1149552001 M * orionpanda bertl: good suggestion. investigating now. So far I've setup test vserver clusters using GFS, Coda, NFS, Intermezzo, AFS, PVFS, and Lustre
1149552020 M * orionpanda Lustre was _so_ close.
1149552043 M * Aiken 2.6.17-rc5.-vs2.1.1-rc21.3 cow on nfs does not work
1149552055 M * Bertl well, I still don't get why there aren't any recent patches against mainline (for lusre)
1149552056 M * Aiken I am now convinved that is why I went aoe
1149552128 M * orionpanda bertl: i don't know. development is active. There's a new beta out every two weeks. production release scheduled for June.
1149552182 M * Bertl anyway, ocfs is on my todo list (as it is in kernel)
1149552213 M * orionpanda yeah? as in full vserver support with ocfs?
1149552221 M * orionpanda cow/iunlink/barrier, etc?
1149552241 M * Bertl yep, I'd say so, if somebody is going to test it, why not
1149552258 M * orionpanda I would test it. Definitely.
1149552331 M * Bertl sounds good to me .. make sure to stay around adn/or pay a visit in the next few days then ..
1149552368 M * orionpanda absolutely. will do.
1149552380 M * orionpanda aiken: are you using the vblade userspace tools for aoe?
1149552389 M * Aiken yes
1149552451 M * Aiken ls -lh shows the files dates from Mar 18 2005.
1149552474 M * Aiken propably out of date by now but it works well enough I have never felt the need to update
1149552483 Q * Thorsten Quit: Leaving
1149552556 J * shedi ~siggi@inferno.lhi.is
1149552597 M * orionpanda if aoe is a simple block device, then it should be possible to use lvm over multiple aoe devices. However, I'm sure this can't scale beyond a certain point.
1149552647 M * orionpanda I guess I could bond multiple gigabit ethernet ports to deal with bandwidth issues.
1149552663 M * Aiken with quick testing I have had aor and iscsi as part of raid 5 arrays
1149552671 M * Aiken aor = aoe
1149552682 M * orionpanda how well does it perform?
1149552906 M * Aiken at the time I just wanted to see if it would work
1149552973 M * orionpanda I guess I could group multiple RAID-5 physical volumes under one giant logical volume using LVM. That would give a unified namespace and redundancy. The only issue (and a big one) is cache coherency across multiple clients.
1149553006 M * Bertl that's where the cluster fs comes in
1149553007 M * orionpanda Which is where something like lustre of ocfs shines.
1149553008 M * orionpanda yes
1149553278 J * MrX ~urk@219.95.14.75
1149553593 Q * lilalinux Remote host closed the connection
1149553747 M * orionpanda Another problem re AoE for hosting vservers: I just realized that mdadm is not cluster aware. So, I couldn't have multiple vserver hosts mount the same 5 aoe devices in a raid-5 configuration.
1149554066 M * Radiance BertI, hiya :) About latest stable releases ... would you advise to upgrade every time a new release is out ? 
1149554107 M * Bertl Radiance: release yes, release candidate .. depends
1149554138 M * Aiken orionpanda I am not sure is aoe will be suitable for multiclient but I am very happy with it with single clients
1149554214 M * Radiance BertI, ok thanks :)
1149554916 Q * derjohn Server closed connection
1149554937 J * derjohn ~derjohn@80.69.37.19
1149555102 Q * sid3windr Server closed connection
1149555105 J * sid3windr luser@bastard-operator.from-hell.be
1149557492 N * sarnold sars
1149557544 M * Bertl okay, I'm off to bed now .. have a good one everyone ... cya tomorrow!
1149557555 N * Bertl Bertl_zZ
1149559490 Q * DreamerC Ping timeout: 480 seconds
1149561297 J * DreamerC ~dreamerc@59-112-8-132.dynamic.hinet.net
1149561297 Q * shedi Read error: Connection reset by peer
1149561406 M * orionpanda Does anyone know to what extent Vserver supports XFS?
1149561448 M * orionpanda I'm looking through the vserver patch and I can see that iunlink and such have been added for xfs. How stable is the XFS support?
