1352334181 Q * Ghislain Quit: Leaving.
1352347984 Q * clopez Ping timeout: 480 seconds
1352349846 M * Bertl off to bed now ... have a good one everyone!
1352349851 N * Bertl Bertl_zZ
1352351788 M * arekm Bertl_zZ: one issue came up. tag_t is used in one driver, conflicting with vserver tag_t type
1352351819 M * arekm Bertl_zZ: otherwise built fine after wiping out xfs changes and disabling cow
1352354989 Q * s_parlane Ping timeout: 480 seconds
1352355004 J * cuba33ci_ ~cuba33ci@114-25-195-201.dynamic.hinet.net
1352355029 M * arekm Bertl_zZ: booted it, so far no suprises, works
1352355121 Q * cuba33ci Read error: Connection reset by peer
1352355129 N * cuba33ci_ cuba33ci
1352356308 Q * sannes Remote host closed the connection
1352358952 J * Ghislain ~aqueos@adsl1.aqueos.com
1352360343 Q * ard Ping timeout: 480 seconds
1352362200 Q * ncopa Quit: Leaving
1352362231 Q * geos_one Quit: ChatZilla 0.9.89 [Firefox 16.0.2/20121029000647]
1352362313 J * ncopa ~ncopa@3.203.202.84.customer.cdi.no
1352362338 Q * ncopa 
1352363129 M * swenTjuln is it possible to do something like "setattr --~hide /proc/partitions" per guest ?
1352363524 M * swenTjuln or even inject custom content ?
1352363688 M * daniel_hozac unhiding is global
1352363698 M * daniel_hozac you can do a bind mount on it though to change the contents
1352363747 M * swenTjuln daniel_hozac: how do I do later? Bind mount single file?
1352363783 M * daniel_hozac like any other bind mount, just specify a file as source and destination
1352363808 M * swenTjuln i didn't know it works for files
1352363864 M * daniel_hozac yeah it's not very commonly used...
1352364000 M * swenTjuln tnx!
1352364193 J * ncopa ~ncopa@3.203.202.84.customer.cdi.no
1352365059 Q * Ghislain Quit: Leaving.
1352367072 J * ard ~ard@shell2.kwaak.net
1352367129 J * Ghislain ~aqueos@adsl1.aqueos.com
1352367583 Q * Ghislain Quit: Leaving.
1352367824 J * Ghislain ~aqueos@adsl1.aqueos.com
1352369744 J * s_parlane ~scott@121-74-248-190.telstraclear.net
1352369837 J * BenG ~bengreen@cpc29-aztw23-2-0-cust144.18-1.cable.virginmedia.com
1352372370 J * clopez ~clopez@fanzine.igalia.com
1352372401 Q * fisted Remote host closed the connection
1352372456 J * fisted ~fisted@xdsl-84-44-239-126.netcologne.de
1352374134 J * geos_one ~chatzilla@85-125-139-162.work.xdsl-line.inode.at
1352374864 Q * ircuser-1 Ping timeout: 480 seconds
1352375590 Q * Aiken Remote host closed the connection
1352376170 Q * nkukard Ping timeout: 480 seconds
1352377872 J * aurel42 ~aurel@sid.a42.de
1352378168 M * aurel42 Hi. I have a couple of LVS servers on obsolete hardware, for organizational reasons it would be easier to migrate them to VMware guests than to new hardware. Are there reasons not to try this? (Does a VMware guest have everything an LVS host needs?)
