1227225946 J * Piet ~piet@86.59.118.153
1227227499 Q * pisco Ping timeout: 480 seconds
1227227537 J * pisco ~pisco@86.59.118.153
1227227934 Q * Piet Quit: Piet
1227228332 M * Bertl okay, off to bed now too .. have a good one everyone!
1227228345 N * Bertl Bertl_zZ
1227228425 J * ntrs__ ~ntrs@77.29.15.194
1227228857 Q * ntrs__ Ping timeout: 480 seconds
1227229882 Q * dowdle Remote host closed the connection
1227231084 Q * geb Remote host closed the connection
1227232754 J * derjohn_mob ~aj@e180208184.adsl.alicedsl.de
1227233243 J * a[u]stin ~austin@hil-100-168.ResHall.Berkeley.EDU
1227233301 M * a[u]stin what more do i have to do after i clone a guest to make the cloned system runnable?
1227233348 M * cehteh setup ip, fstab etc if you didnt yet (with the clone?)
1227233361 M * cehteh add the mark for automatic startup
1227233371 M * a[u]stin okay
1227233375 M * a[u]stin so say i am on debian lenny
1227233384 M * a[u]stin and i build a system with debootstrap
1227233399 M * a[u]stin then clone that system
1227233437 M * a[u]stin why does /etc/vserver/<system name>/vdir point to the /etc/vserver/<system name> of the cloned system
1227233451 M * daniel_hozac how did you clone it?
1227233484 M * a[u]stin vserver my-clone build -m clone -- --source /etc/vserver/system-to-clone
1227233494 M * daniel_hozac well, that'd be why :)
1227233498 M * a[u]stin oh
1227233503 M * a[u]stin i must be misunderstanding then
1227233512 M * daniel_hozac --source should be the name of a guest, or /vservers/<guest>
1227233513 M * a[u]stin i've tried looking for documentation
1227233524 M * a[u]stin oooh
1227233536 M * daniel_hozac such as http://linux-vserver.org/Building_Guest_Systems?
1227233556 M * a[u]stin i know
1227233557 M * a[u]stin is aw that
1227233562 M * a[u]stin and isn't that what i am doing
1227233570 M * a[u]stin my directory is just different
1227233619 M * daniel_hozac you're telling it to clone a configuration directory.
1227233635 M * a[u]stin i don't have a /vservers though
1227233637 M * a[u]stin would that be
1227233652 M * a[u]stin the /var/lib/vservers
1227233664 M * daniel_hozac if you're no Debian, sure.
1227233668 M * a[u]stin okay
1227233670 M * daniel_hozac s/no/on/
1227233685 M * a[u]stin one more question
1227233694 M * a[u]stin can i mount the / directory of the quest read only?
1227233700 M * a[u]stin *guest
1227233707 M * daniel_hozac yes.
1227233720 M * a[u]stin i get problems with /etc/mtab though
1227233781 M * daniel_hozac as expected.
1227233813 M * a[u]stin i'm assuming that is because everything that is mounted attempts to write to it
1227233821 M * a[u]stin can i disable this, or what is a workaround?
1227233827 M * daniel_hozac just ignore it.
1227233831 M * a[u]stin ok
1227233832 M * a[u]stin haha
1227233835 M * a[u]stin easy enough
1227233845 M * a[u]stin thanks a bunch
1227233848 M * a[u]stin i may be back
1227233849 M * daniel_hozac so what class are you taking?
1227233859 M * a[u]stin advanced unix systems administration
1227233863 M * a[u]stin how did you know?
1227233876 M * daniel_hozac a class mate of yours i gues was here earlier .
1227233884 M * a[u]stin probably my partner
1227233888 M * a[u]stin haha
1227233905 M * a[u]stin we've been messing with vservers for the past day or so
1227233962 M * a[u]stin we are having a hard time finding documentation
1227234001 M * daniel_hozac most commands have sufficiently verbose --helps.
1227234036 M * a[u]stin ah
1227234042 M * a[u]stin well
1227234058 M * a[u]stin i feel like the explanation for what the different build methods do is pretty sparse
1227234067 M * a[u]stin a sentence or so
1227234073 M * a[u]stin but i can always look at the scripts
1227234079 M * a[u]stin so i guess its no big deal
1227234084 M * daniel_hozac it builds a guest. if you want details, read the source.
1227234121 M * a[u]stin i just wanted to avoid that :P
1227234124 M * a[u]stin oh well
1227234128 M * a[u]stin thanks for your help
1227234143 Q * a[u]stin Quit: a[u]stin
1227236357 M * bliz42 on a gentoo system, trying to use vemerge.. and it keeps bombing from a read only filesystem error.. any advice here?  seems the gentoo specific documentation is from 2006, but I can't find anything really any more recent
1227236760 J * fb_ fback@red.fback.net
1227236796 J * _Hunger Hunger.hu@Hunger.hu
1227236810 Q * fb synthon.oftc.net larich.oftc.net
1227236810 Q * Hunger synthon.oftc.net larich.oftc.net
1227236810 Q * brc synthon.oftc.net larich.oftc.net
1227236860 M * bliz42 ok, got around the issue by adding a shared packages folder.. which is good anyway, so meh
1227239239 J * dowdle ~dowdle@71-221-8-241.blng.qwest.net
1227240618 Q * hparker Quit: Quit
1227241209 J * hparker ~hparker@2001:470:1f0f:32c:215:f2ff:fe60:79d4
1227241275 Q * dowdle Remote host closed the connection
1227241406 J * testo ~some@202.180.115.27
1227241506 M * testo Where to fina a good documentation about CPU limits?
1227241537 M * daniel_hozac http://linux-vserver.org/CPU_Scheduler
1227241595 M * testo Yes I've seen it but where are examples?
1227241609 J * cybergirl ~cybergirl@212-198-248-34.rev.numericable.fr
1227241722 M * daniel_hozac at the bottom.
1227241744 Q * cybergirl 
1227242006 M * testo i'm idot but can't find  a good example, for example to run 1 vserver at 10% cpu and using only 2CPU from 4 ?
