1193013716 Q * yarihm Quit: Leaving 1193014442 J * TCPIPCHIP ~asm51@200-138-221-219.fnsce702.dsl.brasiltelecom.net.br 1193014453 M * TCPIPCHIP hello 1193014486 M * TCPIPCHIP is there anybody there? 1193014542 Q * TCPIPCHIP 1193016443 J * friendly12345 ~friendly@ppp59-167-69-153.lns1.mel4.internode.on.net 1193018099 Q * hparker Quit: G'nite 1193019793 Q * dowdle Remote host closed the connection 1193021054 Q * ensc Remote host closed the connection 1193022864 Q * friendly12345 Ping timeout: 480 seconds 1193022935 Q * fatgoose Quit: fatgoose 1193023069 J * friendly12345 ~friendly@ppp59-167-76-87.lns1.mel6.internode.on.net 1193023148 M * MooingLemur I've somehow gotten into a position on this new install where vserver-stat doesn't even show "root server" 1193023162 M * MooingLemur and starting a vserver yields no errors 1193023223 M * MooingLemur hmm.. maybe new look of newer util-vserver? 1193025492 Q * rorem- Ping timeout: 480 seconds 1193025590 J * balbir ~balbir@59.178.58.85 1193026278 M * daniel_hozac MooingLemur: the root server should never have been shown in vserver-stat. 1193027043 J * rorem- ~roremtank@bzq-219-46-202.isdn.bezeqint.net 1193027902 J * fatgoose ~samuel@206-248-128-87.dsl.teksavvy.com 1193028920 Q * balbir Ping timeout: 480 seconds 1193029095 Q * fatgoose Read error: Connection reset by peer 1193029102 J * fatgoose ~samuel@206-248-128-87.dsl.teksavvy.com 1193030697 Q * FireEgl Quit: Bye... 1193031598 M * MooingLemur CTX PROC VSZ RSS userTIME sysTIME UPTIME NAME 1193031598 M * MooingLemur 0 73 142.8M 4.4M 0m32s33 48m28s65 40d07h51 root server 1193031598 M * MooingLemur 143 22 270.1M 50.5M 1d16h00 22h09m35 40d07h48 en-gs-03 1193031673 M * MooingLemur that's an older install, with (gentoo) util-vserver 0.30.212-r2 vs 0.30.214 1193031681 M * daniel_hozac yes. 1193031687 M * daniel_hozac it used to be broken. 1193031691 M * MooingLemur ahh.. hah 1193031692 M * MooingLemur :P 1193031789 M * MooingLemur thanks 1193033391 J * DavidS ~david@vpn.uni-ak.ac.at 1193033516 Q * fatgoose Read error: Operation timed out 1193033719 J * virtuoso_ ~s0t0na@pppoe-139.58.110.89-adsl.spbnit.ru 1193034130 Q * virtuoso Ping timeout: 480 seconds 1193034984 J * amax ~a@r5ct9.net.upc.cz 1193035014 M * amax daniel_hozac hi 1193035065 M * amax daniel_hozac 2.3.0.27 got a loopback virtualization or no? 1193035365 J * larsivi ~larsivi@85.221.53.194 1193035789 P * DavidS Leaving. 1193037467 M * tanjix amax: i use 2.3.0.27 since some days and i see loopback device inside a guest, so i would say: yes ;) 1193037503 M * daniel_hozac amax: yes. 1193038178 M * amax hehe thanks 1193038197 M * amax now will try use it on production 1193038209 M * daniel_hozac better use 2.3.0.28 and delta-cow-fix17 then. 1193038243 M * amax hehe ok 1193038249 M * amax making ebuild 1193038249 J * pmenier ~pme@LNeuilly-152-22-72-5.w193-251.abo.wanadoo.fr 1193039323 J * dna ~dna@125-210-dsl.kielnet.net 1193039375 M * tanjix daniel_hozac: is it correct that the localhost device does not rely to pings inside a guest? 1193039498 M * daniel_hozac ping is awful. 1193039512 M * daniel_hozac use TCP/UDP. 1193039545 M * tanjix yes.. but anyway.. why doesn't it reply to icmp? 1193039561 M * tanjix http://paste.linux-vserver.org/7337 1193039574 M * daniel_hozac because ICMP is not interesting enough for anyone to figure out why it doesn't work? :) 1193039621 M * tanjix i am sure that it work but cutomers ask why they cannot ping localhot 1193039946 M * daniel_hozac so are you willing to test it? 1193040141 J * gebura ~gebura@77.192.186.197 1193040245 M * gebura hello 1193040254 M * daniel_hozac hi 1193040435 M * tanjix daniel_hozac: ? sorry? 1193040503 M * daniel_hozac tanjix: should be a pretty simple thing to add... 1193040513 M * daniel_hozac just requires someone do to some testing. 1193040575 M * tanjix then lets do so ;) 1193040590 M * amax daniel_hozac where can I download 2.3.0.28 and delta-cow-fix17 ? 1193040601 M * daniel_hozac amax: usual place, http://vserver.13thfloor.at/Experimental/ 1193040608 M * amax ah, thanks 1193040610 M * amax trying 1193041053 J * Guest175 ~knoppix@p549C7265.dip.t-dialin.net 1193041277 J * JonB ~NoSuchUse@kg1-20.kollegiegaarden.dk 1193041808 Q * JonB Quit: This computer has gone to sleep 1193041809 M * amax 02_delta-cow-fix17.diff got rejects.. hm.. 1193041818 M * amax will try new one 1193041892 J * fb_ fback@red.fback.net 1193041902 M * tanjix daniel_hozac: what shall i dow now / testing? 1193041967 M * daniel_hozac amax: use -l 1193042008 Q * fb Ping timeout: 480 seconds 1193042139 Q * Guest175 Quit: Verlassend 1193042275 M * amax haha 1193042280 M * amax patch with bug _ 1193042282 M * amax ^) 1193042418 M * daniel_hozac it's just whitespace differences. 1193042448 M * amax no 1193042461 M * amax it cant be applied due to 2 rejects 1193042480 M * daniel_hozac works fine for me. 1193042485 M * daniel_hozac (with -l) 1193042485 M * amax bugs in namei.c 1193042492 M * amax ^-) 1193042550 M * amax in fact, i cant setup options to patch inside ebuild ^( 1193042585 M * daniel_hozac really? how do you set the striplevel then? 1193042609 M * amax with -l all ok. 1193042625 M * daniel_hozac tanjix: okay, this is a wild guess... http://people.linux-vserver.org/~dhozac/p/k/delta-pinglo-test01.diff 1193042626 M * amax striplvl in gentoo is auto-probe with dryrun 1193042652 M * amax with -l its works.. ^) 1193042851 M * tanjix hm ok so i should make a new kernel? 