1237421136 Q * dowdle Remote host closed the connection
1237422185 Q * duckx Remote host closed the connection
1237422235 J * saulus ~saulus@c135170.adsl.hansenet.de
1237422396 Q * harobed Ping timeout: 480 seconds
1237422609 Q * saulus_ Ping timeout: 480 seconds
1237423985 Q * FireEgl Ping timeout: 480 seconds
1237424412 N * DoberMann[PullA] DoberMann[ZZZzzz]
1237424553 J * FireEgl ~FireEgl@WTF.4.1.0.c.0.7.4.0.1.0.0.2.ip6.arpa
1237425767 Q * FireEgl Ping timeout: 480 seconds
1237426335 J * FireEgl ~FireEgl@Sebastian.Tcldrop.US
1237427267 J * scientes ~scientes@75-165-16-192.tukw.qwest.net
1237432870 Q * derjohn Ping timeout: 480 seconds
1237433385 J * derjohn ~derjohn@80.69.41.3
1237438716 J * joss ~joss@123.208.42.137
1237438794 M * joss why releases of 2.5.23+ up to 2.6.26 aren't included in patchsets like 2.6.25, is that because vserver is allready included in vanillas?
1237438917 M * cehteh nope
1237438954 M * cehteh there where some major kernel changes which made maintenance for the stable vserver branch unfeasible
1237438984 M * cehteh but there is bertl's 2.3 experimental vserver stuff which many people use for recent kernels
1237438991 M * joss cehteh: i think so too, grepping among those gave too little, ah ok, so 2.6.25 and 2.6.26 aren't supported?
1237439040 M * cehteh dunno about those specifically but 2.6.28 is supported by vserver 2.3
1237439045 M * joss cehteh: ok, but i find the last for 2.6.27.19 i think, experimental was quite ok, only found one bug
1237439070 M * cehteh which is almost complete and we will soon (matter of few months) see a new stable vserver 2.4
1237439102 M * cehteh i didnt found bugs so far (well i dont search, i just run it)
1237439169 M * joss cehteh: there was one when changing vt's (spesifically chaning vt fron guest X to host X ) but it was known one, then Host session was killed
1237439191 M * cehteh ah .. wont happen here, all my vserver are headless :)
1237439255 M * joss cehteh: yup, but that was quite pointless to file it, because wiki said so too , it works when chaning first to console from guest X, and then to Host X..like a vbe stuff or something similar
1237439365 M * joss it won't hang the system only host X graphics are gone:) not restorable, vt is flushed
1237439385 J * doener_ ~doener@i59F5741F.versanet.de
1237439457 M * joss cehteh: ok anyways, is there an experimental vs3 archive also?
1237439486 Q * doener Ping timeout: 480 seconds
1237439503 M * cehteh vs3?
1237439530 M * joss http://ftp.linux-vserver.org/pub/kernel/vs2.3/testing/ has four majors missing 23to26
1237439534 M * cehteh i made a public git repsitory for my own itches which is somewhat unmaintained until i need it again
1237439557 M * cehteh if anyone wants to push new stuff there, it is public pushable without an account
1237439592 M * joss 2.3
1237439601 M * cehteh between .23 to .26 there was much flux on the kernel side
1237439620 M * cehteh and maybe bertl was occupied with other stuff that time too .. ask him
1237439630 M * joss cehteh: ah ok se there aren't patches for those releases at all, yes?
1237439640 M * cehteh dunno
1237439644 M * cehteh maybe nit
1237439646 M * cehteh not
1237439683 M * joss distributions also haven't got them, so i tend to think those were missed out
1237439737 M * joss ah 2.6.26 is there for debian still, yeah
1237439763 Q * hparker Quit: Read error: 104 (Peer reset by connection)
1237439794 M * joss http://vserver.13thfloor.at/Experimental/
1237439811 M * joss ok, thanks, gotta go
1237439998 J * ghislainocfs2 ~Ghislain@LPuteaux-151-41-11-129.w217-128.abo.wanadoo.fr
1237440294 Q * grosgis Ping timeout: 480 seconds
1237440307 J * grosgis ~admin@abeille.lifl.fr
1237440318 Q * ghislainocfs21 Ping timeout: 480 seconds
1237442972 J * sharkjaw ~gab@149-240-82.oke2-bras6.adsl.tele2.no
1237443068 J * scientes_ ~scientes@75-165-16-192.tukw.qwest.net
1237443778 T * * http://linux-vserver.org/ |stable 2.2.0.7, devel 2.3.0.34, grsec 2.3.0.36|util-vserver-0.30.215|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 forget about the minute.