1227242256 Q * derjohn_mob Ping timeout: 480 seconds
1227243275 M * daniel_hozac 10% of each CPU?
1227243651 Q * bliz42 Quit: leaving
1227243931 J * brc bruce@72.20.27.65
1227246051 Q * esa Ping timeout: 480 seconds
1227248147 J * derjohn_mob ~aj@e180193191.adsl.alicedsl.de
1227248756 J * sharkjaw ~gab@149-67-194.231210.adsl.tele2.no
1227249864 J * ntrs__ ~ntrs@77.29.22.25
1227250559 J * ghislainocfs2 ~Ghislain@adsl2.aqueos.com
1227250684 Q * derjohn_mob Ping timeout: 480 seconds
1227250885 M * arekm hm, ipv6 fix wasn't merged into .36?
1227250970 J * mtg ~mtg@dialbs-088-079-143-204.static.arcor-ip.net
1227251449 Q * ntrs__ Ping timeout: 480 seconds
1227252537 Q * arekm Quit: leaving
1227252976 J * arekm arekm@carme.pld-linux.org
1227252979 M * arekm [arekm@carme-pld ~]$ LC_ALL=C sudo ls -l /proc
1227252981 M * arekm ls: cannot open directory /proc: Permission denied
1227252999 M * arekm is this known? (26.7 + vserver 2.3...36)
1227253022 M * arekm 2.6.27.7 of course
1227253078 M * arekm [   47.521257] vxW: [<BB>pidof<AB>,3064:#100|100|100] denied 24 access to proc:ffff88015f410030[#0,1]
1227253362 M * arekm previous working ver was vs2.3.0.35.7
1227253444 M * testo how to set CPU limit? what command to use or where is config files?
1227253466 Q * larsivi Ping timeout: 480 seconds
1227253761 J * derjohn_mob ~aj@p57A6F732.dip.t-dialin.net
1227253761 Q * arekm Quit: leaving
1227253932 N * quinq qzqy
1227254362 J * arekm arekm@carme.pld-linux.org
1227254371 M * arekm back on old vs
1227254504 M * daniel_hozac testo: vsched, /etc/vservers/<guest>/sched
1227254544 M * arekm Hawq: you had the /proc problem, right?
1227254915 M * arekm hm, delta-proc-fix03.diff is applied in .36
1227255133 M * testo what should I put there?  vsched, /etc/vservers/<guest>/sched   ? any docs?
1227255284 M * arekm use google, there are docs
1227255444 M * arekm "you can't use the plain initstyle without the pid virtualization." hm, do plain no longer works in vs2.3?
1227255451 J * dna ~dna@52-200-103-86.dynamic.dsl.tng.de
1227257031 Q * mtg Ping timeout: 480 seconds
1227257344 J * larsivi ~larsivi@85.221.53.194
1227257906 J * mtg ~mtg@dialbs-088-079-143-204.static.arcor-ip.net
1227259125 J * davidkarban ~david@193.85.217.71
1227259151 J * ntrs__ ~ntrs@77.29.14.75
1227260666 J * Mojo1978 ~Mojo1978@ip-88-152-50-100.unitymediagroup.de
1227262292 Q * gnuk Remote host closed the connection
1227262539 J * gnuk ~F404ror@pla93-3-82-240-11-251.fbx.proxad.net
1227262603 J * ghislainocfs21 ~Ghislain@LPuteaux-151-41-11-129.w217-128.abo.wanadoo.fr
1227262664 Q * dna Quit: Verlassend
1227262728 J * seldgehammer a8ac01f7@webchat.mibbit.com
1227262744 Q * ghislainocfs2 Ping timeout: 480 seconds
1227262829 Q * seldgehammer 
1227264988 Q * mtg Quit: Verlassend
1227265300 J * baggins ~baggins@kenny.mimuw.edu.pl
1227265719 Q * derjohn_mob Ping timeout: 480 seconds
1227266153 J * tramjoe_merin ~tramjoe@193.41.238.151
1227266988 Q * ntrs__ Ping timeout: 480 seconds
1227267131 Q * hparker Ping timeout: 480 seconds
1227267211 J * hparker ~hparker@linux.homershut.net
1227268550 N * Bertl_zZ Bertl
1227268553 M * Bertl morning folks!
1227268631 M * baggins hi!
1227268681 M * baggins there is a problem with patch-2.6.27.6-vs2.3.0.36.diff 
1227268690 M * baggins vserver nstest exec ls /proc
1227268694 M * baggins ls: cannot open directory /proc: Permission denied
1227268707 M * baggins dmesg:
1227268709 M * baggins [40122.149314] vxW: [�ls�,5752:#100|100|100] denied 24 access to proc:f742401c[#0,1]
1227268710 A * ard was just about to compile it :-)
1227268729 M * baggins but:
1227268731 M * baggins vserver nstest exec ls /proc/cpuinfo
1227268734 M * baggins /proc/cpuinfo
1227268748 M * baggins weird 
1227268911 M * Bertl what distro are you using?
1227268981 M * baggins pld
1227269001 M * Bertl IIRC, that one is doing a malicious setattr --hide /proc in their scripts
1227269014 M * Bertl get rid of that and it should work fine
1227269061 A * arekm dropped that from latest version of script
1227269076 M * Bertl so an update should be fine too then :)
1227269082 M * baggins arekm: yeah, but did you rebuild the package?
1227269094 M * baggins arekm: answer is no ;)
1227269170 M * baggins Bertl: thanks, removing setattr fixed it :)
1227269187 M * Bertl np
1227269275 M * arekm baggins: "removing" + reboot fixed it? or no reboot?
1227269335 M * baggins setattr -Rx --~hide /proc
1227269360 M * baggins no reboot
1227269375 M * arekm ah, I used the command from "stop" 
1227269395 M * arekm which was broken anyway it seems heh
1227269541 Q * hparker Remote host closed the connection
1227269725 M * arekm Bertl: btw. please merge ipv6 fix for .37
1227269739 J * hparker ~hparker@2001:470:1f0f:32c:215:f2ff:fe60:79d4
1227269784 M * Bertl ah, do we have something working/tested? probably missed that?
1227269804 M * arekm http://people.linux-vserver.org/~bonbons/vs2.3.0.35.10-ipv6-saddr-breakout-fix.diff
1227269836 M * Bertl excellent, you tested it?
1227269958 M * arekm yes, on one machine. Doing reboot now and I'll tell you how it looks on second machine with slightly different env