1149561746 J * Aiken_ ~james@tooax6-195.dialup.optusnet.com.au
1149561780 Q * DreamerC Ping timeout: 480 seconds
1149561872 J * DreamerC ~dreamerc@59-112-8-132.dynamic.hinet.net
1149561978 M * micah orionpanda: as stable as XFS is normally I believe
1149562077 Q * Aiken Ping timeout: 480 seconds
1149562160 M * orionpanda micah: do you know if vserver cow, iunlink, and barriers are fully functional on XFS?
1149562186 J * shedi ~siggi@inferno.lhi.is
1149562196 M * micah orionpanda: no, i can't speak to that
1149562540 Q * shedi Read error: Connection reset by peer
1149563336 J * shedi ~siggi@inferno.lhi.is
1149563735 Q * shedi Read error: Connection reset by peer
1149564694 J * shedi ~siggi@inferno.lhi.is
1149569147 Q * Zaki Ping timeout: 480 seconds
1149569220 Q * shedi Read error: Connection reset by peer
1149569415 Q * sezuan Remote host closed the connection
1149570151 J * shedi ~siggi@inferno.lhi.is
1149570237 J * Zaki ~Zaki@212.118.100.121
1149570759 Q * waldi Server closed connection
1149570779 J * waldi ~waldi@bblank.thinkmo.de
1149571005 Q * hallyn Ping timeout: 480 seconds
1149571521 J * gerrit ~gerrit@c-67-160-146-170.hsd1.or.comcast.net
1149573185 M * jkl anyone have thoughts on gentoo-hardened and vserver?
1149573231 M * jkl specifically can the vserver code run correctly with selinux, grsecurity, pax, PIE,SSP, RSBAC?
1149575260 Q * jake- Server closed connection
1149575276 J * jake- psybnc@murlocs.org
1149575277 Q * FireEgl Ping timeout: 480 seconds
1149575390 Q * ag- Ping timeout: 480 seconds
1149575852 J * ag- ag@caladan.roxor.cx
1149577037 Q * Aiken_ Quit: Leaving
1149578230 Q * shedi Read error: Connection reset by peer
1149579001 J * shedi ~siggi@inferno.lhi.is
1149579777 N * otaku42_away otaku42
1149580200 M * harry 2.6.16.20!!!!!!!!
1149580442 Q * blackfire Server closed connection
1149580442 Q * shedi Read error: Connection reset by peer
1149580446 J * blackfire blackfire@dp70.internetdsl.tpnet.pl
1149580681 J * dna ~naucki@dialer-186-114.kielnet.net
1149580937 J * FireEgl Atlantica@Atlantica.IRCNut.Com
1149581342 J * shedi ~siggi@inferno.lhi.is
1149582587 J * _coocoon_ ~coocoon@p54A07C2F.dip.t-dialin.net
1149582598 M * _coocoon_ morning
1149582917 Q * MrX Ping timeout: 480 seconds
1149583110 J * Viper0482 ~Viper0482@p549755FA.dip.t-dialin.net
1149584025 Q * shedi Read error: Connection reset by peer
1149584557 J * cdrx ~legoater@80.118.33.70
1149584621 J * ||Cobra|| ~cob@146.50.22.204
1149585175 Q * _coocoon_ Ping timeout: 480 seconds
1149585955 Q * eyck Quit: leaving
1149586374 Q * cdrx Read error: Operation timed out
1149586441 Q * sid3windr charon.oftc.net keid.oftc.net
1149586441 Q * notlilo charon.oftc.net keid.oftc.net
1149586596 J * Loki|muh_ loki@satanix.de
1149586612 J * eyck eyck@ghost.anime.pl
1149586718 J * sid3windr luser@bastard-operator.from-hell.be
1149586718 J * notlilo debian-tor@tor-irc.dnsbl.oftc.net
1149586788 J * shedi ~siggi@dsl-og-108-50.du.vortex.is
1149586802 J * _starlein star@fo0bar.de
1149586812 J * cdrx ~legoater@70-33-118-80.kaptech.net
1149586863 Q * starlein Ping timeout: 480 seconds
1149586918 Q * Loki|muh Ping timeout: 480 seconds
1149586918 N * Loki|muh_ Loki|muh
1149587580 Q * cdrx Read error: Operation timed out
1149587651 J * cdrx ~legoater@70-33-118-80.kaptech.net
1149588447 J * DarthVader ~Aniken@203.177.212.163
1149589350 Q * DarthVader Quit: Leaving
1149589378 J * DarthVader ~Aniken@203.177.212.165
1149589378 Q * DarthVader Quit: 
1149589393 J * DarthVader ~Aniken@203.177.212.165
1149589512 Q * cdrx Read error: Operation timed out
1149589605 Q * DarthVader Quit: 
1149589772 J * cdrx ~legoater@70-33-118-80.kaptech.net
1149589794 Q * wenchien Server closed connection
1149589821 J * wenchien ~wenchien@221-169-69-23.adsl.static.seed.net.tw
1149590197 Q * cdrx Read error: Operation timed out
1149590393 J * DarthVader ~Aniken@203.177.212.165
1149590605 J * cdrx ~legoater@70-33-118-80.kaptech.net
1149591332 Q * cdrx Ping timeout: 480 seconds
1149593819 J * cdrx ~legoater@cap31-3-82-227-199-249.fbx.proxad.net
1149593996 N * Bertl_zZ Bertl
1149594000 M * Bertl morning folks!
1149594042 J * harti ~hw@83-215-237-5.seek.stat.salzburg-online.at
1149594052 M * harry hey Bertl 
1149594058 M * harry where's my rc22!? ;p
1149594059 M * Bertl welcome harti! hey harry!
1149594066 M * harry (i;m a real pain in the ass, aren't i :)
1149594075 M * Bertl harry: no, not really :)
1149594076 M * Hollow morning :)
1149594086 M * harry hmm... i'll try harder then :)
1149594087 Q * Viper0482 Ping timeout: 480 seconds
1149594090 M * harry :p
1149594098 M * harry i've seen there is a new grsec too
1149594101 M * harry and a new kernel...
1149594109 M * harry so i've got work todo when rc22 comes out :)
1149594178 M * Hollow Bertl: a question for the kernel expert... when i malloc(2) and pass the malloced memory to fork+execve and free it afterwards in the parent, can the child still use it in its argv?
