1161734418 Q * s0undt3ch Ping timeout: 480 seconds 1161734469 J * s0undt3ch ~s0undt3ch@81.193.56.183 1161734481 J * dmax ~semaj@81.193.56.183 1161734559 M * Bertl welcome dmax! 1161734972 J * Radiance 94e7a38c8b@halt.1984world.eu 1161734983 M * Bertl wb Radiance! 1161735009 M * Radiance thanks mate :) 1161735043 Q * s0undt3ch Quit: leaving 1161735065 J * s0undt3ch ~s0undt3ch@bl4-56-183.dsl.telepac.pt 1161736035 J * Johnnie ~jdlewis@jdlewis.org 1161736606 T * Bertl http://linux-vserver.org/ <- new and shiny | latest stable 2.02.1, exp 2.02.2-rc3, devel 2.1.0, exp 2.1.1-rc43, stable+grsec 2.0.2.1 | util-vserver-0.30.211 | 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 ;) 1161736606 Q * Johnnie Read error: Connection reset by peer 1161736620 M * Bertl *2.1.1-rc43 1161737425 Q * Aiken Quit: Leaving 1161737475 J * Johnnie ~jdlewis@jdlewis.org 1161737942 Q * ms_ Ping timeout: 480 seconds 1161737990 J * FireEgl FireEgl@Sebastian.Atlantica.US 1161738803 Q * Rich_Estill Quit: Leaving 1161741284 J * SMuZZ ~smuzz@monster.dataguard.no 1161741763 M * Bertl okay, I'm off to bed now ... have a good one everyone! cya tomorrow! 1161741769 N * Bertl Bertl_zZ 1161744689 J * ms_ ~ms@arkansas.doc.ic.ac.uk 1161744920 Q * ensc Killed (NickServ (GHOST command used by ensc_)) 1161744930 J * ensc ~irc-ensc@p54B4F709.dip.t-dialin.net 1161745415 Q * ruskie Remote host closed the connection 1161745575 J * ruskie ~ruskie@ruskie.user.oftc.net 1161746290 Q * bronson Ping timeout: 480 seconds 1161746821 Q * Loki|muh Remote host closed the connection 1161746845 J * Loki|muh loki@satanix.de 1161749476 Q * Piet Quit: Piet 1161750984 J * bronson ~bronson@c-71-198-75-160.hsd1.ca.comcast.net 1161753680 M * FaUl *gaehn* 1161757438 J * meandtheshell ~markus@85-125-231-25.dynamic.xdsl-line.inode.at 1161757754 J * bronson_ ~bronson@c-71-198-75-160.hsd1.ca.comcast.net 1161757754 Q * bronson Read error: Connection reset by peer 1161758350 J * Aiken ~james@tooax8-098.dialup.optusnet.com.au 1161759454 Q * sp Ping timeout: 480 seconds 1161760117 Q * ms_ Ping timeout: 480 seconds 1161760447 Q * _are_ Ping timeout: 480 seconds 1161761653 J * _are_ ~are@62.112.159.81 1161762189 J * prae ~Benjamin@host.187.57.23.62.rev.coltfrance.com 1161762541 J * sp sp@ip-87-97.members.virt-ix.net 1161762968 J * dna_ ~naucki@16-206-dsl.kielnet.net 1161764656 J * GhostXz ~Zigara@bas5-kitchener06-1177769094.dsl.bell.ca 1161766139 J * ntrs_ ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com 1161766144 Q * ntrs Read error: Connection reset by peer 1161766672 Q * hardwire Ping timeout: 480 seconds 1161766745 J * hardwire ~hardwire@89-208-58-66.gci.net 1161766846 J * ms_ ~ms@arkansas.doc.ic.ac.uk 1161768884 J * matti_ matti@linux.gentoo.pl 1161769282 Q * _are_ Ping timeout: 480 seconds 1161769312 Q * matti Ping timeout: 480 seconds 1161769312 N * matti_ matti 1161769775 Q * bronson_ Ping timeout: 480 seconds 1161769906 J * coocoon ~coocoon@dslb-084-056-184-085.pools.arcor-ip.net 1161769913 M * coocoon morning 1161772796 Q * coocoon Quit: KVIrc 3.2.0 'Realia' 1161773827 Q * Aiken Ping timeout: 480 seconds 1161774083 Q * virtuoso Ping timeout: 480 seconds 1161774156 Q * brc_ Quit: BitchX: the Cadillac of all clients 1161776745 J * Pazzo ~thomas@dialin-225136.rol.raiffeisen.net 1161776773 M * Pazzo hola :-) 1161776792 M * doener hey Pazzo 1161776814 M * Pazzo Hi doener, how are you? ltns! 1161776843 M * Pazzo Yesterday I compiled 2.6.17.14 + vs2.1.1rc35 + ipv6 experimental patch 1161776853 M * doener quite fine atm, but out-of-time as usual ;) 1161776863 M * Pazzo same here :-) 1161776890 M * Pazzo everything is up and running, ipv4 vservers are ok, ipv6 support (host side) is fine 1161776943 M * Pazzo http://oldwiki.linux-vserver.org/IPv6 seems to be a little bit outdated... is it still necessary to compile vchbind6, to patch the utils etc? 1161777023 M * Pazzo I'm allowed to do some IPv6 vServer tests on a new Dual Xeon... (both of them dual-cored 3.0, /proc/cpuinfo showing 8 cups :-) 1161777045 M * doener ... of coffee? *g* 1161777059 M * doener SCNR 1161777093 M * Pazzo ...and I would like to test some services (apache2, sip proxy etc) inside some vservers running on this host (64bit kernel, btw) 1161777115 M * Pazzo coffee??? all day and all nights :-) 1161777145 M * doener Pazzo: see your typo regarding /proc/cpuinfo ;) (end of sentence) 1161777166 M * Pazzo 8 cups of coffee, great :-) 1161777183 M * doener yeah, would like that in my cpuinfo, too :) 1161777205 M * doener anyway I guess only bonbons and maybe daniel_hozac can tell you about the ipv6 stuff 1161777227 M * Pazzo I have been travelling around with one of our customers this morning, so "only" 4 coffes till now today. But all of them really strong Italian ones ;-) 1161777251 M * doener heh :) 1161777263 M * doener only two for me, third one is in the making 1161777297 J * anonc ~anonc@staffnet.internode.com.au 1161777297 M * doener how's work doing? still happy being an employee? 1161777537 M * anonc evening all - if i want to enable a guest to mount files via loopback (ie mount -o loop /path/to/floppy.img /mnt/loop) with vs 2.1.1-rc31, what more do i need than bcapabilities:CAP_MKNOD and ccapabilities:SECURE_MOUNT ? 