1193042852 M * amax just applied and re-diffed 1193042866 M * tanjix then i cannot testthat right now as i cannot shutdown the server without scheduled maintenance 1193042868 M * amax ^) now patch seems to be ok ^) 1193042959 M * daniel_hozac tanjix: no test server? 1193043045 J * onox ~onox@kalfjeslab.demon.nl 1193043074 M * amax daniel_hozac - do you have a patch to show pid of process have warnings about ? 1193043109 M * amax sometime i got warnings, thats interesting which context/pid did that. 1193043187 M * daniel_hozac the context should already be there. 1193043195 M * amax but pid ? 1193043212 M * amax pid/appname 1193043237 Q * larsivi Quit: Konversation terminated! 1193043241 M * daniel_hozac not yet. 1193043293 J * Piet ~piet@tor.noreply.org 1193043719 J * JonB ~NoSuchUse@kg1-98.kollegiegaarden.dk 1193043806 J * lilalinux ~plasma@dslb-084-058-209-099.pools.arcor-ip.net 1193043831 M * amax daniel :) if you will have 5 min free time, please check, if thats not hard, make this patch for displaying pid, please :) 1193043868 M * amax im bow before you ^) 1193043874 M * amax and will eat hat. ) 1193043893 M * sid3windr enjoy the hat =) 1193043935 M * amax currently vserver not pass our security policy w/o pid displaying in wornings.. 1193043948 M * daniel_hozac so, add it? 1193043949 M * amax pid+processname(owner) 1193043968 M * amax i cant, familiar only with java hardcoding.. 1193043994 N * Bertl_zZ Bertl 1193043999 M * Bertl morning folks! 1193044016 M * amax ye good day 1193044077 M * Bertl daniel_hozac: what is the problem? I would like to see tha hat part too :) 1193044097 M * daniel_hozac that the warning messages don't include pid and process name. 1193044111 M * Bertl yeah, I got that, but _what_ warning? 1193044112 M * daniel_hozac should just be a matter of adding it to vxwprintk... 1193044126 M * Bertl ah, those warnings ... 1193044151 M * Bertl yeah, I think we could do that, would actually make sense 1193044164 M * daniel_hozac yeah. 1193044269 M * sid3windr hat part please! 1193044344 M * Bertl first the patch ... 1193044644 M * amax ^)) 1193044645 M * amax ye 1193044670 M * amax one more trouble.. hm 1193044681 M * amax i have all servers on software raid. 1193044686 M * amax mdadm etc.. 1193044706 M * daniel_hozac Bertl: i think we should output the xid for the cvirt and limit warnings too. 1193044707 M * amax with vserver kernel its unable to disable / device 1193044721 M * daniel_hozac amax: have you stopped all of your guests before rebooting? 1193044725 M * amax yes 1193044732 M * amax even dirty boot 1193044757 M * amax w/o guests started 1193044782 M * amax will explore this more.. 1193044793 M * daniel_hozac so you boot, never start a guest, reboot, and still have problems? 1193044806 M * amax ye, will recheck it again. 1193044835 M * amax and vith vanilla kernel too 1193044902 M * amax please, show in warning actual process name, not only pid. ^) and user owning process 1193044936 M * daniel_hozac uid's too much. 1193046557 J * larsivi ~larsivi@85.221.53.194 1193046582 M * Bertl amax: we will prepare a patch (have to elave right now) you get your hat and webcam ready :) 1193046594 M * amax hehe ^-) 1193046617 M * Bertl okay, off for now .. hopefully back later 1193046622 N * Bertl Bertl_oO 1193046644 M * amax the question about loopback remaping - have you any doc to it ? 1193046659 M * amax for now is it fully isolated ? 1193046671 M * amax lo with 127.0.0.1 from other quests.. 1193046709 M * amax is it personal lo ? all 127.0.0.0/8 ? 1193046732 M * amax seems to be a true 1193046870 M * daniel_hozac yes, all of 127.0.0.0/8 is rewritten to the lback address. 1193046912 M * amax great work.. 1193046921 M * amax the question again... 1193046966 M * amax if was possible to rewrite lo... is it possible to make same for other interfaces ? 1193046980 M * amax or .. wrong q.. 1193046997 M * amax whats an easy way to virtualize ip addresses ? 1193047003 M * amax inside = outside.. 1193047026 M * amax im talking about: if i have move my guest to other IP/server 1193047043 M * amax what the easy way to do it w/o changing toons of configs ? 1193047063 M * amax 0.0.0.0 only remapped to -first- interface. 1193047098 M * amax can we get some special ip's for direct remapping to interface ? 1193047122 M * amax like 10.0.0.1 remaps to eth0 first address assigned in guest 1193047135 M * amax like 10.0.1.1 remaps to eth1 first address assigned in guest 1193047139 M * amax and so on.. 1193047179 M * amax in this case we can use some "network" for mapping to interfaces. and be a happy with guests migration between servers 1193047336 M * amax 10.context.netcard.alias# 1193047395 M * amax seems like im stupid ^) 1193047809 M * amax heh got a better idea ^) 1193047916 M * amax daniel_hozac could you implement this? /etc/vservers/svn2/interfaces/0/remapped_to_ip and ip address inside to remap ? to 192.168.0.123/24 for example 1193047970 M * amax in this case we can use any ip's inside guests assigned manually . 1193048003 M * amax or cant? 1193048035 M * amax hm.. need study network layer. 1193048249 J * mjt ~mjt@81.13.94.2 1193048335 M * mjt there's a small conflict in arch/i386/kernel/doublefault.c for 2.6.22.10 - can the patch on the site be updated please? (I can provide a diff) 1193048409 M * mjt i mean for grsec one 1193048417 M * mjt 2.2+grsecurity 1193048459 M * mjt also, pax stuff should be updated there as well, it seems 1193048474 M * mjt harry: ? 