1237443778 T * Bertl -
1237447110 J * mtg ~mtg@vollkornmail.dbk-nb.de
1237450067 Q * joss Read error: Connection reset by peer
1237450429 Q * FireEgl Ping timeout: 480 seconds
1237450966 J * cga ~weechat@62.196.2.6
1237450981 J * FireEgl ~FireEgl@2001:470:e056:1:4::
1237452444 Q * transacid Remote host closed the connection
1237452688 J * duckx ~Duck@81.57.39.234
1237452877 J * transacid ~transacid@84.19.167.254.nbiserv.com
1237453320 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d
1237453634 J * thierryp ~thierry@home.parmentelat.net
1237453899 J * harobed ~harobed@arl57-1-82-231-110-14.fbx.proxad.net
1237454043 J * wam ~wam@p549D6050.dip.t-dialin.net
1237454163 M * wam Hi! I have a nfs mountpoint in my /etc/vserver/name/fstab. It worked after vserver restart. Now, several weeks later, the mountpoint is called "Stale NFS file handle". ls -l looks like this: d?????????   ? ?    ?        ?                ? backup  - any hints?
1237454283 M * wam *maybe* the admin changed the nfs volume name.
1237454421 M * wam ok, my appologies. While describing the bug I found out that the directory is no longer available. Sorry for the noise.
1237454444 M * wam the remaining question is: Can I remount something while the vserver is up?
1237454960 Q * duckx Remote host closed the connection
1237455201 M * bonbons wam: yes, you can, enter the guest's namespace (not the guest) and do the remounting
1237455326 M * wam bonbons: let me google about that
1237455352 M * wam bonbons: that's vnamespace?
1237455373 M * bonbons wam: yep (at least it was last time I tried)
1237455391 M * wam bonbons: and - let me guess - there are no security restrictions when I do this ;)
1237455478 M * bonbons wam: as you just switch filesystem namespace as host root, you just have the security restrictions that apply to host root
1237455511 M * wam bonbons: thanks a lot. I'll give it a try
1237455543 M * bonbons wam: just be careful on what path you (re)mount, / in host and guest's namespace are the same, guest's are kind of chrooted to their directory
1237455584 M * bonbons and when playing with mount you might want to add '-n' and use complete commandline in order to not edit (the wrong) mtab
1237455651 M * bonbons wam: you might also want to have a look at vmount
1237455698 M * wam bonbons: where do I get that?
1237455724 M * bonbons wam: that's part of util-vserver as is vnamespace
1237455757 M * wam bonbons: I don't have it :/
1237455789 M * bonbons wam: hmm... what version of util-vserver do you have?
1237455789 M * wam what I would try now is: vnamespace -e delivery umount /backup
1237455807 M * wam bonbons: 0.30.214
1237455827 M * bonbons wam: I have 0.30.215
1237455851 M * bonbons wam: first check with vnamespace -e delivery cat /proc/mounts
1237455873 M * wam bonbons: that looks good.
1237455881 M * wam so /backup is translated?
1237455885 M * wam i.e. chrooted?
1237455898 M * bonbons wam: chances are pretty good that /backup is not where you think it is! Probably it's /<path-to-delivery-root>/backup
1237455910 M * wam ok, I'll try that first
1237456014 M * bonbons wam: normally, when guest's are started the host namespace is cloned, fstab for the guest is processed, namespace gets (partially) cleaned from unused mount points from host and finally guest gets started kind of chrooted
1237456034 M * wam bonbons: ok, vnamespace -e delivery umount /vservers/delivery/backup worked. Now in delivery, I have no mount according to /proc/mounts. But /etc/mtab remembers the mount.
1237456139 M * wam hah
1237456140 M * wam worked
1237456141 M * wam mount done
1237456142 M * wam ;)
1237456154 M * wam the only thing that's wrong now is /etc/mtab
1237456163 M * bonbons wam: well, you will have to manually edit that file, mtab is normally managed by mount (though with -n mount ignores it, and in case / != / it might touch the wrong mtab!)
1237456230 M * wam ok
1237456265 M * wam on the host, /proc/mounts doesn't show the new mountpoint and on the vserver /proc/mounts does show it. So all is good. I'll manually remove the mtab entry from the host and add it to the guest. Does that sound ok?
