1115251309 J * cereal ~cereal@217.20.124.153 1115251440 J * SNy ~mfr@bmx-chemnitz.de 1115251525 M * Bertl wb cereal! SNy! 1115251545 M * terr Euh, where is the "util-vserver" archive? 1115251613 M * Bertl http://www.13thfloor.at/~ensc/util-vserver/files/alpha/ 1115251893 J * gabaZzzzleeping ~gaba@protest.net 1115251908 M * Bertl wb gabaZzzzleeping! :) 1115251960 J * ciphernaut ~a@61.88.18.130 1115251980 M * Bertl wb ciphernaut! 1115252001 Q * SNy Read error: Operation timed out 1115252026 Q * cereal Read error: Operation timed out 1115252043 M * ciphernaut hi Bertl! 1115252045 M * Bertl hmm, looks like the internet is flakey again ... 1115252096 J * cereal ~cereal@217.20.124.153 1115252126 J * SNy ~mfr@bmx-chemnitz.de 1115252181 M * terr "configure" complained about not finding "beecrypt". Is it necessary? 1115252208 M * Bertl not if you do not want vhashify 1115252220 M * Bertl (i.e. it should build fine without) 1115252328 M * terr Also should I set "vrootdir" (to the actual value where the vserver "phony" files are), i.e. "/var/lib/vservers" ('cause Debian) 1115252379 M * terr Or is it only when 'build' is called for? 1115252428 M * Bertl you should configure it for _all_ debian defaults ... 1115252462 M * Bertl i.e. look what patches debian uses, and configure them with ./configure 1115252469 M * Bertl s/patches/pathes/ 1115252824 M * terr Euh, do you mean that "prefix" should be "/" ? 1115252846 J * pflanze ~chris@ethlife-dev.ethz.ch 1115252853 M * pflanze Hello 1115252868 M * Bertl hmm, no, IMHO the 'proper' values would be: 1115252876 M * Martin Have to sleep 1115252879 M * Martin Night all 1115252888 M * Bertl night Martin! 1115252890 M * Martin Will try to get that patch tested 1115252892 M * Bertl terr: ./configure --prefix=/usr --sysconfdir=/etc --localstatedir=/var --with-vrootdir=/vservers 1115252905 M * Bertl terr: but that is in disagreement with debian ;) 1115252915 Q * Martin Quit: using sirc version 2.211+KSIRC/1.1 1115252918 M * Bertl hey pflanze! 1115252961 M * pflanze oh, I had running a popupper in this xchat installation which didn't work, 1115252970 M * pflanze I've missed what someone just said to me. 1115252974 M * pflanze Hi in any case;) 1115252994 M * pflanze Bertl: I've got my server back running with vs1.2.x. 1115253005 M * pflanze I'm glad it seems to work fine so far. 1115253008 M * Bertl congrats! 1115253023 M * terr But if I use the Debian values, won't there be a risk of cluttering some data? 1115253052 M * mugwump What was that linux dist that provided functional "bundles" based on Vserver? Anyone remember? 1115253055 M * pflanze So vs1.2+alpha utils seem to be working well together. If there's anything I should look out for, please tell me :). 1115253061 M * mugwump ...box linux or something perhaps? 1115253062 M * Bertl terr: well, you can of course configure it as you like it 1115253085 M * Bertl mugwump: to box I associate blue ;) 1115253105 M * mugwump the site was green I think :) 1115253116 M * Bertl pflanze: nope, 1.2.x and util-vserver should go well 1115253155 M * pflanze Thanks for the good work. 1115253182 M * Bertl you are welcome .. and I guess you know the donations page ;) 1115253209 M * Bertl (if you feel a sudden urge to donate .. or so ;) 1115253231 M * pflanze I'm completely blank right now, so no chance atm.. but I'll keep it in my mind. 1115253233 M * daniel_hozac mugwump: strongbox? 1115253243 M * terr Could I suggest that I upgrade first within Debian; we check, and if something doesn't go as you expect, ... 1115253259 M * mugwump strongbox, that was it, thanks daniel_hozac 1115253265 M * terr then I come back tomorrow with a clean vanilla install 1115253296 M * terr having the time to check all the paths etc. 1115253314 M * Bertl pflanze: don't worry, was more for the irc log ;) 1115253379 M * Bertl terr: well, if you want to do that .. it's fine with me ... 1115253419 M * terr So now I've 0.30.207 1115253529 M * Bertl don't forget to reenable the vshelper 1115253626 M * terr Done. 1115253698 M * terr So: "start" without "--debug", then "stop" with "--debug" ? 1115253708 M * pflanze BTW I'll go change the vserver startup script some time soon so that when booting the host, I'm getting a login prompt without having to wait for the vserver startups to finish. 1115253738 M * pflanze (Would need to redirect the output to some log.) 1115253806 M * Bertl hmm, that is already done, IIRC 1115253833 M * pflanze ah. cool. (I'm not running th latest utils) 1115253845 M * Bertl grep -v '#' /etc/vservers.conf 1115253845 M * Bertl BACKGROUND=no 1115253870 M * Bertl if you change this to yes, they will be started in the background ;) 1115253877 M * pflanze ah coool 1115253886 M * pflanze it's there already 1115254091 M * terr Well... It's going wrong (as expected: etc.). "/proc/virtual/99" still ther after "stop" 1115254176 M * Bertl okay, so we continue tomorrow ... 1115254186 M * terr "UseCnt" was "8" when running, now "2" 1115254201 M * terr "Tasks" was "3" when running, now "0" 1115254238 M * terr So, you are confident that this behaviour is Debian-related??? 1115254270 M * Bertl no idea, but I do not want to figure that out a few hours into testing ;) 1115254276 M * terr Wasn't there something you wanted to check with "vshelper" 1115254297 M * terr ? 1115254301 M * Bertl yup, but I want to se a mainline vshelper for that ... 1115254314 M * Bertl so that I can recreate it on my test host 1115254370 M * terr Is the code different? Or only the locations of the programs? 1115254388 M * Bertl no idea ... you have to compare it ... 1115254525 M * terr OK... I'll prepare some clean setup for tomorrow. 1115254549 M * Bertl please do so ... 1115254642 M * terr See you tomorrow then. What time will you be there? 1115254676 M * Bertl don't know yet, but probably in the afternoon 1115254681 M * Bertl (CET that is) 1115254752 M * terr Fine. Thanks, and have a good night. 1115254837 J * eXplasm2 ~explasm@p549FEB59.dip.t-dialin.net 1115254854 M * Bertl good night to you too! 1115255165 A * pflanze will reboot the router 1115255167 M * pflanze Good night. 1115255172 Q * pflanze Quit: [x]chat 1115255264 Q * explasm__ Ping timeout: 480 seconds 1115255375 Q * ComplexHo Quit: Leaving 1115255835 P * terr 1115256064 Q * lilo Read error: Connection reset by peer 1115256189 Q * cereal Quit: Serverwechsel 1115256209 J * cereal ~cereal@217.20.124.153 1115257394 J * lilo ~lilo@lilo.usercloak.oftc.net 1115258718 Q * lilo Quit: brb 1115259870 M * Bertl okay, I'm off to bed now ... have a nice whatever everyone! 1115259895 N * Bertl Bertl_zZ 1115261137 J * _cereal ~cereal@217.20.124.153 1115261137 Q * cereal Write error: connection closed 1115261142 N * _cereal cereal 1115263946 J * torisa hak@heceta.db.net 1115263973 M * torisa hi there. i'm having trouble running screen as a non-root user in a vserver. is this a known problem? 1115264023 M * torisa screen tells me "No more PTYs" 1115264134 M * mugwump I saw that error being discussed recently 1115264145 M * mugwump Have you checked your /dev/pts is mounted in the vserver? 1115264163 M * mugwump and how are you entering the vserver, ssh or `vserver xxx enter' ? 1115264178 M * torisa yea. /dev/pts seems to be working as expected. i use ssh 1115264210 M * mugwump ok, and you're sure that you've got nowhere near 256 or whatever the default max. ptys is on your system? 1115264245 M * torisa yep. i used strace to try and debug the problem but didn't get very far 1115264290 M * torisa screen uses readlink on /proc/self/fd/0 to open the terminal. root can do this easily 1115264301 M * torisa but non-root returns -1 at readlink 1115264327 M * torisa so i used strace on ls to see if it happned outside of screen, but ls succeeded 1115264420 M * mugwump you normally can't strace screen, it's setuid.. 