1227270002 Q * arekm Quit: reboot
1227270344 J * arekm arekm@carme.pld-linux.org
1227270644 M * arekm Bertl: on second machines also works fine
1227270648 M * arekm s/es/e/
1227270926 M * Bertl okay, just uploaded vs2.3.0.36.1 with that patch added
1227271441 M * pmjdebruijn does that one already include the XFS juju?
1227271464 M * Bertl xfs should be fine again, but not much tested
1227271467 M * pmjdebruijn ok
1227271907 M * pmjdebruijn cool
1227271959 M * ktwilight__ Bertl, but i guess testfs runs fine for xfs?
1227271975 Q * hparker Ping timeout: 480 seconds
1227271988 M * Bertl yep, except for one test case, which currently fails on all filesystems 
1227271998 M * ktwilight__ awesome :)
1227272165 M * ktwilight__ Bertl, mind if i modify the frontpage to include the experimental patches? :)
1227272196 M * Bertl well, they are linked there, IIRC, do you plan to update and list the latest or what?
1227272211 M * Bertl (i.e. do you plan to maintain this?)
1227272222 M * ktwilight__ sure, i don't mind maintaining the list
1227272235 M * Bertl then go ahead, but make it a separate table
1227272266 M * ktwilight__ could i just put it in the same table, but don't link in 2.2 stable?
1227272313 M * ktwilight__ i mean, only add the entry for the Linux kernel and 2.3 Development column.
1227272325 M * Bertl nah, leave the main table as is
1227272351 M * ktwilight__ hm, k, i'll add another table underneath :)
1227272353 J * hparker ~hparker@linux.homershut.net
1227273003 M * ktwilight__ doh, i've no rights to edit it Bertl. but i have prepared the template http://linux-vserver.org/Template:ExperimentalPatchTableMatrix
1227273034 M * ktwilight__ gimme a mo.
1227273164 M * ktwilight__ Bertl, ok, it's done, http://linux-vserver.org/Template:ExperimentalPatchTableMatrix as long as i can modify that template, it's all good. what's needed to do is modify the frontpage to include the template :)
1227273176 J * bliz42 ~ksmith@c-98-193-150-250.hsd1.tn.comcast.net
1227273203 M * ktwilight__ ah sweet! it's done :)
1227273210 M * ktwilight__ but can you please remove "also check out the latest Experimental Releases and let us know if something is broken." since it's repeated actually.
1227273241 M * ktwilight__ i'll monitor Experimental from time to time to include the latest :)
1227273262 M * Bertl yes, but please make the Experimental Releases a link, and avoid the 'here' and 'there' links  :)
1227273280 M * ktwilight__ aight :)
1227273368 M * ktwilight__ ok, done.
1227273392 M * Bertl no, actually I prefer to do that in the main page, i.e. please remove any additional text and stuff, just leave the matrix
1227273436 M * ktwilight__ so just the matrix?
1227273441 M * Bertl yep, please
1227273448 M * ktwilight__ should i remove hte headers as well?
1227273463 M * ktwilight__ uh, header, where it says "Experimental Releases"
1227273468 M * Bertl nah, matrix with headers is fine
1227273472 M * ktwilight__ aight
1227273476 M * Bertl yeah, those too, please
1227273486 M * Bertl I meant the headers of the matrix, not the text stuff
1227273502 M * Bertl i.e. like you had it at the beginning :)
1227273509 M * ktwilight__ ok, refresh please, i hope this is ok :)
1227273529 M * Bertl not yet, remove the Experimental Releases
1227273538 M * ktwilight__ ah right, ya meant that.
1227273544 M * ktwilight__ done.
1227273559 M * Bertl k, tx
1227273562 M * ktwilight__ np :)
1227273687 Q * nkukard Quit: Leaving
1227274113 Q * hparker Quit: Quit
1227275996 Q * independence Ping timeout: 480 seconds
1227276087 J * independence independen@titan.blinkenshell.org
1227276197 J * ntrs__ ~ntrs@77.29.14.75
1227276258 J * whuji ~huji@jem75-8-88-170-103-188.fbx.proxad.net
1227276264 M * whuji hello 
1227276276 M * Bertl hey
1227276308 M * whuji I have a problem with vserver
1227276359 J * hparker ~hparker@linux.homershut.net
1227276361 M * Bertl well, if you tell us what problem, we might even be able to help :)
1227276378 M * whuji I have a debian with vserver, no vm for the moment, but vserver doesn't see my first NICs (I have 2 NICs : eth0 and eth1) it just sees eth1. How could I tell to vserver that I have 2 NICs ?
1227276388 M * whuji (yes I was writing the message :) )
1227276423 M * Bertl well, Linux-VServer does not care about your NICs, it's also no VM or so ...
1227276457 M * Bertl you assign IPs to a guest, and then Linux-VServer will show you all interfaces with those IPs on
1227276474 M * whuji so why I don't see eth0 ? When I boot to a normal kernel (without the vserver patch) it works perfectly (I have eth0 and eth1)
1227276494 M * Bertl are you talking about the host?
1227276523 M * whuji yes 
1227276529 M * whuji I don't have guest for now
1227276536 M * Bertl well, then it is most likely that you are missing a driver
1227276547 M * whuji the 2 NICs are the same
1227276576 M * Bertl what kernels are we talking about and what NIC/driver?
1227276679 Q * ntrs__ Ping timeout: 480 seconds
1227276768 M * whuji well. I have 2 kernels installed : a linux-image 2.6.18 (I use debian etch) and a linux-image-vserver 2.18 (with vserver patch so). When I boot on the first kernel, everything is ok, eth0 and eth1 works. When I boot to the second kernel (vserver) eth0 doesn't work. The 2 NICs are both intel chipset.
