1116720444 M * Doener` hum hum... 1116720474 M * Doener` (watching Next Friday, thus a little slow atm) 1116720478 M * Bertl btw, another question: do we guard set_iattr by CAP_CONTEXT as advertized? 1116720556 M * Bertl Doener`: np @ slow 1116720620 M * Bertl ah, yes, we do, basically first check in switch ... 1116720622 M * Doener` hm? the whole sys_vserver is guarded by CAP_CONTEXT IIRC 1116720713 M * Bertl yup, that's what I meant ;) 1116720943 Q * Shuri Quit: 1116721891 M * Doener` I got no real opinion on a) and b)... maybe tomorrow i get some idea, off to bed now... 1116721928 M * Doener` good night folks (or just "Ey karamba!"? ;) 1116721936 N * Doener` Doener_zZz 1116722031 M * Bertl night Doener_zZz! 1116722821 M * FaUl night all 1116723267 M * Bertl night FaUl! 1116723636 J * eXplasm2 explasm@p549FFB59.dip.t-dialin.net 1116723647 M * Bertl welcome eXplasm2! 1116724073 Q * explasm__ Ping timeout: 480 seconds 1116725744 Q * monrad Quit: Leaving 1116726213 Q * romke Ping timeout: 480 seconds 1116726893 J * duckx ~Duck@mna75-1-81-57-39-234.fbx.proxad.net 1116726900 M * Bertl welcome duckx! 1116727384 M * duckx moin 1116727406 M * duckx Our vservers are rocking ... 1116727418 M * duckx We just migrated to slapd 2.2 ... 1116727428 M * duckx Everything run just fine :) 1116727440 M * Bertl good! 1116727460 M * duckx About 6 servers are running !) 1116727623 M * duckx Man I am god damn tired ... Well need to go to sleep 1116727628 M * duckx See you !) 1116727735 M * Bertl cya! good night! 1116727810 M * daniel_hozac humm. is SIOCGIFMAP deprecated? it doesn't seem to return any reliably valid information. 1116728194 M * Bertl http://lkml.org/lkml/2005/1/31/213 1116728720 M * daniel_hozac what about the other fields though? like base_addr? 1116728985 M * daniel_hozac hmm, seems like that is also not meant to be used. thanks. 1116731618 Q * flock Quit: Expert, n.: Someone who comes from out of town and shows slides. 1116733831 Q * rs Quit: rs 1116739966 M * Bertl off to bed now ... have a nice one everyone! 1116739978 N * Bertl Bertl_zZ 1116747455 Q * micah Ping timeout: 480 seconds 1116747686 J * monrad ~monrad@213083190130.sonofon.dk 1116747698 J * micah micah@micha.hampshire.edu 1116752031 J * [HvD] ~nobody@chello084115137093.5.graz.surfer.at 1116752313 M * [HvD] good morning, yust testing 2.0-rc against 0.30.207-5 (debian) when calling vserver enter i get: "++ exec /usr/sbin/chbind --silent --ip 192.168.231.18/24 /usr/sbin/vnamespace --enter 49153 -- /usr/sbin/vcontext --silent --migrate --chroot --xid 49153 --uid root --" so there is no bash -login called .. any idea ? (works fine with my 2.4.28-vs1.29) .. 1116752369 M * [HvD] kernel 2.6.11.10 1116753554 Q * pflanze Ping timeout: 480 seconds 1116753700 J * rs ~rs@imhotep.rhapsodyk.net 1116754572 Q * rs Quit: rs 1116754892 J * rs ~rs@imhotep.rhapsodyk.net 1116755181 Q * click Ping timeout: 480 seconds 1116755384 Q * yarihm Quit: Leaving 1116756606 Q * rs Quit: rs 1116757734 M * maharaja mhm 1116757758 M * maharaja im running 2.6.9.6 or so with 1.9x 1116757770 M * maharaja never had any problem like this 1116757789 M * maharaja but when the tools are not up2date, youll experience strange problems like these 1116758653 N * Bertl_zZ Bertl 1116758737 M * Bertl morning folks! 1116759095 M * Bertl k, off for breakfast ... 1116759101 N * Bertl Bertl_oO 1116760160 Q * albeiro Ping timeout: 480 seconds 1116760187 J * albeiro albeiro@albeiro.usercloak.oftc.net 1116760309 Q * knoppix_ Ping timeout: 480 seconds 1116760667 N * Bertl_oO Bertl 1116760918 J * knoppix_ ~knoppix@dsl-213-023-147-224.arcor-ip.net 1116761343 J * click click@dsl-static-122-208.aal.tiscali.no 1116761409 M * Bertl welcome click! 