1115264445 M * torisa oh that might explain it 1115264450 M * torisa hm 1115264482 M * mugwump does `cd /proc/self/fd && ls -lL' work? 1115264532 M * torisa i see 4 devices, 0, 1, 2, 255 1115264537 M * mugwump note you also can't use screen if have used "su" 1115264552 M * torisa right 1115264593 M * mugwump I have no idea how screen actually goes about creating ptys tho 1115264622 M * mugwump you have a /dev/tty as well? 1115264635 M * mugwump and /dev/pmtx? 1115264652 M * torisa tty exists, but pmtx does not 1115264691 M * mugwump needs to be c 5 2, perms 666 1115264695 M * mugwump root:root 1115264718 Q * SNy Read error: Connection reset by peer 1115264746 M * mugwump (and, of course, it needs to be created outside the vserver) 1115264856 M * torisa how do i update the vserver /dev entries once pmtx has been created on the host? 1115264875 M * mugwump just copy it into the vserver's /dev from the outside 1115264882 M * mugwump (cp -a) 1115264882 M * torisa okay 1115264924 M * torisa same error 1115264931 M * mugwump damn 1115264953 M * mugwump so, you've checked it looks right using ls -l /dev/pmtx inside the vserver? 1115264959 M * torisa yes 1115264997 M * torisa (stupid) it doesn't have to do with a lack of CONFIG_DEVPTS_FS_XATTR does it? 1115265140 M * mugwump that's turned on on my system, it surprises me that such an obscure option would be needed for such a basic function though 1115265151 M * mugwump I also have CONFIG_DEVPTS_FS_SECURITY=y 1115265156 Q * cereal Ping timeout: 480 seconds 1115265204 M * torisa hmm 1115265219 M * torisa i'll poke around and see what happens hehe 1115265224 M * torisa thanks for trying 1115265232 M * mugwump np 1115267002 J * cereal ~cereal@217.20.124.153 1115267024 J * lilo ~lilo@lilo.usercloak.oftc.net 1115267242 J * SNy ~mfr@bmx-chemnitz.de 1115271139 Q * lilo Quit: bbiab new kernel 1115271339 J * lilo ~lilo@lilo.usercloak.oftc.net 1115271532 Q * lilo Quit: 1115271756 J * lilo ~lilo@lilo.usercloak.oftc.net 1115272053 J * lilo_ ~lilo@lilo.usercloak.oftc.net 1115272053 Q * lilo Read error: Connection reset by peer 1115276621 N * ciphernaut ciphernautZZ 1115286960 N * Doener|gone Doener 1115286972 M * Doener morning folks! 1115287260 M * eyck I use su and use screen.. ;) 1115287296 M * Doener torisa: still around? 1115287692 M * Pazzo hi Doener! 1115287700 M * Doener hey Pazzo! 1115287717 M * Doener how are you? 1115287784 M * Pazzo everything is fine, thnx! sun is shining, I was sitting half an hour in the sun, drinking coffe, orange juice and reading some docs... what about you? 1115287839 M * Doener wurd: those steps i wrote down for you were meant to be done with a pristine sources.list for fc3 as supplied by the tools. also all steps are meant to be done from the start (i.e. if you already built a vserver, you cannot use it for this...). And finally: I told you that the tools would tell you when to start/stop the vserver and that i thus left those steps out... 1115287912 M * Doener slowly getting awake ;) seems to get normal for me not to sleep on tuesday nights... but at least that makes me sleep really good on wednesday night then :) 1115287973 M * Doener sun is shining here, too... might become a good day... 1115288001 M * Pazzo hehe... wurd still struggling with his full-featured vserver-desktop linux? :-) 1115288043 M * Doener guess so... 1115288096 M * Doener maybe xen would be the better way to go for him... but as i never used xen, i won't recommend anything ;) 1115288425 M * Pazzo Doener: did you read my post to the ml? about the debian stuff? 1115288500 M * Doener just saw it... 1115289458 M * Doener Pazzo: hmm... why do we need that -s switch? 1115289635 M * Doener ah ok... the -- doesn't enable argument passing the debootstrap, but to the build script that uses debootstrap 1115289716 M * Doener Pazzo: but -s just passes a 'script' argument to debootstrap, right? 1115289720 M * Pazzo yep 1115289750 M * Pazzo I've sent a new mail with an example as requested by "Gilles" 1115289761 M * Doener ok, what about the following: 1115289787 M * Doener we add the -s switch and a -v switch... 1115289810 M * Doener the -v switch is used for debootstraps --variant... 1115289872 M * Doener the -s is then used for _really_ special setup variants and -v is the usual "it is 'sarge' but i did some modifications" 1115289887 M * Doener (where 'sarge' may also be 'woody', 'sid', etc.) 1115289923 M * Doener then, we could add debian to the .distributions directory and allow to put various scripts there 1115290069 M * Pazzo Doener: you're right, -v is "better" than -s! the -s switch is already there in sarge/sid packages as I have asked kilian time ago to add it - I don't know if it is also to be found in upstream sources 1115290086 M * Pazzo but we should add both switches there! 1115290198 M * Pazzo you are right about the .distributions directory - 'til now I had to care just about my own bullshit and left the scripts in the first place I found to be useful 1115290260 M * Pazzo but as already written on irc/ml if we want to have such things to be already in util-vserver the right place will be somewhere in /usr/lib/util-vserver/distributions/... 1115290331 M * Pazzo ... and /etc/vservers/.distributions/... for the default sources.list and maybe some other stuff 1115290388 J * prae ~prae@sherpadown.net 1115290440 M * Doener yep... 1115290451 M * Doener guess i'll add some stuff to the ossf repo, soon... 1115290533 M * Doener but i'll need to figure out how exactly that debootstrap wrapper works first ;) 1115290569 J * Viruz Viruz@85.206.205.84 1115290582 M * Doener welcome prae, Viruz 1115290591 M * prae Mr Doener 1115290662 M * Doener Pazzo: what remount hack? the proc remount? is that still needed? 1115290712 M * Doener ah ok... you asked the same question ;) 1115290732 A * Doener never experienced those problems anyways... 1115290800 Q * Viruz Quit: 1115290820 A * Pazzo experienced those problems and will not remove the remount hack unless some "vserver guru" tells him that he can do so because the problem is solved! 1115290827 M * Pazzo ;-) 1115290856 M * Pazzo I have enough trouble all days so I'm not gonna search for it *gg* 1115290922 M * Pazzo (search for it -> search for trouble) 1115290935 M * Pazzo I'm gonna go eat something, cu later! 1115290942 N * Pazzo PazZzzzooo 1115291045 M * FaUl *gaehn* 1115291109 M * Doener hmm... we could even implement that external package management for debian vservers... 1115291197 M * Doener our dogs need a walk... later! 1115291200 N * Doener Doener|gone 1115292270 N * Doener|gone Doener 1115293545 J * terr ~nobody@ip-213-49-106-244.dsl.scarlet.be 1115293681 M * Doener welcome terr 1115293688 Q * rs Remote host closed the connection 1115294807 M * terr Hello. 1115295119 N * BobR_oO BobR 1115295235 Q * terr Quit: I am going away 1115295492 N * BobR BobR_oO 1115295730 J * terr ~nobody@ip-213-49-106-244.dsl.scarlet.be 1115296007 Q * Doener Remote host closed the connection 1115296036 J * Doener ~doener@p54875D6D.dip.t-dialin.net 1115296372 M * terr Waiting for Herbert? 1115296811 N * PazZzzzooo Pazzo 1115296814 M * Pazzo re 1115297042 M * Doener wb Pazzo 1115297274 M * Pazzo Doener! 1115297344 M * Pazzo Doener: external package management for debian vservers? you really want to build debian vservers unaware of what packages are installed? or do I missunderstand you? 1115297431 M * Doener well... if you want a really barebone vserver, internal package management might be annoying... also you can probably use a shared package cache and such things with external package management... 