1149594185 J * mire ~mire@98-167-222-85.COOL.ADSL.VLine.Verat.NET
1149594212 M * Hollow malloc(3) even
1149594231 M * harry Hollow: i would say: yes
1149594235 M * Adrinael Aren't all memory pages copied anyway
1149594244 M * Adrinael Including stack, and execve params
1149594245 M * harry since it copies the task struct etc...
1149594250 Q * harry Server closed connection
1149594250 J * harry ~harry@d54C2508C.access.telenet.be
1149594266 M * harry so the fork duplicates the memory "pointer"
1149594282 M * Hollow and execve duplicates it in the user mode stack
1149594283 M * harry free only frees once, one pointer to that mem still exists, so it's not free'd
1149594285 M * harry (i think)
1149594296 M * Hollow well a pointer is only a pointer
1149594302 M * Bertl Hollow: depends on _how_ the fork is done, with fork() it should work quite fine, with clone() only if you copy the memory properly
1149594304 M * Hollow if you free momry it points to the momry is gone
1149594308 M * harry yeah, but free should sanitize that :)
1149594326 M * harry (imho)
1149594340 M * Hollow Bertl: ok, thanks.. CLONE_VM that would be, right?
1149594351 M * harry i wouldn't bet anything on that tho...
1149594354 M * harry ;)
1149594424 M * Hollow harry: guess you're wrong :)
1149594441 M * Hollow simple test: http://phpfi.com/122224
1149594459 M * Hollow will bail out with: double free or corruption
1149594468 M * harry awtsj
1149594485 M * harry hmm...
1149594505 M * harry that could lead to quite interesting race conditions! :)
1149594586 M * Hollow but in the case of CLONE_VM the usermode stack will be duplicated (same pointer, different physical adress)
1149594633 M * harry that would seem normal... clone duplicates stuff
1149594650 M * harry hence the name :)
1149594655 M * Hollow yay!
1149594656 M * Hollow :)
1149594677 M * Hollow but it's different memory, that#s what i wanted to say.
1149594682 M * harry uhu
1149594688 M * harry and fork doesn't do that
1149594697 J * Viper0482 ~Viper0482@p5497780F.dip.t-dialin.net
1149594699 M * Hollow it does..
1149594712 M * Hollow at least bertl said it does :o
1149594818 M * Hollow hm.. but fork is only with SIGCHLD as clone flags
1149594834 M * Hollow *confused*
1149594849 M * Hollow argh
1149594852 M * Hollow <- dumb
1149594863 M * harry <-- dito
1149594864 M * Hollow CLONE_VM means they _share_ virtual memory
1149594892 M * Bertl yup
1149595043 M * harry Hollow:        Under Linux, fork is implemented using copy-on-write pages, so the only penalty incurred by fork is the time and memory required to duplicate the parent’s  page  tables, and to create a unique task structure for the child.
1149595093 M * Adrinael That's just implementation detail
1149595097 M * Hollow yup, CoW is clear
1149595097 M * Adrinael Semantically they are copied.
1149595215 M * harry so, with fork, you can execve with the malloc'ed area in argv
1149595223 M * harry so, with clone, you can't execve with the malloc'ed area in argv
1149595234 M * harry (if you free that space in main)
1149595250 M * Adrinael s/with clone/with clone and flags thisandthat/
1149595285 M * harry Adrinael: yeah, ok :)
1149595295 M * Hollow what about execve without fork?
1149595302 M * Hollow man page says stack will be overwritten
1149595312 M * Hollow does that mean free?
1149595496 M * Adrinael execve without fork? How do you suppose to be able to call free() after execve?
1149595520 M * harry execve just changes the context of the current running process afaik
1149595529 M * Adrinael execve doesn't return
1149595531 M * harry Adrinael is right
1149595539 M * harry that case is irrelevant
1149595540 M * harry :)
1149595572 M * Hollow so that means, i can pass malloced mem to execve's argv and don't care
1149595584 M * Bertl yup
1149595593 M * Hollow good
1149598031 Q * DarthVader Quit: Leaving
1149598120 P * click [IRSSI]
1149598313 J * click click@ti511110a080-0437.bb.online.no
1149598534 M * Bertl wb click!
1149598734 M * Adrinael Three easy numeric questions:
1149598736 M * Adrinael What is the latest usable version, what kernel version does it use, and what tools version does it require?