1116761448 M * click rebooted and back to dualcpu again 1116761550 M * Bertl [HvD]: so you have trouble with 'enter'? 1116761573 M * [HvD] yep right .. 1116761579 N * Doener_zZz Doener 1116761590 M * Bertl morning Doener! 1116761596 M * Doener greetings (ems) folks! ;) 1116761609 M * Doener scnr 1116761620 M * Bertl lol 1116761671 M * Bertl [HvD]: what does 'vserver --debug exec true' do? 1116761682 M * Doener sleeping definately helped :) I just actually understood a) (route virt.) 1116761703 M * [HvD] well give it a try with 0.30.204-5 before starting debuggig ;) 1116761728 M * [HvD] well 0.30.204-5 gives me the same .. so its a user problem *grml* 1116761761 M * [HvD] --debug gives: ++ exec /usr/sbin/chbind --silent --ip 192.168.231.18/24 /usr/sbin/vnamespace --enter 49153 -- /usr/sbin/vcontext --silent --migrate --chroot --xid 49153 --uid root -- 1116761770 M * [HvD] but should: 1116761770 M * [HvD] ++ exec /usr/sbin/chbind --silent --ip 192.168.231.18/24 /usr/sbin/vnamespace --enter 49153 -- /usr/sbin/vcontext --silent --migrate --chroot --xid 49153 --uid root -- bash -login 1116761795 M * Doener config style? 1116761803 M * [HvD] alpha-utils 1116761815 M * Bertl please be so kind and upload the complete trace somewhere (e.g. pastebin.com) 1116761824 M * [HvD] k 1116761834 M * Doener well, they support both ;) but i guess that means new style config... 1116762040 M * [HvD] http://pastebin.com/287834 1116762051 M * [HvD] new style yes .. sorry ;) 1116762141 M * [HvD] maybe i should not mount / in /fstab .. but this works fine with 2.4.28-ctxxxx 1116762335 Q * aba Ping timeout: 480 seconds 1116762336 M * [HvD] yust found it .. ;) 1116762340 M * Bertl ahem, 49153? 1116762349 M * [HvD] ok you should not mount rootfs in fstab .. 1116762380 M * [HvD] seems like /dev/raid/qa-www-0_rootfs / ext3 defaults does not work in 2.0-rc /fstab .. ;) 1116762408 M * Bertl well, doesn't make sense, no? 1116762433 M * Bertl or do you want to mount root at startup? 1116762468 M * [HvD] well does make sens .. you dont need to mount vserver in masts fstab .. which is fine .. 1116762477 M * [HvD] s/sens/sense/ 1116762491 M * Doener hm, namespace issue... 1116762514 M * Doener the mounts from /fstab (in the configuration that is, right?) happens in the new namespaces 1116762525 M * [HvD] let me look .. 1116762540 M * Doener the checks to find a shell for "vserver xxx enter" happen before the namespace is entered 1116762549 M * Bertl [HvD]: but you are aware that this requires the guest to be 'trusted' 1116762551 M * [HvD] ahh .. thats why ;) 1116762562 M * [HvD] well its all trusted here ;) 1116762570 M * Bertl okay ;) 1116762581 M * Doener Bertl: hmmm... why? 1116762591 M * Bertl [HvD]: so probably nonamespace will allow this 'trick' too ... 1116762614 M * Doener (i'm assuming that /fstab means /etc/vservers//fstab) 1116762619 M * Bertl Doener: because of /dev/raid/qa* ? 1116762637 M * [HvD] Doener: yes it does 1116762658 M * [HvD] well i will give it a try with nonamespace 1116762674 M * Doener [HvD]: you can also just provide the shell on your own 1116762680 M * [HvD] well how ? 1116762691 M * [HvD] apps/shells ;) (yust guessing) 1116762694 M * [HvD] suexec does not work .. 1116762702 M * Doener echo "/bin/bash -login" > /etc/vservers//shell 1116762708 M * [HvD] bad guess ;) 1116762718 M * [HvD] this isnt in the man is it ? 1116762748 M * Doener well, it's in the trace at pastebin ;) 1116762754 M * [HvD] hehe .. 