1115297448 M * Doener of course this isn't something you would want to use in a vps hosting environment ;) 1115297472 M * Doener and {in,ex}ternalizing the package management should be doable on the fly 1115297483 M * Doener it's just copying a few files around... 1115297494 M * Doener s/copying/moving/ 1115297570 Q * prae Quit: Pwet 1115297927 M * Pazzo yep 1115297934 M * terr Could you be more explicit as to what should go and shouldn't go in a "barebone" vserver? 1115297956 M * Pazzo Doener: and how do you keep the vservers "state" info about the installed packages? 1115298009 M * Pazzo terr: the term "barebone" vserver isn't clear to me too. Doener? 1115298068 M * terr In fact I'm looking for such a minimal setup... 1115298102 M * daniel_hozac Pazzo: sed uses POSIX regexes, IIRC. (see regex(7)) 1115298120 M * Pazzo thnx daniel_hozac! 1115298123 M * Doener hm, not sure if this term (barebone) is really valid in this context... 1115298151 M * Pazzo now I have no more excuse not to use sed ;-p 1115298163 M * Doener anyway, what i mean is 'you just got the absolutely essential stuff to run exactly what you need in this vserver' 1115298232 M * terr Just figured out: Pazzo == Tomas Gelf (These nicknames can be annoying: terr = Gilles on th ML) 1115298275 M * Doener if you just f.e. use your vserver just to run an apache, you don't need much things in there... and package management is one of those things you can do from the host if you want to thus keeping your vserver smaller 1115298320 M * Pazzo hehe - hi Gilles! 1115298343 M * terr Excuse me: Which distribution(s) are you using? 1115298344 M * Pazzo maybe I should switch my nick ;-) 1115298355 M * terr Hi. 1115298362 M * Pazzo terr: debian. Debian. and DEBIAN ;-) 1115298372 M * terr And Doener? 1115298413 M * Doener host: Gentoo, vservers: debian... recently i setup a fc3 test vserver... but primarily to see how it works (a vserver build with apt-rpm that is) 1115298470 A * Doener only got a single box, thus only a few vservers... 1115298501 M * terr So, to answer my question to the ML, could you briefly comment on the difficulties/advantages of external vs internal management? 1115298563 M * Doener internal: vserver is self-contained, you can do everything from within the vserver... 1115298599 J * pflanze ~chris@unk-110.ethz.ch 1115298600 M * terr I would assume (not knowing much about it) that the Debian people won't take vserver into account when devising their management procedure... 1115298611 M * Doener external: smaller vservers, possibly shared package caches... 1115298620 M * pflanze Hello. 1115298626 M * Pazzo hi pflanze! 1115298644 M * terr Hence the difficulties to keep state-of-the-art for the given distribution. 1115298645 M * Doener i didn't spent that much thought on this stuff yet... enrico probably knows a lot better than me 1115298680 M * Doener terr: dpkg already allows to specify a different state directory and installation base directory... 1115298691 M * pflanze Has anyone of you ever used, or heard of, possibilities to log tty's, depending on the context id / only of the host context ? 1115298693 M * terr I like "vserver is self-contained" 1115298701 M * Pazzo Doener: these are things always depending on the specific situation - and we should provide the possibility to do both: internal and external package management 1115298738 M * Doener Pazzo: exactly... and for rh/fc util-vserver already does this... but for debian, there's just debootstrap, i.e. internal package-management... 1115298752 M * pflanze (s/tty/pts/) 1115298753 M * Doener i don't intend to replace that, just to _add_ external pm 1115298774 M * Pazzo For example I'm absolutely not interested in external package management - why should I? Debian's package management is great, I won't miss it ... 1115298792 M * Pazzo ...a shared cache doesn't interest me as I mantain my personal mirrors... 1115298798 Q * eyck Read error: Operation timed out 1115298807 M * Pazzo ...and if I wouldn't I would use an apt-proxy 1115298901 M * Pazzo apt-get clean creates free space - but I don't care about space as hard disks are too cheap today ;-) 1115298941 M * Pazzo but that's my case - and we have to provide a solution for everybody and his dog, right? 1115299014 M * terr If you (Doener) need it (external) fine. But I think that providing a distribution-specific method that create a flawless vserver (i.e. working, self-contained, within the policy of the contained distribution) 1115299017 M * Pazzo how should external package management work with debian? 1115299026 M * Pazzo what about /var/cache/debconf/ ? 1115299122 M * Pazzo and what about /var/lib/dpkg/ ? 1115299152 M * Pazzo I have absolutely no idea how this should work - maybe it's possible and really easy - but don't ask me how! 1115299170 M * Doener 15:11:19 Doener terr: dpkg already allows to specify a different state directory and installation base directory... [ for Pazzo this time ;) ] 1115299190 M * Doener (where /var/lib/dpkg is the state directory... ) 1115299198 M * Pazzo :) sorry Doener, didn't see this! 1115299232 M * Pazzo ok, so implementing this would be probably very easy 1115299249 M * Pazzo Doener: and debconf? 1115299258 M * Pazzo -> Pazzo what about /var/cache/debconf/ ? 1115299285 M * Pazzo can this safely be ignored? 1115299336 M * Doener i'm still figuring out the details... ;) 1115299347 M * Pazzo sorry, I half to leave for something like half an hour - keep on discussing this! 1115299354 N * Pazzo PazZzzzooo 1115299368 M * terr Someone (Bjorn I think) on the ML implied that external package might induce conflicting tools. No? 1115299382 M * Doener Bjoern = me 1115299396 M * PazZzzzooo :-) 1115299398 M * PazZzzzooo cya 1115299427 M * terr OK! Then so you know what you said ;-) Did I understand correctly? 1115299482 M * Doener external package management only works if you got the used package management tool on your host 1115299511 M * Doener and for a debian host there will be a name clash with apt-rpm, as both use 'apt-get' 1115299538 M * Doener thus this is at least harder to get working... AFAIK noone tried to get this working yet 1115299584 M * Doener external package management doesn't induce conflicts, it simply doesn't work because of conflicts ;) 1115299599 M * terr So is this only to save (disk) space? 1115299631 M * terr How much do you expect to be saved? 1115299674 M * Doener not much, and as i said i didn't spent much thought on it yet... enrico probably knows more... 1115299711 M * Doener (as all the existing rpm [i.e. rpm, apt-rpm and yum] vserver build methods use external packagemanagement) 1115299767 M * terr Sorry for being so slow (maybe), but what is the advantage if not disk space?? 1115299840 M * terr It seems to me that it doesn't fit with the vserver philosophy somehow as a self-contained world. 1115299886 M * terr I.e. trying to install it differently than if it was the host system... 1115300002 M * Doener well... if you just want to chroot your mail server, do you provide it a full blown system, or just that stuff that is absolutely needed? 1115300087 M * Doener now the Linux-VServer stuff can be seen as an extended chroot... you don't only chroot into that minimal "system", but also chbind, add some other limitations... 1115300103 J * Fibbs ~chris@p83.129.207.131.tisdip.tiscali.de 1115300111 M * Fibbs Good morning everybody1 1115300161 M * terr Why is there cause-effect reltionship between "internal" and "full-blown" ? 1115300182 M * terr Good afternoon! 1115300222 M * Fibbs Can anybody tell me why i have files that i can't access from inside a vserver? They have no attributes (lsattr ) and are root:root 540, for example it is the apache2.conf and /etc/hosts. 1115300286 M * daniel_hozac do you use xid tagging? 