1352378201 M * daniel_hozac shouldn't be a problem.
1352378250 M * aurel42 daniel_hozac: that's good news, thanks. :)
1352378312 J * ircuser-1 ~ircuser-1@35.222-62-69.ftth.swbr.surewest.net
1352378465 J * starfish ~Chatzilla@659AABU56.tor-irc.dnsbl.oftc.net
1352378660 Q * starfish Remote host closed the connection
1352378708 J * starfish ~Chatzilla@82VAAHSIV.tor-irc.dnsbl.oftc.net
1352379268 Q * starfish Remote host closed the connection
1352379281 J * starfish ~Chatzilla@659AABU67.tor-irc.dnsbl.oftc.net
1352380213 Q * geos_one Quit: ChatZilla 0.9.89 [Firefox 16.0.2/20121029000647]
1352381566 Q * fisted Remote host closed the connection
1352381602 Q * starfish Killed (tomaw (No reason))
1352381617 N * Bertl_zZ Bertl
1352381617 M * Bertl morning folks!
1352381618 N * Bertl Bertl_xX
1352381618 N * Bertl_xX Bertl
1352381626 J * fisted ~fisted@xdsl-84-44-239-126.netcologne.de
1352382609 M * BlackPanx hello guys
1352382667 J * starfish ~Chatzilla@83TAACFAL.tor-irc.dnsbl.oftc.net
1352382750 M * BlackPanx a quick question... if i have my host have an external ip on bridged interface br0 and all my vservers on this host are routed thru internal interface on another server... is it possible to somehow nat the host's port to vserver no matter if his default gateway is thru other internal ip (and consequentially external ip is different from host's)?
1352382751 M * starfish a quick question... if i have my host have an external ip on bridged interface br0 and all my vservers on this host are routed thru internal interface on another server... is it possible to somehow nat the host's port to vserver no matter if his default gateway is thru other internal ip (and consequentially external ip is different from host's)?
1352382809 M * BlackPanx vservers as guests... 
1352382810 M * starfish vservers as guests...
1352382821 M * BlackPanx can you stop repeating ? 
1352382822 M * starfish can you stop repeating ?
1352383001 F * ChanServ +o Bertl
1352383016 M * Bertl starfish: what's the problem?
1352383017 M * starfish starfish: what's the problem?
1352383030 K starfish Bertl starfish
1352383106 M * Bertl BlackPanx: shouldn't be a problem for iptables
1352383132 M * Bertl if I understand correctly, it should be similar to a transparent proxy setup
1352383296 M * BlackPanx hmm
1352383421 M * BlackPanx if i have fully virtualized guest with qemu or libvirt, then these rules fix this problem:
1352383422 M * BlackPanx iptables -t mangle -A PREROUTING -i br0 --dst xxx.xxx.xxx.xxx -p tcp --dport 80 -j MARK --set-mark 0x1
1352383424 M * BlackPanx iptables -t nat -A PREROUTING -m mark --mark 0x1 -j DNAT --to-destination 10.40.1.91
1352383425 M * BlackPanx iptables -t nat -A POSTROUTING -m mark --mark 0x1 -j SNAT --to-source 10.40.1.11 -> this being host's internal ip having default gateway thru public interface br0 and not 10.40.1.1 as guests.
1352383440 M * BlackPanx but this doesnt want to work with vserver
1352383457 M * BlackPanx maybe i should have iptables rule inside guest and not host ?
1352383462 M * BlackPanx that one for postrouting
1352383464 M * BlackPanx ?
1352383470 M * BlackPanx but probably not...
1352383529 M * daniel_hozac what are you actually trying to do?
1352383553 M * daniel_hozac make the host appear as if it's using its own IP address to the guests?
1352383669 J * geos_one ~chatzilla@80.123.185.198
1352383694 M * Bertl sounds a little strange, but should still be doable ...
1352383746 M * Bertl although the purpose is beyond my imagination
1352383787 M * Bertl off for now .. bbl
1352383791 N * Bertl Bertl_oO
1352383989 M * BlackPanx the thing is, provider binds external ip on host's mac. and i want to use that ip for let's say webserver. The only way to do that is to have host routed thru it's own public interface. Guests are routed thru internal interface to internal gateway for purpose of ospf... but i would want to still use this single ip of host, so i dont have to buy let's say a whole range of ip's on the gateway... if i have more webserver guests.
1352384037 M * BlackPanx so i simply routed host thru external interface and all guests thru internal gateway
1352384077 M * BlackPanx but logically, when i forward port from host's public ip to guest, then guest want to reply thru it's internal gateway... which is wrong
1352384136 M * BlackPanx it's just hetzner's limitations of public ip's... i'd prefer if they gave me all these public ips and i'd add them to one server and have that for gateway, have nat there and problem solved.... but reality is different.