1193048638 M * daniel_hozac amax: what? what is it that you want to achieve? 1193048716 Q * ay Quit: leaving 1193049550 J * meandtheshell ~markus@85.127.117.23 1193050179 Q * onox Quit: leaving 1193050679 N * pmenier pmenier_off 1193050983 Q * lilalinux Read error: Connection reset by peer 1193051031 J * lilalinux ~plasma@dslb-084-058-230-209.pools.arcor-ip.net 1193051116 J * yarihm ~yarihm@84-75-130-73.dclient.hispeed.ch 1193051470 Q * JonB Quit: This computer has gone to sleep 1193051685 Q * rob-84x^ Ping timeout: 480 seconds 1193051903 M * amax daniel_hozac, let me explain. 1193051922 M * amax imagine, huge count of virtual hosts/guests 1193051945 M * amax sometime need to change network 1193051951 M * amax to other 1193051992 M * amax in guests many services, using more then 1 ip's/aliaces 1193052006 M * amax so, its used by direct pointing to. 1193052035 M * amax how can i easy MOVE guest w/o config change inside to other network space ? 1193052059 M * harry mjt: yes? 1193052074 M * daniel_hozac amax: i think an example would help... 1193052124 M * amax hm.. 1193052134 M * amax example from me ? 1193052141 M * daniel_hozac yes. 1193052159 M * amax ok, example. i have 2 web servers in one guest 1193052191 M * amax on different ip's 1193052201 M * amax apache+nging/ligghttpd 1193052202 M * harry mjt: i'll update the grsec patch now 1193052217 M * harry now as in: i'll make a new patchset :) 1193052225 M * mjt heh. 1193052225 M * amax on eth0 i got 2 aliaces for apache+lighttpd 1193052243 M * amax so on interface 0 in guest i have 2 -fixed- ips 1193052248 M * amax ip addr sh 1193052249 M * mjt there's a small conflict in kernel/sys.c with 2.6.22.10 - trivial #include fix 1193052258 M * mjt not in arch/i386/kernel/doublefault.c 1193052275 M * harry mjt: i'm gonna take the latest grsec patch 1193052287 M * harry which has some other bugfixes/... 1193052293 M * amax daniel_hozac after migration to other network space i should edit all configs and replace old network addresses with new one. 1193052310 M * amax with single ip configuration thats clear. i use 0.0.0.0 and happy 1193052318 M * amax due to remapping 1193052330 M * amax but how with 2 or more ips per guest ? 1193052341 M * daniel_hozac you can still use 0.0.0.0.... 1193052353 M * amax for ONE ip 1193052361 M * amax but I have 2 in guest 1193052410 M * daniel_hozac 0.0.0.0 still works fine. 1193052417 M * amax for one ip, right ? 1193052422 M * daniel_hozac no, for all of them. 1193052427 M * amax how for all ? 1193052429 M * daniel_hozac just like on a regular system... 1193052436 M * amax waaait ^)) 1193052451 M * amax interface 0. assigning 2 ip 1193052462 M * amax inside guest we can see both. ip addr sh 1193052505 M * amax remapping working to first assigned address to guest, not to aliase 1193052548 M * amax partially I can use 0.0.0.0 in configs, but how to second addresses ? is it possbile to virtualize/remap them same as 0.0.0.0 & 1193052550 M * amax ? 1193052554 M * amax like /etc/vservers/svn2/interfaces/0/remapped_to_ip and ip address inside to remap ? to 192.168.0.123/24 for example 1193052593 M * daniel_hozac i have no idea what that means. 1193052603 M * daniel_hozac if your guest has two IP addresses, 0.0.0.0 will bind to both of them. 1193052627 M * daniel_hozac (and lback) 1193052635 M * amax i see, but I cant use 0.0.0.0 in this case 1193052652 M * amax i have 2 ip. i should use in configs different ips for 80 port 1193052656 M * amax for bind to 1193052686 M * amax so apache and lighttpd should bind to they own addresses and 80 ports 1193052700 M * amax do you understand me ? or my english too bad ? 1193052705 M * daniel_hozac okay, so 0.0.0.0 has absolutely nothing to do with anything. 1193052773 M * amax well, its just a some common bind point 1193052788 M * amax i can use now 0.0.0.0 in configs 1193052794 M * amax with single ip guests 1193052808 M * amax and moving guests let me not rewrite many configs. 1193052824 M * amax due to unification and remapping 0.0.0.0 to assigned ip 1193052842 M * amax but. with 2 or more ip configuration i unable to use 0.0.0.0 1193052851 M * amax for binding services. 1193052860 M * amax im talking about one simple thing.. 1193052868 M * mjt harry: what's the "usual procedure" you're using to cook up the patch? I mean, are there some "usual" conflicting areas between pax/grsec and vserver? 1193052921 Q * yang Remote host closed the connection 1193052949 M * daniel_hozac amax: just use NAT. 1193052967 M * harry mjt: there are a few rejects which have to be fixed 1193052972 M * harry some are harder than others 1193052975 Q * yarihm Quit: Leaving 1193052987 M * harry at first you loose a lot of time figuring stuff out 1193053004 M * harry now i'm getting used to it, so it's quite easy to fix them 1193053005 M * amax daniel_hozac nat, ok whats is pattern to make multiplie 192.168.0.1 on server ? 1193053010 M * harry still need to be careful tough... 1193053046 M * amax daniel_hozac just for let many guests use 192.168.0.1 (same ip virtualized like lo ) ? 