1115300301 M * Fibbs daniel_hozac: I did, but i'm actually not anymore using them 1115300313 M * Fibbs But possibly there is still something active, that could be. 1115300330 M * terr Most important (in my view) is to provide a debootstrap way to create a minimal *minimal* vserver setup. 1115300333 M * Fibbs ufff. how i will have to read how to deactivate... it's long time ago 1115300403 M * terr So that root can log in and install whatever (mail, apache, ...) is required. How does that make it full-blown? 1115300442 M * Doener no, i've been exaggerating there ;) 1115300494 M * Doener but if you want to just have your mail server there, you don't even need a way to login there... 1115300566 M * terr Oh, so that's *the* advantage of external management (?) 1115300584 M * Doener that's one thing i can up with at the moment... 1115300611 M * Doener i don't say "Hey, use it, it will solve all your problems, fill your bank account and make you look like Elvis!" 1115300629 M * Fibbs hey sorry, but can anyone tell me how to fastly deactivate all set context disc limits? 1115300644 M * terr Note that I don't want to criticize "external" :-} 1115300663 M * Doener i've just seen that this is possible with the rpm stuff and that it is possibly doable with debian, too. 1115300691 M * terr It's just that I've had problems with "internal"... 1115300693 M * Doener then i've thought that there has to be something about it and that others may find it useful 1115300732 M * terr ... already (stuff that should have been disabled and all that) 1115300789 M * terr ... So that I think the "internal" kind should be functional and easy before trying to do something that's more complicated. 1115300808 M * Doener hm... internal works without problems here.... 1115300856 M * terr So, you didn't have to remove some "init.d" links? 1115300896 M * terr Like mounting filesystem, activating swap, hardware clock... 1115300905 M * Doener that's not packagemanagement related ;) that's just the way debootstrap builds the system... it does it like it would do on a real host... 1115300929 M * Doener and yes i had to remove those... 1115300979 M * terr But THAT's what I mean: The vserver should be ready to work (built in a sensible way knowing it is a vserver). 1115301008 M * terr THEN, you can start using it as if it was a normal system... 1115301058 M * terr ATM that's not the case with debian guests, which I find more annoying than not having yet another way to install software. 1115301421 Q * albeiro Ping timeout: 480 seconds 1115301551 J * albeiro albeiro@albeiro.usercloak.oftc.net 1115301564 M * terr Talking about something else: Is there a definitive/authoritative guide on how to make vserver work on Gentoo? [As I've read there were problems.] 1115301604 M * Doener Gentoo as host? no problem... Gentoo as guest? no idea 1115301627 M * terr Gentoo host, on AMD64. 1115301732 M * maharaja whats the problem if the light on my mobo is working, the fans are spinning - it beeps if there is no ram plugged in 1115301757 M * maharaja but there is no output from the onboard graphics card 1115301861 Q * Fibbs Quit: Verlassend 1115302463 N * PazZzzzooo Pazzo 1115302464 M * Pazzo re 1115302823 M * wurd gle.ca/-:- SignOff `Sk8t3r_boy`: #blackmetal (Ping timeout) 1115302835 M * wurd um. didnt mean to paste that 1115302924 M * Pazzo terr: a base fully-working debian vserver (without ssh, installed as in my mail on the ml) "weights" about 92M 1115302946 M * Pazzo contains a base sarge & a working package management 1115303013 M * Pazzo maharaja: who cares about the graphic card - today blinking, spinning & beeping is enough to be cool ;-) 1115303133 M * Pazzo maharaja: let mb, cpu and ram be the only parts in your system, put all other cables, disks, floppy & co away ... 1115303158 M * Pazzo ...to see if it you are able to recieve a signal (bios screen) then 1115303177 M * Pazzo if yes: plug in step by step one device after the other 1115303205 M * Pazzo if no: cpu or mb problem - try to change cpu, if it doesn't help through away your mb 1115303311 M * Pazzo maharaja: did this system ever work? or is it completely new hardware? 1115304056 M * wurd Doener are you still there ? 1115304913 J * openMosixview ~matt@p50885CB0.dip.t-dialin.net 1115305327 Q * BWare Read error: Operation timed out 1115305584 J * BWare ~bware@office.intouch.net 1115308540 M * maharaja Pazzo: the mobo is new 1115308553 M * maharaja and the power supply is a different one but it should work 1115308555 M * maharaja no cables attached 1115308743 Q * Doener Quit: Leaving 1115308955 Q * openMosixview Quit: Client Exiting 1115309108 J * Doener ~doener@p54875D6D.dip.t-dialin.net 1115309754 M * torisa .away 1115309760 M * torisa oops 1115311206 N * Bertl_zZ Bertl 1115311218 M * Bertl morning folks! 1115311229 M * Bertl torisa: hmm, away? 1115311370 M * torisa not anymroe :) 1115311374 M * torisa how are you? 1115311443 M * Bertl fine thanks! and you? 1115311474 M * torisa okay 1115311518 M * torisa i get to start my new job today. i'm excited 1115311567 M * Bertl sounds good, what is your position there? 1115311586 M * Bertl (i.e. what do they expect you to do ;) 1115311601 M * torisa coding PEAR/PHP for a university library 1115311620 M * terr Hello Herbert... and others. 1115311626 M * torisa hi terr 1115311629 M * Bertl hey terr! 1115311670 M * Bertl ready to continue where we left of yesterday? 1115311683 M * terr Hopefully, I have now an acceptable setup ;-} 1115311707 M * Bertl excellent! did you already try with the mainline tools? 1115311747 M * terr Yes. But same strangeness. 1115311794 M * terr I've deinstalled the debian package, 1115311806 M * terr compiled 0.30.207, 1115311816 M * Pazzo hi Bertl! 1115311907 M * terr and installed everything under "/usr/local/u-vs/etc" "/usr/local/u-vs/usr" and "/usr/local/u-vs/vservers" ... 1115311912 M * Pazzo terr: if you need it, there is 0.30.207 in sid - and now it's installable ;-) 1115311950 M * terr Pazzo: Ask Bertl ;-/ ... 1115311965 J * mep_ mep@p5091EAC3.dip.t-dialin.net 1115311990 M * Pazzo Bertl: tell me! =;-p 1115312039 M * terr ... so that removing "/usr/local/u-vs/" you get something that suits you better: "/etc", "/usr" and "/vservers" 1115312104 M * terr I've also re-created the problematic vserver. 1115312135 M * terr But by using the "debootstrap" build method... 1115312204 A * Bertl is reading up now ... 1115312219 M * terr ... so you can still blame Debian ;-) 1115312245 M * Bertl terr: no, debootstrap is fine ... 1115312275 M * Bertl (and just for the record, I'm not blaming debian) 1115312308 M * Pazzo *gg* 1115312313 M * Bertl http://vserver.13thfloor.at/Experimental/vshelper-debug.diff 1115312327 M * Bertl terr: please apply this to your vshelper 1115312346 M * Bertl then get the system into the 'working' state 1115312376 M * Pazzo Bertl, terr: what are you trying to fix? as it seems to be debian-related I'm very curious! 1115312404 Q * mep__ Ping timeout: 480 seconds 1115312407 M * Bertl start _and_ stop the vserver .. save the information of vps and the files created in /tmp/ starting with a pid and having two _underscores ... 1115312455 M * Bertl Pazzo: it seems that terr somehow kills his host-init process by stopping a vserver guest 1115312474 M * terr More slowly please. 1115312483 M * Bertl terr: did already ask you to upload the guest somewhere (maybe as dump or tar?) 1115312496 M * terr No. 1115312526 M * terr Should I apply the patch then recompile all util-verser? 1115312544 M * Bertl no, this patch is just against the vshelper file 1115312560 M * terr Ah, OK. 1115312568 M * Bertl so with the 0.30.207 tools installed, do 1115312579 M * Bertl patch -p0 t :-) 1115317488 M * Bertl lol, maybe you should not do that by default ... 