1149598749 M * Adrinael Usable means here "does not crash and burn"
1149598767 M * daniel_hozac what version does that?
1149598790 J * DarthVader ~Aniken@203.177.212.163
1149598802 M * Adrinael I don't know.
1149598811 M * Adrinael If I knew such things, I wouldn't ask!
1149598816 M * Bertl lol
1149598834 M * Bertl latest usable version is 2.1.1-rc21
1149598850 M * Bertl for kernel 2.6.16.17+
1149598876 M * Bertl it requires util-vserver 0.29 or later, but I'd use 0.30.210 or later
1149598913 M * Adrinael I seem to have 0.30.209
1149598925 M * Adrinael Whee, so today's todo: New kernel.
1149600324 M * harry i want my rc22! :)
1149600391 M * Bertl harry: you know the magic word?
1149600401 M * harry tonight?
1149600412 M * Bertl hmm, nope :)
1149600429 M * harry hmm.... fuck off, harry?
1149600430 M * harry ;)
1149600441 M * Bertl hmm, wrong again :)
1149600443 M * harry quit your bitchin', harry?
1149600445 M * harry ;)
1149601539 Q * tokkee Server closed connection
1149601562 J * tokkee tokkee@ssh.faui2k3.org
1149601582 M * Bertl harry: well, not really ....
1149601650 M * Bertl harry: but wikipedia can help you :)
1149601656 M * harry hmm.... 
1149601666 M * harry if i only know what to look for :)
1149602051 M * sid3windr wikipedia.org/wiki/magic_word ;)
1149602072 M * sid3windr haha
1149602073 M * sid3windr it exists
1149602187 M * Bertl :)
1149602552 Q * mountie Quit: LUNCK!
1149603087 M * harry ahaaaaaaa
1149603093 M * harry i want my rc22 please!
1149603163 M * Bertl wow! here you go!
1149603177 T * Bertl http://linux-vserver.org/ | latest stable 2.01, 1.2.10, 1.2.11-rc1, devel 2.1.0, exp 2.{0.2,1.1}-rc22 | util-vserver-0.30.210 | 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 ;)
1149603185 M * phedny :)
1149603186 M * Loki|muh . o O ( was ahaaaaaaa the magic word? )
1149603188 M * Hollow heh
1149603207 M * phedny Loki|muh: I guess you'd have to read 15:54:11 < sid3windr> wikipedia.org/wiki/magic_word ;)
1149603233 M * Loki|muh phedny: thats why I ask, I cannot find it there ;)
1149603249 M * phedny Loki|muh: just below the horizontal line
1149603254 M * phedny "The term magic word may also refer to the word please when used by adults to teach children politeness:"
1149603267 M * Loki|muh ah ;)
1149603298 M * harry wiiiiiiii :)
1149603311 M * Bertl harry: enjoy!
1149603314 M * harry tnx
1149603320 M * harry time for 0xc0ffee
1149603320 M * Bertl you're welcome!
1149603333 M * harry then... backups, then, ask where my network cards are
1149603338 M * harry then... fix da patch!
1149603343 M * harry new grsec, new kernel, new vserver
1149603351 A * harry living la vida loca
1149603352 M * harry ;)
1149603356 M * Bertl lol
1149603744 J * _Roey ~katz@h-69-3-4-130.mclnva23.covad.net
1149603745 N * _Roey Roey
1149603747 M * Roey hi!!
1149603749 M * Roey Hey Bertl!
1149603759 M * Roey woohoo.
1149603766 M * Roey I'm still registered w/the system.
1149604065 M * Bertl hey Roey!
1149604147 M * Roey =)
1149604189 M * Roey Bertl,
1149604198 M * Roey how 's ngnet coming do you know?
1149604207 M * Bertl slowly ...
1149604256 M * Roey ok.
1149604258 M * Roey I dunno.
1149604263 M * Roey I'm sorta getting tired of managing this.
1149604275 M * Bertl hmm?
1149604281 M * Roey My original plan was to segment all these services (web, mail, openvpn, dns) into separate guests.
1149604284 M * Roey but:
1149604291 M * Roey openvpn doesn't work with vserver
1149604299 M * daniel_hozac ... yes it does.
1149604301 M * Roey vserver needs kernel patches
1149604302 M * Roey etc.
1149604304 M * Roey too much! who cares.
1149604322 M * Roey daniel_hozac:  in a weird way yes? I looked at the directions and they looked pretty complicated.
1149604343 M * Bertl we will probably simplify that in the near future
1149604370 M * Roey how so?
1149604407 M * Bertl well, assumed that an address is known (beforehand) tun/tap is not a big deal
1149604664 M * daniel_hozac Bertl: looks like you lost delta-saddr-fix01.diff for 2.0.2-rc22.
1149604681 Q * harti Quit: Leaving
1149604687 M * daniel_hozac s/01/02/
1149604702 M * daniel_hozac which happens to be missing a C in the first place :)
1149604712 M * daniel_hozac +#ifdef ONFIG_VSERVER_REMAP_SADDR ;)
1149604755 J * Zaki_ ~Zaki@88.213.16.126
1149604864 M * Roey oh
1149604866 M * Roey and one more thing
1149604881 M * Roey vserver I had to install on a debian-unstable server.
1149604886 M * Roey and I don't wanna do that :)
1149604897 M * daniel_hozac "had to"?
1149604942 Q * Zaki Ping timeout: 480 seconds
1149604953 M * Roey I set this thing up some time ago.
1149604953 M * daniel_hozac Bertl: and isn't delta-osfsys-fix01 relevant for stable as well?
1149604987 M * Bertl Roey: you do not have to explain your decisions, whatever they are ...
1149605001 M * Roey <daniel_hozac> "had to"?
1149605002 M * Roey the point is,
1149605011 M * Roey I feel like I'm on a rocky but slipper cliff
1149605027 M * Bertl daniel_hozac: ah, yes, will fix it in place, just updated the wrong branch
1149605027 M * Roey thankfully I only have one of the three guests actually switched over
1149605045 M * Roey Bertl:  will vserver become part of mainstream kernel anytime?
1149605068 M * Bertl no, but a generic virtualization will (which will be used by Linux-VServer and others)
1149605078 M * Roey what do you mean?
1149605091 M * Roey this generic virtualization using xen? or VT/pacifica?
1149605136 M * Bertl daniel_hozac: ad delta-saddr-fix01.diff+02, I don't think we should put that into stable, do you?
1149605144 M * Bertl Roey: neither nor ...
1149605152 M * daniel_hozac Bertl: well, it's already in stable.
1149605163 M * Roey daniel_hozac:  what is?
1149605178 M * Bertl hum, right ...
1149605201 M * Bertl ah, well, it's probably fine, but I just realized that the fix02 is broken :)
1149605219 M * daniel_hozac yep.
1149605383 M * Bertl but 2.1.x is fine ...
1149605392 M * harry rc22 probs?
1149605407 Q * DarthVader Quit: Leaving
1149605458 M * Bertl harry: will be fixed up in place in a minute
1149605468 M * harry kinkyness!!
1149605471 M * daniel_hozac and only for stable.
1149605560 M * Bertl okay, done
1149605802 Q * notlilo autokilled: A user from your address has violated network policy. Please email support@oftc.net if you feel this is in 