1227276808 M * Bertl sounds weird, but more like a debian issue ...
1227276815 M * whuji ok
1227276816 Q * independence Ping timeout: 480 seconds
1227276825 Q * sharkjaw Remote host closed the connection
1227276828 M * Bertl in general, I'd try a newer kernel, e.g. at least 2.6.22
1227276843 M * whuji I've already written to debian mailing list. Waiting for replies.
1227276861 M * whuji ok I will try with a newer kernel from backports
1227276863 M * whuji thanks.
1227276864 M * Bertl Linux-VServer does not change anything in the way the network drivers work
1227276881 M * Bertl so I would be surprised if the Linux-VServer patches would affect this
1227276909 M * Bertl but, if you like to make sure, get a kernel.org 2.6.18 kernel, compile and test this (without patches)
1227276923 M * Bertl and then apply the Linux-VServer patch, and see if anything changes
1227276936 M * Bertl (I'm pretty sure it will not :)
1227276949 M * whuji I will try with a newer kernel first 
1227276972 M * Bertl let us know how it goes ...
1227276979 M * whuji yep. I stay here
1227276995 M * whuji thanks a lot
1227276998 M * Bertl yeah, no problem, feel free to hang around as long as you like
1227277233 M * pmjdebruijn hmm cool
1227277244 M * pmjdebruijn I've built 2.3.0.36.1 against 2.6.27.7 (instead of .6)
1227277247 M * pmjdebruijn works just fine
1227277470 M * Bertl okay, off to grab some groceries .. bbl
1227277479 N * Bertl Bertl_oO
1227277640 M * pmjdebruijn btw, this is an excellent patch as well (unrelated to vserver): http://jengelh.medozas.de/files/kernel/linux-2.6.25.8-jen67/vt-colored-kernel-message-outp-1.txt
1227277667 Q * Mojo1978 Read error: Connection reset by peer
1227277732 J * independence ~independe@titan.blinkenshell.org
1227277782 M * whuji Bertl > it works with 2.6.26 from backports ! thank you !!
1227277850 M * daniel_hozac pmjdebruijn: i don't see why you would want that in the kenrel.
1227277899 M * pmjdebruijn daniel_hozac: makes seperating kernel messages from userland messages really easy
1227277910 M * pmjdebruijn daniel_hozac: it's definately not critical...
1227277915 M * pmjdebruijn daniel_hozac: but convenient
1227277991 M * daniel_hozac you could just have klogd do that.
1227278484 Q * independence Remote host closed the connection
1227278628 J * ncopa ~ncopa@149-13-151.oke2-bras2.adsl.tele2.no
1227278644 M * ncopa hi
1227278692 M * daniel_hozac hello
1227278852 J * independence independen@titan.blinkenshell.org
1227278929 M * whuji with the command 'newvserver', in the '--interface' arg, I must specify a virtual interface ? or a real one ?
1227278942 M * daniel_hozac don't use newvservre.
1227278959 M * daniel_hozac use vserver ... build.
1227278966 Q * larsivi Remote host closed the connection
1227278988 M * whuji why ?
1227279103 M * whuji there is no method for debian
1227279109 M * whuji oh 
1227279114 M * whuji sorry, there is one
1227279397 M * whuji what's mean 'ncontext: vc_net_create(): Invalid argument' I've configured my guest to use eth1 interface (which is a physical interface). Is that a wrong way ?
1227279436 M * daniel_hozac it means you're using really old utils and you didn't specify a context id.
1227279578 M * whuji and the context id is always '42' ?
1227279832 M * daniel_hozac the context id is the numerical identifier of your guest.
1227279848 M * daniel_hozac like uids and usernames.
1227279849 M * ard vserver <newvservername> build -m debootstrap --interfac lo:10.10.10.10/32 --context 1234 -- -d etch
1227279862 M * ard is the preferred debian way :-)
1227279877 Q * matti Remote host closed the connection
1227279896 M * daniel_hozac assuming you spell --interface correctly :)
1227279913 M * ard yeah, well, I have problems with my keyboard. Really!
1227279962 M * whuji okey, thank you I've already created my guest on debian with 42 and id. thanks you !
1227279992 M * whuji now I have to read documentation to understand how vserver handle the network
1227280032 M * whuji but I don't see anything about that :/
1227280078 M * daniel_hozac it's simple. guests get a subset of the IP addresses.
1227280090 M * daniel_hozac those are all they can use.
1227280143 Q * davidkarban Quit: Ex-Chat
1227280222 M * whuji I see that. I can't give public ip to guest ?
1227280245 M * whuji oh ok, I think I must create a new route 
1227280257 M * whuji and assign ips to physical iface on the host
1227280271 M * ard the host can give any address to the guest to use
1227280284 M * ard but the host chooses which addresses, the guest is not allowed to
1227280294 M * daniel_hozac anything the host can use, the guest can use too.
1227280309 M * daniel_hozac iff the IP has been assigned to the context.
1227280368 M * whuji ok, and when I shutdown a guest, vserver can't umount because 'must be superuser to umount' is that a problem ?
1227280393 M * ard no :-)...
1227280426 M * ard guests are not allowed to mount or umount, the host will take care of that, you are seeing the remains of an rc script that does not have to run
1227280452 M * whuji ok
1227280773 M * whuji so ips of guests are alias on host's NICs
1227280823 M * whuji hm. I don't have internet on guest. 
1227280842 M * daniel_hozac if you assigned that IP to the host, would it work?
1227280855 M * daniel_hozac and, does ping -I <guest's IP> google.com on the host work?
1227280869 M * whuji yes
1227280903 M * whuji oh I can't ping google.com
1227280911 M * whuji sorry i can
1227280936 M * whuji but not from the guest
1227280961 M * daniel_hozac did you setup /etc/resolv.conf?
1227280990 M * whuji yes
1227281018 M * whuji ping: icmp open socket: Operation not permitted
1227281024 M * whuji when I ping an IP
1227281237 Q * independence Remote host closed the connection
1227281238 J * independence independen@titan.blinkenshell.org
1227281270 M * daniel_hozac what IP did you give the guest?
1227281341 M * whuji 88.191.91.19
1227281387 M * daniel_hozac and the host?
1227281469 M * whuji 88.191.89.19
1227281532 M * ard 88.191.89.19 is routable with traceroute, but 88.191.91.19 not
1227281532 M * daniel_hozac so it's a /22?
1227281556 M * ard Hmmm, /22 is very big ;-)
1227281577 M * daniel_hozac not really.
1227281610 M * daniel_hozac especially not for a network where you're expecting a lot of virtualization to be going on.
1227281617 M * whuji it's 2 public ip
1227281640 M * daniel_hozac sure, but are they on the same network?
1227281643 M * whuji no
1227281656 M * whuji the netmask is 24 
1227281676 M * daniel_hozac then you'll have to setup the guest to use whatever gateway the 88.191.91 network needs.
1227281677 M * ard heh, here in holland most companies start with a /23, unless you are really able to fill out more
1227281704 M * daniel_hozac i'd expect any datacenter to have more than one /23.
1227281706 M * ard (we got a /20 )
1227281717 M * whuji ok so I can do this using /etc/network/interface of the guest as usal ?
1227281721 M * whuji usual
1227281724 M * daniel_hozac no.
1227281730 M * daniel_hozac the guest cannot mess with networking at all.
1227281743 M * ard  /etc/vservers/guestname/interfaces/*/ip
1227281778 M * ard the host will setup the networking before it starts the guest
1227281861 M * whuji and what do I write into /etc/vservers/guestname/interfaces/0/ip ?
1227281883 M * ard change it to an ip address belonging to your network after shutting down the running guest
1227281943 M * whuji oh ok. But I can't choose my IPs. I rent a server from a datacenter and they give me just this 2 IPs.