1352384203 J * nkukard ~nkukard@41-133-237-213.dsl.mweb.co.za
1352387328 Q * BenG Quit: I Leave
1352387992 M * Bertl_oO sounds like you want connection tracking there
1352388030 M * Bertl_oO i.e. mark those incoming from the host IP and return the replies accordingly
1352388180 M * BlackPanx hmm
1352388602 J * Walex ~Walex@78-86-80-54.zone2.bethere.co.uk
1352389353 Q * fisted Quit: leaving
1352390070 Q * ensc Ping timeout: 480 seconds
1352391918 Q * kshannon Ping timeout: 480 seconds
1352393527 J * kshannon ~kris@kris.shannon.id.au
1352393701 J * jasat ~Chatzilla@1GLAAATFK.tor-irc.dnsbl.oftc.net
1352393932 N * jasat jasoom
1352394552 J * bonbons ~bonbons@2001:960:7ab:0:7de3:deff:7ebc:57ce
1352394565 J * tuxmania ~bonbons@2001:960:7ab:0:7de3:deff:7ebc:57ce
1352395445 N * jasoom barsoom
1352395836 J * fisted ~fisted@93.232.249.7
1352395876 Q * barsoom Killed (Ganneff (sod off))
1352395955 J * barsoom ~Chatzilla@659AABVJV.tor-irc.dnsbl.oftc.net
1352395999 Q * barsoom Killed (Ganneff (No reason))
1352396972 Q * tuxmania Quit: Leaving
1352399593 Q * jokar Ping timeout: 480 seconds
1352399685 Q * ensc|w Remote host closed the connection
1352399696 J * ensc|w ~ensc@www.sigma-chemnitz.de
1352401343 J * sannes ~ace@cm-84.211.87.28.getinternet.no
1352402903 Q * clopez Ping timeout: 480 seconds
1352404044 Q * Walex Read error: Connection reset by peer
1352405493 Q * fisted Remote host closed the connection
1352405504 J * fisted ~fisted@93.232.249.7
1352405662 Q * fisted Read error: Connection reset by peer
1352405727 J * fisted ~fisted@p5DE8F907.dip.t-dialin.net
1352405857 J * Aiken ~Aiken@2001:44b8:2168:1000:21f:d0ff:fed6:d63f
1352407031 M * s_parlane has anyone got HA vserver working ? and how did you do it ?
1352407119 M * Bertl_oO define 'HA vserver'
1352407387 M * s_parlane any means of surviving a node crash/failure
1352407416 M * Bertl_oO I guess that is the default by almost all providers
1352407422 M * s_parlane (even if the vserver does have the appearance of being rebooted during the failure)
1352407456 M * Bertl_oO rather simple, either use a shared filesystem or make sure to keep a recent backup if the data is static
1352407485 M * s_parlane ok, so shared file system approach and i repeatidly get kernel oops
1352407487 M * Bertl_oO then use whatever means (heartbeat, cron, etc) to detect a failure and start the backup guest
1352407507 M * Bertl_oO what kernel do you use? what filesystem?
1352407527 M * s_parlane currently i have: gfs2 + pacemaker + cman + gentoo ~amd64
1352407543 M * s_parlane on dell PE-2950-III's
1352407558 M * Bertl_oO okay, what kernel/patch version is that?
1352407573 M * s_parlane 3.5.4-vs2.3.4.3
1352407609 M * Bertl_oO and what kernel oopses do you get? could you upload one for me?
1352407611 M * s_parlane also gfs2 stops all I/O if a node goes away
1352407628 M * Bertl_oO yeah, gfs2 is probably not the best choice
1352407742 M * s_parlane better options ?
1352407779 M * Bertl_oO drbd for example
1352407779 J * hijacker_ ~hijacker@cable-84-43-134-121.mnet.bg
1352407797 M * Bertl_oO i.e. doesn't need to be active-active for a failover
1352407860 M * s_parlane ok, there is a san behind them
1352407874 M * s_parlane so better to ext4 and mount as required ?
1352407941 M * Bertl_oO for example, especially if the oops is gfs2 related (which might be possible)
1352407955 M * s_parlane its def gfs code
1352407968 M * s_parlane how do i get less not to go into binary mode ?
1352408032 M * Bertl_oO hmm, by not 'less'ing binary data? :)
1352408059 M * s_parlane its /var/log/messages, its not binary data
1352408086 M * Bertl_oO I guess there is some non ascii there then, try to remove it with e.g. 'tr' first
1352408107 M * sannes s_parlane: You have more than two nodes in your gfs2 setup so that quorum is maintained if a node takes a dive right? (and fencing set up?)