1237456331 M * wam bonbons: ^^
1237456377 M * bonbons wam: yep
1237456390 M * wam bonbons: thank you very much. You really helped me out!
1237456418 M * bonbons wam: you're welcome
1237456691 J * dna ~dna@53-215-103-86.dynamic.dsl.tng.de
1237458950 N * DoberMann[ZZZzzz] DoberMann
1237458979 Q * xdr Quit: leaving
1237459255 N * Bertl_zZ Bertl
1237459260 M * Bertl morning folks!
1237459612 M * maio good morning :)
1237460249 M * hijacker_ afternoon
1237460561 J * dna_ ~dna@53-215-103-86.dynamic.dsl.tng.de
1237460981 Q * dna Ping timeout: 480 seconds
1237461284 J * davidkarban ~david@88.86.104.103
1237461500 Q * sharkjaw Remote host closed the connection
1237461763 J * sharkjaw ~gab@149-240-82.oke2-bras6.adsl.tele2.no
1237463051 M * bonbons Bertl: is it wanted to have most recent patches return PPID=1 for guest's init?
1237463125 M * bonbons Bertl: this might be the reason why "ps -auxf" fails to return anything... (-f asks ps to built process tree, and ps probably does not know how to handle init...)
1237463169 M * Bertl ah, I knew I forgot something ...
1237463195 M * Bertl i.e. no, it is not on purpose, it is more the result of mainline changes
1237463198 M * bonbons Bertl: well, that issues exists for some time already on 2.3.xxx
1237463229 M * Bertl do you have a fix at hand or do you want to do some testing?
1237463310 M * Bertl I presume it should be easy to fix, we already special case the pid there, so ppid shouldn't be a big deal, we expect 0 there in ps auxf ?
1237463376 J * gnuk ~F404ror@pla93-3-82-240-11-251.fbx.proxad.net
1237463403 M * bonbons Bertl: in 2.6.22.19-vs2.2.0.7 we return 0, also 0 on host (all versions I looked at)
1237463426 M * bonbons Bertl: If you have a fix I can test it, not tried to write one yet
1237463486 M * Bertl okay, give me a minute for a test patch, the thing is, we should test all 3 setup cases, i.e. init-less, init, and init with pid=1
1237463546 M * bonbons hm, what's the difference between init and init with pid=1?
1237463571 M * Bertl we can run guests with init at its real pid, but still being the init process/child reaper
1237463597 M * bonbons ah, never noticed that feature...
1237463604 M * Bertl that's mainly for cases where it isn't required to make init pid=1
1237463621 M * bonbons well, all my guests have init with pid=1 ...
1237464014 M * Bertl do you have Linux-VServer debugging enabled?
1237464044 M * bonbons yes, I do
1237464081 M * Bertl excellent, let's enable the following: cvirt: bit 2
1237464116 M * Bertl then, inside a guest, try to only retrieve the proc entry containing the ppid
1237464138 M * Bertl (from the fake init)
1237464203 M * bonbons hm, echo 2 > /proc/sys/vserver/debug_cvit
1237464212 M * bonbons cat /proc/<1>/status
1237464233 M * bonbons and nothing did show up in dmesg
1237464306 M * bonbons or am I numbering the bits in the wrong way?
1237464423 M * Bertl yep bit 0=1, bit 1=2 bit2=4
1237464577 M * bonbons ok, that produces output (lots of it), for plain init
1237464614 M * Bertl okay, try to reduce it to a minimum for one 'cat'
1237464627 M * Bertl should be like 3-4 lines, I think
1237464869 M * bonbons 20 lines for cat /proc/1/status
1237464924 M * Bertl okay, please upload them
1237465129 M * bonbons http://paste.linux-vserver.org/12790
1237465206 M * Bertl hmm, the init inside the guest shows 1 for the ppid (in your setup, right?)
1237465316 M * bonbons Bertl: yep, that's also what I read in /proc/1/status inside the guest
1237465356 M * bonbons but it should be read as 0 according to what I read on vanilla linux and on host
1237465613 M * Bertl hmm, the interesting part is that the following doesn't show 1, it shows the real pid instead (which is wrong too btw)