1116762767 M * Doener i didn't know that file before, so let's check the flowerpage... 1116762772 M * [HvD] should open my eyes before aksing other people 1116762795 M * [HvD] any idea whe i get EPERM when openeing /dev/null in vservre ? 1116762809 M * [HvD] (same vserver) 1116762822 M * Doener (shell is on the flowerpage...) 1116762842 M * Doener hm, is it really a /dev/null? check major/minor number 1116762851 M * [HvD] lets see .. 1116762858 M * Doener should be 1,3 1116762862 J * meebey_ ~meebey@booster.qnetp.net 1116762888 M * [HvD] yes its 1,3 .. 1116763137 M * Doener Bertl: AFAICT the purpose of the secure-mount program is to allow such mounts in the configuration's fstab file... and after all, the mounts happen before the vserver is running, so no chroot'ing did happen yet and thus the dev-node is not required inside the vserver... am i wrong? 1116763179 M * Bertl no idea, lost the overview of what the tools do :/ 1116763291 M * [HvD] the dev node is not required inside the vserver .. 1116763310 M * Bertl okay, good to know ... 1116763311 M * [HvD] can prove this with 2.4.28-ctx .. 1116763343 M * [HvD] and yes setting up the shell manually works .. 1116763414 M * Doener ok, with 2.4 we have no namespace support in the vserver part of the kernel, so that explains it 1116763427 M * Bertl try with nonamespace 1116763474 Q * hellekin Quit: BitchX: try our lowfat flavor too! 1116763600 M * [HvD] well seems like mounting rootfs in vservers/fstab is not a good idea anyway with namespace enabled: 1116763601 M * [HvD] grz-1:/etc/vservers/qa-www-0# vserver qa-www-0 stop 1116763601 M * [HvD] No command given; use '--help' for more information. 1116763713 J * hellekin ~hellekin@v41.ath.cx 1116763729 M * Doener hm, you probably have to provide all commands then, because the namespace is entered too late and the tools can't determine what they should use... 1116763741 M * Doener guess you should file that as a bug report at savannah... 1116763753 M * [HvD] ok will do .. 1116763856 M * Bertl Doener: did you sleep well tonight? 1116763856 M * [HvD] qa-www-0:/dev# find -type c -exec cat {} \; 1116763857 M * [HvD] cat: ./full: Permission denied 1116763857 M * [HvD] cat: ./null: Permission denied 1116763857 M * [HvD] cat: ./ptmx: Permission denied 1116763857 M * [HvD] cat: ./random: Permission denied 1116763858 M * [HvD] cat: ./tty: Permission denied 1116763858 M * [HvD] cat: ./urandom: Permission denied 1116763860 M * [HvD] cat: ./zero: Permission denied 1116763860 M * [HvD] mhh strange 1116763875 M * Doener Doener: too long, but well :) 1116763887 M * Bertl hmm, i.c. ;) 1116763892 M * [HvD] ;-) 1116763909 M * Bertl Doener: so you're ready for some major testing/code reading? 1116763915 M * Doener [HvD]: to stop the vserver, you could try to provide the stop command yourself... echo "my-stop-command" > /etc/vservers//apps/init/cmd.stop 1116763943 M * [HvD] something like "rc 6" i gues s. 1116763952 M * Doener [HvD]: probably... 1116763972 M * Doener Bertl: I'm trying to find some better font for my xterm atm, my current one is a little hard on the eyes... 1116764057 M * Bertl -misc-fixed-medium-r-semicondensed 1116764075 Q * meebey_ Quit: leaving 1116764144 M * Doener Bertl: as *faceName? 1116764159 A * Bertl doesn't use xterm ... 1116764172 M * Bertl it's mgt (multi-gnome-terminal) here 1116764176 M * Doener ah, *font it is... 1116764194 M * Doener eek, no, looks ugly on my tft... 1116764205 M * Doener thanks anyways 1116764207 M * Bertl -misc-fixed-medium-r-semicondensed-*-*-12 1116764222 M * Bertl (to be precise) and it looks good here on tft, wanna screenshot? 