1227281967 M * ard ah, ok
1227281984 M * ard so you just got those 2.
1227281989 M * whuji yes.
1227281999 M * ard That means that 88.191.91.19 should be as pingable as 88.191.89.19
1227282019 M * ard do me a favor:
1227282025 M * whuji yes ?
1227282025 M * ard shutdown the guest, and do:
1227282037 M * ard ip a add 88.191.91.19/32 dev lo
1227282057 M * ard on the host
1227282071 M * whuji done
1227282090 M * ard 88.191.89.19 is reachable and 88.191.91.19 is not
1227282099 M * ard try:
1227282111 M * ard tcpdump -n -i eth0 -c 100 host 88.191.91.19
1227282125 M * whuji u add it on lo not on eth0, is it ok ?
1227282125 M * ard if eth0 is your public interface
1227282131 M * ard it's ok ;-)
1227282156 M * whuji ok
1227282167 M * ard do you see anything?
1227282176 M * whuji no : my NIC is not in prom
1227282195 M * ard that's ok..
1227282222 M * ard if 88.191.91.19 belongs to you according to the routers of your provider, you should have seen an arp for that address
1227282266 M * ard if you change the "host 88.191.91.19" to "host 217.196.45.116"
1227282273 M * ard do you see any icmp traffic?
1227282282 M * ard (the latter is my office ip)
1227282293 M * whuji no
1227282307 N * Bertl_oO Bertl
1227282310 M * Bertl back now ...
1227282315 M * ard is eth0 your public interface?
1227282321 M * whuji yes
1227282327 M * Bertl what ard is trying to tell you, is that the provider lied to you :)
1227282329 M * whuji eth0 : 88.191.89.19
1227282343 M * ard that's weird:
1227282348 M * Bertl (or you got the second ip wrong in some way)
1227282352 M * ard  tcpdump -n -i eth0 -c 100 host 217.196.45.116
1227282364 M * ard should definitely show something ;-)
1227282372 M * whuji 16:46:09.353862 IP 88.191.89.19 > 217.196.45.116: ICMP echo reply, id 24447, seq 3, length 64
1227282372 M * ard 64 bytes from 88.191.89.19: icmp_seq=2 ttl=56 time=16.5 ms
1227282375 M * ard yes!
1227282379 M * ard that's it :-)
1227282398 M * whuji I'm very sorry I've written host 217.196.45.11 :/
1227282407 M * whuji instead of 116
1227282419 M * ard once in a while you will see an arp for 88.191.89.19, that's a way for the provider to know where your server is
1227282436 M * ard you should see an arp for 88.191.91.19 too...
1227282440 M * ard try:
1227282447 M * ard  tcpdump -n -i eth0 -c 100 arp
1227282454 M * whuji done
1227282471 M * ard do you see anything?
1227282479 M * whuji yep
1227282480 M * whuji 16:47:36.389690 arp who-has 88.191.89.131 tell 88.191.89.1
1227282480 M * whuji 16:47:37.499457 arp who-has 88.191.89.205 tell 88.191.89.1
1227282480 M * whuji 16:47:48.880143 arp who-has 88.191.89.1 tell 88.191.89.95
1227282480 M * whuji 16:47:53.857544 arp who-has 88.191.89.252 tell 88.191.89.1
1227282489 M * whuji and some more
1227282516 J * geb ~geb@4.4.82-79.rev.gaoland.net
1227282527 M * ard if you watch that, and everything stays within: 88.191.89.0/24, your provider lied to you ;-)
1227282545 M * whuji lied about what ?
1227282547 M * ard and they should give you an ip within 88.191.89.0/24
1227282560 M * whuji 'cause everything stays within 88.191.89.0/24 -_-
1227282560 M * ard about 88.191.91.19 being your second ip
1227282587 M * Bertl double check that it actually is .91. and not .89
1227282596 M * ard exactly ;-)...
1227282605 M * daniel_hozac and that you have the correct interface.
1227282606 M * Bertl looks very much like a typo, i.e. 191.91 (repeated 91)
1227282612 M * whuji I give u my provider's configuration :
1227282616 M * whuji auto eth0
1227282616 M * whuji iface eth0 inet static
1227282616 M * whuji 	address 88.191.89.19
1227282616 M * whuji 	netmask 255.255.255.0
1227282616 M * whuji 	network 88.191.89.0
1227282616 M * whuji 	broadcast 88.191.89.255
1227282616 M * whuji 	gateway 88.191.89.1
1227282626 M * Bertl (please use paste.linux-vserver.org for everything longer than 3 lines)
1227282629 M * whuji #auto eth1
1227282629 M * whuji iface eth1 inet static
1227282629 M * whuji         address 88.191.91.19
1227282629 M * whuji         netmask 255.255.255.0
1227282629 M * whuji         network 88.191.91.0
1227282630 M * whuji         broadcast 88.191.91.255
1227282630 M * whuji         gateway 88.191.91.1
1227282689 M * whuji sorry. used paste.linux-vserver.org
1227282689 M * daniel_hozac there you go.
1227282756 M * ard aaarghhh
1227282773 M * ard multi-homed *and* multi routed *and* multi-nic-d
1227282797 M * whuji meaning ?
1227282832 M * ard if that's your providers configuration, it means they gave you a second cable to your computer
1227282860 M * ard do you lease or rent a server that you have never seen, or how is this construction possible?
1227282884 M * whuji I rent a server that I've never seen
1227282890 M * whuji it's a dedibox
1227282892 M * ard Ok, clear, than it's possible
1227282911 M * ard try doing the tcpdumps on eth1 instead of eth0
1227282934 M * ard if eth1 is not up, ifup it first, *after* you have made sure you can log in from a console
1227282954 J * dowdle ~dowdle@scott.coe.montana.edu
1227283003 M * ard but now I have to leave you in the hands of Bertl and daniel_hozac because somebody forced social commitments on me :-(