1237465625 M * Bertl chcontext --xid 42 --flag fakeinit -- bash
1237465647 M * bonbons btw, 3160 is the guest's init, 3422 is the shell inside the guest
1237465820 M * bonbons well so n both cases it show the read pid of parent process
1237465833 M * Bertl okay, try the following patch:
1237465845 M * Bertl http://vserver.13thfloor.at/Experimental/delta-initppid-fix01.diff
1237466103 J * takeru ~takeru@nttkyo584177.tkyo.nt.ftth.ppp.infoweb.ne.jp
1237466355 M * bonbons ok, that makes ps work as expected and /proc/1/status display expected ppid==0
1237466603 M * Bertl great! could you test it with the other two init setups as well please?
1237466621 M * Bertl I think it should be fine, but better save than sorry
1237466695 M * bonbons for the fake-init case it looks goot with "chcontext --xid 42 --flag fakeinit -- bash"
1237466742 M * Bertl carefull, that is basically the same as you tested
1237466751 M * Bertl i.e. the 'plain' init style
1237466832 M * bonbons need to re-read doc to know how to run those other modes...
1237467163 Q * cga Quit: WeeChat 0.2.6
1237467227 M * bonbons chcontext --xid 42 -- bash     -> cat /proc/$$/status: Pid:    3822, PPid:   3818
1237467235 M * bonbons and there is no /proc/1/
1237467267 M * bonbons not sure what PPid should look for this one... 3818 is process on the host
1237467424 M * Bertl yeah, that should be right, I think
1237467633 M * bonbons it does not disturb ps, but might be it's not best to show the host process PID either...
1237467671 M * Bertl well, if you set the bash as init process, you should get the 'correct' behaviour
1237467687 M * Bertl vcontext --create --xid 42 -- vattribute --set --secure --flag info_init -- vcontext --migrate-self -- bash
1237467694 M * Bertl this is for the blend through init case
1237467736 M * Bertl vcontext --create --xid 42 -- vattribute --set --secure -- vcontext --migrate-self --initpid -- bash
1237467744 M * Bertl and this one for the init with 'real' pid
1237467927 M * bonbons for blend-through init: /proc/1/status: ppid=0, /proc/$$/status: ppid=host_process (shouldn't it show up as '1'?)
1237467936 Q * mtg Ping timeout: 480 seconds
1237468073 M * bonbons for init with 'real' pid: /proc/1/status: ENOENT, /proc/$$/status: ppid=0
1237468119 Q * sharkjaw Remote host closed the connection
1237468187 J * sharkjaw ~gab@149-240-82.oke2-bras6.adsl.tele2.no
1237468205 M * Bertl well, we can think about adjusting the ppid for such processes, but we need a criterion which would identify them
1237468215 M * Bertl (in the blend through init case)
1237468335 M * Bertl but I'm not sure this is a problem in a guest with init style 'sysv'
1237468346 M * Bertl (because the parent already died there)
1237468548 M * bonbons a criterion could be that all ppids outside of own context get set to 1 (who gets the SIGCHLD if $$ exists prior to its real parent on host?)
1237468611 M * Bertl yes, but that is unrealistically to test for
1237468642 M * Bertl i.e. we would have to walk all pids or lookup processes for a given pid to check if they are inside or outside
1237468706 M * Bertl we could kill the parent pid when the process enters a context, but I'm not sure that is what we want either
1237468746 M * bonbons could we change parent-ship on entering context?
1237468779 M * Bertl in theory, yes, but changing it to the host init (which would be the sensible thing to do) is not the best idea
1237468799 M * Bertl i.e. you would lose connection to the parent shell on enter for example
1237468850 M * bonbons don't know what impact that would have (is it possible for a thread to change context individually?)
1237468889 M * Bertl the kernel only knows about processes, so kernel threads per se are handled like a process, where userspace threads do not matter
1237468954 M * bonbons well, kernel knows some hierarchy as it either shows processes in /proc/x/ or /proc/y/tasks/z/
1237468993 M * Bertl all in all, I think the advantage of reparenting would be easily outweighted by the disadvantages of changing the parent for simple creates or enter/migrate
1237468998 M * bonbons s/tasks/task/
1237469037 M * Bertl it is shown that way in proc, but there is no real difference in the kernel
1237469134 M * bonbons so just a matter of what is shared between processes for the choice of putting it right in /proc or in /proc/z/task/?
1237469169 M * Bertl yeah, basically taskgroup and session id
1237469275 M * bonbons in this case it gets non-trivial to do proper reparenting which remains sensible at all levels... Do you know how that has been solved for mainline contexts?