1161777618 M * doener what fails? 1161777661 M * Hollow VXC_ADMIN_LOOP? 1161777760 M * anonc 'mount -o loop FreeBoot.img /mnt/loop' gives 'mount: Could not find any loop device'. 'mount -o loop=/dev/loop/0 FreeBoot.img /mnt/loop' gives '/dev/loop/0: Permission denied' 1161777856 M * anonc mounting via the second command in the host using the loop device node that exists inside the guest successfully mounts the image 1161777931 M * anonc ie 'mount -o loop=/vservers/[vserver]/dev/loop/0 /vservers/[vserver]/path/to/FreeBoot.img /mnt/loop' works on the host 1161778324 M * Pazzo doener: yeah, still happy! I'm still working hours and hours like before - but it really makes fun, we are a great team - and I'm allowed to do a lot of funny things on a lot of cute servers :-) 1161778937 J * Milf ~Miranda@141.12.8.129 1161779099 J * Piet hiddenserv@tor.noreply.org 1161779581 M * daniel_hozac anonc: nodev filesystem? 1161779626 M * doener daniel_hozac: nodev /dev? 1161779632 M * doener sounds unlikely 1161779632 M * daniel_hozac Pazzo: yes, it's still necessary to use chbind6 and patch the utils. 1161779636 M * daniel_hozac doener: i agree. 1161779853 M * anonc daniel_hozac: what's a nodev filesystem? 1161779872 M * daniel_hozac a filesystem mounted with the nodev option. 1161779879 M * anonc oh :) 1161779994 M * anonc cat /proc/mounts doesn't show nodev being set (either inside or outside the guest) and I can mknod inside the guest due to the CAP_MKNOD entry in bcaps 1161780083 M * anonc /vservers is bind mounted from a subdirectory of a large local filesystem - let me try bypassing the bind mount... 1161780390 M * Pazzo daniel_hozac: thnx 1161780539 M * anonc nope - same problem. i'm still wondering about the nodev though...hmm 1161780579 M * daniel_hozac if it's not in /proc/mounts, it's not a problem. 1161780713 M * doener anonc: you did follow Hollow's advice to add VXC_ADMIN_LOOP, right? 1161780725 M * anonc /dev/root / ext3 rw,data=journal 0 0 - but later bind mounts do have the nodev set: /dev/root /usr/portage ext3 ro,nodev,data=ordered 0 0 1161780772 M * doener that's expected and does not affect /dev (unless it is on a bindmount) 1161780782 M * doener but all kinds of other things would also bail out then 1161780833 M * anonc doener: i had a grep through the util-vserver 0.30.211 source and the kernel source and couldn't find any reference to it... 1161780853 M * doener the nodev? 1161780862 M * doener should be in secure-mount.c IIRC 1161780885 M * anonc no - the VXC_ADMIN_LOOP flag 1161780889 M * doener ah 1161780912 M * daniel_hozac i think it's pretty new and only in devel. 1161780920 M * doener not in rc42.5 1161780972 M * daniel_hozac VXC_ADMIN_CLOOP is in -rc43, at least. 1161780972 M * doener ah, VXC_ADMIN_CLOOP 1161780994 M * anonc ah - i'm way behind that 'cos I haven't started testing 2.6.18 yet 1161781013 M * daniel_hozac seems CLOOP is only for cryptoloop though, no? 1161781035 M * anonc aha - admin_cloop *is* in 2.6.17-vs2.1.1-rc31 1161781122 N * Bertl_zZ Bertl 1161781127 M * Bertl morning folks! 1161781133 M * daniel_hozac morning Bertl! 1161781134 M * anonc morning bertl 1161781158 M * Bertl hey anonc, how's going? 1161781193 M * anonc Bertl: not bad. rolling out a number of active-active drbd/vserver machines and very happy with them 1161781276 M * Bertl is drbd up to date kernel wise? 1161781301 M * Bertl (just curious, haven't checked yet) 1161781368 M * anonc i'm running 2.6.17-vs2.1.1-rc31-gentoo amd64 with an svn checkout or drbd as of build time :) 1161781396 M * anonc checkout *of* drbd i mean 1161781438 M * Bertl i.c. still in testing phase or already deploying? 1161781612 M * anonc well - i use 0.7.22pre and it seems very stable. 0.8 isn't out of testing yet. by active-active i don't mean using a clustered fs and multiple writing nodes - just 2 drbd disks replicated between two hosts. a bunch of vservers running on one host on its active drbd and a bunch on the other. all managed by heartbeat. 1161781907 M * Bertl ah, i.c. .. makes sense 1161782107 Q * Pazzo Ping timeout: 480 seconds 1161782107 Q * ms_ Ping timeout: 480 seconds 1161782216 M * anonc with heartbeat2 (but not with version 1 from memory - I haven't tried yet) you can make a ring of servers like this as you expand. so have drbd resource X (and thus the vservers stored on it) replicated between machines A and B. Have resource Y replicated between machines B and C, and resource Z replicated between C and A. Keep expanding the ring and you maintain failover for every machine. at least thats the theory. 1161782342 J * virtuoso ~s0t0na@80.253.205.251 1161782434 M * Bertl welcome virtuoso! 1161782615 J * Pazzo ~thomas@dialin-225136.rol.raiffeisen.net 1161782630 M * Bertl hey Pazzo! LTNS! 1161782944 Q * virtuoso Read error: Operation timed out 1161783372 M * Pazzo hi Bertl!!! 