1116764362 M * Bertl http://vserver.13thfloor.at/Stuff/font.png 1116764446 M * Doener hm, ok, it's just the bold text that looks ugly 1116764704 M * hellekin aterm*font:-*-clean-medium-r-normal-*-12-*-*-*-*-*-iso8859-15 1116764704 M * hellekin aterm*boldFont:-*-bitstream vera sans mono-medium-r-normal-*-12-*-*-*-*-*-iso8859-15 1116764732 M * hellekin i'm on a 800x600 display. Need that. clean bold sucks 1116764753 M * hellekin both fonts work pretty well together 1116764804 M * Doener hm, Terminus it is for now... 1116764942 M * Doener ok Bertl, so what do you have for me? 1116765262 M * Bertl I have a bunch of changes for 2.0 1116765278 M * Bertl sec preparing a diff ... 1116765294 M * [HvD] tortilla chips with tacco souce anyone ? (getn sick) 1116766287 M * Bertl Doener: just a second, I'm splitting it up to make reviewing easier 1116766446 M * Doener np 1116766825 M * Bertl Doener: okay, first one ready ... just cleanups 1116766826 M * Bertl http://vserver.13thfloor.at/Experimental/delta-cleanup-01.diff 1116766848 M * Bertl (well, except for the BUG_ON()) 1116767726 M * Bertl and here is the second one ... 1116767727 M * Bertl http://vserver.13thfloor.at/Experimental/delta-cleanup-02.diff 1116768190 M * Bertl Doener: looks good so far? 1116768239 M * Bertl next one is the switch to claim/release for network ;) 1116768241 M * Bertl http://vserver.13thfloor.at/Experimental/delta-claim-01.diff 1116768784 M * Doener i'm (as usual) a little stuck at the preprocessor stuff ;) 1116768805 M * Bertl just ask ... 1116768884 Q * cryo Remote host closed the connection 1116769013 M * Bertl and here is the last one in this series: 1116769015 M * Bertl http://vserver.13thfloor.at/Experimental/delta-cleanup-03.diff 1116769052 M * Bertl after that, the network stuff should use the same mechanisms as the context stuff 1116769082 M * Bertl (if did not miss anything, of course ;) 1116769458 M * Doener ah, you're preparing for ngnet :) 1116769546 M * Bertl well, it was more the best way to make the LEGACYNET default and only allow to disable it with experimental, but yes ;) 1116769634 M * SiD3WiNDR ngnet++ 1116769661 M * Doener first two look good 1116769877 M * Doener back in a few... haven't voted yet... (election today...) 1116769880 N * Doener Doener|gone 1116774637 N * Doener|gone Doener 1116774658 M * Doener took longer than I expected... (we decided to have some food...) 1116774680 M * SiD3WiNDR heyhey 1116774686 M * SiD3WiNDR DO NOT FORSAKE IRC AND LINUX FOR FOOD 1116774690 A * SiD3WiNDR puts up a big sign 1116774691 M * SiD3WiNDR ;) 1116774712 M * Doener primary reason was the election... i guess that's valid, right? ;) 1116774849 M * Bertl wb Doener! 1116775024 M * SiD3WiNDR hehe 1116775031 M * SiD3WiNDR well, barely 1116775031 M * SiD3WiNDR ;) 1116775099 J * lilo_ ~lilo@lilo.usercloak.oftc.net 1116775205 Q * lilo Ping timeout: 481 seconds 1116775336 J * ntrs ntrs@Dardeene-68.188.50.87.charter-stl.com 1116775345 M * Bertl welcome ntrs! 1116775447 M * Doener third also looks good... 1116775870 M * Doener and so does the last one. 1116776139 J * rs ~rs@80.214.248.1 1116776521 Q * rs Read error: Connection reset by peer 1116777252 M * Bertl Doener: okay, could you do some network related testing to those patches? 1116777285 M * Bertl i.e. start a guest, open a few connections, kill the vserver ... something like that 1116777309 M * Bertl just to verify that the new BUG_ONs() do not trigger (i.e. we didn't miss anything there) 1116777434 M * Doener sure 1116777725 J * mep_ mep@p5091B839.dip.t-dialin.net 1116777734 M * Bertl evening mep_! 