1227283007 M * whuji ok. eth1 is up now. do I start my guest ?
1227283016 M * ard 64 bytes from 88.191.91.19: icmp_seq=2 ttl=56 time=16.9 ms
1227283021 M * ard it works ;-)
1227283027 M * whuji ha :) ok thank you a lot 
1227283033 M * ard well, daniel_hozac will probably explain some things ;-)
1227283042 M * Bertl you want to setup multiple routing tables to make that work
1227283049 M * whuji yep. But if I do a ssh on this ip I'm in the host, not in the guest
1227283054 M * ard depends on the route filtering done by his provider
1227283060 M * geb hi
1227283071 M * Bertl whuji: yes, that is expected, your host's sshd is not restricted yet
1227283079 M * ard but yes, you have a challenge ;-)
1227283081 M * ard O/~
1227283108 M * Bertl whuji: http://linux-vserver.org/Frequently_Asked_Questions#When_I_try_to_ssh_to_the_guest.2C_I_log_into_the_host.2C_even_if_I_installed_sshd_on_the_guest._What.27s_wrong_here.3F
1227283120 M * whuji ok... So I don't understand where vserver handle packets
1227283145 M * Bertl networking happens on the host, nothing virtual, no bridging no additional network stack
1227283156 M * Bertl Linux-VServer uses IP isolation, which is fast and lean
1227283184 M * Bertl i.e. you assign a bunch of IPs to a guest (or a single one), and the guest apps are only allowed to use that set of IPs
1227283184 N * qzqy quinq
1227283203 M * Bertl the host is not restricted and can use all IPs
1227283213 M * whuji yep. But packets are received by the host, who does the routing ? vserver ?
1227283260 M * Bertl the host does all the routing
1227283275 M * Bertl there is no routing between host and guest
1227283321 M * Bertl so, to make your guest work as expected, you basically start on the host
1227283327 M * whuji so when I do a ping on 88.191.91.19, who respond ? host or guest ?
1227283335 M * Bertl always the host
1227283370 M * Bertl you should start (on the host) to make e.g. the following work:
1227283373 M * whuji ok (sorry, I use to use xen, so I must handle vserver system)
1227283380 M * Bertl ping -I 88.191.91.19 www.google.com
1227283408 M * Bertl whuji: yes, I expected something VM like :) just forget all about routing and bridging and network stacks you learned there :)
1227283449 M * whuji I see that :) 
1227283452 M * Bertl just see the host and guests as a single entity, with certain restrictions on the guests (i.e. subset of host IPs)
1227283459 M * whuji but now, when I ping on the guest I have a ping: icmp open socket: Operation not permitted
1227283493 M * Bertl what kernel are we on now?
1227283503 M * whuji 2.6.26 
1227283510 M * whuji you were right Bertl :)
1227283546 M * Bertl did you configure something in ccapabilities?
1227283548 M * whuji (so where does 
1227283557 M * whuji capabilities ?
1227283605 M * Bertl can you upload the output of 'ip addr ls' inside the guest, and the ping command you are currently using?