1193053087 M * harry and then there are some things that i do differently 1193053108 M * harry (as in: i dont patch the makefile, i just use the localversion thing that's in the kernel vanilla source 1193053126 M * harry and some optimisation in do_brk (mmap stuff) 1193053140 M * harry which i'm sure Bertl will do too soon :) 1193053147 M * daniel_hozac amax: we're not into IP virtualization, we do isolation. 1193053152 M * daniel_hozac it's much more efficient... 1193053166 M * mjt harry: which sounds like a good candidate for upstream when :) 1193053182 M * daniel_hozac mainline are working on level 2 virtualization, most of which should be getting merged in 2.6.24, IIUC. 1193053257 M * harry mjt: it's there in vanilla kernel, the vserver only patch just doesn't use it yet 1193053258 M * mjt by the way, is there any usage of vserver(+grsec) on x86-64? 1193053284 M * mjt i'm trying it right now (compiling stuff), wonder how wide it's usage is ;) 1193053312 M * mjt as in: should i expect commonly recognized probs, or be at my own? :P 1193053335 M * mjt kidding really, but not without a reason 1193053582 M * harry i use that on my server in germany :) 1193053605 M * harry no problems so far 1193053613 M * harry it's a athlon64 machine 1193053630 M * harry and i have 2 quad dualcores over here , running grsec+vserver stuff 1193053646 Q * mire Read error: Operation timed out 1193053678 M * mjt so it should be at least runnable ;) 1193053731 M * mjt what's "level 2 virtualization" anyway? 1193053808 M * mjt hmm 1193053816 M * mjt quad dualcore? 1193053957 M * mjt Google: Your search - "level 2 virtualization" - did not match any documents. 1193054015 M * daniel_hozac commonly referred to as netns. 1193054046 M * daniel_hozac level 2 virtualization means virtual interfaces, with bridging or routing to the host. 1193054105 M * mjt ahh, you mean NETWORK level 2!.. 1193054179 M * daniel_hozac as we were talking about networking, yes... 1193054208 M * mjt -EOUTOFCONTEXT 1193054217 M * mjt ;) 1193054222 N * pmenier_off pmenier 1193054309 N * fb_ fb 1193054451 J * Yvo yvonne@vpn230.rz.uni-mannheim.de 1193054476 P * Yvo 1193054586 Q * Aiken_ Quit: Leaving 1193054813 Q * meandtheshell Quit: Leaving. 1193055536 Q * larsivi Quit: Konversation terminated! 1193055789 J * onox ~onox@145.94.219.223 1193055898 J * hparker ~hparker@linux.homershut.net 1193055929 Q * snooze Remote host closed the connection 1193055953 J * snooze ~o@1-1-4-40a.gkp.gbg.bostream.se 1193056170 J * ema ~ema@rtfm.galliera.it 1193057140 M * bzed mjt: I'm using it on my server, too 1193057160 M * bzed works very well, took some time to figure out which grsec features not to activate, though 1193057232 M * daniel_hozac isn't that what harry's README tells you? 1193057287 P * friendly12345 1193057402 M * bzed no clue :) I somehow prefer to figure such things out by myself, starting with all stuff enabled 1193057405 M * bzed ;) 1193057721 J * JonB ~NoSuchUse@kg1-98.kollegiegaarden.dk 1193058041 M * harry Bertl_oO: not back yet? 1193058179 M * daniel_hozac such is the definition of _oO... 1193058286 M * harry i know... 1193058290 M * harry maybe he forgot ;) 1193058423 M * JonB Bertls Open Office? 1193058472 J * Yvo yvonne@vpn-9-038.rz.uni-mannheim.de 1193058502 M * harry daniel_hozac: do you have any idea if bertl looked at the len >> PAGE_SHIFT vs charged thingy ? 1193058526 M * daniel_hozac you told him about it, no? 1193058538 M * harry i did, but haven't heared anytyhing since 1193058560 M * daniel_hozac where's the delta? 1193058571 M * harry nowhere yet ;) 1193058574 M * harry i'll make one 1193058886 J * rob-84x^ rob@submarine.ath.cx 1193058992 Q * pmenier Read error: Connection reset by peer 1193059019 J * pmenier ~pme@LNeuilly-152-22-72-5.w193-251.abo.wanadoo.fr 1193059088 M * harry http://harry.enzoverder.be/charged.diff 1193059157 M * harry adds sizeof unsigned long in memory, but removes 8 bitshifting operations 1193059656 J * the-dude ~martijn@senturparks.xs4all.nl 1193059709 A * harry has no testmachine for my latest patch 1193059874 J * AndrewLe1 ~andrew@flat.iis.sinica.edu.tw 1193059880 J * sid3wind1 luser@bastard-operator.from-hell.be 1193059881 P * the-dude 1193059893 J * Adrinael_ adrinael@rid7.kyla.fi 1193059900 J * eviljonn1 ~eviljonny@loki.eviljonnys.com 1193059900 J * _Medivh ck@paradise.by.the.dashboardlight.de 1193059902 J * maddoc_ maddoc@social.ostruktur.com 1193059911 J * phreak``_ ~phreak``@deimos.barfoo.org 1193059920 J * [Guy] ~korn@elan.rulez.org 1193059921 J * click_ click@ti511110a080-4620.bb.online.no 1193059921 J * snooze_ ~o@1-1-4-40a.gkp.gbg.bostream.se 1193059925 Q * snooze cation.oftc.net kinetic.oftc.net 1193059925 Q * onox cation.oftc.net kinetic.oftc.net 1193059925 Q * Piet cation.oftc.net kinetic.oftc.net 1193059925 Q * amax cation.oftc.net kinetic.oftc.net 1193059925 Q * Medivh cation.oftc.net kinetic.oftc.net 1193059925 Q * click cation.oftc.net kinetic.oftc.net 1193059925 Q * phreak`` cation.oftc.net kinetic.oftc.net 1193059925 Q * Guest2104 cation.oftc.net kinetic.oftc.net 1193059925 Q * grobie cation.oftc.net kinetic.oftc.net 1193059925 Q * eviljonny cation.oftc.net kinetic.oftc.net 1193059925 Q * opuk cation.oftc.net kinetic.oftc.net 1193059925 Q * Guy- cation.oftc.net kinetic.oftc.net 1193059925 Q * mnemoc cation.oftc.net kinetic.oftc.net 1193059925 Q * maddoc cation.oftc.net kinetic.oftc.net 1193059925 Q * ex cation.oftc.net kinetic.oftc.net 1193059925 Q * Adrinael cation.oftc.net kinetic.oftc.net 1193059925 Q * ard cation.oftc.net kinetic.oftc.net 1193059925 Q * sid3windr cation.oftc.net kinetic.oftc.net 1193059925 Q * AndrewLee cation.oftc.net kinetic.oftc.net 1193059928 J * opuk ~kupo@c213-100-138-228.swipnet.se 1193059929 J * onox ~onox@wlan-145-94-219-223.wlan.tudelft.nl 1193059930 J * mnemoc ~amery@kilo105.server4you.de 1193059930 J * ex ex@valis.net.pl 1193059984 J * grobie ~grobie@master.schnuckelig.eu 1193059990 J * Guest200 ~ensc@www.sigma-chemnitz.de 1193060088 J * DavidS ~david@vpn.uni-ak.ac.at 1193060097 M * sid3wind1 ohnoes 1193060114 N * AndrewLe1 AndrewLee 1193060165 P * DavidS 1193060229 J * Piet ~piet@tor.noreply.org 1193060398 M * harry bloody virgi^H^H^H^H^HHELL! 