1116777884 Q * mep__ Read error: Operation timed out 1116778178 M * Bertl okay, off for now ... more changes to come ... (probably tomorrow) 1116778184 N * Bertl Bertl_oO 1116778738 M * mep_ hi 1116778896 Q * Doener Quit: Leaving 1116778926 J * Doener ~doener@p54874F2C.dip.t-dialin.net 1116780882 J * rs ~rs@imhotep.rhapsodyk.net 1116781638 Q * micah Quit: leaving 1116781647 J * micah micah@micha.hampshire.edu 1116783102 Q * Loki|muh Ping timeout: 480 seconds 1116783467 Q * ntrs Quit: Leaving 1116785954 J * Loki|muh loki@satanix.de 1116787319 Q * rs Quit: rs 1116788424 J * Thorsten ~Thorsten@dsl-084-058-044-019.arcor-ip.net 1116790343 J * stephen5 ~steve@user-229.lns2-c7.dsl.pol.co.uk 1116790350 M * stephen5 hey all 1116790381 M * stephen5 does anybody know how I can set resource limits [cpu,ram] but allow a vserver to burst to the full capability of the machine sometimes? 1116791092 J * Doener` ~doener@p54876C38.dip.t-dialin.net 1116791102 M * stephen5 uh..also, I'm trying to stop a vserver but it's hanging at "Shutting down interface eth0:" .. how can I force it? 1116791520 Q * Doener Ping timeout: 480 seconds 1116793193 M * FaUl Doener`: you are from nrw too? 1116793724 M * Doener` yep 1116794095 M * FaUl Doener`: where exactly? 1116794122 M * FaUl btw. democracy sucks :-( 1116794150 M * daniel_hozac i agree. power to me! :) 1116794180 M * FaUl daniel_hozac: hehe 1116794461 M * stephen5 anybody know? 1116794468 M * stephen5 > I'm trying to stop a vserver but it's hanging at "Shutting down interface eth0:" .. how can I force it? 1116794525 M * daniel_hozac stephen5: first you should disable those scripts in your vserver. 1116794536 M * daniel_hozac stephen5: vkill should be able to kill the it off though. 1116794599 M * stephen5 how do I find the vserver's pid? 1116794614 M * [HvD] stephen5: remove network shutdown from rc6.d 1116794625 M * Doener` FaUl: north-north-north ;) I guess "Rahden" tells you absolutely nothing... 1116794680 M * daniel_hozac stephen5: there would likely be multiple processes still running in the context. check with vps. 1116794720 M * stephen5 [HvD]: which file would that be in? 1116794881 M * FaUl Doener`: i guess dortmund will tell you something :-) 1116795043 M * [HvD] stephen5: find /etc/rc6.d/|grep -e "network\|shutdown" 1116795247 M * stephen5 [HvD]: got it, what should I change in that file? 1116795271 M * [HvD] remove it .. 1116795277 M * [HvD] but only in the vserver .. ;-) 1116795285 M * [HvD] i mean .. remove the symlink .. 1116795306 M * stephen5 ah ok =) 1116795316 M * [HvD] either witch chkconfig, update-rc.d, or rm .. what ever you prefer ;) 1116795337 M * [HvD] same accounts for hwclock and random 1116795358 M * stephen5 now it's saying this 1116795360 M * stephen5 "Device eth0 has different MAC address than expected, ignoring." 1116795399 M * [HvD] which distro is your guest ? 1116795411 M * stephen5 centos 1116795411 M * stephen5 4 1116795584 M * stephen5 shouldn't the vserver script assign the IP addresses to the vserver though? from .conf? 1116795591 M * daniel_hozac yes, it does. 1116795611 M * stephen5 so the vserver's init scripts for eth0 should be removed? 1116795622 M * daniel_hozac right. 1116795784 M * daniel_hozac chkconfig network off; rm -f /etc/rc[06].d/S01* 1116796580 J * rs ~rs@imhotep.rhapsodyk.net 1116796611 Q * [HvD] Read error: Connection reset by peer 1116796636 N * BobR_zZ BobR 1116796681 N * BobR BobR_zZ 1116797547 Q * rs Quit: rs 1116798671 N * Bertl_oO Bertl 1116798804 M * Bertl evening folks! 1116798814 A * FaUl goes dieing 1116798825 M * Bertl hmm? 1116798894 M * eyck dieting maybe? 