1237469382 M * Bertl it simply hasn't been addressed at all
1237469395 M * Bertl i.e. mainline pid contexts cannot be entered
1237469416 M * Bertl and they have to have an init process
1237469531 M * bonbons and how does it work at creation?
1237469564 M * Bertl the kernel spawns the init process and leaves it parentless
1237469588 M * Bertl (similar to the spawning of the host init)
1237469631 M * bonbons so differently from all the other namespaces (uid, fs, uts, ...)!?
1237469661 M * Bertl yes, well, most of the newer namespaces are 'different' in some way
1237469675 M * Bertl just consider the cgroup and CFS 'space'
1237469740 M * bonbons hm, except for the name those two are unknown to me...
1237470388 A * bonbons out to lunch
1237470399 M * Bertl enjoy!
1237470453 M * bonbons thanks
1237470772 J * cga ~weechat@62.196.2.6
1237471105 J * mtg ~mtg@vollkornmail.dbk-nb.de
1237471160 M * Bertl welcome mtg!
1237471460 Q * sharkjaw Quit: Leaving
1237471636 M * Bertl okay, off for now .. bbl
1237471640 N * Bertl Bertl_oO
1237471915 Q * takeru Quit: takeru
1237473477 Q * scientes Ping timeout: 480 seconds
1237475704 Q * mtg Quit: Verlassend
1237476609 J * dowdle ~dowdle@scott.coe.montana.edu
1237477974 J * pmjdebru1jn pascal@jester.pcode.nl
1237477977 J * SpComb^ terom@zapotek.paivola.fi
1237477982 J * _Hunger Hunger.hu@Hunger.hu
1237477984 J * sladen paul@starsky.19inch.net
1237477985 J * Wonka_ produziert@chaos.in-kiel.de
1237477988 J * Hollow_ ~bene@shiva.xnull.de
1237477992 J * maio_ ~maio@217.196.213.98
1237477994 J * oo_ ~oo@64.28.12.166
1237477995 J * _Radiance ~Radiance@193.16.154.187
1237478008 J * [Guy] ~korn@elan.rulez.org
1237478015 Q * wam reticulum.oftc.net galapagos.oftc.net
1237478015 Q * Hunger reticulum.oftc.net galapagos.oftc.net
1237478015 Q * Guy- reticulum.oftc.net galapagos.oftc.net
1237478015 Q * Bertl_oO reticulum.oftc.net galapagos.oftc.net
1237478015 Q * sladen__ reticulum.oftc.net galapagos.oftc.net
1237478015 Q * Wonka reticulum.oftc.net galapagos.oftc.net
1237478015 Q * oo reticulum.oftc.net galapagos.oftc.net
1237478015 Q * Hollow reticulum.oftc.net galapagos.oftc.net
1237478015 Q * maio reticulum.oftc.net galapagos.oftc.net
1237478015 Q * Radiance reticulum.oftc.net galapagos.oftc.net
1237478015 Q * phedny reticulum.oftc.net galapagos.oftc.net
1237478015 Q * yang reticulum.oftc.net galapagos.oftc.net
1237478015 Q * SpComb reticulum.oftc.net galapagos.oftc.net
1237478015 Q * pmjdebruijn reticulum.oftc.net galapagos.oftc.net
1237478015 N * Wonka_ Wonka
1237478020 J * Bertl_oO herbert@IRC.13thfloor.at
1237478021 N * _Radiance Radiance
1237478021 J * phedny_ ~mark@phedny.vps.van-cuijk.nl
1237478021 J * yang yang@static-ip-62-75-255-125.inaddr.server4you.de
1237478631 M * maio_ I'm getting a lot of: [498462.114720] vxW: [�ps�,32367:#1|0|0] did lookup hidden f59b1ca4[#6,8] �/dev/pts�.
1237478634 M * maio_ is it ok?
1237478690 J * wam ~wam@p549D6050.dip.t-dialin.net
1237480142 Q * maio_ Quit: sax
1237481916 Q * harobed Ping timeout: 480 seconds
1237482183 Q * cga Quit: WeeChat 0.2.6
1237482260 Q * wam Quit: All your base are belong to us!
1237482415 J * hparker ~hparker@2001:470:1f0f:32c:290:96ff:fe50:40fa
1237482853 Q * balbir_ Ping timeout: 480 seconds
1237483806 J * maio ~maio@217.196.213.98
1237486121 N * DoberMann DoberMann[PullA]
1237486159 Q * ghislainocfs2 Quit: Leaving.
1237486561 J * xdr ~xdr@gote2.244.cust.blixtvik.net
1237487007 Q * xdr Quit: Lost terminal
1237487465 J * cga ~weechat@94.36.110.25
1237487659 N * Bertl_oO Bertl
1237487664 M * Bertl back now ...
1237487686 M * Bertl maio: basically means that _something_ inside the guest is looking up a pts entry
1237487702 M * Bertl harmless, but I suspect you arre using an older kernel (debian)
1237488313 Q * gnuk Quit: NoFeature
1237488672 Q * thierryp Quit: ciao folks
1237488707 Q * Piet Ping timeout: 480 seconds
1237489200 J * BenG ~bengreen@94-169-110-10.cable.ubr22.aztw.blueyonder.co.uk
1237489501 J * Piet ~piet@asteria.debian.or.at
1237489907 Q * davidkarban Quit: Ex-Chat
1237490572 M * critch Bertl: still hanging around?
1237490584 M * Bertl yup
1237490606 M * critch solved my routing problem in a way to make me happy. 
1237490614 M * Bertl excellent!
1237490621 M * critch Created a new route table in /etc/iproute2/rt_tables
1237490633 M * critch Added a rule for my private network to use the new table
1237490642 M * critch then created a default route in that table.
1237490660 M * critch Now my vserver guests act just like my vmware guests.
1237490662 M * Bertl yep, that's multiple routing tables
1237490683 M * critch It took me a while to remember my solution for a multiple WAN router that I made for the office.
1237490739 M * critch I have a firewall here doing 1-1 NAT for a DSL PPPoE connection and a NAT for a cable modem connection. All that working fine and allows the workstation decide which outbound network to place the traffic on.
1237491348 J * ghislainocfs2 ~Ghislain@LPuteaux-151-41-11-129.w217-128.abo.wanadoo.fr
1237492335 J * dreisbac ~rosewell@61.148.115.250
1237492335 P * dreisbac 
1237492811 M * maio Bertl: I'm using 2.6.26-1-vserver-686
1237492832 M * Bertl as expected :)
1237492843 M * Bertl if you update, the messages should disappear
1237492860 M * Bertl but as I said, they are harmless
1237493165 M * maio ok, thanks. btw I'm also getting some kernel oops messages in my dmesg (something about write_tty) - is that harmless too? :)
1237493202 M * Bertl nope, oopses are never harmless, but please upload some (paste.linux-vserver.org) to check if it actually is an oops
1237493352 M * maio http://paste.linux-vserver.org/12791
1237493371 J * duckx ~Duck@81.57.39.234
1237493454 M * Bertl yep, that's an oops, another reason to update the kernel
1237493473 M * maio hm that's debian stable :)
1237493492 M * Bertl yeah, stable == broken and won't get fixed :)
1237493531 M * maio hehe
1237493576 M * Bertl that's the burden debian users have to carry ...
1237493666 M * critch only for a little while. Should soon have a new kernel in testing and unstable. Learn about apt pinning so you can get specific parts from upstream and still make it work. Or just create a kernel the debian way.
1237493782 J * Keeper longhorn40@90.151.15.68
1237493982 M * Keeper Hello all!
1237494004 M * Bertl wb Keeper!
1237494135 M * Keeper Bertl: How are you?
1237494145 M * Bertl fine, thanks! and you?
1237494173 M * Keeper I am fine
1237494262 M * Keeper Bertl: I have some questions...
1237494266 Q * hparker Ping timeout: 480 seconds
1237494299 M * Bertl let's hear ..
1237494484 M * Keeper Bertl: Swap in the guest environment possible?
1237494520 M * Bertl yes and no, swapping is something which happens host wide, i.e. ther eis no point in having 'guest specific' swapping
1237494530 M * Bertl *there is
1237494554 M * Bertl but, we support a virtualization which shows memory limits as guest memory and swap space
1237494569 M * Bertl i.e. you can make it look like a guest had memory and swap
1237494675 M * Keeper Bertl: How can I do?
1237494701 M * Bertl the RSS soft and hard limits can be used for that
1237494722 M * Bertl the soft limit will show up as memory, and the hard limit as memory + swap
1237494784 M * Bertl (just enable virt_mem)
1237494980 M * Keeper Bertl: And will show the swap in a guest environment?
1237495000 M * Bertl yep
1237495037 M * Keeper Bertl: Good
1237495150 M * Keeper Bertl: I wanted to put VSserver on Gentoo, but there is a bug
1237495162 M * Bertl hmm?
1237495268 M * Keeper Bertl: Error while creating the guest environment with a template gentoo
1237495301 M * Bertl are we talking gentoo as host distro, or gentoo as guest distro, or both?
1237495366 M * Keeper Bertl: Gentoo host > gentoo guest
1237495388 M * Bertl should work perfectly fine, what error did you get, and what template did you use?
1237495402 M * Bertl (note: there is a gentoo build method :)
1237495577 M * Keeper Bertl: Standard template is not altered stage3, the error later, I said now is no server to set Vserver
1237495718 M * Bertl could you upload the output to paste.linux-vserver.org?
1237495731 M * Bertl (make sure to use export LC_ALL=C LANG=C :)
1237496079 M * Keeper Bertl: I do not understand here should send a message paste.linux-vserver.org, yes?
1237496102 M * Bertl yes, do whatever causes your error, and upload the output there
1237496115 M * Bertl we'll have a look and probably can help you
1237496246 M * Keeper Bertl: I will send a message describing the error
1237496521 M * _Shiva_ Bertl: about that /dev/pts lookup earlier... I have those, too ... @2.6.28.7-vs2.3.0.36.8 
1237496561 M * _Shiva_ Bertl: however the guest is Debian Lenny.. ;-) 
1237496619 M * _Shiva_ Bertl: but a gentoo guest does produces this, too
1237496630 M * Bertl util-vserver version?
1237496641 M * Bertl and when do you get it?
1237496686 Q * BenG Quit: I Leave
1237496722 M * _Shiva_ uhhm... util-vserver still is gentoo-portage stable.. Kernel compiled @Fri Mar 6 21:22:32 CET - using the patches that were available just b4
1237496749 J * ViRUS ~mp@p579B4A91.dip.t-dialin.net
1237496757 M * _Shiva_ util-vserver 0.30.215
1237496796 M * Bertl could you try with a 0.30.216pre, preferably the latest one?
1237496817 M * Bertl also, do you get that message when you enter the guest, or during normal guest behaviour?
1237496859 M * _Shiva_ hmm.. mostly at "vserver <XID> stop"
1237496877 M * _Shiva_ i'll give latest util-vserver a try
1237496892 M * _Shiva_ (not right now, though)
1237496917 M * Bertl it should not happen when the guest is running normally, only on external guest manipulation (like enter or such)
1237496940 M * Bertl i.e. in cases where the host brings a pty/pts into the guest
1237496982 M * _Shiva_ like starting a task within "screen" within guest ? :-)
1237497005 M * Bertl nah, that shouldn't cause it, when you didn't enter the guest
1237497013 M * Bertl i.e. when you ssh-ed in for example
1237497062 M * _Shiva_ ah - and maybe it's Asterisk trying to get a console to spit out it's console messages..
1237497097 M * Bertl acquiring new ptys from inside the guest should be fine
1237497360 M * _Shiva_ Bertl: first column is the processes' name, that dit the lookup, right?
1237497626 M * Bertl task name:address[#tag,ino] path
1237497663 M * _Shiva_ ha! I'm able to trigger two lookups at once :-)
1237497685 M * _Shiva_ within the guest: # ps waux
1237497705 M * Bertl but you didn't enter it, just used ssh or so to access it?
1237497749 M * _Shiva_ I've entered the guest using "vserver <XID> enter", which does not trigger anything
1237497762 M * Bertl well, that _is_ what causes it
1237497780 M * _Shiva_ then enter "ps waux" in the shell - THAT triggers two lookups
1237497789 M * Bertl you are bringing a pty from the host into the guest
1237497803 M * Bertl but it still belongs to the host, so it cannot be looked up inside the guest
1237497823 M * Bertl the ps waux inside the guest looks up the ptys (for whatever reason)
1237497872 M * _Shiva_ you're right (<- sure you are ;-)) .. when I use ssh to enter the guest it does not trigger anything..
1237498045 M * Keeper Bertl: All you need to modify the Guest OS?
1237498108 M * Bertl basically you do not need to modify the guest OS, but it helps to clean up hardware related scripts and such
1237498209 M * Keeper Bertl: Remove all excess
1237498378 M * Keeper Bertl: what kernel.vshelper?
1237498414 M * Bertl that is a script or program, which gets called from the kernel to do certain things (in userspace, on behalf ot the kernl)
1237498423 M * Bertl e.g. when a reboot or shutdown is called
1237498512 M * Keeper Bertl: Thank you very much
1237498666 M * Bertl no problem
1237498921 M * Keeper Bertl: Thank The new version of VServer is possible to transfer the device from the Host system in the Guest system?
1237498970 M * Bertl you can always 'copy' a device from the host into the guest, the guest will then be able to use it ... recent kernels allow you to do device mapping too
1237498992 M * Bertl i.e. map certain devices to other devices and/or allow to create those device node
1237499061 M * Bertl (from insid ethe guest, that is)
1237499164 M * Keeper Bertl: So it will be possible to install the Guest system to LVM partition and transmit it to the Guest system?
1237499196 M * Bertl yes, putting the guest on an LVM logical volume is a common practice
1237499210 M * Bertl but note, that this way you cannot benefit from unification
1237499295 Q * zugowaq Remote host closed the connection
1237499366 M * Keeper Bertl: Still Guest system to LVM is very good
1237499394 M * Bertl there are advantages and disadvantages as usual
1237499473 M * Keeper Bertl: I agree, but each of its...
1237499589 M * Keeper Bertl: Correct me if I'm wrong, but I noticed that a lot of opportunities for Vserver, but not all documented
1237499613 M * Bertl that's right, there a limitless possibilities :)
1237499677 M * Bertl and applications not even I could think up ... for example some time ago Linux-VServer was used to create a multi seated desktop setup ...
1237499702 M * Bertl i.e. running several X11 instances with separate keyboards and mice on the same machine
1237499722 M * Bertl (but isolated from eachother)
1237499839 M * Keeper Bertl: Cool
1237499977 M * Keeper Bertl: Recently I have had this idea ... whether it is possible to Guest system via VNC?
1237500067 Q * cga Quit: WeeChat 0.2.6
1237500542 M * Bertl sure, works in several ways, easiest approach is to run a vnc x11 server in the guest
1237500675 M * Bertl e.g. Xvnc
1237500712 M * Keeper Bertl: likely because you are talking about the restrictions in this technology is not...
1237500785 M * Keeper Bertl: I mean VServer...
1237501065 N * DoberMann[PullA] DoberMann[ZZZzzz]
1237501190 M * Keeper Bertl: I decided to put CentOS 5.2 Host and Build Centos 5 Guest, so now there is one drawback... When installed RPM packages for the assembly of Guest yum does not show a fall, he will download at the moment...
1237501567 J * harobed ~harobed@arl57-1-82-231-110-14.fbx.proxad.net
1237501699 M * Bertl well, I'm sure there are yum options to achieve that, not using yum here, so I don't know
1237502016 M * Keeper Bertl: Ïîêàçûâàåò Total download size: 36 M íî íå ïîêàçûâàåò ÷òî èìåííî ñåé÷àñ downloading
1237502039 M * Keeper Bertl: sorry
1237502056 M * Keeper Bertl: Shows Total download size: 36 M but does not show that it is currently downloading
1237502073 M * Bertl maybe it is not reaching anything?
1237502089 M * Bertl make sure that the guest would be able to reach the internet
1237502141 M * Bertl (double check with tcpdump on the host if necessary)
1237502218 M * Keeper Bertl: 	Everything works, just does not show protssss...
1237502342 M * Keeper Bertl: Everything works, just does not show the status
1237502523 M * Bertl well, then I presume it is a missing yum option
1237502556 M * Bertl like --show-progess or so
1237502658 Q * bonbons Quit: Leaving
1237502661 M * Keeper Bertl: 	Maybe, okay it does not matter at all, I do not like CentOS just for interest, I put up my street in the little
1237502739 M * Keeper Bertl: So sorry for banal questions
1237503098 M * Keeper Bertl: Sorry for stupid questions, just I collect information for the site
1237503144 Q * doener_ Quit: quit quit quit quit quit
1237503315 J * larsivi ~larsivi@70.84-48-63.nextgentel.com
1237503371 M * Bertl I think with recent util-vserver it is possible to specify additional yum options and/or set the defaults used .. daniel_hozac knows more there
1237503473 M * Bertl but I also presume that this 'silent' install is what he actually prefers, as he probably is using yum
1237503546 Q * maio Quit: Lost terminal
1237503789 M * Keeper Bertl: Just in my opinion, it would be if yum by default only shows the process, because not everyone will understand that getting the right package is yum but this just does not show
1237504446 M * Keeper Bertl: Be sure to mount /vservers on a separate partition?
1237504521 M * Keeper daniel_hozac: hi
1237504538 J * thierryp ~thierry@home.parmentelat.net
1237504542 Q * thierryp 
1237504708 M * Bertl it makes sense to make /vservers a separate partition if you want to use unification and/or tagging (guest limits on a shared partition)
1237504723 M * Bertl but it also works if it is one the same partition as the host
1237504946 M * Keeper Bertl: Now running CentOS 5 Host and Guest and again to make sure that your project is worth attention because a lot of opportunities and manage easily
1237506006 M * Keeper Thank you very much that the answers to my questions, even though they were sometimes banal. It was nice to you, talk to learn many new and interesting. In the near future will be the documentation for the site...
1237506193 M * Bertl you're welcome! I'm looking forward to the upcoming documentation :)
1237506387 M * Keeper Bertl: I wanted to ask, but how would you call a domain for a site on Russian Vserver?
1237506532 J * hparker ~hparker@2001:470:1f0f:32c:290:96ff:fe50:40fa
1237506904 J * ktwilight_ ~ktwilight@91.178.144.54
1237507138 Q * ktwilight Ping timeout: 480 seconds