1352408234 M * s_parlane http://pastebin.com/UFhPqJY0
1352408254 M * s_parlane sannes: not currently, i really need to fix that, and yes i understand why its a problem
1352408281 M * s_parlane all my oops are at that same line of C
1352408303 M * s_parlane and its triggerable whenever the guest runs env-update (specifically the ldconfig causes it)
1352408309 M * Bertl_oO interesting, you might want to run the strack trace addresses through addr2line
1352408325 M * Bertl_oO addr2line -e vmlinux (in the kernel source tree)
1352408367 M * s_parlane i did and it gave me a stupid answer, so ill see if i can find a better one
1352408385 M * Bertl_oO make sure the kernel is built with debug information for that
1352408503 M * s_parlane yeah, im trying to find one recent enough that my kernel is the same config
1352408723 M * s_parlane http://pastebin.com/Y36bgcRg
1352408738 M * s_parlane that one has the same appearance as the other one, just i can successfully debug it
1352408982 M * Bertl_oO so the state isn't LM_ST_EXCLUSIVE
1352409038 M * Bertl_oO have you checked with the gfs2 folks yet?
1352409126 M * s_parlane not yet
1352409159 M * s_parlane i was looking for ideas of what i could run in the meantime
1352409192 M * Bertl_oO definitely ext3 on nas should do
1352409252 M * Bertl_oO check where LM_ST_EXCLUSIVE is set and where gl_state gets changed
1352409262 M * s_parlane ok
1352409270 M * s_parlane ill bbs, need to go to work
1352409272 M * Bertl_oO (of course, the gfs2 folks should know that by heart)
1352409277 M * s_parlane thanks for your help
1352409281 M * Bertl_oO you're welcome!
1352409758 Q * s_parlane Ping timeout: 480 seconds
1352410285 Q * hijacker_ Quit: Leaving
1352410601 Q * cuba33ci Read error: Connection reset by peer
1352410677 J * cuba33ci ~cuba33ci@114-36-236-249.dynamic.hinet.net
1352410730 J * s_parlane ~scott@202.49.72.37
1352410782 M * s_parlane has anyone got vserver and ceph working together ?
1352410936 M * Bertl_oO shouldn't be a problem ... IMHO
1352411056 M * s_parlane can i mount each vservers rootfs separately ?, in /etc/vserver/name/fstab ?
1352411934 Q * bonbons Quit: Leaving
1352412112 J * BenG ~bengreen@cpc29-aztw23-2-0-cust144.18-1.cable.virginmedia.com
1352412300 Q * sannes Remote host closed the connection
1352412378 M * s_parlane oh yay, i got a different crash
1352412400 M * s_parlane wait, no, i got the same bug, but passwd caused it this time
1352412660 M * Bertl_oO https://bugzilla.redhat.com/show_bug.cgi?id=810713
1352412696 M * Bertl_oO the fix at the end might be interesting to try out
1352412859 M * s_parlane ok, i'll apply that if possible
1352413031 M * s_parlane it does apply to my kernel
1352413595 M * s_parlane bug went away, yay, thanks
1352413603 M * Bertl_oO \o/
1352413720 M * Bertl_oO again, an excellent time to point at: http://linux-vserver.org/Donations
1352413903 Q * fisted Ping timeout: 480 seconds
1352413985 Q * FireEgl Remote host closed the connection
1352414938 Q * BenG Quit: I Leave
1352414977 J * FireEgl FireEgl@2001:470:e5ad:1:50c4:2de8:9bdf:851d
1352415338 J * fisted ~fisted@xdsl-84-44-239-126.netcologne.de
1352415600 Q * fisted Remote host closed the connection
1352415650 J * fisted ~fisted@xdsl-87-78-12-66.netcologne.de
1352416080 Q * Harzilein Ping timeout: 480 seconds
1352416088 J * Harzilein ~harzi@harzilein.eu.org
1352416223 Q * Ghislain Quit: Leaving.
1352417136 Q * trippeh Read error: No route to host
1352417188 Q * FireEgl Ping timeout: 480 seconds
1352417391 J * trippeh atomt@t-1000.ugh.no
1352417567 Q * Aiken Quit: Leaving
1352417731 J * FireEgl ~FireEgl@173-25-83-57.client.mchsi.com
1352418284 Q * FireEgl Quit: Leaving...