1116798904 M * FaUl Bertl: just stupid democracy that destroyed my plan for the next 2 years and declared the last 2 years of my life as wasted time 1116798915 M * Bertl eyck: or maybe dyeing ;) 1116798941 M * eyck hmm, how does dyeing your hair fixes democracy? 1116798954 M * Bertl red, green, blue ;) 1116798981 M * eyck FaUl: what happened? 1116798997 M * Bertl they had elections ... 1116799025 M * eyck they usually do 1116799116 M * FaUl eyck: well, they elected our right party, which want to enforce tuition fees, I wasted my last 2 years making up my "abitur" 1116799178 M * FaUl and now the chance for me to go to university is very small because i've no money 1116799223 M * eyck cool, you can work and start your studying process in the background, 1116799245 M * eyck although me myself, I got my degree almost free, 1116799313 M * FaUl well, i've to pay eur 500 per semester and i've still pay rental-fee for my flat and buy something to eat. 1116799354 M * eyck that would be about 2kzl, 1116799365 M * FaUl kzl? 1116799404 M * eyck local money 1116799422 M * FaUl where are you from? ;0 1116799438 M * eyck sounds about right, that's how much some of the cheaper semesters cost around here, when you're studying on 'night shift' 1116799442 M * eyck Poland, 1116799480 M * FaUl is poland one of that euro-countries right now? 1116799487 M * eyck unfortunatelly yes. 1116799488 M * FaUl do you still have szloties? ;) 1116799495 M * eyck we do. 1116799522 A * FaUl have a few million "old" szloties somwhere 1116799553 M * eyck million? nice. 1116799581 M * FaUl well that exchange-rate was not very high IIRC 1116799587 M * eyck I don't know what happend with transition to euro 1116799611 M * eyck we had a denomination 1 new zl == 10_000 old zl ;) 1116799634 M * FaUl and 1 zl are how many ct? 1116799646 M * eyck 1 euro =~ 4zl 1116799653 M * eyck 1 $ =~ 3zl 1116799674 M * FaUl are you sure? 1116799693 M * eyck about those aproximations? 1116799711 M * FaUl maybe i'm wrong and that were'nt millions but tenthousends 1116799721 M * FaUl IIRC around 20DM in this times 1116799736 M * eyck maybe, million old zl would be ~20 euro 1116799789 M * FaUl i've to look for it when i visit my parants next time 1116799795 M * eyck oh, here it is, we're expecting euro in 2009. 1116799803 M * FaUl ok 1116799830 J * badiane ~chatzilla@user-0cdf0g4.cable.mindspring.com 1116799855 M * badiane hello every one 1116799879 M * eyck hello, 1116799888 M * eyck we're discussing democracy today, 1116799901 M * badiane I'm having some problems with my vservers 1116799918 M * badiane I am running debian 1116799928 M * eyck good for you, 1116799933 M * eyck what kind of problems? 1116799939 M * badiane util-vserver 0.30.207-5 1116799951 M * eyck and 2.6.12? 1116799952 M * FaUl however n8 1116799957 M * badiane 2.6.10 1116799971 M * eyck cool, I know nothing about it, ask Bertl. 1116799979 M * eyck nighty night, 1116799981 M * badiane ok 1116799995 M * Doener` i'm off to bed now... good night folks! 1116799998 M * badiane it's still late afternoon her 1116799998 N * Doener` Doener_zZz 1116800000 M * badiane but night 1116800032 M * badiane hello bertl 1116800178 P * Thorsten Leaving 1116801587 J * rs ~rs@imhotep.rhapsodyk.net 1116801654 M * Bertl hey badiane! 1116802050 Q * knoppix_ Quit: Verlassend 1116802591 M * badiane I'm back 1116802593 M * badiane hello 1116802600 M * Bertl :) 1116802605 M * badiane :-) 1116802622 M * badiane as I stated before I'm running debian 1116802627 M * badiane 2.6.10 1116802629 M * Bertl congrats! 