1193060440 N * sid3wind1 sid3windr 1193061082 T * harry http://linux-vserver.org/ | latest stable 2.2.0.4, 2.0.3-rc3, devel 2.3.0.27, stable+grsec 2.2.0.4 | util-vserver-0.30.214 | libvserver-1.0.2 & vserver-utils-1.0.3 | He who asks a question is a fool for a minute; he who doesn't ask is a fool for a lifetime -- share the gained knowledge on the Wiki, and we'll forget about the minute ;) 1193061188 J * FireEgl ~FireEgl@4.0.0.0.1.0.0.0.c.d.4.8.0.c.5.0.1.0.0.2.ip6.arpa 1193061237 Q * JonB Quit: This computer has gone to sleep 1193061506 Q * Piet Remote host closed the connection 1193061801 Q * lilalinux Remote host closed the connection 1193062566 J * ard ~ard@gw-tweakb16.kwaak.net 1193062928 J * speedy ~speedy@home.speedy.org 1193064112 J * dowdle ~dowdle@scott.coe.montana.edu 1193064278 J * Julius ~julius@p57B25985.dip.t-dialin.net 1193064295 N * snooze_ snooze 1193064534 J * JonB ~NoSuchUse@kg1-98.kollegiegaarden.dk 1193064880 Q * Julius Ping timeout: 480 seconds 1193064886 P * Yvo 1193065344 Q * baggins Ping timeout: 480 seconds 1193065375 N * maddoc_ maddoc 1193065581 J * baggins ~baggins@kenny.mimuw.edu.pl 1193066373 J * fatgoose ~samuel@76-10-151-220.dsl.teksavvy.com 1193066474 Q * fatgoose 1193067045 M * snooze hello, so i'm still having the same prob as before with 2.6.22.6-vs2.2.0.3, works with vanilla but not with patched version (.config is the same), updated to latest udev but didnt make any difference... error is: 1193067049 M * snooze No filesystem could mount root, tried: ext3 ext2 msdos vfat iso9660 xfs 1193067051 M * snooze Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(3,0) 1193067110 M * snooze any ideas? 1193067193 Q * gebura Quit: Quitte 1193067360 M * snooze oh and fs is xfs (which it also says it's trying) 1193067386 M * JonB snooze: yes 1193067391 M * JonB snooze: ideas that is 1193067401 M * JonB snooze: where is your root fs located? 1193067408 J * fatgoose ~samuel@76-10-151-220.dsl.teksavvy.com 1193067439 Q * onox Quit: gtg 1193067442 M * snooze JonB: well /dev/hda1 1193067533 M * JonB snooze: is that a pata or sata device? 1193067562 M * snooze that'd be pata i guess, it's a rather old laptop 1193067573 A * mjt wonders why the hell "standard" initramfs things are so damn limited... 1193067594 M * mjt ("standard" = used by debian, redhat, ...) 1193067609 M * daniel_hozac hmm? 1193067610 M * mjt hda is an IDE device 1193067648 M * snooze yup 1193067699 M * mjt my initramfs present an option to debug any error found during system startup, including dropping into shell, where all the usual unix programs are available (thanks to busybox) 1193067720 M * mjt i wonder why the same can't be adopted by "mainstream" 1193067743 M * mjt it helped me countless number of times 1193067754 A * mjt runs away: some urgent work... 1193067825 Q * phreak``_ Quit: Reconnecting 1193067853 J * phreak`` ~phreak``@deimos.barfoo.org 1193067950 M * snooze http://xn--smrmedsocker-5ib.se/config-2.6.22.6-v .. my kernel config if that would do any good 1193068012 M * daniel_hozac that sounds pretty disgusting... 1193068253 M * snooze oh you're swedish haha ;) 1193068909 J * balbir ~balbir@59.178.35.88 1193069264 Q * JonB Quit: This computer has gone to sleep 1193069579 J * bragon ~bragon@2001:7a8:aa58::1 1193069965 Q * balbir Ping timeout: 480 seconds 1193070177 J * yarihm ~yarihm@vpn-global-dhcp3-207.ethz.ch 1193070317 J * halfdog ~halfdog@chello213047155237.5.sc-graz.chello.at 1193070729 J * balbir ~balbir@59.178.59.28 1193070865 M * halfdog Any vserver developers online at the moment? 1193071349 J * JonB ~NoSuchUse@kg1-98.kollegiegaarden.dk 1193071369 Q * balbir Read error: Operation timed out 1193071376 Q * fb Read error: Connection reset by peer 1193071392 J * fb fback@red.fback.net 1193071430 M * halfdog Any developers online? 1193071512 M * daniel_hozac why? 1193071543 M * halfdog Might have found some security problem and searching for someone to talk about it? 1193071590 M * halfdog Tried Herbert Poetzl via mail, but seems that I did not come through. 1193071632 M * daniel_hozac what's the problem? 1193071714 Q * yarihm Quit: Leaving 1193071741 J * onox ~onox@kalfjeslab.demon.nl 1193071746 M * JonB halfdog: daniel_hozac is pretty knowledgeable 1193071787 M * halfdog yes I know, I'm just searching for some signature informatin 1193071788 Q * JonB Quit: This computer has gone to sleep 1193072214 J * balbir ~balbir@59.178.58.93 1193072273 J * Freax ~Lucifer@bl7-141-98.dsl.telepac.pt 1193072365 Q * Linus Ping timeout: 480 seconds 1193072374 Q * Freax Read error: Connection reset by peer 1193072427 Q * pmenier Quit: Konversation terminated! 