1115317495 M * FaUl ack 1115317497 M * Bertl and after all, it's just your fault ;) 1115317505 M * FaUl *g* 1115317536 M * Pazzo for german-speakers: one of my italian teachers always told me: "Thomas, tu sei zu stink zu faul!" 1115317550 M * FaUl hehe 1115317587 M * FaUl Bertl: i wonder why you implemented ip-restrictions by addresses, not by interfaces? that would result in less troubles and it would enable ipv6 without much work 1115317619 M * Bertl well, the linux kernel is not based on interfaces ... it is based on IPs 1115317657 M * Pazzo FaUl: and maybe someone would like to share interfaces between vservers 1115317658 M * Bertl of course you could 'restrict' some things to interfaces, but that would certainly give you a bunch of suprises ... 1115317689 M * FaUl ah, ok, then i'll don't try to patch this, thx ;-) 1115317795 M * FaUl you said on the website that the ipv6-support is mostly stupid cut&paste-work, but i am not quite sure what has to be done - is there somebody who tries to implement that already? or is there simply noone other then me who needs ipv6-support? 1115317820 M * Bertl you are using ngnet? 1115317837 M * Bertl do you need ipv6 on the host or with the guests too? 1115317872 M * FaUl on the guests - i want to implement a shell-vserver for irc and things, and without v6 it sucks ;) 1115317877 M * Bertl FaUl: btw, this might be interesting for you: http://archives.linux-vserver.org/200311/0470.html 1115317890 M * Bertl (regarding interfaces ;) 1115318007 M * FaUl Bertl: already read that 1115318010 M * FaUl but thx 1115318034 M * Bertl maybe you should read it again, if you want restrictions based on interfaces ;) 1115318219 M * FaUl *read* 1115318411 M * FaUl hmm, i'm not quite sure if i understood that correctly - why just let connect chose (with his magic) from the list of configured ips? 1115318416 M * FaUl configured on the vserver? 1115318488 M * Bertl well, it will choose, you just have to _know_ the magic ;) 1115318512 M * Bertl what you want is more what ngnet does (i.e. virtualized interfaces and routing) 1115318722 M * FaUl but ngnet is only avilable for linux 2.6, isn't it? 1115318738 M * Bertl yep, correct ... 1115318777 M * FaUl well, i don't want 2.6 on the server because it isn't stable with ipv6, vlans and ethernet-bonding 1115318848 M * Bertl well, you can start backporting ngnet then .. should not be too 2.6 specific ... 1115319113 M * Pazzo Bertl: what does /usr/lib/util-vserver/vshelper-sync do? 1115319136 M * Pazzo seems to be new and it happens that I recieve a "A timeout or other error occured while waiting for the synchronization" 1115319158 M * Pazzo "The vserver will be killed nevertheless..." - error 1115319163 M * Bertl it synchronizes certain things like vserver shutdown ... 1115319220 M * Pazzo what kind of synchronization is to be done at vserver shutdown? 1115319227 M * wurd after doing this: vserver fc3-base pkg install apt , i get the following error : /bin/chown: cannot access `/var/lib/rpm/[A-Z]*': No such file or directory 1115319227 M * wurd Done. 1115319351 M * Bertl sounds like a bug I discovered some time ago (with the internalization fo the package management) 1115319381 M * Bertl IIRC, enrico did provide a solution ... *looking now* 1115319692 M * FaUl Bertl: but as far as i've seen there isn't any ipv6-support in ngnet either, is it? 1115319722 M * Bertl no, but it's pretty easy to add there ... 1115319733 M * Bertl (because there are no IP related checks at all) 1115319782 M * Bertl I already stated, that if somebody is willing to test ipv6 on ngnet, I do the implementation ;) 1115319844 M * Pazzo re 1115319864 M * Pazzo uname -a: 2.6.11.8-vs2.0-pre4+vserver 1115319906 M * FaUl Bertl: so if i get 2.6.11.* with vs2.0-pre4 up and running, you will do the implementation? 1115319924 M * Bertl for ngnet on 2.6 (ipv6) yes ... 1115319926 M * Pazzo I'm not very convinced by this sync stuff - what exactly does it do? 1115319933 M * Bertl (but you have to do the testing) 1115319949 M * FaUl mh, i would do my very best in testing ;) 1115319967 M * Bertl Pazzo: you ahve to ask enrico or read the source code ... 1115320005 M * FaUl but beware, i'm not full-time pc-worker but pupil 1115320018 M * Bertl wurd: 1115320018 M * Bertl oh, I see. /etc/rpm/macros contains still the 1115320019 M * Bertl | %_dbpath /.rpmdb 1115320019 M * Bertl Removal of this line should fix it. 1115320029 M * Bertl (from enrico) 1115320106 M * wurd well, in /etc/rpm there are 4 files beginning with the word "macros" 1115320116 M * wurd but none of them names are just "macros" 1115320122 M * wurd it's all macros. 1115320130 M * Pazzo hmmm... thnx Bertl 1115320145 M * wurd = jpackage, prelink, specspo & up2date . 1115320245 M * Bertl you are loogin in the guest I hope? 1115320253 M * Bertl s/looging/looking/ 1115320266 M * wurd what do you mean in the gues ? 1115320291 M * Doener wurd: that message is ok... 1115320308 M * Doener and the macros file will appear _after_ you internalized the package management... 1115320321 M * wurd Done. 1115320330 M * wurd /bin/chown: cannot access `/var/lib/rpm/[A-Z]*': No such file or directory 1115320330 M * Doener (and the document even tells you that the above line has to be removed...) 1115320333 M * wurd this is ok ? 1115320339 M * Doener 21:11:31 Doener wurd: that message is ok... 1115320344 M * wurd ok thanks 1115320351 M * FaUl Bertl: is there somthing more then applying patch-2.6.11.8-vs2.0-pre4.diff on the kernel to enable ngnet? 1115320394 M * Doener afk again... 1115320404 M * Bertl FaUl: yes, the ngnet patch itself ;) 1115320412 M * wurd by the way doener, what did you mean by "also all steps are meant to be done from the start" 1115320418 M * FaUl Bertl: where is it? 1115320419 M * wurd does this mean i have to delete all existing vservers 1115320425 M * wurd or does it mean i simply have to build a new one 1115320437 M * Doener the latter... 1115320453 M * Pazzo How can I find out why I'm not able to cleanly shut down my vservers? 1115320456 M * Pazzo Vserver '/etc/vservers/test21' still running unexpectedly; please investigate it manually.. 1115320472 M * Doener otherwise you might have to do some other steps to get the vserver into the 'right' state to continue... 1115320475 M * wurd i don't understand why my build didnt work, because i followed all your steps correctly, and didnt break any rules you gave me in the message you wrote today... 1115320483 M * FaUl ah, in ngnet-directory obvously ;) 1115320488 M * Pazzo hmmm... with last versions they have been killed - now they are still in vserver-stat 1115320512 M * Pazzo all processes 1115320542 M * Bertl FaUl: yeah, but let me ease your pain by updating 9.4 to the latest kernel ... 1115320548 M * Bertl (and uploading a patch ;) 1115320551 M * wurd just to be 100% sure, the /etc/vservers/.distributions/fc3/apt/sources.list file contains the following line uncommented?:rpm http://sunsite.informatik.rwth-aachen.de/ftp/pub/Linux/fedora/fedora 3/i386 os updates extras 1115320553 M * Pazzo any idea? 1115320577 M * Bertl Pazzo: sounds strange, let's investigate ... 1115320593 M * FaUl hmm, not 2.0rc4? i've allready applied that ;) 1115320597 M * wurd i believe this file doesnt affect my apt in any way, so i dont understand why you told me to verify this file, while the error i get is concerning apt 1115320615 M * Bertl FaUl: 2.0-pre4 is fine .. the ngnet is not up to date ... 1115320626 M * Pazzo Bertl: kernel 2.6.11.8-vs2.0-pre4, utils .207 - with 1.9.5.5.x everything seemed to be fine 1115320626 M * FaUl ah,ic 1115320650 M * Bertl Pazzo: the way helpers work have changed to eliminate some races 1115320651 M * FaUl seems that only tcp_minisocks needs updating... 1115320694 M * Pazzo testme.sh-0.12 runs fine 1115320727 M * Pazzo Bertl: where should I start investigating? 