1116802658 M * badiane with util-vserver 207-5 1116802712 M * badiane when I start the vserver I receive the /proc/uptime message 1116802720 M * Bertl which is? 1116802724 M * badiane I have run the vprocunhide script 1116802738 M * badiane but I still cannot start the vserver 1116802762 M * badiane AwH-ui- /proc/uptime 1116802769 M * badiane is what I get from showattr 1116802805 M * badiane How do I get some of the proc dirs and files to show in some of the vservers 1116802820 M * Bertl what is 'the /proc/uptime message'? 1116802858 M * badiane "/proc/uptime can not be accessed. Usually, this is caused by" 1116802870 M * badiane "procfs-security. Please read the FAQ for more details" 1116802874 M * Bertl okay ... 1116802903 M * Bertl vprocunhide runs fine or complains? 1116802912 M * badiane runs without problems 1116802926 M * badiane "Fixing /proc entries visibility...." 1116802937 M * badiane well I assume that it fine 1116802940 M * badiane since that's all I get 1116803038 M * Bertl search for a file called vprocunhide-files 1116803044 M * badiane ok 1116803064 M * Bertl upload it somewhere (i.e. pastebin.com) 1116803085 M * badiane hold on 1116803198 M * badiane ok it's up 1116803203 M * badiane on pastebi 1116803205 M * Bertl url? 1116803206 M * badiane pastebin 1116803221 M * badiane http://pastebin.com/288152 1116803228 M * Bertl tx 1116803292 M * Bertl which path has your vprocunhide = 1116803293 M * Bertl ? 1116803433 M * badiane hold on 1116803505 M * badiane "/usr/lib/util-vserver/defaults/vprocunhide-files" 1116803536 M * Bertl okay, this is the data file, what about the vprocunhide 'script'? 1116803574 M * badiane "/etc/init.d/vprocunhide" 1116803607 M * Bertl okay, so you did /etc/init.d/vprocunhide start, no? 1116803636 M * badiane yes 1116803687 M * Bertl and if you do now showattr /proc/uptime 1116803692 M * Bertl it returns? 1116803728 M * badiane "AwH-ui- /proc/uptime" 1116803814 M * Bertl somehow your kernel or tools are broken ... 1116803825 M * Bertl let's check with the following: 1116803840 M * Bertl setattr --~hide /proc/uptime 1116803844 M * Bertl showattr /proc/uptime 1116803937 M * badiane "Awh-ui- /proc/uptime" 1116803963 M * Bertl so it seems the setattr itself works, but the script fails ... 1116803975 M * Bertl let's change that back and modify the script to see more ... 1116803985 M * Bertl setattr --hide /proc/uptime 1116803998 M * badiane the server started 1116804010 M * Bertl then add '-x' to the shebang line of /etc/init.d/vprocunhide 1116804055 M * badiane I don't understand the syntax fo setattr very well 1116804075 M * Bertl just use the command I used ... 1116804079 M * Bertl setattr --hide /proc/uptime 1116804083 M * badiane what do the -- and the ~ mean 1116804100 M * Bertl --hide means set the hide flag, --~hide means clear it 1116804143 M * badiane so the tilde in front of the options whether they be admin wath or hide mean remove 1116804152 M * badiane and without the tilde mean enable 1116804161 M * Bertl yep 1116804181 M * badiane "AwH-ui- /proc/uptime" 1116804183 M * badiane it's back 1116804284 M * Bertl yup, now please add the -x to the vprocunhide script 1116804320 M * Bertl first line shows #!/bin/bash change that to #!/bin/bash -x 1116804353 M * badiane ahh ok tracing 1116804461 M * badiane http://pastebin.com/288159 1116804464 M * Bertl thanks 1116804613 M * badiane :-) 1116804659 M * Bertl okay, now we do the same to /usr/lib/util-vserver/vprocunhide 1116804676 M * badiane ok 1116804681 M * Bertl and check if the /proc/uptime changed ... 