I can give you my .config 1266395018 M * Bertl okay, so let me recap: 2.6.32.8 works fine, no problems whatsoever 1266395050 M * Bertl 2.6.32.8-vs2.3.x boots fine, you can logon, but once you do vserver-stat, the machine is dead? 1266395070 M * incd Yeap, with 2.6.32.8-vs2.3.x containers work 1266395088 M * incd But the vmserver-stat and the machine is dead 1266395117 M * Bertl okay, you have the ILOM configured as serial console? 1266395135 M * incd Yea 1266395148 M * Bertl and it logs e.g. kernel boot messages? 1266395197 M * incd Yea, but it just froze, no messages on tty 1266395213 M * Bertl okay, did you try to send a magic sysrq? 1266395245 M * Bertl also, could you check that kernel debugging and the softlockup check are activated in the kernel config? 1266395460 M * incd I'll try, bit busy at work atm 1266395471 M * Bertl np, thanks! 1266395567 M * incd CONFIG_DETECT_SOFTLOCKUP=y 1266395841 M * Bertl so that's on, but doesn't trigger? 1266395855 J * balbir ~balbir@122.248.163.1 1266395881 M * Bertl do you know how to send a magic-sysrq via ILOM? did you try it when the kernel was responsive? 1266395895 M * Bertl sysrq-h should give you a help line 1266395943 M * incd Didn't try that, java ilom applet has that as shortcut in menu tho :) 1266396191 M * Bertl okay, would be interesting (if you find the time) to a) check that the magic sysrq works via ILOM (when the kernel is fine) and b) to see if you can get a stack trace/process dump once the machine is unresponsive 1266396502 M * _Shiva_ do SUN machines not have "Stop-A" debugging consoles any more..? 1266398742 J * bobnormal ~irc@87-194-32-179.bethere.co.uk 1266398747 M * bobnormal mornin' 1266400035 Q * balbir Ping timeout: 480 seconds 1266400216 J * _franklin_ ~quassel@95.108.8.74 1266400647 J * gnuk ~F404ror@pla93-3-82-240-11-251.fbx.proxad.net 1266402127 J * ktwilight__ ~keliew@91.179.214.40 1266402438 J * ktwilight ~keliew@254.4-240-81.adsl-dyn.isp.belgacom.be 1266402505 Q * ktwilight_ Ping timeout: 480 seconds 1266402568 Q * _franklin_ Remote host closed the connection 1266402714 Q * ktwilight__ Ping timeout: 480 seconds 1266403655 J * thierryp ~thierry@lns-bzn-47f-62-147-212-202.adsl.proxad.net 1266404151 J * balbir ~balbir@122.248.163.1 1266404241 Q * thierryp Remote host closed the connection 1266404858 Q * balbir Ping timeout: 480 seconds 1266405134 J * hijacker__ ~hijacker@213.91.163.5 1266405134 Q * hijacker_ Read error: Connection reset by peer 1266406874 M * biz Bertl: thanks for investigating :) 1266407103 J * balbir ~balbir@122.248.163.1 1266407688 Q * balbir Ping timeout: 480 seconds 1266408043 J * thierryp ~thierry@lns-bzn-47f-62-147-212-202.adsl.proxad.net 1266408135 Q * thierryp Remote host closed the connection 1266410823 J * thierryp ~thierry@lns-bzn-47f-62-147-212-202.adsl.proxad.net 1266411101 Q * thierryp Remote host closed the connection 1266411689 J * taenzerme ~Adium@static-87-79-237-223.netcologne.de 1266413126 J * rickytato ~freezer@host100-164-dynamic.41-79-r.retail.telecomitalia.it 1266413798 J * balbir ~balbir@122.172.149.6 1266413812 J * thierryp ~thierry@lns-bzn-47f-62-147-212-202.adsl.proxad.net 1266417002 Q * sharkjaw Remote host closed the connection 1266417458 Q * ghislain Quit: Leaving. 1266417467 J * ghislain ~AQUEOS@adsl2.aqueos.com 1266417676 Q * hijacker__ Quit: Leaving 1266417716 Q * thierryp Remote host closed the connection 1266417769 Q * dna Quit: Verlassend 1266417883 J * dowdle ~dowdle@scott.coe.montana.edu 1266418199 M * Bertl nap attack ... bbl 1266418204 N * Bertl Bertl_zZ 1266418898 Q * elesouef Quit: leaving 1266419000 J * hijacker ~hijacker@213.91.163.5 1266420025 J * thierryp ~thierry@lns-bzn-47f-62-147-212-202.adsl.proxad.net 1266420240 Q * thierryp Remote host closed the connection 1266421332 Q * balbir Read error: Connection reset by peer 1266422048 J * balbir ~balbir@122.172.63.183 1266422850 Q * ncopa Quit: Ex-Chat 1266423533 J * barismetin ~barismeti@zanzibar.inria.fr 1266426819 Q * jrklein Quit: jrklein 1266426943 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d 1266427490 J * imcsk8 ~ichavero@148.229.1.11 1266427756 J * thierryp ~thierry@lns-bzn-47f-62-147-212-202.adsl.proxad.net 1266427777 Q * bonbons Read error: No route to host 1266427877 Q * taenzerme Quit: Leaving. 