1115320754 M * FaUl Bertl: i think i can apply that patch by my own, you don't have to do that work 1115320769 M * Pazzo btw: I installed 3 new vservers with my enhanced debootstrap, went fine 1115320787 M * Bertl FaUl: okay, go ahead, we compare the results later 1115320805 M * Pazzo I was installing first one single vs, then two of them at the same time - went perfectly 1115320840 M * Pazzo I started one single new vs, then two of them together - no problem 1115320855 M * Pazzo stopping one of them failed - stopping another one also 1115320880 M * Pazzo vservers are still there, I can enter them and see only defunct processes 1115320892 M * Bertl it's probably the same issue terr and romke observed 1115320900 M * Bertl (just slightly different effects) 1115320924 M * Pazzo PROC 9, VSS 0, VSZ 0 for all of them 1115320937 M * Bertl so still processes running there? 1115320942 M * Bertl try to vkill them 1115321030 M * Pazzo tried signal 15 and 9 (vkill --xid 120 -s 15) 1115321034 M * Pazzo no way 1115321052 M * Bertl are those zombies? 1115321074 M * Pazzo yes 1115321077 M * Bertl is your vserver init based? 1115321088 M * Pazzo running top: Tasks: 11 total, 1 running, 1 sleeping, 0 stopped, 9 zombie 1115321098 M * Pazzo initstyle plain 1115321127 M * Bertl Doener: guess here we have an example for missing vserver reparenting .. no? 1115321202 M * Pazzo starting another vs - running fine, inside it everything seems ok, now I'll try a shutdown from inside... 1115321261 M * Pazzo [S90halt] 1115321275 M * Pazzo the others have [S90reboot] 1115321282 M * Pazzo but same thing 1115321299 M * Pazzo hey... 1115321309 M * Pazzo after doing halt I re-entered the vserver 1115321335 M * Pazzo vserver-stat still showed me a value for VSZ/RSS 1115321385 M * Pazzo after some seconds it wrote "Killed" to the konsole (this was not the konsole where I wrote halt) - and I was thrown out of the vserver 1115321399 M * Pazzo VSZ/RSS ar now both 0 1115321403 M * Bertl okay, could you wrap up a minimal guest too (which has this behaviour) I'm going to investigate this in half an hour or so ... 1115321412 M * Pazzo (and I can re-enter the vs 1115321426 M * wurd apt-get groupinstall base 1115321426 M * wurd E: Invalid operation groupinstall 1115321434 M * Pazzo what do you mean by "wrap up a minimal guest"? 1115321449 M * Bertl (with three vserver guests showing similar behaviour I should be able to locate and fix it) 1115321449 M * wurd IIRC you found the solution for this yesterday Bertl ? 1115321480 M * Pazzo this have been 3 ones... 1115321491 M * Bertl wurd: never used groupinstall ... 1115321499 M * wurd oh 1115321514 M * wurd my memory has errors then 1115321514 M * Pazzo ...other two debootstrap installations are running right now 1115321522 M * Bertl Pazzo: yes, but I have others too ... so take the smalelst one, make a tar.bz2 and upload it somewhere ;) 1115321531 M * FaUl Bertl: http://do.bundessicherheitsministerium.de/diff-2.6.11.8-vs2.0-pre4-ng9.0.diff 1115321552 M * Bertl hmm, close but no banana, we are at ng9.4 ;) 1115321568 M * Pazzo Bertl: ok, so you need a .tar.gz of a freshly installed guest system, right? 1115321569 M * FaUl fuck! 1115321582 M * Bertl FaUl: but there should be deltas 1115321657 M * FaUl ah 1115321746 M * FaUl ok, just have to build that damn diff again :) 1115321850 J * knoppix_ ~knoppix@dsl-213-023-129-109.arcor-ip.net 1115321875 M * Bertl welcome knoppix_! 1115321883 M * FaUl *wait* 1115321966 M * Bertl btw, did you test compile your patch? 1115322013 M * FaUl not yet, wait 1115322095 M * Pazzo Bertl: dsl upload is painful :-( 1115322168 M * Bertl don't tell me, I'm on ISDN ;) 1115322178 M * Pazzo md5sum: 4d9edd34de63234aece0ff458eb21d3e, size: 31637481 - should be finished in something like 30 minutes 1115322548 M * Pazzo Bertl: searching the list I discovered that Gilles (terr) seems to have the same problem - is it this what you have been talking about before? 1115322869 Q * terr jupiter.oftc.net kinetic.oftc.net 1115322993 M * Pazzo more than 50% finished, I have to leave now but the link should be ok - if you have ISDN you can already start downloading :-) 1115323120 M * Pazzo within max 15 minutes upload should be finished - cya Bertl & thank you! 1115323365 M * Bertl okay, cya! 1115323401 M * Bertl Pazzo: yep, gilles == terr 1115323766 M * wurd Doener please say something if youre there ? 1115323768 M * wurd ;) 1115323784 M * wurd Bertl what do you think this means ? 1115323786 M * wurd apt-get update 1115323786 M * wurd (enter: 1115323786 M * wurd 1,2 1115323786 M * wurd c 1115323786 M * wurd 3 1115323786 M * wurd c 1115323786 M * wurd 3 1115323788 M * wurd c 1115323790 M * wurd )...? 1115323832 M * wurd http://www.13thfloor.at/~doener/wurd 1115323833 M * Bertl as I stated yesterday, for me this is the input to the questions ... 1115323839 M * wurd yeah i guess 1115323843 M * wurd but i receive no questions.. 1115323848 M * wurd no questions are asked 1115323888 M * Bertl probably you are doing something different than Doener! 1115323900 M * wurd blah 1115323902 M * wurd :( 1115323964 J * prae ~prae@sherpadown.net 1115323977 M * wurd earlier today he told me that : 1115323978 M * wurd wurd: those steps i wrote down for you were meant to be done with a pristine 1115323978 M * wurd sources.list for 1115323978 M * wurd fc3 as supplied by the tools. also all steps are meant to be done from the start (i.e. if you 1115323978 M * wurd already built a vserver, you cannot use it for this...). And finally: I told you that the 1115323982 M * wurd tools would tell you when to start/stop the vserver and that i thus left those steps out... 1115324011 M * wurd but really, that doesnt help me much 1115324019 M * daniel_hozac wurd: you wanted an FC3 guest with X, right? 1115324024 M * wurd yes... 1115324037 M * wurd heres what Doener told me to do daniel_hozac : http://www.13thfloor.at/~doener/wurd 1115324058 M * daniel_hozac wurd: i know, i've read that. 1115324059 M * wurd i really dont understand the part just below "apt-get update" .. at the bottom of the page... the part that starts with (enter: 1115324062 M * Bertl well, then let's do that ... 1115324090 M * daniel_hozac i'm not really familiar with apt, but with yum, this should be very simple. 1115324092 M * wurd Bertl do what ? 1115324106 M * Pazzo Bertl: all days the same story :-( I want to go home, order a pizza by phone, prepare me to start, get a phone call - and will for sure be blocked unless the pizza is cold... 1115324116 M * Bertl wurd: did you uncomment the sunsite.informatik.rwth-aachen.de entry as only one? 1115324126 M * Pazzo ...upload is finished, md5sum is ok, have phun with my vserver! 1115324133 M * Bertl thanks! 1115324142 M * wurd what do you mean "as only one" ? ... it is the only uncommented line in the whole file, yes 1115324168 M * Bertl excellent ... 1115324187 N * Pazzo PazZzzzzooo 1115324192 M * Bertl now make sure that you do not already have a vserver called fc3-base 1115324200 M * wurd as for the 'apt-get update' command, it only works if i modify this file: /etc/apt/sources.list . because the lines that are there by default are "ayo.freshrpms.net" and doener told me not to use that 1115324209 M * wurd so i put 'sunsite.informatik' instead. 1115324226 M * wurd i dont know if its right though 1115324231 M * wurd he didnt answer me on that yet 1115324240 M * wurd but its the only solution i found 1115324275 M * wurd the 'apt-get update' command DID work properly, using 'informatik' .... BUT i do not get asked the questions that would be answered by 1,2,c,3,c etc. 1115324309 M * wurd now make sure that you do not already have a vserver called fc3-base >>> yeah thats fine i've just built a new one with a different name 1115324316 Q * PazZzzzzooo Quit: maybe my workstation would like to fell asleep too ;-) have a good night! 