1193072443 J * Freax ~Lucifer@bl7-144-187.dsl.telepac.pt 1193072820 Q * balbir Ping timeout: 480 seconds 1193072932 J * mire ~mire@57-170-222-85.adsl.verat.net 1193073014 J * Linus ~Lucifer@bl7-139-254.dsl.telepac.pt 1193073055 Q * Freax Ping timeout: 480 seconds 1193073284 J * Freax ~Lucifer@bl7-134-20.dsl.telepac.pt 1193073494 J * balbir ~balbir@59.178.35.198 1193073635 Q * Linus Ping timeout: 480 seconds 1193073717 N * Bertl_oO Bertl 1193073722 M * Bertl halfdog: hmm? 1193073918 J * Satriani ~Lucifer@bl7-128-81.dsl.telepac.pt 1193073995 M * Bertl wb Satriani! 1193074134 Q * ema Quit: leaving 1193074295 Q * Freax Ping timeout: 480 seconds 1193074305 J * JonB ~NoSuchUse@kg1-98.kollegiegaarden.dk 1193074415 Q * Satriani Ping timeout: 480 seconds 1193075936 J * _gh_ ~gerrit@c-67-169-199-103.hsd1.or.comcast.net 1193076719 J * Julius ~julius@p57B2537F.dip.t-dialin.net 1193077523 Q * FireEgl Ping timeout: 480 seconds 1193077982 M * mjt hmm. vserver+grsec patch become larger. 1193077995 M * mjt from 178kb to 190kb. 1193078045 J * FireEgl ~FireEgl@4.0.0.0.1.0.0.0.c.d.4.8.0.c.5.0.1.0.0.2.ip6.arpa 1193078114 M * hparker More of the good stuff 1193078356 Q * onox Quit: leaving 1193078606 J * ema ~ema@rtfm.galliera.it 1193079054 M * mjt i wonder... do vserver patches conflicts with xen stuff? I mean, is it possible to have virtualized xen guest? 1193079073 M * mjt to run vserver servers inside xen guest, that is 1193079075 M * Bertl nom abd yes 1193079077 M * daniel_hozac yeah. 1193079078 M * Bertl *no 1193079092 M * Bertl hmm, my typing is waaay off 1193079096 A * hparker swaps more of Bertl's key caps around 1193079098 M * daniel_hozac there are a couple of rejects, but they're easy to fix. 1193079107 M * mjt ok, i'll try... ;) 1193079113 M * daniel_hozac (IIRC, been a while since i patched a Xen kernel...) 1193079125 M * daniel_hozac what are they at now, 2.6.20? 1193079142 M * mjt i don't like xen, but it turned out this our NEW machine don't have vm support in hardware. Oh well. 1193079167 M * daniel_hozac what kind of cheap-ass machine is that? :) 1193079186 M * mjt it IS a cheap-ass - a server from ibm for.. $600! 1193079213 M * JonB mjt: cant you run KVM? 1193079216 Q * balbir Ping timeout: 480 seconds 1193079223 M * daniel_hozac JonB: he just said that... 1193079243 M * mjt "IBM System x3200" from dmidecode 1193079251 M * JonB daniel_hozac: i thought all new AMD and INTEL cpu had that special thing 1193079259 M * mjt yup 1193079276 M * mjt but this is pentium-d. They added hw virtualisation in core or core2 1193079287 M * JonB okay 1193079412 M * snooze slab/slub.. any of those two recommended to be used with vserver? 1193079495 M * Bertl I think slab is fine for now (for typical applications) 1193079565 Q * halfdog Ping timeout: 480 seconds 1193079696 M * snooze okey, was just thinking if choosing slub instead of slab might have been why the vserver-kernel wouldnt boot 1193079777 M * Bertl well, have you tried with a mainline kernel and slub? 1193079791 M * snooze yup :p 1193079796 M * snooze it worked 1193079807 M * Bertl so I don't think that is related 1193079812 M * snooze prolly not 1193079833 M * snooze mainline 2.6.22.10 wouldnt work, but 2.6.22.6 and .9 is 1193079851 M * Bertl interesting .. mainline no patches? 1193079851 M * snooze same conf for all of them.. im getting kinda annoyed ;) 1193079859 M * snooze no patches 1193079875 M * Bertl sounds even more like a race issue with your udev/initramfs 1193079902 M * snooze i updated udev to latest version as you recommended tho 1193079914 M * snooze and not using any initramfs 1193079935 A * mjt hates udev... 1193079972 M * Bertl snooze: ahem? how is your userspace getting the device right then? 1193079998 J * balbir ~balbir@59.178.33.152 1193080004 M * snooze or hm, initramfs = initrd no? 1193080005 M * Bertl snooze: I mean, without initramfs, you have to be lucky to have the devices there in time 1193080026 M * JonB snooze: yes, = 1193080060 M * snooze well i just know i'm not using any initrd, never been 1193080068 M * snooze using slackware with lilo 1193080118 M * JonB snooze: using modules? 1193080125 M * Bertl well, then I'm not surprised you have troubles with recent kernels 1193080136 M * snooze JonB: ah no, never 1193080161 M * snooze Bertl: enlighten me why it's needed, please 1193080194 M * Bertl device initialization got lazy (because of udev) 1193080217 M * Bertl so, basically what happens when you boot up, is that the devices 'slowly' fall into place 1193080233 M * JonB Bertl: how does initramfs help that? Does initramfs run a command that makes sure that udev is populated? 1193080240 M * Bertl normally initramfs setup ensures that the 'real' root device is present, before changing into the rootfs 1193080267 M * Bertl if there is no initramfs, you have to be lucky :) 1193080270 M * snooze oh, i had no idea 1193080280 M * Bertl won't work on usb and most scsi at all 1193080291 M * Bertl could be fast enough for ide and sata 1193080307 M * JonB Bertl: until that time you upgrade and reboot it remotely 1193080319 M * snooze well i have been successfully booting kernels from time to time.. patched with vserver, vserver+grsec etc 1193080340 M * snooze but most times not 1193080370 M * Bertl yeah, not unexpected for me ... I would say you can fix most if not all of them by using the initramfs :) 1193080388 M * snooze okey 1193080405 M * snooze does this "problem" only exist with recent kernels? 1193080410 M * Bertl note: I'm no fan of udev either ... :) 1193080412 M * snooze because i really never had it before 2.6.22 1193080441 M * Bertl wit is basically there since 2.6.16 or so (when devfs was removed) 1193080511 Q * Julius Remote host closed the connection 1193080520 M * Bertl just checked, it seems to be 2.6.14 1193080521 M * snooze compiled lots and lots of kernels before 2.6.22 however 1193080527 M * snooze without trouble 1193080547 M * Bertl give it a try, and see if initramfs fixes your 'broken' kernels 1193080560 M * mjt that's a large change 1193080569 M * mjt to "give it a try" 1193080582 M * Bertl not really, just get the mkinitrd from your distro 1193080597 M * Bertl run it, and specify the initramfs created as initrd 1193080628 M * mjt ahh.. provided slackware has proper mkinitrd/whatever 1193080642 M * snooze i believe it has 1193080645 M * Bertl I'm sure it has, unless they are still at 2.6.14 1193080658 M * Bertl (or probably 2.6.12 :) 1193080670 M * snooze default slack 12 is 2.6.21 or something 1193080688 M * Bertl so probably the tools are already installed 1193080733 M * snooze and there is actually a symlink to a "Slackware initrd mini HOWTO" found in /boot, hehe 1193080777 M * snooze just never thought i'd have to touch initrds when i didnt use any modules 1193080801 M * snooze will give it a try now though ;) 1193080807 M * Bertl actually there is a kernel option, IIRC, which can make the kernel 'wait' some time 1193080823 M * Bertl so you might try that one with a reasonable large amount first 1193080886 M * snooze hmm what do you think that option is called? 1193080922 M * Bertl rootdelay=[KNL] Delay (in seconds) to pause before attempting 1193080932 M * Bertl from Documentation/kernel-parameters.txt 1193080944 M * Bertl rootwait[KNL] Wait (indefinitely) for root device to show up. 1193080954 M * dilinger Bertl: i updated the bug; hopefully you got the trac email this time? 1193081016 M * Bertl was it cc to bertl or Bertl? 1193081023 M * dilinger bertl 1193081041 M * mjt is there a difference? 1193081066 M * Bertl unfortunately yes :) 1193081082 M * Bertl dilinger: yes, seems so, errors are fine if you run it several times in a loop or so 1193081086 M * dilinger our trac.. not so smart 1193081100 M * dilinger Bertl: that seems indicative of a race, no? 1193081102 M * mjt that's bizzare - email addys are usually case-insensitive, and all sorts of issues come off when it suddenly becomes case-sensitive... 1193081117 M * Bertl dilinger: otherwise the fill script should not throw any ln errors, just 'no such file' 1193081122 M * dilinger i mean, when you're creating and deleting files in a certain order (and we are), you should always get the same results.. 1193081151 M * Bertl dilinger: you are 'breaking' links inbetween 1193081174 M * Bertl dilinger: which involves splice and is not atomic (never will be, I guess) 1193081280 Q * mire Ping timeout: 480 seconds 1193081289 M * Bertl dilinger: if you can convince Al Viro to implement kernel side copy (preferable inode to inode) we can make it atomic 1193081493 J * larsivi ~larsivi@101.84-48-201.nextgentel.com 1193081895 Q * phreak`` Ping timeout: 480 seconds 1193082220 J * phreak`` ~phreak``@deimos.barfoo.org 1193082227 Q * JonB Quit: This computer has gone to sleep 1193082280 Q * balbir Ping timeout: 480 seconds 1193082937 J * balbir ~balbir@59.178.53.191 1193083090 M * mjt like sys_cp(char __user_ *from_filename, char _user_ *to_filename) ?? 1193083119 M * daniel_hozac doesn't have to be a syscall. 1193083135 M * daniel_hozac we want to use it in the kernel. 1193083159 M * mjt as far as i remember, usual mantra is "don't mess with files from kernel" 1193083186 M * mjt s/from kernel/from within kernel/ 1193083224 M * mjt that'd be sys_cp(int from_fd, int to_fd, int flags) - like sendfile() 1193083260 M * Bertl yes, but the important part is, that we have to do it in the kernel 1193083277 M * daniel_hozac splice already does that... 1193083576 M * Bertl the perfect kernel side mechanism for our purpose would be an inode level dumplication 1193083593 A * mjt imagines a multi-gigabyte COW file which gets copied atomically - thus, no other filesystem access during the time... and a user finally hits "reset" button thinking his system went havoc.. and gets severe filesystem corruption as a result... 1193083625 M * Bertl sure, but we have that, to some extend already, just think lower layer I/O issue 1193083646 M * Bertl (e.g. bad disk, or NFS with missing filier) 1193083652 M * Bertl *filer 1193083661 M * mjt "we" don't have severe fs corruption, at least not that often ;) 1193083689 M * Bertl not talking filesystem corruption, talking about the I/O layer 1193083895 J * onox ~onox@kalfjeslab.demon.nl 1193083899 M * onox I get /usr/lib64/util-vserver/vserver.functions: line 919: -s: command not found 1193083904 M * onox when I try to start a vserver 1193083926 M * mjt . o O { bashisms } 1193083959 J * southtel ~southtel@68-114-23-151.dhcp.gwnt.ga.charter.com 1193083965 M * daniel_hozac onox: when running trunk, use the latest revision before reporting problems :) 1193083989 M * onox good advice :) 1193084047 Q * ruskie Quit: Caught sigterm, terminating... 1193084070 M * Bertl okay, off to bed now .. had a long day, and have to get up early tomorrow ... 1193084077 M * Bertl have a good one everyone! cya! 1193084082 M * daniel_hozac good night! 1193084083 N * Bertl Bertl_zZ 1193084133 J * DavidS ~david@vpn.uni-ak.ac.at 1193084196 M * onox daniel_hozac: now I get: vmemctrl: vc_set_badness(): Function not implemented 1193084237 M * onox /lib/rcscripts/sh/init-vserver.sh: No such file or directory 1193084277 M * daniel_hozac ah, hmm. 1193084448 M * onox daniel_hozac: do you have an idea? 1193084474 M * daniel_hozac for the latter? looks like you're using a Gentoo guest with the gentoo initstyle that was created with really old utils. 1193084502 M * onox damn, you really are knowledgable 1193084539 M * onox hmm 1193084539 M * onox no 1193084549 M * onox i can also not start vservers i created last week 1193084677 M * daniel_hozac i don't see how you get to the latter error though... 1193084691 M * daniel_hozac the first should cause an immediate failure 1193084841 M * onox hmm, normal 0.30.214 ebuild does work 1193084851 M * onox i can start vservers again 1193084883 Q * balbir Read error: Operation timed out 1193085000 Q * southtel Ping timeout: 480 seconds 1193085656 J * ruskie ruskie@ruskie.user.oftc.net 1193085701 J * balbir ~balbir@59.178.37.66 1193085797 M * daniel_hozac onox: 2627 should be fine. 1193086404 M * onox didn't work for me 1193086436 M * daniel_hozac you've tested it already? 1193086454 M * daniel_hozac what's the error this time? 1193086506 M * onox I tested it 30 minutes ago 1193086538 M * daniel_hozac i committed it 12 minutes ago... you have some sort of psychic test method? :) 1193086557 M * onox oh, oops 1193086577 M * onox i didn't see U src/vmemctrl.c :) 1193086580 M * onox i'll test it :+ 1193086651 J * southtel ~southtel@68-114-23-151.dhcp.gwnt.ga.charter.com 1193086689 M * southtel I have a host server with two nics in it, is there some magic iptables config I need to do to get a guest to be able to route outside the local subnet? 1193086742 M * daniel_hozac the guest will route the same way your host does. 1193086758 M * daniel_hozac only difference is that the guest can't use all of the IP addresses. 1193086828 M * onox daniel_hozac: 2627 works :) 1193086833 M * onox what did you change? 1193086843 M * southtel I believe that I know on a base level what you mean, but I just don't understand the dynamics enough to fix our issue... 1193086863 M * daniel_hozac onox: i fixed the bug ;) 1193086893 M * daniel_hozac southtel: so i guess we'll need a paste of ip a; ip r to be able to help you... 1193086946 M * southtel Good times...I'll go find a pastebin... 1193086953 M * daniel_hozac paste.linux-vserver.org. 1193086968 Q * dna Quit: Verlassend 1193087482 M * southtel http://paste.linux-vserver.org/7382 1193087482 M * southtel I included a little bit of explanation about what I'm trying to accomplish... 1193087582 M * onox does somebody know the difference between vmware-server and vmware-workstation? 1193087607 M * daniel_hozac southtel: the problem is that your default route is via the 192.168.x.y network. 1193087710 M * daniel_hozac southtel: you can use source-based routing to change that, e.g. ip rule add from 10.50.50.0/24 table 42; ip route add 10.50.50.0/24 dev eth0 table 42; ip route add default via 10.50.50.1 dev eth0 table 42 1193087714 M * daniel_hozac or something like that. 1193087755 M * southtel Gotcha...I think I'd seen somebody else who had to do something similar, but I hadn't yet come across that today. 1193087778 M * southtel Daniel: Thanks a ton, I'll try that now... 1193087792 M * southtel ...btw, should that be run from the host or the guest? 1193087801 M * daniel_hozac http://archives.linux-vserver.org/200311/0470.html 1193087810 M * daniel_hozac the host. the guest cannot change anything networking related. 1193088276 Q * onox Quit: zzzz 1193088436 Q * speedy Quit: Read error: 1.414213562 (Excessive square root of 2) 1193088634 J * Aiken ~james@ppp121-45-206-11.lns1.bne1.internode.on.net 1193089495 Q * larsivi Quit: Konversation terminated! 1193089923 J * yarihm ~yarihm@84-75-130-73.dclient.hispeed.ch 1193090981 Q * DavidS Quit: Leaving. 1193091291 Q * balbir Ping timeout: 480 seconds 1193092016 J * balbir ~balbir@59.178.42.190 1193092070 Q * ema Quit: leaving 1193092394 J * mire ~mire@136-168-222-85.adsl.verat.net 1193093649 Q * southtel Read error: Operation timed out 1193093850 Q * yarihm Quit: Leaving 1193094386 Q * balbir Read error: Operation timed out 1193094580 J * southtel ~southtel@68-114-23-151.dhcp.gwnt.ga.charter.com 1193094753 M * micah arg, I just realized that my queries for resolution on the why the other arches haven't built on backports was falling on deaf ears because I wasn't identified with nickserv :P 1193095134 J * balbir ~balbir@59.178.32.22 1193095817 Q * balbir Read error: Operation timed out 1193096536 J * balbir ~balbir@59.178.36.28 1193096675 J * tanjix2 ~tanjix@dslb-084-058-015-224.pools.arcor-ip.net 1193097002 Q * tanjix Ping timeout: 480 seconds 1193097061 M * micah daniel_hozac: do you have any historical information about what was in /etc/vservers.conf besides what is there now, and VSERVERS_ROOT and VSERVER_ROOT? 1193097067 M * micah as far as I can tell, this was it 1193097126 M * micah specifically I am interested in what was there in legacy configs 1193097155 Q * tanjix2 Ping timeout: 480 seconds 1193097195 M * micah (or could have been in there) 1193097523 Q * eyck Quit: leaving