1149605837 M * Bertl hmm, seems like lilo has still troubles?
1149605853 M * daniel_hozac apparently.
1149605953 M * daniel_hozac what did you change in devel? or did you just regenerate the patch?
1149605966 M * Bertl the time stuff should be there
1149605994 Q * ray6 Server closed connection
1149606022 M * daniel_hozac s/should/shouldn't/?
1149606038 M * Bertl no, it is supposed to be in devel
1149606044 J * Zaki[] ~Zaki@212.107.127.127
1149606052 M * Bertl it wasn't in rc21 AFAIK
1149606061 M * daniel_hozac no, but it was in the previous rc22 patch.
1149606098 M * Bertl color me confused, what are you talking about?
1149606139 M * daniel_hozac ok, nevermind. the 2.1.1-rc22 patches don't differ.
1149606167 M * Bertl okay, I didn't upload a new one, just touch the time :)
1149606334 Q * ||Cobra|| Remote host closed the connection
1149606402 Q * Zaki_ Read error: Operation timed out
1149606595 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com
1149606652 Q * mountie Remote host closed the connection
1149607194 J * stefani ~stefani@tsipoor.banerian.org
1149607417 M * Bertl welcome stefani!
1149607424 M * stefani hola
1149607482 M * stefani I had a vserverguest bring down one of my servers.  the  VSZ was huge. :\
1149607674 M * Bertl really? maybe VM limits next time?
1149607759 M * stefani that is something i have to master, yes 
1149608045 J * s_ sp@83-215-237-4.seek.stat.salzburg-online.at
1149608117 J * harti ~harti@85-124-100-86.dynamic.xdsl-line.inode.at
1149608137 M * Roey stefani:  I thought vserver was rather safe
1149608137 M * Roey ?
1149608139 M * Roey and stable?
1149608204 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com
1149608239 N * s_ sascha
1149608244 M * stefani Roey:   the issue could well be due to the comment spam on the vserver taking over resources, and since that vserver did not have any special limits on it, 
1149608339 M * stefani it did lead to 'out of memory' on the host
1149608435 Q * gerrit Ping timeout: 480 seconds
1149608444 M * Roey in any case I don't know what to do now.
1149608464 M * Roey I have four services I want to host on one system:  mail, dns, web, openvpn.
1149608479 M * Roey should I use vserver? xen? uml? chroot?
1149608614 Q * sascha Quit: time to go
1149608658 M * Bertl well, chroot is the default linux approach
1149608681 M * Bertl xen and uml will require you to configure a complete guest (with kernel) for each service
1149608708 M * Roey right.
1149608709 M * Bertl Linux-VServer can be used to enhance the chroot approach as far as having separate guests for each service
1149608713 M * Roey but
1149608723 M * Roey only chroot does not require recompiling any kernel.
1149608729 M * Roey you can use stock kernel if you want.
1149608737 M * Roey I chose vserver
1149608738 M * Bertl yes, but it's insecure ...
1149608738 M * Roey but
1149608749 M * Roey I do NOT like that the host is debian/unstable.
1149608753 M * Roey I want to change it to debian/stable.
1149608756 M * Roey and ofr athat matter
1149608768 M * Bertl well, go ahead, it doesn't matter what distro you have on the host
1149608770 M * Roey I want to change it back from amd64 -> ia32.
1149608774 M * Roey amd64 causes too many probelms.
1149608786 M * Bertl again, feel free to do so .. no problem there either
1149608820 M * Roey I'm trying to think out how to do it
1149608824 M * Roey the guests are 32-bit.
1149608831 M * Roey the host is 64-bit and Unstable.
1149608858 M * Bertl sounds fine to me, but as I said, feel free to change that
1149608914 M * Roey :)
1149609006 M * harry can someone explain this?: http://pastebin.com/762582
1149609027 J * lilalinux ~plasma@dslb-084-058-236-114.pools.arcor-ip.net
1149609105 M * Bertl do a pvscan 
1149609114 M * Bertl then probably vgscan and lvscan
1149609128 M * harry pc-10-33-20-084:~# pvscan Found duplicate PV 0aTlT6pnZ0ThblnbVJ8hj7p13qEeXolL: using /dev/sdb1 not /dev/sda1 Found duplicate PV 0aTlT6pnZ0ThblnbVJ8hj7p13qEeXolL: using /dev/sdb1 not /dev/sda1 PV /dev/md1    VG sysvg   lvm2 [32.59 GB / 608.00 MB free] PV /dev/sda1              lvm2 [68.36 GB] Total: 2 [100.95 GB] / in use: 1 [32.59 GB] / in no VG: 1 [68.36 GB]
1149609141 M * harry how do i remove that lvm stuff on /dev/md2?
1149609193 M * Bertl dd if=/dev/zero of=/dev/md2 
1149609242 M * harry nono... it's in lvm.conf... i have to remove sd[ab] checks
1149609774 M * Bertl okay, off for now .. back later this evening ...
1149609779 N * Bertl Bertl_oO
1149609990 J * bonbons ~bonbons@83.222.39.166
1149610133 A * harry rules!!!!!!!
1149610772 Q * jkl Ping timeout: 480 seconds
1149610854 Q * trasher Server closed connection
1149610887 J * trasher manu@217.160.128.201
1149611331 Q * stefani Remote host closed the connection
1149611343 J * _coocoon_ ~coocoon@p54A07478.dip.t-dialin.net
1149611410 Q * Medivh Server closed connection
1149611416 J * Medivh ck@paradise.by.the.dashboardlight.de
1149611585 N * sars sarnold
1149612125 J * ray6 ~ray@vh5.gcsc2.ray.net
1149612128 M * ray6 reee
1149612134 Q * _coocoon_ Quit: KVIrc 3.2.0 'Realia'
1149612375 Q * harti Quit: Leaving
1149613053 Q * mire Ping timeout: 480 seconds
1149613092 Q * mnemoc Ping timeout: 480 seconds
1149613572 J * mnemoc ~amery@216.241.24.10
1149613602 J * doener ~doener@i5387D2A1.versanet.de
1149613605 Q * meebey Server closed connection
1149613616 J * mire ~mire@109-167-222-85.COOL.ADSL.VLine.Verat.NET
1149613620 J * meebey meebey@booster.qnetp.net
1149615142 N * otaku42 otaku42_away
1149616236 J * doener_ ~doener@i5387FDB1.versanet.de
1149616396 Q * doener Read error: Operation timed out
1149617133 Q * cdrx Quit: Leaving
1149617446 M * tokkee How can I limit network resourcen in a guest system?
1149617468 M * daniel_hozac just like you would on a regular Linux box.
1149617471 M * tokkee resources even ;-)
1149617488 M * tokkee daniel_hozac: Okay... thx.
1149617740 Q * daniel_hozac Server closed connection
1149617754 J * daniel_hozac ~daniel@c-2d1472d5.010-230-73746f22.cust.bredbandsbolaget.se
1149617969 J * michal` ~michal@www.rsbac.org
1149618178 J * stefani ~stefani@tsipoor.banerian.org
1149618494 Q * shedi Quit: Leaving
1149618996 Q * stefani Quit: I Quit!
1149619437 J * mkhl ~mkhl@200-153-181-192.dsl.telesp.net.br
1149619661 M * bragon hello
1149619674 M * bragon i have a problem with the new debian system packages
1149619675 M * harry hey bragon 
1149619678 M * bragon Diff
1149619679 M * harry that's not good
1149619682 M * bragon in my guest
1149619703 M * harry sry... gf here... gotta go
1149619704 M * bragon my guest don't want to upgrade with the new debian system
1149619717 M * micah bragon: what do you mean?
1149619719 M * bragon people have a isue ?
1149619737 M * micah bragon: no...
1149619742 M * bragon micah before we have packages.gz for packages to upgrade
1149619749 M * bragon no we have Diff
1149619761 M * bragon with Diff i don't manage to upgrade my vserver with apt
1149619797 M * micah bragon: this sounds like a debian issue, not a vserver issue...
1149619815 M * bragon when i'm not in a guest my upgrade is ok
1149619828 M * bragon i have this problem only with vserver
1149619843 M * micah bragon: what commands are you running?
1149619859 M * bragon apt-get update && apt-get upgrade
1149619876 M * micah and what happens?
1149619894 M * bragon i never have package to upgrade
1149619908 M * micah you run these commands _inside_ the vserver?
1149619935 M * micah and what version of debian are you running?
1149619946 M * bragon in the host sarge
1149619951 M * bragon in the guest unstable
1149619971 M * micah you run these commands _inside_ the vserver?
1149619977 M * bragon yes
1149619980 M * bragon of course
1149620009 J * enet ~jpduyx@146.50.228.22
1149620009 M * micah then, either your mirror that you use in your apt/sources.list is not updating, or there are no new packages for what you have installed
1149620022 M * micah this is not a vserver issue
1149620104 M * bragon i have try with many mirror
1149620121 M * micah bragon: do you know a package that you have installed that there is an update for?
1149620142 M * bragon yes
1149620146 M * micah which one?
1149620161 M * bragon in unstable you have upgrade every days
1149620170 M * micah which package, specifically
1149620179 M * bragon but since 3 weeks i don't manage to upgrade my guest
1149620296 M * micah can name a package that you have installed in your guest that you know there is a package update for?
1149620318 M * bragon wesnoth-server
1149620327 M * bragon for exemple
1149620337 M * micah ok, what does apt-cache policy wesnoth-server show in your vserver guest?
1149620352 M * tokkee bragon: I just upgraded unstable Debian guests yesterday...
1149620383 M * tokkee bragon: Do you have an AMD64 box?
1149620384 M * bragon tokkee no soucy with the Diff ?
1149620395 M * bragon yes my guest is x86_64
1149620408 M * bragon (sorry for my english)
1149620424 M * tokkee bragon: Ah... you should change your sources.list to the default debian mirrors. No need for debian-amd64 any more.
1149620440 M * bragon hum
1149620442 M * bragon i try
1149620520 M * bragon \o/
1149620530 M * bragon tokkee thx
1149620537 M * tokkee bragon: No problem ;-)
1149620562 M * bragon sorry for the disturb
1149620567 M * tokkee bragon: Just Sarge still needs debian-amd64...
1149620630 M * bragon ok
1149621109 J * stefani ~stefani@tsipoor.banerian.org
1149621495 Q * derjohn2 Ping timeout: 480 seconds
1149621511 J * derjohn2 ~aj@dslb-084-059-011-117.pools.arcor-ip.net
1149622247 Q * Viper0482 Remote host closed the connection
1149622372 J * shedi ~siggi@inferno.lhi.is
1149622851 Q * doener_ Remote host closed the connection
1149623200 J * CoBoLt ~countcobo@d54C498B4.access.telenet.be
1149623205 M * CoBoLt oi all
1149623247 M * daniel_hozac hello
1149623255 M * CoBoLt hi daniel
1149623264 M * CoBoLt currently rebuilding my alpha
1149623281 M * CoBoLt had a disk crash about 30" after I configured everything and no back-ups
1149623283 M * CoBoLt :(
1149623392 Q * bogus Server closed connection
1149623407 J * bogus ~bogusano@fengor.net
1149625261 J * Cocobu ~Cocobu@trash.mana.pf
1149625786 J * s0undt3c1 ~s0undt3ch@82.155.71.91
1149625786 Q * s0undt3ch Read error: Connection reset by peer
1149626374 Q * mkhl Quit: 
1149626562 Q * lilalinux Ping timeout: 480 seconds
1149627055 Q * otaku42_away Server closed connection
1149627078 J * otaku42_away ~otaku42@legolas.otaku42.de
1149627086 J * lilalinux ~plasma@dslb-084-058-204-063.pools.arcor-ip.net
1149627139 N * otaku42_away Guest169
1149627283 Q * bonbons Quit: Leaving
1149629352 J * doener ~doener@i5387FDB1.versanet.de
1149629535 Q * derjohn2 Ping timeout: 480 seconds
1149629885 Q * bragon Server closed connection
1149629885 J * bragon ~bragon@sd866.sivit.org
1149630412 J * hallyn ~xa@adsl-75-2-81-182.dsl.chcgil.sbcglobal.net
1149630434 P * stefani I'm Parting (the water)
1149630603 M * Cocobu hi i've just compiled the latest vserver but when I try this command : vserver lfs build -m skeleton it end with error
1149630605 M * Cocobu chattr: Inappropriate ioctl for device while reading flags on /etc/vservers/.defaults/vdirbase/lfs
1149630646 M * doener did all testme.sh tests succeed?
1149630651 M * Cocobu yes
1149630670 M * doener hm, which file system is that?
1149630683 M * Cocobu reiserfs
1149630813 M * doener did you enable xattrs?
1149630836 M * Cocobu no my root partition is mounted with the option default
1149630864 M * Cocobu should I added the xattrs option and reboot ?
1149630869 M * doener in kernel configuration i mean
1149630899 M * doener (might also need a mount option though, I know almost nothing about reiserfs)
1149630928 M * Cocobu ok i'll try that and back if it doesn't work, thx for the tips
1149630955 M * doener yw
1149631107 M * Loki|muh I wonder if udev could work inside a vserver if enough capabilities would be given to it
1149631178 M * doener Loki|muh: sure... but that vserver would then also have access to all device nodes
1149631204 M * Loki|muh yeah
1149631224 M * Loki|muh so its possible but totally stupid (in most cases) ;)
1149631243 M * doener if you actually just want _that_, (a complete, managed /dev), I'd consider doing a bind mount of the udev tmpfs instead
1149631259 M * doener ... a complete, managed /dev in a vserver that is
1149631337 M * Loki|muh I want a initscripts-vserver package for ubuntu dapper and I am just over removing unnecessary initscripts ;)
1149631373 M * Loki|muh because the dapper upgrade fails because of initscripts-postinstall (which tries to convert /var/lock & run to tmpfs)
1149631375 Q * CoBoLt Quit: Leaving
1149631581 N * Bertl_oO Bertl
1149631584 M * Bertl evening folks!
1149631588 M * doener hi Bertl 
1149631751 M * Bertl how's it going?
1149631758 M * orionpanda bertl: do you know if anyone has written an fuser filesystem (layered ontop of ext3) that supports vserver?
1149631806 M * Bertl hmm, fuser as in fuse?
1149631817 M * orionpanda err. yes. right
1149631855 M * Bertl well, I'd say most of them will work
1149631858 M * orionpanda i'm thinking I might just have to write my own crude distributed fs (reliable udp multipcast over ethernet)
1149631871 M * orionpanda how about iunlink and barrier?
1149631880 M * Bertl hehe, so did you have a look at ocfs2?
1149631918 M * orionpanda I have, yes. I have three opteron's sitting here that i'm installing it on. haven't tried vserver yet, but the vserver kernel patch doesn't support it.
1149631949 M * Bertl yes, Linux-VServer does not support it yet :)
1149632027 M * orionpanda well, i have a guinea pig setup here to test ocfs2 patches for you
1149632074 M * Bertl did you already try it out?
1149632075 M * orionpanda patching lustre turns out to be extremely complicated (client kernel patch, server kernel patch, client kernel module patch, server kernel module patch).
1149632104 M * Bertl would be helpful for me to know about setup and issues/test cases
1149632105 M * orionpanda i have not run vserver on ocfs2 yet. should I? I thought it wasn't supported?
1149632121 M * orionpanda oh. ok. I'll see wht happens
1149632123 M * Bertl no, ocfs2 (ignoring the vserver specific features)
1149632199 M * Bertl especially interesting would be the simplest way to get ocfs2 working on a single node (for testing)
1149632210 M * orionpanda oh. i have three ocfs2 volumes mounted. It's not a unified namespace.
1149632239 M * Bertl what about the cluster manager? how much config does it need?
1149632273 M * orionpanda That's simple. I enabled the ocfs2 and aoe kernel options. Then I exported a local ide block device over aoe. SO that could be done for a single node. There's a single config you need to distribute
1149632305 M * Bertl what does that config contain, and what services do the cluster manager require?
1149632355 M * orionpanda the config file is very simple. node names,  block devices, and ip addresses
1149632369 M * orionpanda there's a cluster manager daemon that has to run on each node. one is master
1149632375 M * Bertl can you upload an example to paste.linux-vserver.org please?
1149632380 M * orionpanda tjem tjere are sp,e ,pdi;es tp ;pad
1149632388 M * orionpanda sorry
1149632398 M * orionpanda ok. will do. 
1149632400 M * Bertl is that actually some kind of language?
1149632419 M * orionpanda sorry, i was typing while looking at another screen
1149632427 M * Loki|muh :D
1149632455 M * Bertl .o( hmm, which doesn't answer my question :)
1149632531 J * Aiken ~james@tooax7-205.dialup.optusnet.com.au
1149632532 M * orionpanda I'll put together  a writeup of what I'm doing and upload.
1149632542 M * Bertl orionpanda: excellent! tx!
1149632545 M * Bertl welcome Aiken!
1149632554 M * Aiken hello
1149632560 M * doener hmm... "then there are some modules to load"
1149632591 M * Bertl ah, yes, key shifting :)
1149632666 M * Bertl that's something I do not immediately think of, as I never learned to type with a proper finger system
1149632684 M * orionpanda yes, ocfs2 consists of (1) the kernel patch (included in 2.6.16); (2) kernel modules (3) management tools
1149632707 M * Bertl you do not need 1 as it is in mainline now
1149632712 M * orionpanda right
1149632724 M * Bertl you do not need 2 either as far as I understood
1149632760 M * Bertl the management stuff consists of two parts, a set of lowlevel tools and some gtk control gui
1149632770 M * orionpanda don't know. I just followed the user's guide and installed them anyway -- wasn't sure what version was included.
1149632807 M * Bertl so we'll focus on the bare metal part of ocfs2 (i.e. kernel + low-level tools for now)
1149633041 Q * click Remote host closed the connection
1149633428 J * click ~click@ti511110a080-0437.bb.online.no
1149633476 P * click 
1149633736 J * click click@ti511110a080-1902.bb.online.no
1149633958 Q * mire Quit: Leaving
1149634166 Q * click Remote host closed the connection
1149635115 J * jkl eric@c-71-56-216-223.hsd1.co.comcast.net
1149635442 M * doener daniel_hozac: just for the record, hashify only works for running vservers ;)
1149635585 M * anonc doener: I certainly didn't know that about hashify - very useful information
1149635631 M * doener well, daniel, cehteh and myself had a discussion a few days ago, about hashify being safe to be done with a running vserver
1149635633 M * Bertl doener: very interesting indeed ...
1149635649 M * anonc incidentally - is it just me or is the memory split option missing from patch-2.6.17-rc5-vs2.1.1-rc21.3.diff
1149635668 M * doener mem split is in vanilla now AFAIK
1149635668 M * Bertl anonc: yes, I removed it, mainline got something very similar
1149635689 M * Bertl (not that many options, but sufficient for most systems)
1149635744 M * daniel_hozac doener: really? so the guest has to be running? whY?
1149635766 M * doener let me put it this way:
1149635767 M * doener host:~# vserver www hashify
1149635767 M * doener 'vserver ... suexec' is supported for running vservers only; aborting...
1149635769 M * doener ;)
1149635779 M * daniel_hozac how odd.
1149635799 M * doener i didn't care to investigate
1149635821 M * cehteh doener: "works for running vservers"?
1149635887 M * daniel_hozac doener: hmm, i don't get that here.
1149635900 M * daniel_hozac hashifying stopped guests works fine.
1149635916 M * doener that's strange
1149635949 M * daniel_hozac i don't see how that warning would even be triggered for hashify...
1149636007 M * anonc i always use the 2-2 split - this commit referenced here http://kerneltrap.org/node/6554 suggests its now dependent on config_embedded
1149636049 M * daniel_hozac         depends on EXPERIMENTAL && !X86_PAE
1149636115 M * doener + prompt "Memory split" if EMBEDDED
1149636139 M * daniel_hozac ... more ways than one to depend... sigh.
1149636159 M * anonc grr - where's X86_PAE set?
1149636161 M * daniel_hozac that syntax suggests that you can set it anyway, just not through the menus.
1149636166 M * daniel_hozac IMO.
1149636187 M * doener maybe... Bertl? ;)
1149636187 M * daniel_hozac HIGHMEM64G
1149636196 M * daniel_hozac == X86_PAE
1149636228 M * anonc ah - just found that - but i've disabled highmem and the embedded option still doesn't show up
1149636314 M * anonc ah - found it
1149636696 M * Bertl make menuconfig + '/' is your friend :)
1149636709 M * anonc oh - and you can't compile nfs server support into patch-2.6.17-rc5-vs2.1.1-rc21.3.diff - see http://paste.linux-vserver.org/63
1149636756 M * Bertl hum, that's a bug then
1149636769 Q * dna Quit: Verlassend
1149636792 M * Bertl anonc: but that's not true for 2.6.16.20-rc2.1.1-rc22 yes?
1149636817 M * Bertl or is it? .. checking now
1149636886 M * anonc that's with just nfserver support enabled - no tcp, no nfsv3, no acls
1149636967 M * daniel_hozac not true for -rc21 at least, and -rc22 didn't change those files.
1149636994 M * anonc i would expect a bunch of nfs stuff from trond would be merging into 2.6.17...
1149637029 M * Bertl yeah, that's quite fine, but I obviously got that one wrong on the port
1149637157 M * anonc Bertl: sorry that i haven't been able to patch binutils to track down the cow bug - if i can find time i'll see if it persists in rc22 or 2.6.17-rc5
1149637219 M * Bertl anonc: np, that would be great!
1149637295 M * daniel_hozac anonc: did you have CONFIG_DEBUG_INFO enabled in the first place?
1149637480 M * anonc doesn't look like it - though i did have vserver debuggin enabled...
1149637611 M * daniel_hozac so probably a patched binutils won't help.
1149637633 M * anonc perhaps a cluebat for me might be more beneficial
1149637915 Q * lilalinux Remote host closed the connection
1149638101 Q * weasel Quit: Reconnecting
1149638109 J * weasel weasel@asteria.debian.or.at