1115324328 M * daniel_hozac wurd: you're running apt-get update inside the vserver, right? 1115324393 M * wurd no i'm not.... :( i should be ? 1115324446 M * daniel_hozac i would assume so. 1115324475 M * wurd i'll try.. 1115324515 M * wurd blah 1115324520 M * wurd Fetching list of repositories/mirrors... 1115324521 M * wurd E: Failed to fetch mirror list file: 1115324539 M * daniel_hozac can you resolve hostnames in the vserver (i.e. did you setup /etc/resolv.conf)? 1115324565 M * wurd youre right.. no i didnt since its a freshly built vserver 1115324570 M * wurd i dont even have "vi" nor "vim" 1115324669 M * wurd thanks for your help everyone.. i gotta go :/ 1115324685 M * wurd if somebody finds the solution while i'm gone, please note it down, thanks !!! 1115324697 M * wurd (or simply email it to me: csdlaval@hotmail.com) 1115324720 M * wurd have a good evening 1115324729 Q * wurd Quit: a 1115324781 M * Bertl hmm .. interesting ... 1115325084 M * FaUl Bertl: ah, does not compile because VS_CONTEXT_CREATED is undeclared in kernel/vserver/context.c, which file should i include to fix that? 1115325315 M * FaUl hmm VS_CONTEXT_CREATED is only mentioned in kernel/vserver/context.c :) 1115325430 M * FaUl ah 1115325616 M * Bertl not that simple, eh? 1115325623 M * FaUl just fixing ;) 1115325717 J * terr ~nobody@ip-213-49-106-244.dsl.scarlet.be 1115325719 Q * terr Server closed connection 1115326121 M * FaUl well, i've patched so much things around, i guess your patch will look different ;) 1115326227 M * Bertl http://vserver.13thfloor.at/Experimental/NGNET/diff-vs2.0-pre4-ng9.5.diff 1115326233 M * Bertl check it for differences ... 1115326257 M * FaUl ng9.5? 1115326265 M * Bertl (best done by applying my patch to a new tree, then doing a diff -NurpP between them ... 1115326293 M * Bertl updated the version number, as I had to change a few things ;) 1115326343 M * FaUl at least it seems that my patch compile so far ;) 1115326355 M * Bertl mine too, mine too ;) 1115326580 M * FaUl apply what against what? applying your patch agaainst a 2.6.11.8 with vs2.0-pre4 says that Reversed (or previously applied) patch detected! Assume -R? [n] 1115326613 M * FaUl first time ath 8390.c 1115326617 M * FaUl -h 1115326700 M * Bertl that's at least strange .. let me verify that ... 1115326740 M * FaUl mh, seems that that #if 0 ... #endif-foo is already in my version of 8390.c 1115326763 M * Bertl well, it's not in mine of vs2.0-pre4 ;) 1115326768 M * FaUl but i unpacked a clean version of 2.6.11.8 and only applied that vs-2.0-pre4 1115326772 M * FaUl thats strange 1115326775 M * FaUl i'll try again 1115327121 M * FaUl i guess i know what happened 1115327140 M * FaUl i tried to apply that patch without applying the vserver-patch before 1115327142 Q * jm Remote host closed the connection 1115327259 M * FaUl so i guess it applied a few patches before stopping 1115327321 M * FaUl it applied cleanly now 1115327567 M * Doener back... 1115328066 M * FaUl Bertl: ok, you won, your patch is different ;-) 1115328146 M * Bertl well, it's not about different ... could you upload the differences? 1115328188 M * FaUl narf, if you said that 2 seconds before i hadn't removed the other stuff 1115328210 M * Doener well, never delete anything ;) 1115328225 M * FaUl any way to recover trees on an ext3? :) 1115328227 A * Doener has to re-fetch init sources atm... :/ 1115328248 M * Bertl FaUl: nope ext3 is final ;) 1115328274 M * FaUl Doener: if you have only a 60gb-hard-disk, your mind will change ;) 1115328296 M * Doener i have 2 80GB disks as raid-1 array... not so much more space.. 1115328303 M * FaUl but i've deleted the config for the server, fsck 1115328316 M * Bertl $ cat /proc/ide/hda/capacity 1115328317 M * Bertl 19541088 1115328319 M * Doener and still abount 30GB free... 1115328330 M * Doener s/abount/about/ 1115328373 M * FaUl Bertl: looks like my server, but there it is /proc/scsi :{) 1115328426 M * FaUl ls 1115328428 M * FaUl oops 1115329388 M * FaUl Bertl: which options should i enable in vserver-config? ext-gen networking, nfsd user auth proc security? 1115329409 M * FaUl have to go, bye 1115329415 M * Bertl k, cya! 1115329456 M * FaUl ill read backlog, so if you have desired options, just write it or query me 1115329546 M * Bertl k 1115329653 M * Bertl FaUl: keep as close as possible to the example configs in the NGNET dir .. I'll upload mine for the vs2.0-pre4 1115329667 P * Doener Leaving 1115329672 J * Doener ~doener@p54875D6D.dip.t-dialin.net 1115329691 M * Doener and once more i hit ^w to delete a word... 1115329717 M * Bertl hmm, guess I should not try that on irssi now ... 1115329721 M * Doener (which means close channel tab in xchat...) 1115329734 M * Bertl ah, back to xchat ;) 1115329782 M * Doener yep, i finally abadoned my bnc and xchat already had configurations for my servers/channels without bnc ;) 1115329823 M * Bertl so not happy with irssi? 1115329874 M * Doener no, just didn't have the time to configure it now... and xchat was just there 1115329885 M * Doener (i.e. configured) 1115329890 M * Bertl k, just curious ;) 1115330036 M * Doener what i missed was a) nick list (i've seen that this is possible with irssi), b) channel switching probably needs some time to get used to (can i get a channel list, too?) c) mouse scrolling!... for just reading, i don't like using the keyboard to scroll up and down... 1115330090 M * aba Doener: You get the next active channel with Alt+A, and the first 19 channels with Alt+1 ... Alt+9 Alt+Q .. Alt+o 1115330103 M * Bertl ad b), the tabs have shortcuts ... 1115330137 M * Doener aba: yep... but i don't remember which channel is which ;) and i often don't want to go to every active one... 1115330139 M * Bertl yup, and a) what do you have in mind there? 1115330151 M * aba Doener: you're getting used to that :) 1115330172 M * Bertl Doener: you can 'sort' the channels to your preference 1115330253 M * Doener i know ;) i even found out that i can use split windows (i like that!)... but i'm not good at remembering such stuff in short time... 1115330375 M * Doener ad a) http://angelic.flexnet.org/angelic.png -- that list on the right... not that i'd need it (i guess), but it will make the transition easier ;) 1115330550 M * Bertl http://wouter.coekaerts.be/site/irssi/nicklist maybe? 1115330739 M * Doener yep, looks like it :) 1115331614 Q * sebd Remote host closed the connection 1115332033 N * ciphernautZZ ciphernaut 1115332180 M * mugwump to_tz('morning') all 1115332182 M * Doener morning ciphernaut 1115332192 M * Doener morning mugwump 1115332212 M * ciphernaut hi all 1115332268 M * mugwump I've written one more message to enrico about the svn thing, but I don't really have the energy to try to convince someone that they should be an *open* source developer rather than a closed open source developer 1115332271 M * Doener Bertl: do you have any experience with arch/monotone or can explain what's the difference in svn's and arch's changesets? 1115332289 M * Doener or mugwump, can you? ;) 1115332308 M * mugwump ok, the idea is with arch, bitkeeper, monotone, darcs, etc that the code base is change-set based 1115332317 M * mugwump ie, it could be represented by a series of patches. 1115332324 M * Doener i always thought that what i get with svn diff -r 1:2 basically is a changeset... 