1266431176 Q * barismetin Quit: Leaving... 1266431437 Q * _are_ Remote host closed the connection 1266431450 J * _are_ ~quassel@h1417489.stratoserver.net 1266432009 Q * gnuk Quit: NoFeature 1266432092 Q * thierryp Remote host closed the connection 1266432396 J * hijacker_ ~hijacker@87-126-142-51.btc-net.bg 1266432870 J * sur5r sur5r@courante.etherkiller.de 1266432895 M * sur5r hi@ll 1266433081 M * sur5r has anyone ever had the phenomenon that barriers apper out of nowhere, without being created knowingly? 1266433086 M * sur5r *appear 1266433144 M * daniel_hozac Debian? 1266433149 M * sur5r yes 1266433156 M * daniel_hozac the initscript sets them for you 1266433169 M * sur5r barriers in the middle of the vserver? 1266433175 M * daniel_hozac no. 1266433192 M * sur5r i just found one inside of a vserver in /var/spool/postfix/... 1266433225 M * daniel_hozac what kernel? 1266433231 M * daniel_hozac and what filesystem? 1266433251 M * sur5r 2.6.32-trunk-vserver (from debian), has vserver 2.3 inside, reiserfs 1266433284 M * Bushmills most likely a file system issue 1266433296 M * Bushmills did you let your partition run full once? 1266433308 M * daniel_hozac try with a vanilla kernel first. 1266433316 M * sur5r don't think so, but possible in theory 1266433333 M * Bushmills reiserfs is not very forgiving with that consition 1266433336 M * Bushmills condition 1266433342 M * sur5r being full? 1266433353 M * sur5r never had any problems with reiserfs being full 1266433355 M * Bushmills yes. zero space left. then deleted some files to make space again 1266433364 M * sur5r outside of vservers at least 1266433398 M * sur5r happens to me every now and then on my notebook 1266433441 M * sur5r daniel_hozac: vanilla as from kernel.org plus which vserver patch? 1266433447 M * Bushmills in fact, reiser was one of the least robust one when i tested a number of fs 1266433502 M * daniel_hozac sur5r: yes 1266433532 M * sur5r Bushmills: let's not get this into an fs discussion. i'm using reiserfs as my only filesystem on a decent amount of machines since 2003 at least. i never lost a single file on it. even with power outages (notebook) 1266433587 M * sur5r daniel_hozac: ok, will try. unfortunately i don't know what exactly triggers the problem 1266433600 M * Bushmills it is still not unlikely that fs is the cause of your problemsa 1266433627 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d 1266433635 M * sur5r Bushmills: accepted. i will also look in that direction 1266433647 M * daniel_hozac sure the fs is the problem. 1266433659 M * daniel_hozac most likely not in the way you think though. 1266434929 J * taenzerme ~Adium@static-87-79-237-223.netcologne.de 1266437585 Q * hijacker_ Quit: Leaving 1266437858 J * TheWildgoose ~TheWildgo@office.nippynetworks.com 1266437956 M * TheWildgoose Hi, I just upgraded my kernel from 2.6.22 to 2.6.32 (grsec+vs) and now my 32bit guests fail to start, 64bit are fine. Restarting with the old kernel and all is well. Anyone got any ideas where to look please? 1266438046 M * daniel_hozac did you enabke IA32 emulation in your kernel? 1266438077 M * TheWildgoose .config says yes... 1266438125 M * TheWildgoose error reads: /usr/sbin/chbind: line 135: 9240 Killed "${create_cmd[@]}" "${chain_cmd[@]}" -- "$@" 1266438155 M * TheWildgoose I can't rule out some grsec interaction here mind... 1266438169 M * TheWildgoose (I have very little of the grsec/pax stuff enabled though) 1266438254 M * TheWildgoose I had util-vserver 0.30.216_pre2849 (in gentoo naming convention). I just upgraded to _pre2855 (untested) - could this be a factor? 1266438453 Q * taenzerme Quit: Leaving. 