1227283619 M * Bertl (the command line I mean :)
1227283758 M * whuji the ping command I use in the guest ?
1227283782 M * daniel_hozac whuji: upgrade your utils.
1227283782 M * Bertl yep
1227283783 M * whuji oups
1227283787 M * whuji inside the guest
1227283806 M * whuji do u received ?
1227283816 M * whuji (I've used paste.linux-vserver.org)
1227283823 M * Bertl yep, tx
1227283857 M * Bertl whuji: what util-vserver version?
1227283877 M * whuji ‪0.30.212-1
1227283910 M * Bertl yep, I guess you want to update that, but not to the pre2772 version, which is 'known broken' :)
1227283921 M * Bertl daniel_hozac: what in detail is the problem here?
1227283939 M * daniel_hozac it's using the first pid namespace API.
1227283949 M * daniel_hozac the one with new vc_ctx_create/migrate versions.
1227283958 M * whuji I have the  0.30.216~r2772-4~bpo40+2 0 in backports
1227283968 M * Bertl yep, avoid that one :)
1227283972 M * daniel_hozac if the VS-API is >= 0x00020304
1227283980 M * daniel_hozac should be fine on the old kerenel whuji is runnig.
1227283982 M * daniel_hozac +n
1227283984 M * Bertl daniel_hozac: okay, and how does that affect ping?
1227283990 M * daniel_hozac it doesn't.
1227284003 M * Bertl hmm, so how is that related?
1227284005 M * daniel_hozac i was talking about the known brokenness.
1227284033 M * Bertl ah, okay, well, if we talk about that, a 2.6.27 or 2.6.22 kernel would be better too
1227284034 M * daniel_hozac default ncaps weren't added until 0.30.214, i think
1227284082 M * whuji I've upgraded my utils-vserver
1227284101 M * whuji I must delete and create again my guest ?
1227284108 M * Bertl nope
1227284114 M * Bertl just restart it
1227284129 M * whuji done
1227284145 M * whuji it works !
1227284177 M * Bertl let's see if you can reach the outside now (from the guest)
1227284183 M * whuji yep I can
1227284198 M * Bertl excellent, then restrict the sshd (if not already done) _on_the_host_
1227284201 M * whuji and I can connect by ssh on the guest too
1227284206 M * whuji it's ok
1227284211 M * Bertl great so, you're done :)
1227284217 M * whuji thanks a lot u 2 :)
1227284225 M * Bertl you're welcome!
1227284244 M * whuji the new release of debian will correct this issues
1227284248 M * whuji those issues
1227284303 M * whuji ahh great. 
1227284384 M * whuji because etch have too old packets of vserver to work fine
1227284464 M * whuji so for the firewall, I must do this on the host
1227284469 M * Bertl yep
1227284492 M * Bertl if you want logical separation, just make a separate table for each guest
1227284502 M * whuji yep thanks ;)
1227284521 M * whuji so when vserver is a virtualisation ?
1227284535 M * Bertl it is more an isolation technology
1227284551 M * Bertl but nowadays it is called 'OS level virtualization'
1227284571 M * Bertl mainly because the interface to userspace (between OS and apps) is virtualized
1227284650 M * whuji ok
1227284679 M * whuji so I can mirgate my guests without any hardware constraints ?
1227284709 M * Bertl within binary compatibility
1227284712 M * whuji I just copy my guest from a server to another with its config file and 'hop'
1227284726 M * Bertl e.g. from i586 to a x86_64
1227284732 M * whuji sure
1227284745 M * whuji as guests share the kernel with host
1227284752 M * Bertl correct
1227284766 M * whuji ok. I do understand vserver better now.
1227284782 M * Bertl it is like a chroot (or bsd jail) on steroids
1227284790 M * whuji so it's radically different from xen
1227284809 M * Bertl not just filesystem is isolated, but also pids, users, networking ...
1227284833 M * whuji ok. I just through it was a real virtualization but with a sharing system for the kernel 
1227284843 M * whuji ok
1227284863 M * Bertl you won't get that performance with virtualization :)
1227284872 M * whuji sure :)
1227284884 M * Bertl another advantage is resource sharing
1227284890 M * whuji and vserver is stable enough to be used in prod ?
1227284905 M * Bertl it is used in production since 2002
1227284926 M * whuji well ok :D
1227284953 M * whuji so just another question
1227284958 M * whuji about networking again
1227284982 M * whuji if I want to do some nat, from the host to a guest, can I ?
1227284997 M * Bertl sure
1227284998 M * whuji for example I want to nat the port 80 to an guest
1227285007 M * whuji and the port 25 to another etc...
1227285017 M * whuji so I give my guest a private ip
1227285018 M * whuji ok
1227285020 M * Bertl just add the proper S/DNAT rules to the proper chains
1227285020 M * whuji stupid question
1227285034 M * whuji of course
1227285058 M * whuji so is there a real booting process on guest ?
1227285080 M * Bertl no kernel boot, just userspace
1227285088 M * whuji ok
1227285096 M * Bertl i.e. runlevel scripts get executed
1227285097 M * whuji no kernel, no initrd
1227285101 M * whuji ok
1227285105 M * Bertl (or init is started, and takes over)
1227285126 M * whuji so I can delete /boot safely
1227285132 M * Bertl yep
1227285169 M * daniel_hozac upgrading the kernel/boot loader packages might get sort of upset though.
1227285234 J * doener ~doener@i577B8A1D.versanet.de
1227285235 M * whuji sure.
1227285258 Q * independence Ping timeout: 480 seconds
1227285320 M * whuji and I can remove every mount* scripts in init.d too ?
1227285329 M * Bertl yep
1227285329 M * whuji and umount*
1227285331 M * whuji ok
1227285337 M * Bertl and everything hardware related
1227285341 Q * doener_ Ping timeout: 480 seconds
1227285352 M * whuji networking too ?
1227285364 M * Bertl yep
1227285394 M * whuji hostname too ?
1227285432 M * Bertl that one is virtualized, but you could live without
1227285439 M * Bertl i.e. you can set a default outside
1227285456 M * whuji ok
1227285503 M * whuji but I can't have different kernel modules in my guest 
1227285518 M * Bertl you can't have kernel modules in guests at all
1227285535 M * whuji ok
1227285639 M * bliz42 hey Bertl, what linux dist do you use vserver on?
1227285660 M * Bertl mostly a customized mandriva
1227285750 J * independence independen@titan.blinkenshell.org
1227285760 M * Bertl welcome independence!
1227286106 M * whuji just for theory, I can't make partitions on guest, right ?
1227286126 M * Bertl partitions as in disk slices?
1227286126 M * whuji I must do it on host ok
1227286129 M * whuji thanks !d
1227286136 M * whuji yep
1227286144 M * whuji we don't say partition in english ?
1227286149 M * Bertl well, in theory, you can give access to the disk to a guest
1227286166 M * Bertl but usually you want to avoid that for security reasons
1227286187 M * whuji of course, yes
1227286208 M * Bertl and yes, partition is correct, just ambiguous
1227286232 M * whuji ok
1227286250 M * whuji what's the other meaning ?
1227286278 M * Bertl hardware virtualization is called 'partitioning' too
1227286299 M * Bertl (in the sense of older IBM mainframe, not VT)
1227286321 M * whuji oh, ok
1227286334 M * independence Bertl: hehe, thanks :)
1227286350 M * independence my shell got DDoSed :/
1227286358 J * JonB ~NoSuchUse@130.227.63.19
1227286360 M * Bertl  ouch :)
1227286375 M * JonB hey Bertl 
1227286428 M * Bertl hey JonB! how's going?
1227286486 M * JonB Bertl: doing great, i defend my master thesis monday
1227286497 M * whuji If I want to compile something which need kernel header sources in my guest, I need to install kernel header sources, right ?
1227286505 M * Bertl JonB: congrats!
1227286517 M * Bertl whuji: yep
1227286527 M * whuji so I must have the vserver patched one ?
1227286531 J * larsivi ~larsivi@9.80-202-30.nextgentel.com
1227286533 M * JonB Bertl: thanks
1227286548 M * Bertl whuji: no, not really, whatever sources you need for the compiled result
1227286583 M * whuji but it will always work with the patched one
1227286679 M * Bertl userspace should not depend on real kernel headers, only on API (sanitized) headers (like glibc)
1227286691 M * Bertl unless you are compiling kernel modules, you should be fine
1227286712 M * whuji ah ok. thanks again :)
1227286775 M * whuji and I can 'virtualize" a centOS on a debian 
1227286792 M * Bertl sure, all kinds of distros
1227286928 M * whuji ok so it's a distribution level virtualization. 
1227286933 M * whuji everything clear now
1227286943 M * Bertl OS level to be precise :)
1227286944 M * whuji thanks to everybody :)
1227286977 M * whuji the kernel is a part of the os, no ?
1227286998 M * daniel_hozac the kernel is the OS.
1227287019 A * mnemoc disagrees
1227287021 M * whuji and vserver doesn't virtualize the kernel
1227287039 M * Bertl it 'virtualizes' the kernel/userspace interface
1227287117 M * whuji ok
1227287343 Q * tramjoe_merin Quit: using sirc version 2.211+KSIRC/1.3.12
1227287438 J * Pazzo ~ugelt@reserved-225136.rol.raiffeisen.net
1227287647 Q * JonB Quit: Leaving
1227287730 M * whuji an last thing and I must go.
1227287768 M * whuji hmm, actualy no :)
1227287770 M * whuji it's ok
1227287784 M * whuji thank you ! 
1227287860 M * whuji good evening  !
1227287864 M * Bertl cya!
1227287876 M * whuji (it's evening in France)
1227287905 Q * whuji Quit: Leaving.
1227289124 Q * kir Quit: Leaving.
1227290378 M * bliz42 anyone famliar with the vserver gentoo portage setup?
1227291690 J * chI6iT41 ~chigital@tmo-100-114.customers.d1-online.com
1227293974 Q * chI6iT41 Ping timeout: 480 seconds
1227295468 J * chI6iT41 ~chigital@tmo-100-63.customers.d1-online.com
1227295771 M * Bertl bliz42: Hollow definitely is
1227295951 J * DavidS ~david@85.125.165.34
1227296870 Q * DavidS Quit: Leaving.
1227297306 Q * arthur Ping timeout: 480 seconds
1227297914 Q * FireEgl Ping timeout: 480 seconds
1227298505 Q * gnuk Quit: NoFeature
1227298570 J * arthur ~arthur@pan.madism.org
1227300180 Q * Pazzo Quit: Ex-Chat
1227301909 J * esa ~esa@ip-87-238-2-45.static.adsl.cheapnet.it
1227302138 Q * chI6iT41 Ping timeout: 480 seconds
1227302806 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d
1227302985 M * Hawq hello
1227303024 M * Hawq sorry for being so late Bertl.
1227303459 M * PowerKe bliz42: What do you want to know?
1227303652 J * nkukard ~nkukard@196.212.73.74
1227303807 M * bliz42 just had some issues with some permissions within the vservers when emerging with the shared portage tree being read-only.. worked through some of the issues, but others are still a problem
1227303831 M * PowerKe the tree should be read-only, but distfiles read-write
1227303891 M * bliz42 well, using the guide on the system the portage tree was read only, but mounted a shared packages folder read-write to save binaries into once built once
1227303901 M * bliz42 that made things work nicely for most builds i've tried
1227303915 M * bliz42 but when trying to build coreutils, it wants to write '/usr/portage/sys-apps/coreutils/Manifest' -- which it doesn't have access to do
1227303934 M * bliz42 and this is using the vemerge util
1227303956 M * PowerKe Hmm, it's odd that it wants to write there if you didn't change anything in the ebuild
1227303995 M * bliz42 ya.. so updated main portage and sync'd and even went through a local emerge on the host hoping it would fill out anything it needed in the shared portage tree, then tried again but same error
1227304069 M * bliz42 figured it might be something odd about the setup within the vserver template image i was using, got it off another site, so rebuilding my host right now, then i'll try to build my own template image and see if i can get things to work any cleaner
1227304071 M * PowerKe Maybe you can try: ebuild /usr/portage/sys-apps/coreutils/coreutils-6.10-r2 manifest
1227304093 M * PowerKe and: ebuild /usr/portage/sys-apps/coreutils/coreutils-6.10-r2.ebuild digest
1227304107 M * bliz42 on the host or the guest?
1227304108 M * PowerKe (also add .ebuild on the previous one) But that shouldn't happen on a properly synced tree
1227304116 M * PowerKe host, guest is read-only
1227304123 M * bliz42 ya, ok
1227304130 M * bliz42 when i get things setup i'll try that and see what happens
1227305438 J * Walex ~Walex@82-69-39-138.dsl.in-addr.zen.co.uk
1227305683 M * Bertl welcome Walex!
1227305824 M * Hawq Bertl: any patches to test? :)
1227305910 M * Bertl yep
1227305941 M * Bertl but we have to put them together, I haven't compile tested anything yet, just prepared the pieces
1227305978 M * Bertl the first thing I'd like you to try is to start with your 'working' version and the following patches
1227305983 J * ntrs ~ntrs@77.29.199.52
1227306068 M * Bertl http://paste.linux-vserver.org/12620
1227306082 M * Bertl i.e. you want to revert them (-R)
1227306477 Q * Walex Read error: Connection reset by peer
1227306627 M * Hawq Bertl: kernel/signal.c:1164: error: 'struct vx_info' has no member named 'vx_initpid'
1227306641 M * Hawq after reverting the part from link
1227306678 J * chI6iT41 ~chigital@tmo-100-164.customers.d1-online.com
1227306683 M * Bertl okay, leave the last part, the initpid thingy out
1227306692 M * Bertl i.e. make that
1227306697 M * Bertl if (vx_check(vx_task_xid(p), VS_ADMIN|VS_IDENT) &&
1227306710 M * Bertl ->pid > 1 && !same_thread_group(p, current)) {
1227306728 M * Bertl (missing p)
1227307165 M * Hawq it will take a while. its recompiling whole tree
1227307180 M * Bertl okay
1227307350 Q * bonbons Quit: Leaving
1227308567 Q * ntrs Ping timeout: 480 seconds
1227309084 J * cga ~weechat@94.36.113.17
1227309171 J * Radiance ~Radiance@193.16.154.187
1227309767 Q * cga Quit: WeeChat 0.2.6
1227310226 M * Hawq Bertl: booted kernel. switching stoppped working
1227310274 M * Bertl good :)
1227310322 M * Bertl now we do the crosscheck, get the Linux-VServer version and apply the patch you just reverted
1227310425 M * Hawq you mean, untouched vanilla + vs and apply the patch?
1227310457 M * Bertl kernel.org kernel, with normal Linux-VServer patch, but those two hunks (from the pastebin) applied
1227310496 M * Bertl that will give you full proc isolation, but no signalling isolation
1227310520 M * Bertl the next step is to add a bunch of debug messages to those signalling checks, and see which signal is the critical one 
1227311138 M * Hawq its recompiling whole tree again. it usually takes ~45 minutes so I guess its better to finish checks tomorrow. You'll be online?
1227311170 M * Bertl yep, I should be (at least in the evening)
1227311184 M * Hawq err... finish checks today but later as its after midnight :)
1227311189 M * Hawq at least here
1227311195 M * Bertl hehe, yeah, fine :)
1227311292 M * Hawq what exactly should I change in next step? I'd check as much as I can before evening.
1227311330 M * Bertl I presume applying the hunks will make it work
1227311362 M * Bertl if that is true, we need some debug messages there to see _what_ signals from _where_ exactly get rejected
1227311410 M * Hawq in signal.c?
1227311441 M * Bertl yep
1227311591 M * Hawq hm. I just realized that just bzImage is needed for our tests. I should get one a lot faster than full rebuild
1227311595 M * Hawq :)
1227311661 M * Hawq heh. already have one. booting
1227311898 M * Hawq it works properly
1227311999 M * Hawq so... time for debug messages :)