1161783435 M * Pazzo from time to time I get some free time to test some new things - and whenever I do so it happens that I found myself to be back in this channel ;-) 1161783447 M * Pazzo this time I'm giving IPv6 a try 1161783488 M * Bertl i.c., maybe you also want to give the brand new 2.1.1-rc43 a spin? 1161783518 M * Pazzo Bertl: sure, n.p. 1161783566 M * Pazzo btw: is there a changelog to be found somewhere? wiki stops at rc28... 1161783577 M * daniel_hozac nobody has updated it since ;) 1161783582 M * Pazzo should I test rc43 with or without ipv6? 1161783598 M * Pazzo :) 1161783617 M * Bertl whatever you prefer 1161783621 M * Pazzo k 1161783626 M * Pazzo is 64bit ok? 1161783631 M * Bertl excellent 1161783636 M * Pazzo perfect. 1161783750 M * Pazzo I'm about to teach some of our sysadmins some basics regarding OpenSER in the next 2-3 hours, I'll compile and test it afterwards 1161783786 M * Pazzo Bertl: link? 1161783793 M * Bertl http://vserver.13thfloor.at/Experimental/patch-2.6.18.1-vs2.1.1-rc43.diff 1161783812 M * Pazzo thnx 1161783844 M * Bertl and for ipv6 1161783846 M * Bertl http://people.linux-vserver.org/~bonbons/ipv6/patch-2.6.18.1-vs2.1.1rc42-ipv6k.diff 1161783866 J * _are_ ~are@gprs-pool-1-022.eplus-online.de 1161783867 M * Bertl (should apply to rc43 too) 1161783967 Q * wenchien Ping timeout: 480 seconds 1161784739 M * Pazzo Bertl: Kernel is compiling, should be finished within 20 minutes. I'll be back in something like two hours... 1161784765 Q * Pazzo Quit: ... 1161784801 J * lilalinux ~plasma@dslb-084-058-207-083.pools.arcor-ip.net 1161784980 J * Pazzo ~thomas@dialin-225136.rol.raiffeisen.net 1161785957 M * Hollow Bertl: do you have one or two minutes? 1161786220 M * Bertl sure 1161786342 M * Hollow i need a bit of help with a exercise... could you take a look at exercise 4 in http://home.xnull.de/work/study/TI3/Exercises/S01.pdf especially if i have explained correctly questions 1 and 2, and help me with question 7, i'd say yes, but have no clue why that is so 1161786496 M * Bertl hmm, as this is in german, let's take it off the channel 1161786521 M * Hollow ok 1161786598 J * pagano ~pagano@131.154.5.21 1161786664 J * Wenix ~wenix@81.7.189.11 1161787907 Q * Milf Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org 1161787978 J * bon ~bon@217.118.101.36 1161788002 M * bon hello .) 1161788015 M * Bertl welcome bon! 1161788027 M * Bertl wb pagano! Wenix! 1161788034 M * bon anyone using vs on rhel4? 1161788053 M * Bertl probably, do you encounter issues? 1161788071 M * bon i haven't found any rpms for rhel x86_64, that's why i'm asking :) 1161788082 M * bon should i give the fedora ones a try then? 1161788096 M * bon (or should i buy glasses? :) 1161788101 M * Bertl you mean for userspace? 1161788122 M * bon no, for the whole vserver setup 1161788133 M * bon i am using vs succesfully on slackware and would like to run it on rhel too 1161788172 M * Bertl well, kernel should not require an rpm, just compiling :) 1161788190 M * Bertl for userspace, I think the fc rpms should be close, maybe not perfect 1161788263 M * bon indeed 1161788283 M * bon hm, first tome kernel compile on rhel ;) 1161788287 M * bon time even 1161788308 J * hossein ~hossein.z@85.185.146.117 1161788380 Q * Johnnie Quit: G'bye! 1161788393 J * Johnnie ~jdlewis@jdlewis.org 1161788405 P * hossein Kopete 0.12.2 : http://kopete.kde.org 1161788444 M * bon i should probably turn off selinux, right? 1161788446 J * djrise ~djrise2b@194.3.30.212 1161788453 M * Bertl bon: can't hurt :) 1161788463 M * Bertl welcome djrise! 1161788463 M * djrise hi everyone 1161788503 M * djrise i have a question for all person in the channel 1161788510 M * bon give it a try ;) 1161788546 M * phedny are there known problems with vservers having a dash in the name (eg. first-aid) 1161788552 M * phedny like: 1161788557 M * phedny tenerife:~# vserver first-aid enter 1161788557 M * djrise does someone use vrrp or ucarp in a vserver ? 1161788557 M * phedny No command given; use '--help' for more information. 1161788570 M * Bertl djrise: what's that? 1161788590 M * bon phedny: vserver 'first-aid' enter didn't work? 1161788590 M * Bertl phedny: which version? 0.30.211? 1161788614 M * phedny 0.30.210 (Debian package) 1161788619 M * phedny problem is only with this vps 1161788619 M * djrise it's virutal ip addresse for high availibility server 1161788631 M * bon linux-ha? 1161788638 M * Bertl djrise: userspace? 1161788659 M * djrise yes 1161788707 M * Bertl how is the kernel involved? i.e. what has it to do? 1161788726 M * phedny hmm, this makes 'vserver first-aid stop' not work either 1161788743 M * Bertl phedny: use vkill 1161788779 M * djrise bert: linux 2.6.17.13 with vserver 2.1.1-rc35 1161788836 J * ms_ ~ms@arkansas.doc.ic.ac.uk 1161788852 M * Bertl djrise: okay, maybe you can explain it to me somewhat .. i.e. what it does, and why ... 