1115332454 M * mugwump yes, it is. the thing svn doesn't do is remember that you merged a given changeset when you use `svn merge' 1115332479 M * mugwump however, svk layers that stuff on top of svn 1115332640 M * mugwump http://superbeast.ucsd.edu/~landry/ArX/codecon/img22.html 1115332698 M * mugwump That guy is really scraping the barrel with those "cons" :) 1115332865 Q * knoppix_ Quit: Verlassend 1115332926 M * mugwump I love that one "Written in Perl, so gets no respect" 1115332932 M * Doener yep, me too :) 1115333082 J * terr ~nobody@ip-213-49-106-244.dsl.scarlet.be 1115333104 M * Bertl wb terr! 1115333112 M * terr Hi. 1115333136 M * terr I just sent you a mail, 1115333180 M * terr because my IRC client kept trying to connect to server which was brought down for maintenance. 1115333211 M * Bertl ah, i.c. 1115333222 M * terr Anyway... Did you find something new? 1115333238 M * terr I saw other people might encounter similar problems. 1115333240 M * Bertl not yet, but I'm sure I will tonight ... 1115333317 M * terr Something I can do? 1115333359 M * Bertl yeah, you could upload your config for that guest too ... 1115333372 A * Bertl forgot about that ... 1115333463 M * terr It should be included (the "./etc" directory, you mean?) 1115333489 M * Bertl hmm, nope the /etc/vservers/ dir ... 1115333567 M * terr Yes, indeed: It's in there: "./etc/vservers/template/" 1115333591 M * Bertl ah, okay ... didn't notice yet ... 1115333600 M * terr But I've just seen that there are absolute symbolic links... 1115333612 M * Bertl that's probably fine though ... 1115333623 M * terr E.g.: ./etc/vservers/template/run -> /usr/local/u-vs/usr/var/run/vservers/template 1115333644 M * terr which might cause you a problem :-{ 1115333682 M * Bertl guess I don't need those ... 1115333773 M * terr Nevertheless, for future releases of the tools, it might be more "portable" to have those links changed to relative. No? 1115333811 M * FaUl re 1115333835 M * Bertl terr: well, maybe an option, not sure, as the relative structure might change too 1115333909 M * Bertl wb FaUl! 1115333957 M * Doener terr: problems happen on 2.6.11.8-vs2.0pre4, too. right? 1115333977 A * Doener will go the hard way once more... ;) 1115333991 M * terr I didn't try that one yet. 1115334003 M * Doener ok, on which ones does it happen? 1115334019 M * Doener (just the latest one that was tested...) 1115334026 M * terr Currently I have 2.6.11.7-vs2.0-pre3 1115334047 M * terr It happened also on 2.6.11.7-vs2.0-pre2 1115334050 M * Bertl Doener: want a delta pre3-pre4? 1115334100 M * Doener if it's on your server anyways ;) otherwise i can get one myself... after all you showed me how to keep a lot of trees lying around ;) 1115334134 M * Bertl okay, then go ahead ... 1115334236 M * FaUl distcc rocks 1115334597 M * Bertl sometimes, yes ... 1115334623 M * Doener ah ok, just nfs and uts patches.. 1115334949 M * Doener terr: did you get any kernel messages? oopses? 1115334969 M * terr No, nothing. 1115334979 M * Bertl dmesg is clean? 1115335025 M * terr IIRC there was a message mentioning vshelper. 1115335087 M * Doener that counts as a kernel message ;) 1115335110 M * terr Sorry. 1115335122 M * Bertl np, that's why I asked ... 1115335127 M * terr ---> /usr/lib/util-vserver/vshelper: (startup 45678) returned sync with 256 1115335154 M * Doener that showed up when the errors started, right? 1115335166 M * Bertl hmm, 45678 ... could you use static context id for a try? 1115335190 M * Doener that is static ;) 1115335202 M * Bertl ah, oops, yep ;) 1115335229 M * Bertl that is the test xid I use in testme.sh *G* 1115335238 M * terr Some of these messages appeared at boot! 1115335258 M * Bertl where are the brown paper bags again? 1115335291 M * terr The vserver were built with the --context option 1115335312 M * terr namely 99 (or 98) 1115335332 M * terr aonther example: /usr/local/util-vserver/lib/util-vserver/vshelper: (startup 99) returned sync with 256 1115335374 M * terr and (with debian package util-vserver): localhost kernel: /sbin/vshelper: (startup 99) returned sync with 256 1115335376 M * Bertl hmm, why do mainline tools cause 256 exit on startup? 1115335455 M * Doener non-existant startup delegation script? 1115335698 M * Bertl hmm, I don't get those on my test machines ... 1115335715 M * mep_ how can i deny some users to view all processs running on the box 1115335745 M * Bertl no, I get those too ... thought this was already in mainline *sigh* 1115335764 M * Bertl mep_: what do you mean? 1115335788 M * mep_ with ps -ef for example you see wich processes running on a box 1115335799 M * mep_ how can i disallow this to some users 1115335831 M * Bertl you are not talking about vserver guests, you are talking about linux in general, no? 1115335877 M * mep_ yes ;) 1115335891 M * Bertl you simply can't without a kernel patch 1115335894 M * mep_ but it runs ina a vserver ;) 1115335898 M * mep_ narf 1115335902 M * mep_ grsecurity so :/ 1115335907 M * mep_ okay thx 1115335915 M * Doener why would you want to do so, anyways? 1115335968 M * Bertl to confuse folks adding up the process resources ;) 1115335978 M * FaUl have i already mentioned that i like distcc? 1115336008 M * Bertl well, yes, my question is just, what are you compiling _that_ long? ;) 1115336043 M * FaUl Bertl: I debugged some wired scsi-controler-related problems and now rebuild the kernel ;) 1115336073 M * FaUl but its ready right now, i hope i have not to go the server again ;) 1115336074 M * mep_ doener not every user should see my started processes ) 1115336079 M * Doener hmm... rebuilding a kernel usually doesn't take that long... 1115336100 M * Bertl ./testme.sh 1115336100 M * Bertl Linux-VServer Test [V0.12] (C) 2003-2005 H.Poetzl 1115336100 M * Bertl chcontext is working. 1115336100 M * Bertl chbind is working. 1115336100 M * Bertl Linux 2.6.11.8-pa4-vs2.0-pre4 parisc64/0.30.207/0.30.207 [Ea] 1115336102 M * FaUl Doener: right, but looking what option causes the panic may take :) 1115336103 M * Bertl VCI: 0001:0025 273 03110076 1115336106 M * Bertl --- 1115336108 M * Bertl [000]# succeeded. 1115336110 M * Bertl [001]# succeeded. 1115336113 M * Bertl [011]# succeeded. 1115336115 M * Bertl [031]# succeeded. 1115336118 M * Bertl [101]# succeeded. 1115336120 M * Bertl [102]# succeeded. 1115336123 M * Bertl [201]# succeeded. 1115336125 M * Bertl [202]# succeeded. 1115336133 M * FaUl parisc64? nice ;) 1115336138 M * Doener not bad ;) 1115336148 M * Doener that parisc64 is you own box, right? 1115336155 M * Doener your even 1115336230 M * Bertl yup 1115336255 A * FaUl has only alpha, ultrasparc and x86 1115336270 M * Bertl hmm, did you say alpha? 1115336310 M * FaUl yes, 500mhz one processor on an evaluation-board and that alhpaxl-266 and an alphaserver 2100a :) 1115336317 A * Doener has only an x86... 1115336342 M * Bertl FaUl: *you feel an urge to try linux-vserver on your alpha* 1115336386 M * FaUl Bertl: hmm, i'll do as far as i find a second of time to play around with that 1115336584 M * Bertl that would be really nice ... 1115336831 M * FaUl Bertl: my problem is that the firmware on the 500mhz-alpha, which is the only one i use actually, is broken and i need an disk-drive to fix that :) 1115336870 M * Bertl what about net booting it? 1115337097 M * terr Is there some point in my staying around? Otherwise, I might well go to sleep |-} 1115337121 M * Bertl atm not really, so I'd suggest going to sleep now ... 1115337131 M * Bertl we probably know more tomorrow ... 1115337189 M * terr Good news. Good night then, and see you tomorrow. 1115337209 M * Bertl c'ya and sweet dreams! 1115337237 M * terr Thanks. Bye... 1115337241 P * terr 1115337530 Q * prae Quit: Pwet 1115337537 M * FaUl Bertl: thats the problem - it has no hard-drive and that fscking broken firmware does not recognize the dec-nic 1115337550 M * Doener Bertl: hmm... Gilles/terr said his host's init was defunct, right? 1115337551 M * FaUl well, ill go to bed now 1115337552 M * FaUl n8 1115337561 M * Doener godd night FaUl 1115337585 M * Bertl night FaUl! 1115337594 M * Bertl yep, we have the following patterns so far: