1312330729 J * kbad ~Adium@pool-108-38-208-174.lsanca.fios.verizon.net
1312330795 M * kbad whenever I build a vserver kernel it always has a "+" at the end of it and it's driving me batshit because my tftp setup doesn't like it
1312330799 M * kbad does anyone know how to undo that?
1312330836 M * Bertl huh? how do you build it?
1312330858 M * kbad git pull from kernel.org and then apply with a patch
1312330861 M * kbad build it with make
1312330879 M * Bertl and where would the + come from?
1312330906 M * kbad I thought it might be in the vserver patch somewhere
1312330942 M * kbad maybe it's something about my build environment..
1312330953 M * Bertl not really, but did you try with one of the tarballs instead of the git pull?
1312330978 M * kbad I haven't, I didn't think it would matter since I was using a tagged release 
1312330984 M * kbad I'll give it a go though
1312331025 M * Bertl I can assure you, Linux-VServer doesn't add a '+' in the kernel version or anywhere nearby, so either that or the build environment causes this
1312331042 M * kbad you would know :)
1312331049 M * kbad thanks
1312331300 M * Bertl np
1312331973 M * kbad it was a git thing
1312331980 M * kbad in scripts/setlocalversion
1312332000 M * kbad :)
1312332006 M * MooingLemur kbad: I was about to say or check the EXTRAVERSION assignment near the top of the main kernel's Makefile and hax it yourself.
1312332028 M * kbad ahh, it appends to the end of that
1312332028 M * kbad he
1312332033 M * kbad heh
1312332052 M * MooingLemur kernel's main*
1312337635 Q * hparker Quit: Quit
1312344077 M * Bertl off to bed now .. have a good one everyone!
1312344090 N * Bertl Bertl_zZ
1312344711 P * kbad 
1312347650 Q * tokkee Remote host closed the connection
1312347651 J * tokkee tokkee@osprey.tokkee.org
1312347695 J * michal_ ~michal@www.rsbac.org
1312347780 Q * michal Read error: Connection reset by peer
1312347783 N * michal_ michal
1312348700 Q * uranus Quit: Verlassend
1312354087 Q * derjohn_mob Ping timeout: 480 seconds
1312355327 J * derjohn_mob ~aj@213.238.45.2
1312358901 J * BenG ~bengreen@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com
1312358916 M * BenG morning all
1312358926 M * BenG sorry this is a bit lazyweb of me, but...
1312358964 M * BenG I'm build a 3.0 kernel, but remember that there's a config option that needs disabling to get vserver working on the 3.0 series
1312358980 M * hijacker morning
1312358983 M * BenG can anyone remember what that option is?
1312358990 M * BenG morning hijacker 
1312358992 M * hijacker user namespace
1312358998 M * BenG okay
1312359012 M * BenG what's the CONFIG title?
1312359041 M * BenG CONFIG_USER_NS ?
1312359334 M * hijacker i am not sure, this is what was mentioned in here recently
1312359340 M * hijacker should be that one, yes
1312362687 M * BenG cheers hijacker, compiling now with that option disabled
1312362727 M * hijacker BenG, HTH
1312363618 J * ant-1 ~ant-1@80.78.6.61
1312363705 P * ant-1 
1312364472 M * ser "chrt --idle 0" is good to put processes in the 'cpu' background - how to do the same with "I/O" background, as e.g. mdadm is checking arrays?
1312364764 M * daniel_hozac ionice
1312365312 M * cehteh well wotn work for mdadm as it triggers the kernel to do the work
1312365336 M * daniel_hozac sure, mdadm is already a background task.
1312365346 M * cehteh there are the sync speed knobs in /proc and /sys but they are often inprecise and broken 
1312365376 M * cehteh daniel_hozac: the raid resyncing is not done by mdadm but in kernel by a kernel thread, mdadm only triggers it
1312365408 M * cehteh you can trigger it by writing to /sys too without mdadm
1312365463 M * cehteh ls /sys/block/md1/md/  (or whatever your array is)
1312365794 M * ser ionice manual is not clear, -c 3 means idle?
1312365833 M * cehteh yes
1312365869 M * ser ok, cheers, guys
1312365873 M * cehteh but see what i've saied, reniceing mdadm wont do the job, you have to reduce sync_speed_max
1312365947 M * ser no, it was an example of a perfectly 'idle' job
1312366050 M * ser can i mix these two, eg. "ionice -c 3 chrt --idle 0 command" ?
1312366069 M * daniel_hozac yes
1312366985 M * cehteh nice 19   also puts processes into a 'batch' priority 
1312366998 M * ser is it possible to put a one whole vserver to ionice idle ?
1312367002 M * cehteh yes
1312367008 M * ser how?
1312367029 M * cehteh uhm i am doing that .. but hey i set that up years ago and never cared  :P
1312367041 M * ser :)
1312367062 M * cehteh cat /etc/vservers/devel/ionice/class 
1312367062 M * cehteh 3
1312367081 M * cehteh  cat /etc/vservers/devel/nice 
1312367081 M * cehteh 11
1312367086 M * ser thanks, i was aware of cgroups only
1312367110 M * cehteh years ago i saied :P meanwhile there are prolly better ways to do that
1312367124 M * daniel_hozac not really
1312367131 M * cehteh Linux rootserver 2.6.28.3-vs2.3.0.36.4 #2 SMP Wed Feb 4 21:09:59 CET 2009 x86_64 GNU/Linux
1312367144 M * cehteh .. still didnt build a new kernel, never touch a working system
1312367702 M * ser no security updates? :D
1312367810 M * cehteh psst :P
1312368981 M * ser ok another question: can i run the whole vserver with chrt --idle 0 ?
1312369020 M * cehteh certainly .. there are script hooks
1312369046 M * ser where should i seek for them?
1312369052 M * daniel_hozac the great flower page
1312369055 M * daniel_hozac as always
1312369060 M * cehteh google for "great flo...
1312369062 M * cehteh :)
1312369065 M * ser OK
1312369072 M * cehteh chrt --idle 0 -p $$   
1312369088 M * cehteh (where you have to ensure that $$ resolves to the right pid)
1312369089 M * ser i was sure there are only /etc/vservers parameters
1312369105 M * ser not scripts
1312369202 M * ser and i was not wrong, there are no examples
1312369234 M * daniel_hozac because you can do anything you want?
1312369257 M * ser i know i can do enything but the only question is: how :) :)
1312369259 M * daniel_hozac man bash i'm sure has examples.
1312369275 M * ser which script should i use - initialize?
1312369281 M * ser or post-start?
1312369283 M * daniel_hozac did you read about them?
1312369304 M * ser ok, reading, sorry
1312369449 M * ser ok, still do not get it, there re pre, post - but which script runs the vserver processes themselves? 
1312369474 M * ser should i chrt existing processes after they are run?
1312369489 M * ser in post- script?
1312369492 M * cehteh its initial process
1312369511 M * cehteh the init (if you use that) or whatever you use to start it
1312369527 M * cehteh chrt .. vserver foo start 
1312369610 M * ser OK i see, i will try :)
1312372160 M * daniel_hozac or a non-executable pre-start script that sets VSERVER_EXTRA_CMDS=( /path/to/chrt --idle 0 )
1312374082 M * ser i see, thanks!
1312375576 M * cehteh are the EXTRA_CMDS sourced?
1312375602 M * cehteh otherwise you need the -p $$
1312376034 N * Bertl_zZ Bertl
1312376040 M * Bertl morning folks!
1312379307 Q * yang Ping timeout: 480 seconds
1312379723 Q * BenG Remote host closed the connection
1312379925 J * ghislain ~AQUEOS@adsl2.aqueos.com
1312380911 Q * ghislain Quit: Leaving.
1312381102 J * jrdnyquist ~jrdnyquis@slayer.caro.net
1312381275 J * kir ~kir@swsoft-msk-nat.sw.ru
1312381593 P * kir Leaving.
1312381923 J * BenG ~bengreen@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com
1312383563 J * fisted ~fisted@p50883EFD.dip.t-dialin.net
1312383660 J * hparker ~hparker@2001:470:1f0f:32c:beae:c5ff:fe01:b647
1312384052 J * SkyNet2000 ~SkyNet200@71-81-25-51.dhcp.gwnt.ga.charter.com
1312384184 Q * ncopa Quit: Leaving
1312385474 J * strub ~vincent@ASt-Lambert-151-1-32-17.w82-124.abo.wanadoo.fr
1312385673 M * strub Bertl: read-only bind mount support has apparently been dropped from the latest patches (2.6.39/3.0)
1312385678 M * strub is this intentional ?
1312385700 M * strub i.e I used to be able to do mount -o bind,ro /foo /bar, and /bar would be readonly
1312385717 M * daniel_hozac that's been the case for a long time now.
1312385727 M * strub with the latest patches, /bar is read-write and I need a mount -o remount,ro /bar to make it read-only
1312385732 M * daniel_hozac yes.
1312385746 M * daniel_hozac mainline decided that was the better interface.
1312385769 M * strub vserver patches used to (up til 2.6.38) add a mnt->mnt_flags = mnt_flags in fs/namespace.c:do_loopback()
1312385818 M * daniel_hozac it's been a mainline thing since 2.6.26.
1312386013 M * strub daniel_hozac: I know, but it used to work without needing a remount on vserver kernels until 2.6.38 (included)
1312386068 M * daniel_hozac that was most likely not intentional.
1312386085 M * strub OK, so it's time to update my scripts :)
1312386147 Q * fisted Ping timeout: 480 seconds
1312386787 M * Bertl yeah, was kind of an accident, we don't want to change mainline behaviour without good reason
1312386913 M * Bertl daniel_hozac: does util-vserver handle the ro bind case yet? i.e. with a single line, or does one need two lines in fstab?
1312386928 M * strub OK, no problem - I think I'll just add a one-liner patch to my kernels to retain the old behaviour
1312386960 M * Bertl always an option as well
1312387077 M * daniel_hozac single line works
1312387090 M * Bertl thought so, thanks for the info!
1312387242 J * rickytato ~rickytato@2-225-96-236.ip175.fastwebnet.it
1312387550 Q * BenG Quit: I Leave
1312387829 Q * padde Remote host closed the connection
1312387988 Q * DLange Remote host closed the connection
1312388006 Q * nox Remote host closed the connection
1312388011 Q * Marillion Remote host closed the connection
1312388025 Q * mnemoc Remote host closed the connection
1312388054 Q * biz Remote host closed the connection
1312388096 Q * monrad-51468 Quit: bla
1312388109 Q * Phoner_ Remote host closed the connection
1312388354 J * Marillion ~dirk@178.63.150.30
1312388354 Q * Marillion Remote host closed the connection
1312388423 Q * nospoonuser_ Remote host closed the connection
1312388756 J * nospoonuser ~nospoonus@shell.net23.de
1312388925 J * Phoner phoner@administrat.org
1312389303 J * nox ~nox@host.noxlux.de
1312389589 J * DLange ~DLange@dlange.user.oftc.net
1312389803 Q * derjohn_mob Ping timeout: 480 seconds
1312389974 J * standiik 5ab691ba@ircip4.mibbit.com
1312390235 M * standiik hello. i have strange network problem on vserver guest. both on the host and guest are ip addresses from the same range with the same gateway. but from guest i cant ping gateway
1312390243 M * standiik running on debian squeeze
1312390293 M * standiik i followed the instruction from http://www.howtoforge.com/linux_vserver_debian_etch
1312390706 J * padde ~padde@patrick-nagel.net
1312390731 M * daniel_hozac did you use the same IP as is used somewhere else in the network?
1312390739 M * daniel_hozac can the host ping the gateway?
1312390810 M * standiik yes, host can ping gateway. from the guest i can ping host ip
1312390927 Q * strub Remote host closed the connection
1312390944 J * biz biz@baze.de
1312390956 M * ser # ip r 
1312390962 M * ser in guest, please
1312390988 M * daniel_hozac doesn't really matter if it's in the guest or not, it'll just be obscured.
1312391012 Q * SkyNet2000 Quit: Leaving
1312391020 M * standiik 81.31.32.0/24 dev eth0  proto kernel  scope link  src 81.31.32.168
1312391021 M * standiik default via 81.31.32.161 dev eth0
1312391036 M * standiik 168 is host, 169 is guest, 161 gateway
1312391099 M * Bertl try ping -c 1 -I 81.31.32.169 www.google.com
1312391109 M * Bertl (on the host)
1312391126 M * standiik no response
1312391128 Q * dowdle Remote host closed the connection
1312391135 M * Bertl now try the same with .168
1312391149 M * standiik works
1312391163 M * daniel_hozac you might also want to try arping -I eth0 -s 81.31.32.168 81.31.32.169
1312391174 M * Bertl so, your router (or maybe a firewall on the host) blocks your traffic
1312391233 M * standiik where is no firewall on the host, its a clear installation
1312391270 M * Bertl well, any sensible distro would install a firewall nowadays (in the default install)
1312391282 J * bonbons ~bonbons@2001:960:7ab:0:c15d:788a:5d9c:7e7e
1312391286 M * standiik i installed minimal network version of debian
1312391294 M * Bertl anyway, the problem is with the .169 IP not being handled properly
1312391329 M * Bertl i.e. either the gateway (.161) doesn't like the IP, or the reply packets do not reach your host or something blocks them on the way
1312391399 J * dowdle ~dowdle@scott.coe.montana.edu
1312391404 M * standiik once i tried from guest run /sbin/route and then i pinged gateway and suddenly it worked. but after restart occured the same problem. now its impossible to repeat this situation
1312391449 M * standiik i know, its very strange
1312391462 M * daniel_hozac i'd try the arping
1312391469 M * daniel_hozac sounds like you have an address conflict.
1312391480 M * Bertl yeah, chances are good you have a different machine with the same IP
1312391711 M * standiik ok and how can i test, if the conflict exists?
1312391734 M * standiik sorry for my questions, iam not advanced user. iam still learning
1312391768 M * Bertl with the arping :)
1312391794 M * Bertl arping -I eth0 -s 81.31.32.168 81.31.32.169
1312391875 M * standiik  arping: Weird MAC addr 81.31.32.168
1312391901 M * daniel_hozac sounds like you have a strange arping, but i guess that's expected with debian....
1312391928 M * standiik now iam sure, there is a conflict. because i stopped guest and tried http://81.31.32.169 and there is a response from apache
1312391947 M * standiik on the host there is no opened port 80
1312391980 M * standiik so it cant be host. is it possible to retrieve MAC of some IP address?
1312391995 M * Bertl with arp?
1312392022 M * standiik i have never used arp before. i try to read something about that
1312392037 M * daniel_hozac i am positive that you have :)
1312392047 M * Bertl :)
1312392061 M * Bertl try (on the host) ping -c 1 81.31.32.169
1312392066 M * Bertl then execute 'arp'
1312392107 M * Bertl should list you a bunch of IP/MAC associations, including the MAC for that IP
1312392115 M * Bertl (if it can be found)
1312392175 M * standiik i have to install arping on the host and i cant access internet
1312392283 M * Bertl how so, I thought the host IP did work?
1312392285 M * standiik arping have other dependencies (pcap) so it will take some time
1312392296 M * standiik sorry, i tried from guest
1312392300 M * standiik mmt
1312392329 M * Bertl the guest shouldn't even be up when you investigate IP collisions
1312392331 M * standiik i tried arping -I eth0 -s 81.31.32.168 81.31.32.169 and got arping: Weird MAC addr 81.31.32.168
1312392337 M * standiik ok
1312392359 M * Bertl well, weird distro, use 'man arping' to figure out how that arping is supposed to work
1312392409 M * sid3windr yea
1312392413 M * standiik there is -S not -s
1312392414 M * sid3windr in debian there are 2 arpings I think
1312392422 M * sid3windr one in iputils-ping the other in netutils-ping
1312392425 M * sid3windr one sucks, the other doesn't
1312392435 M * Bertl get the other :)
1312392546 M * standiik now i see different MAC assigned to 81.31.32.169 than MAC of host interface
1312392550 M * standiik so its collision
1312392564 M * Bertl yup
1312392590 M * standiik i really thank You all
1312392679 M * Bertl np
1312392753 M * standiik 60 bytes from 00:11:d8:31:e5:ba (81.31.32.168): index=0 time=267.982 usec
1312392759 M * standiik 42 bytes from 00:14:85:3f:54:5c (81.31.32.168): index=1 time=979.893 msec
1312392766 M * standiik root@master:~# arping -I eth0 -S 81.31.32.168 81.31.32.168
1312392794 M * standiik when i run it respond from two MACs
1312392826 M * Bertl so, fix that and everything should work
1312392829 M * standiik second one is right MAC of interface
1312393159 Q * transacid Remote host closed the connection
1312393544 M * ser let us back to ionice/chrt pair, i set everything and it works, but it prints something like that
1312393549 M * ser # vserver uni enter
1312393549 M * ser ionice: ignoring given class data for idle class
1312393562 M * ser (but it works)
1312393609 M * Bertl what is your ionice command?
1312393651 M * Bertl I suspect you specify something besides -c3
1312393671 M * Bertl (while the idle class doesn't have any priority or other data)
1312393698 M * ser $ cat uni/ionice/class
1312393699 M * ser 3
1312393705 M * ard standiik : are they both the same host?
1312393715 M * daniel_hozac util-vserver unconditionally adds -n.
1312393728 M * Bertl ah, so there is the problem
1312393737 M * ser ah, ok - it explains everything, cheers :)
1312393749 M * Bertl patches are welcome, I guess :)
1312393755 M * ard standiik : reading backlog: obviously not, you have 1 interface ;-)
1312393759 M * standiik ard: no, host have only one interface
1312393766 M * ser Bertl: I'll have a look
1312393806 M * standiik ard: so it means that there are two interfaces on the network with the same address?
1312393824 M * ard standiik : Then it's a question of somebody has to lart the owner of the other machine. You can BTW bypass it by unicasting arp-updates to the router, the other one will not be reachable anymore ;-0
1312393871 M * ard standiik : yes: another host has taken your IP ;-)
1312393872 M * Bertl depends on how smart the router is
1312393907 M * standiik i wrote to my provider. thank you again for help
1312393950 M * ard Are there routers that don't accept arp-updates?
1312393970 M * daniel_hozac fairly sensible thing to reject.
1312393972 M * ard or maybe some security configuration?
1312393989 M * daniel_hozac "oh look, i'm on the same network as <big website>, let's take it down!"
1312394011 M * ard Heh, you really don't want to know... I've seen it all...
1312394031 M * ard usually cheap ass providers :-)
1312394075 M * ard Any decent provider will put you in a private vlan or in a private DMZ, whatever the terminology
1312394089 M * daniel_hozac yup
1312394163 M * ard standiik has a weird gateway btw... 161, sounds like he doesn't have a /24 but more like a /28 or so, and the other .168 is a friend of his :-)
1312394189 J * transacid ~transacid@84.19.167.254
1312394216 N * transacid Guest4623
1312394261 N * Guest4623 transacid
1312395018 M * Bertl yeah, the /24 is probably wrong
1312395871 Q * standiik Quit: http://www.mibbit.com ajax IRC Client
1312396885 J * mnemoc ~amery@shell.opensde.net
1312403477 J * fisted ~fisted@p50883EFD.dip.t-dialin.net
1312403688 N * ntrs ntrs_away
1312404255 Q * bonbons Quit: Leaving
1312407770 J * derjohn_mob ~aj@d004230.adsl.hansenet.de
1312410138 Q * derjohn_mob Ping timeout: 480 seconds
1312410530 J * imcsk8 ~ichavero@148.229.1.11
1312411223 J * derjohn_mob aj@tmo-024-87.customers.d1-online.com