1161788885 M * phedny Bertl: tkz 1161788902 M * phedny anyway, renaming it to firstaid without dash doesn't work either, so I guess problem is somewhere else :) 1161788932 M * Bertl phedny: I'd suggest to test with 0.30.211 first 1161788942 M * djrise in a normal server 1161788951 M * Bertl phedny: (and if that fixes it, file a bug report to debian) 1161788962 M * djrise when you use ucarp you can add a virutal adress 1161788987 M * djrise one in the master server 1161789002 M * djrise and one in another (backup) 1161789016 M * djrise when the first server fail down 1161789035 M * phedny Bertl: I'll try .. but could the problem be related to the fact that I have in the fstab file: /dev/vg/vps1 / ext3 rw,dev 0 0 1161789039 M * djrise the second become the master and take the virtual ip address 1161789062 M * djrise it's a technologi use for fail over server 1161789090 M * djrise Bertl: sorry my english 1161789097 M * djrise Bertl : it's ok? 1161789113 Q * Piet Ping timeout: 480 seconds 1161789149 M * Bertl djrise: yeah, let me ask a few questions though 1161789166 M * djrise ok ;-) 1161789195 M * Bertl djrise: how is that different from simply using one and the same ip on two hosts (or from bringing up the specific ip on fail over)? 1161789199 J * Piet hiddenserv@tor.noreply.org 1161789268 M * Bertl vrrp seems to be the protocol standard, as far as I could figure 1161789290 M * Bertl but it refers to 'routers' not hosts 1161789319 M * djrise yes vrrp it's a protocol not gnu but like hpfs (cisco) 1161789326 M * djrise ucarp it's gnu 1161789354 M * djrise it's for routers but now you can use it with host 1161789402 M * Bertl okay, so let's assume, you have ucarp running on a host 1161789420 M * Bertl what does it do normally, what does it do on failover? 1161789440 M * djrise ucarp run on two host 1161789449 M * djrise one is the master and 1161789455 M * djrise another backup 1161789456 M * doener Bertl: I guess the trick is that vrrp covers ip and mac address 1161789482 M * djrise the master have a virtual ip and the bakckup two 1161789489 M * doener the backup box also changes its mac address if the master fails (read on wikipedia ;) 1161789493 M * Bertl doener: how is that different from, let's say, a heartbeat setup? 1161789495 M * djrise when the master fail over the backup become the master 1161789527 Q * Johnnie Ping timeout: 480 seconds 1161789538 M * djrise it's a virtual macaddress for both 1161789552 M * djrise i don't know heartbeat 1161789558 M * Bertl djrise: okay, well, should work as long as the mac change happens for the entire host 1161789572 M * Bertl djrise: i.e. you will not be able to change the mac per guest 1161789595 M * Bertl but I see no problem with two hosts in active-passive mode 1161789625 M * Bertl usually folks prefer an active-active setup though, which typically is realized with e.g. heartbeat 1161789725 M * djrise acitve-passive mode ? 1161789753 M * Bertl one host is 'in use' (active), the other one is in 'stand by' (passive) 1161789798 M * djrise ok so i should try hearbeat maybe ? 1161789815 M * Bertl well, that was just a suggestion (because we already know that :) 1161789829 M * Bertl djrise: the active-active setup would look like this: 1161789851 M * Bertl - you have two hosts, with a bunch of ip addresses for the guests 1161789868 M * Bertl - you put half of the guests on host A and the other alf on host B 1161789898 M * Bertl - you use a shared filesystem or blockdevice to mirror guests on both systems 1161789915 M * Bertl - when one system goes down, you start the guests on the other system 1161789935 M * Bertl - send some arp packets to 'update' the routers 1161789943 M * Bertl - and that's it, failover complete 1161790012 M * Bertl if your 'guests' can run more than once without producing inconsistancies, you can have them runnning with a private ip, and jsut 'map' them to the public at any time 1161790065 M * djrise do you know a link with how to please 1161790069 M * djrise have 1161790093 M * Bertl http://oldwiki.linux-vserver.org/Vserver+DRBD 1161790128 M * Bertl http://oldwiki.linux-vserver.org/advanced+DRBD+mount+issues 1161790198 M * derjohn Bertl, did rc43 only fix the problem I confirmed for AMD64 ? 1161790203 M * derjohn or more ? 1161790204 M * Bertl IIRC, derjohn2 has/had something running like this and a few other folks too 1161790216 M * Bertl ah, hey derjohn (without 2) :) 1161790232 M * Bertl derjohn: only? did it fix it? :) 1161790243 M * derjohn My kernel packerls provide drbd patches, but I do do actively use it (yet) 1161790253 M * derjohn _are_ has experience with it 1161790282 M * Bertl djrise: see, so maybe conteact _are_ here if you need more details 1161790286 M * derjohn linux-vserver.derjohn.de 1161790293 M * derjohn if anyone wants to try 1161790316 M * djrise ok thank you Bertl and everyone 1161790337 Q * _are_ Ping timeout: 480 seconds 1161790339 M * djrise i must read now and try 1161790343 M * Bertl djrise: you're welcome! 1161790343 M * djrise and come back! 1161790350 M * Bertl feel free to hang around 1161790379 Q * djrise Quit: Quitte 1161790605 J * rgl Rui@217.129.151.190 1161790607 M * rgl hello 1161790625 M * Bertl wb rgl! 1161790670 M * rgl why does linux have this thing called HIGHMEM and LOWMEM? ie, I had to enable the HIGHMEM kernel option for it being able to use all available mem 1161790692 M * derjohn Bertl, should I enable CONFIG_BLK_DEV_COW_COMMON ? 1161790706 M * Bertl rgl: before I answer that, how much memory do you have (and what arch)? 1161790711 M * derjohn I do not use Cow or unification yet, but would it have any drawbacks ? 1161790722 M * rgl Bertl, 2G on i386 1161790749 M * Bertl rgl: okay, first: you do _not_ have to enable highmem for that 1161790764 M * Bertl derjohn: no idea, not linux-vserver related 1161790816 M * Bertl rgl: the highmem is required because of the limited address space of the x86 arch (4GB) 1161790830 M * Bertl rgl: which has to be shared/splitted between kernel and userspace 1161790855 M * Bertl rgl: for 2GB a split of 2/2 instead of the 3/1 default will give you the full 2GB 1161790863 M * rgl Bertl, humm so I should change the split instead? 1161790866 M * rgl ah ok :D 1161790878 M * Bertl it is prefereable to enabling highmem 1161790889 M * Bertl because highmem means more overhead on memory access 1161790893 M * rgl Bertl, what is the caveat with the highmem thing? 1161790902 M * Bertl (highmem is mapped in and out when required) 1161790902 M * doener rgl: you're too slow ;) 1161790918 M * rgl doener, man you bet! :D 1161790930 M * rgl Berlt answers faster than I write english hehe 1161790960 M * rgl I see, thx Bertl :) 1161790982 M * Bertl you're welcome! 1161791007 M * rgl lets hope none of my apps needs 2G of address space *G* 1161791037 M * rgl err more than 2g hehe 1161791078 M * Bertl unlikely 1161791227 M * rgl :) 1161791242 M * rgl you guys prefer grsec over selinux? 1161791251 M * Bertl se-what? 1161791277 M * Bertl nah, seriously, if configured properly all of them will work 1161791308 M * rgl hehe 1161791800 M * matti Hi Bertl. 1161791902 M * Bertl hey matti! everything fine? 1161791946 Q * phedny_ Quit: System going out of service 1161792014 M * matti Bertl: Yes, thanks. I hope you too. 1161792055 M * Bertl yeah, everything fine here 1161792280 Q * rgl Quit: Fui embora 1161792369 J * bonbons ~bonbons@83.222.36.111 1161792459 J * rgl Rui@217.129.151.190 1161792461 J * Johnnie ~jdlewis@jdlewis.org 1161792860 Q * prae Quit: Quitte 1161794330 J * the_hydra ~a_mulyadi@202.59.168.29 1161794340 M * Bertl hey the_hydra! :) 1161794349 M * the_hydra hello Bertl! 1161794351 M * the_hydra :) 1161794359 M * the_hydra what a welcome greetz :) 1161794386 J * debugger Rui@217.129.151.190 1161794423 M * the_hydra Bertl: how is life today? 1161794425 M * meandtheshell Bertl: disabling highmem - see above - you wrote enabling ... 1161794439 M * Bertl the_hydra: fine, can't complain ... 1161794478 M * the_hydra Bertl: nice to hear 1161794480 M * Bertl meandtheshell: hmm? 1161794508 M * meandtheshell Bertl: well the overhead comes with highmem enabled 1161794598 M * Bertl yep, so? 1161794673 M * meandtheshell Bertl: for rgl it would be better to disable highmem support - no - am I confused? 1161794681 M * meandtheshell :) 1161794690 M * Bertl yes, isn't that what I said? 1161794703 M * meandtheshell (18:44:10) Bertl: it is prefereable to enabling highmem 1161794763 M * Bertl yep, using a different split is preferable to enabling highmem, so? 1161794818 M * meandtheshell ah - ok - my fault - in that context of course ... leaving the instead of splitting out turns the meaning :) 1161794857 Q * rgl Ping timeout: 480 seconds 1161794891 M * Bertl hmm, considering the typos and my bad english, I guess yes :) 1161794972 M * meandtheshell Bertl: let's forget about it - the thing is cleared up now that's the most important thing ... :-) 1161795060 M * Bertl my words ... 1161795446 N * debugger rgl 1161795451 M * daniel_hozac phedny: that bug should be fixed in 0.30.211. 1161795483 M * phedny daniel_hozac: yes, it is 1161795491 M * phedny sorry for me not noticing it :) 1161795969 M * Pazzo (back) 1161795989 M * Pazzo Bertl: 2.6.18.1-vs2.1.1-rc43ipv6 is up and running 1161796030 M * Pazzo vservers starting, sip proxy and web server running fine 1161796034 M * Bertl excellent! 1161796042 J * Johnsie ~jdlewis@jdlewis.org 1161796051 M * Bertl wb Johnsie! 1161796052 Q * Johnsie Remote host closed the connection 1161796108 M * Pazzo IPv6 support on the host is ok 1161796131 M * Pazzo now way to test ipv6 within vservers - didn't have time to build patched utils 1161796150 M * Pazzo (and will for sure not find enough time today, maybe tomorrow) 1161796155 M * Bertl okay 1161796191 M * Pazzo (tested 64bit host, 64bit kernel, 32bit vservers 1161796220 M * Pazzo sys: 2x dual-core xeon 3.0 1161796234 M * Bertl could you run a few tests for me too? 1161796240 M * Pazzo for sure, tell me 1161796244 Q * matti oxygen.oftc.net osmosis.oftc.net 1161796244 Q * Loki|muh oxygen.oftc.net osmosis.oftc.net 1161796244 Q * ensc oxygen.oftc.net osmosis.oftc.net 1161796244 Q * Radiance oxygen.oftc.net osmosis.oftc.net 1161796244 Q * bragon oxygen.oftc.net osmosis.oftc.net 1161796316 M * Bertl first, the obligatory testme.sh and testfs.sh (if possible) 1161796358 J * matti matti@linux.gentoo.pl 1161796358 J * Loki|muh loki@satanix.de 1161796358 J * ensc ~irc-ensc@p54B4F709.dip.t-dialin.net 1161796358 J * Radiance 94e7a38c8b@halt.1984world.eu 1161796358 J * bragon ~weechat@sd866.sivit.org 1161796361 M * Pazzo where do I get the latest testme.sh from? 1161796381 M * Bertl http://vserver.13thfloor.at/Stuff/SCRIPT/ 1161796396 M * Pazzo thnx 1161796417 Q * Johnnie Ping timeout: 480 seconds 1161796444 M * Pazzo 000,001,011,031,101,102,201,202 -> succeeded 1161796465 M * Bertl could you upload the output to paste.linux-vserver.org? 1161796524 M * Pazzo done 1161796548 M * Pazzo I've never seen testfs.sh before: what does it do / how should I run it? 1161796728 M * Pazzo Interested in dmesg too? 1161796730 M * Bertl it tests the xid tagging and attributes 1161796739 M * Bertl Pazzo: yep, sure 1161796765 M * Bertl to run the testfs.sh, you need the following: 1161796799 M * Bertl a device (real, lvm or loopback) and a mount point, and the filesystem utilites 1161796809 M * Pazzo ok, found http://oldwiki.linux-vserver.org/TestScripts 1161796819 M * Pazzo I'll create a new lvm device. size? 1161796821 M * Bertl note: the test will format the device with all checked filesystems 1161796854 M * Bertl so it must not contain any valuable data and it should be large enough to hold the filesystems 1161796999 M * Pazzo just run: ./testfs.sh /dev/vspace/lvol0 ? 1161797048 J * virtuoso ~s0t0na@80.253.205.251 1161797074 M * Pazzo hmm... params? -l? -t? 1161797077 M * Pazzo -v? 1161797088 M * Bertl testfs.sh -D -M /mnt/test -tx 1161797104 M * Pazzo ok, it's running 1161797110 M * Pazzo (10gigs lvm) 1161797161 M * Pazzo http://paste.linux-vserver.org/562 1161797199 M * Bertl okay, 116 is supposed to fail with altest devel 1161797203 M * Pazzo Bertl: sorry, no jfs/reiser/xfs support in this kernel 1161797211 M * Bertl np, thanks for testing! 1161797215 M * Pazzo what does 116 do? 1161797236 M * Bertl it changes properties of a CoW link 1161797264 M * Bertl as the CoW was improved to handle that (by breaking the link) it succeeds, which is considered a failure in this script 1161797272 M * Pazzo k 1161797395 J * bronson_ ~bronson@66.160.177.224 1161797531 M * Pazzo Bertl: I'm running 2.1rcXY (mostly 2.6.17.13) on all productional servers, it's running great. 1161797559 M * Pazzo how "stable/reliable/whatever" do you consider the experimental IPv6 patch? 1161797604 M * Bertl well, bonbons is very comfortable with his patches 1161797629 M * Bertl I have not spent too much time on it yet, but it looks okay to me 1161797648 M * Pazzo thnx! 1161797693 M * Pazzo I would like to add it to all of my kernels - I would like to (seriously) start playing around with different services :-) 1161797750 M * Bertl well, ipv6 integration is on my todo list 1161797762 M * Pazzo It's going to be a new era for the famous ping of death: 1161797767 M * Pazzo ping6 dead::dead:dead:dead:dead 1161797770 M * Pazzo :-) 1161797772 M * Bertl so once 2.1.1 is out, I'll focus on that 1161797778 M * Pazzo great 1161797844 J * stefani ~stefani@tsipoor.banerian.org 1161797856 M * Bertl hey stefani! 1161797868 M * stefani salut. 1161797913 M * stefani Bertl: i was invited to be interviewd by some IBM-sponsored group about my virtualization use. they had not heard of vservers. :( 1161797946 M * Bertl well, many folks haven't heard yet, so spread the word :) 1161798006 M * Bertl stefani: but for IBM folks involved in virtualization, it's a shame 1161798041 M * Bertl (but you said, IBM sponsored, so maybe no IBM folks there :) 1161798092 M * stefani we shall see. i have one of your papers which i will give them. 1161798113 M * Bertl k, maybe also point them to the wiki/url and the channel ehre 1161798115 M * Bertl *here 1161798198 Q * FireEgl Ping timeout: 480 seconds 1161798232 M * stefani bien sur. 1161798633 M * Pazzo I'm gonna go - cya @ll, bye Bertl! 1161798643 M * Bertl cya! 1161798655 Q * Pazzo Quit: ... 1161798997 J * Johnnie ~jdlewis@jdlewis.org 1161801108 Q * derjohn2 Ping timeout: 480 seconds 1161801139 J * derjohn2 ~aj@dslb-084-058-255-014.pools.arcor-ip.net 1161801579 Q * Adrinael Ping timeout: 480 seconds 1161801586 J * debugger Rui@217.129.151.190 1161801654 N * debugger rgl_ 1161802052 Q * rgl Ping timeout: 480 seconds 1161804237 Q * ms_ Ping timeout: 480 seconds 1161804566 Q * the_hydra 1161805429 P * Wenix 1161806062 J * dreamind ~dreamind@C2107.campino.wh.tu-darmstadt.de 1161806139 M * dreamind Hi folks :) 1161806151 M * Bertl hey dreamind! 1161806158 M * dreamind hi Bertl :D 1161806239 M * dreamind anybody working on a grsecurity + vserver patch for 2.6.18.1? 1161806310 M * Bertl hmm, IIRC there is one already :) 1161806366 M * Bertl ah, no, my fault, harry didn't update yet (AFAICT) 1161806407 M * dreamind hm... maybe I could merge the 2 patches and send them to harry? or something similar? 1161806458 M * dreamind I've already merged the patches for 2.6.17.13 on my own (some weeks ago) 1161806787 M * dreamind Bertl: or is there any other way I could prob. participiate in the project? 1161806917 M * daniel_hozac i don't think harry would mind. 1161807307 M * Bertl of course, you should spend some time on testing the patches too 1161807492 M * dreamind Bertl: of course. :) 1161808004 J * DreamerC_ ~dreamerc@59-115-49-99.dynamic.hinet.net 1161808035 Q * bonbons Quit: Leaving 1161808374 Q * DreamerC Ping timeout: 480 seconds 1161808788 J * debugger Rui@217.129.151.190 1161808971 N * debugger rgl 1161809067 M * rgl you guys known why we have to create a nameserver record on the registrar when we register a domain? 