1266438506 M * TheWildgoose Hi Daniel - grateful for any suggestions on getting some useful error messages/debug out of this thing please? It's a live server so I need to bounce it as few times as possible.... Thanks 1266438593 M * daniel_hozac does dmesg have anything to say? 1266438653 M * TheWildgoose zero in dmesg... (which suggests it's not grsec) 1266438684 M * daniel_hozac nothing at all? 1266438704 M * TheWildgoose well, just the end of normal bootup - I get a bunch of mount messages in my case 1266438793 M * TheWildgoose config used largely a make oldconfig from the old kernel - I did add some extra iptables stuff, possibly tweaked KVM options and a few things I can't recall now - seemed safe at the time 1266438847 M * daniel_hozac so on the host, can you run /vservers//bin/true? 1266438896 M * TheWildgoose I can't retest at this moment, but I think not. I tried ...../bin/awk which just said "killed" 1266438919 M * TheWildgoose this was not in a chroot and I assumed it was just using wrong shared libraries 1266438973 M * TheWildgoose going to briefly reboot into duff kernel 1266439102 M * TheWildgoose hmm, actually might abort that - I can see immediately that on my older kernel I can run ...//bin/awk without a problem - seems like my kernel is not really letting me run 32bit binaries... Any thoughts? 1266439162 M * TheWildgoose .config shows CONFIG_IA32_EMULATION=y and CONFIG_IA32_AOUT=y 1266439211 M * TheWildgoose curiously though when I built this kernel I find that the kernel is in arch/x86/boot/bzImage and not /boot/x86_64/boot/bzImage - I had assumed some change in defaults? 1266439367 M * daniel_hozac x86_64 and i386 were merged some kernel versions ago 1266439407 M * TheWildgoose ok, just retested and even guest/bin/true is "killed" 1266439457 M * fback don't you need full 32bit userland support to run 32bit guest binaries on the host? 1266439489 M * TheWildgoose I think I may have multilib binaries on this host - point being that it works if I flip back to old kernel and not on new kernel 1266439493 M * fback ie glibc and other libraries, not just the kernel 1266439513 M * fback ah, yes, you're right 1266439607 M * fback (shouldn't it be "unknown binary" without 32bit support and not "killed"?) 1266439632 M * TheWildgoose quite possibly - killed should mean segfault I think? 1266439688 M * daniel_hozac no, segfaults are Segmentation fault. 1266439776 M * TheWildgoose ok, I have now repro'd it on another machine running a very similar new kernel. This machine is slightly less critical so I can try reverting to a non grsec kernel and see if that's a factor 1266441268 Q * TheWildgoose Ping timeout: 480 seconds 1266441799 Q * jrdnyquist Quit: Leaving 1266442532 Q * bonbons Quit: Leaving 1266443692 J * jrdnyquist ~jrdnyquis@slayer.caro.net 1266444395 Q * ghislain Quit: Leaving. 1266444967 J * kjj_ ~kjj@pool-74-107-128-126.ptldor.fios.verizon.net 1266445129 Q * kjj_ 1266445332 Q * kjj Ping timeout: 480 seconds 1266445523 M * sur5r daniel_hozac: i think a made a mistake when setting up the whole thing... reiserfs is mounted without the attrs option. could that be the source of my problem? 1266445751 M * daniel_hozac sur5r: yeah, i think that's required. 1266445758 M * sur5r uh ok 1266445764 M * sur5r will fix that on all my vservers 1266445818 M * sur5r interestingly, the testfs.sh script does not fail when i remove the attrs option from the test 1266445872 M * sur5r well no.. i'm not sure whether it does any checks at all 1266445919 M * daniel_hozac did you specify -x -y -z? 1266445927 M * sur5r ah, i have to specify which tests to run... should've read the usage more carefully 1266446266 M * sur5r ok, complains. looks like there is no COW attr on reiserfs 1266446550 M * sur5r ah no, the test script does not know about the COW flag 1266446563 M * sur5r so it does not expect the c/C to be there 1266446577 M * daniel_hozac try a newer version of it. 1266446595 M * sur5r will do 1266446669 M * sur5r uh, does that exist? i'm already on 0.30.216-pre2864 1266446679 M * daniel_hozac of the test script. 1266446747 M * sur5r uh, test script is 0.23 already, packed is only 0.18. i see 1266446859 M * sur5r ah, much better 1266450808 Q * Piet Remote host closed the connection 1266450874 J * Piet ~Piet__@04ZAAAHLO.tor-irc.dnsbl.oftc.net