1161809137 M * cehteh registering a domain does not need to imply a namserver entry/service for you 1161809177 M * cehteh some registrars offer that .. maybe just parking the domain .. or redirecting it .. or full access 1161809186 M * cehteh often they charge extra for that service 1161809200 Q * bronson_ Ping timeout: 480 seconds 1161809212 J * bronson_ ~bronson@adsl-64-161-106-11.dsl.snfc21.pacbell.net 1161809252 Q * rgl_ Ping timeout: 480 seconds 1161809361 Q * dna_ Quit: Verlassend 1161809481 M * rgl cehteh, I mean, I don't want to use the registrar DNS server, I want to use my own, but still I have to register its address and IP on the registrar :| 1161809520 M * Bertl address _and_ ip? 1161809535 P * stefani I'm Parting (the water) 1161809565 M * rgl Bertl, the name and ip 1161809594 M * Bertl hmm, and you have to register name/ip from the registrar? 1161809609 M * cehteh rgl: yes the upstream nameserver has to point to your nameserver 1161809620 M * Bertl rgl: I mean, could you give me an example 1161809635 M * cehteh means if you have foo.org. and your nameserver is ns.foo.org 1161809642 M * rgl for example, I register the domain name, example.com, with its registration I have to also supply a nameserver. 1161809659 M * Bertl rgl: that's normal 1161809664 M * cehteh then the .org nameserver has to point to ns.foo.org for each query about foo.org 1161809678 M * cehteh rgl: which registrar btw? 1161809678 M * Bertl rgl: how should name resolving work without a nameserver? 1161809689 M * rgl I also want that nameserver to ne, a.ns.example.com, so I also have to create that a.ns.example.com name in the registrar 1161809708 M * rgl I also want that nameserver, eg, a.ns.example.com, so I also have to create that a.ns.example.com name in the registrar 1161809719 M * Bertl rgl: nah, once you have example.com, you can add that yourself 1161809723 M * cehteh rgl: if your nameserver is withhin your domain then you have to have a glue record from your registrar 1161809735 M * cehteh i learned that the hard way recently ;) 1161809756 M * rgl cehteh, I'm also learning that today! 1161809767 M * cehteh check that http://www.dnsreport.com/tools/dnsreport.ch?domain=yourdomainname ... 1161809805 M * cehteh and if you need such a glue record push your registrar to add one if they dont let you do that via web interface or automatically 1161809805 M * rgl I learn it because I wanted another nameserver, eg: a.ns.other.example.com, so I had to register that one on the registrar too! and I was suprised by that! 1161809812 M * doener rgl: assume that a client looks for the A record for www.example.com. the .com nameserver says "ask ns1.example.com" 1161809838 M * doener rgl: then the client goes and asks about "ns1.example.com", without the glue record, what would the .com nameserver answer? 1161809885 M * rgl doener, the thing is, I had another domain, eg: example.net, and the registrar didnt let me use the nameserver a.ns.other.example.com unless I registered that name in the example.com registrar :| 1161809924 M * rgl what I didn't understand, I still don't understand, is why I need to register the a.ns.other.example.com name! 1161809996 M * doener so the nameserver was in a different domain? shouldn't require a glue record then AFAIK 1161810036 M * rgl cehteh, I did use that tool, and it says everything was fine, but still I had to register the a.ns.other.example.com with the example.com registrar :| 1161810054 M * rgl doener, I'm puzzled too :| 1161810142 M * rgl cehteh, that is, the a.ns.other.example.com has to appear in this site database: http://registrar.verisign-grs.com/whois 1161810171 M * rgl (in the "nameserver" search that is) 1161810966 J * ms_ ~ms@arkansas.doc.ic.ac.uk 1161811448 M * kevinp if /tmp is setup as tmpfs, does that reserve space in RAM for that? So if I make the /tmp 150 MB, do I loose 150 MB of available RAM? 1161811473 M * Bertl not immediately 1161811491 M * Bertl but if somebody puts a 150MB file there, then it will use up RAM/swap 1161811492 M * daniel_hozac Bertl: btw, have you gotten around to the PLM results yet? looks like we're missing a #include in any files using linux/vserver/debug_cmd.h 1161811510 M * Bertl yep, should be fixed in rc43 1161811525 M * kevinp thanks Bertl 1161811531 M * daniel_hozac oh, ok, hehe. 1161811533 M * Bertl (at least I fixed it .. not sure that got in rc43 though) 1161811543 M * kevinp is there a way to specify whether it is RAM or swap? 1161811578 M * daniel_hozac Bertl: hmm, it's not in the -rc43 patch i have here. 1161811588 M * Bertl nah, seems I didn't include it there yet 1161811631 J * Aiken ~james@tooax6-006.dialup.optusnet.com.au 1161811738 M * daniel_hozac strange that it only shows up on some of the 64-bit arches though. 1161812029 M * Bertl yep 1161812154 M * dreamind nite :) 1161812159 Q * dreamind Quit: dreamind 1161812432 M * ray6 reee 1161812437 M * ray6 NE1 on Systems, BTW? 1161812916 M * Bertl hey ray6! hmm? 1161812961 M * ray6 hmm what? 1161812972 M * Bertl NE1? Systems? 1161812978 M * Bertl what's systems? 1161813001 M * ray6 Bertl: the second largest IT expo in germany.. (or europe? :) 1161813027 M * Bertl hmm, interesting, when/where? 1161813038 M * ray6 Bertl: www.systems.de... right now in Munich 1161813059 M * Bertl never heard of it ... seriously :) 1161813078 M * ray6 bertl: oh, interesting. It's "that other one besides cebit" :) 1161813106 M * SNy Well, since CeBIT seems to die, it's just "the other one". 1161813111 M * SNy Or something. 1161813113 M * ray6 :-) 1161813220 M * ray6 SW Soft is there... %-) And NetBSD is demoing Xen... :) 1161813259 M * kevinp Is it possible to run vmstat in a guest? 1161813335 M * Bertl you can run it, but procfs is missing certain infos IIRC 1161813393 M * Bertl daniel_hozac: ah, finally found the fix in my ppc64 kernel :) 1161813395 M * Bertl http://vserver.13thfloor.at/Experimental/delta-compat-fix01.diff 1161813500 M * daniel_hozac looks good. 1161813557 M * Bertl I tried to work around the duplicate inclusion too, but I gave up on that, I think we should remove the duplicate constrain for vs_base.h 1161813656 M * Bertl maybe we should make the duplicate inclusion checks some kind of debug compile feature? 1161813701 M * daniel_hozac that would make sense. 1161813752 M * daniel_hozac i haven't looked in to what's causing the duplicate inclusion. isn't vs_base.h just included from sched.h? 1161813758 M * Bertl yeah, just I need to figure a way to get that tested in PLM 1161813775 M * Bertl include/asm-alpha/uaccess.h : #include 1161813784 M * Bertl that, for example, causes it 1161813815 M * daniel_hozac recursive include of sched.h? 1161813825 M * daniel_hozac or multiple inclusions, i mean. 1161813849 M * Bertl well, it seems that most archs require sched in uaccess, but not x86 1161813880 M * Bertl unfortunately sched.h requires vs_base 1161813914 M * daniel_hozac what else requires vs_base? 1161813924 M * Bertl anyway ... I'm off to bed now .. kind of tired ... have fun folks! 1161813930 M * daniel_hozac ok, good night! 1161813937 N * Bertl Bertl_zZ 1161814659 M * derjohn daniel_hozac, any issues with rc43 (didnt follow this evening) - or can I "provide" mit rc43 to public? 1161814707 M * daniel_hozac derjohn: compile time failures on ppc64 and ia64 without the patch above, no other issues that i'm aware of. 1161814825 M * derjohn ia64 ? pcc ? well, I am lucky : only x86 and x86_64. should I include the delta ? 1161814842 M * derjohn or is it of no use then? 1161814858 Q * michal` Ping timeout: 480 seconds 1161814868 M * daniel_hozac i guess you won't need it. 1161814929 M * derjohn Dimmu, your guesses are fine most time ;) so I stay with the rc43 ;) 1161815013 M * doener oh! a new field in top :) %st 1161815083 M * daniel_hozac for? 1161815092 M * doener no clue yet... 1161815114 M * daniel_hozac heh 1161815118 M * doener searching the manpage for "st" is pointless ;) 1161815132 M * daniel_hozac yeah, i noticed. 1161815174 J * michal` ~michal@www.rsbac.org 1161815315 Q * rgl Quit: Fui embora 1161816355 M * bronson_ Is there any easy way to create a private network interface common to all machines running on a system? 1161816387 M * bronson_ Or just an IP address is fine. 1161816414 M * daniel_hozac you'll have to assign it to each and every one. 1161816416 M * bronson_ I'd like all my virtual machine to be able to access a private HTTP server running on the main machine. 1161816423 M * daniel_hozac (if by machines you mean guests) 1161816429 M * bronson_ Yes, guests. 1161816437 M * daniel_hozac why wouldn't the host's IP address work? 1161816438 M * bronson_ So create another loopback address and then assign it to each machine? 1161816446 M * bronson_ The host doesn't have a single IP. 1161816458 M * bronson_ Oh, and I don't want the HTTP server listening on any public IP. 1161816493 M * daniel_hozac just adding a private IP address would work fine. 1161816502 M * daniel_hozac why would you assign it to the guests though? 1161816512 M * bronson_ (the host has 5 IPs, each of which is distributed to a single guest, and a few guests that have no public interface) 1161816527 M * bronson_ Oh, right. 1161816533 M * bronson_ No need to assign anything to the guests. 1161816539 M * bronson_ Just so long as the private IP is routable. 1161816552 M * bronson_ I'll try that. 1161816556 M * bronson_ daniel_hozac: thanks. 1161816559 M * daniel_hozac no, just so long as the address is on the host. 1161816721 A * mnemoc use dummy0 for the guests 1161816942 M * doener bronson_: anything on the same box is automatically routable via lo 1161816979 M * doener bronson_: take a look at "ip route show table local" 1161817359 Q * meandtheshell Quit: exit (0); 1161817718 J * lilalinux_ ~plasma@dslb-084-058-200-061.pools.arcor-ip.net 1161817826 Q * lilalinux Read error: Operation timed out 1161820008 Q * mire Remote host closed the connection 1161820380 J * kir_home ~kir@213.152.157.70 1161820555 M * derjohn daniel_hozac, still there? 1161820590 M * derjohn Starting vservers of type 'default'...chroot-sh: open("/etc/mtab"): Read-only file system 1161820598 M * derjohn but i can write to /etc ... 1161820604 M * derjohn since 311 uograde