1116804819 M * badiane http://pastebin.com/288160 1116804857 M * badiane "AwH-ui- /proc/uptime" 1116804863 M * Bertl ah, we have an alternative file ;) 1116804870 M * Bertl what's in # 1116804871 M * Bertl /etc/vservers/.defaults/apps/vprocunhide/files 1116804887 M * badiane I was just going to mention that to you 1116804901 M * badiane I moved the vprocunhide-files in there 1116804914 M * Bertl why? 1116804917 M * badiane hopping to enable 1116804941 M * Bertl okay, please undo those changes ... 1116804959 M * badiane well it's stated (on the flower page) that I can put it there (I seem to remember) 1116804965 M * badiane ok 1116805003 M * badiane I removed the files inside the directory 1116805009 M * badiane should I also remove the directory 1116805033 M * Bertl remove the directory too 1116805049 M * badiane what if I wanted to enable certain parts of the proc filesystem exposed to some vserver and not to others, is that possible 1116805054 M * badiane ok 1116805058 M * Bertl later ... ;) 1116805073 M * badiane :-) 1116805076 M * Bertl and no, it's not possible ... 1116805123 M * badiane ok 1116805184 M * badiane the namespaces concept as opposed to 000 chmod is automatically enabled or has to be enabled. 1116805216 M * Bertl 000 chmod should not be used since a year or so ;) 1116805298 M * Bertl and yes, namespaces are default 1116805335 M * badiane "+ /usr/sbin/setattr -x '--!hide' /proc/uptime" 1116805347 M * badiane this is the uptime line from the script 1116805348 M * Bertl looks much better, no? 1116805358 M * badiane what does 1116805360 M * Bertl now check /proc/uptime ... 1116805416 M * badiane it started 1116805431 M * Bertl of course, it's now doing what it's supposed to do ;) 1116805431 M * badiane "Awh-ui- /proc/uptime" 1116805506 M * badiane procs are ok now 1116805531 M * Bertl so where did that dir/file come from? did you create them first? 1116805543 M * badiane just did a ifconfig and nothing showed which is better than before when every interface showed 1116805547 M * badiane I created it 1116805555 M * badiane I think its mentioned in the flower page 1116805587 M * badiane I was reading on of the old irc records online where you mentioned that maybe the instructions for the new dir structure 1116805595 M * badiane might not be understood so easily 1116805603 M * Bertl :) 1116805606 M * badiane and it brought a smile to my face 1116805632 M * Bertl # /etc/vservers/.defaults/apps/vprocunhide 1116805632 M * Bertl * files 1116805632 M * Bertl A list of files which will be made visibly by vprocunhide. Wildcards are allowed and anything ending in '/' will be processed recursively. When this file exists, it overrides the defaults in SYSDEFAULTDIR/vprocunhide-files. The entries there must be absolute filenames inclusive the leading '/proc'. 1116805647 M * badiane because the first time I saw it I couldn't make the difference between what was a file or a directory 1116805648 M * Bertl note the 'override' part ;) 1116805679 M * badiane ahhh 1116805695 M * Bertl btw, that's from the flower page ;) 1116805723 M * badiane I thought that it overrode the main file and became a replacement 1116805747 M * Bertl yes, that's basically what it does ... 1116805760 M * badiane I didn't realize that the override meant some sort of annulment 1116805783 M * Bertl maybe it should read 'replaces' there ... 1116805807 M * badiane so why didn't it work if it's used instead 1116805820 M * Bertl well, I guess you broke it somehow ... 1116805845 M * Bertl there is a simple bash loop iterating over the entries 1116805845 M * badiane hmmm 1116805872 M * Bertl in your trace, it exists after the first read 1116805886 M * badiane in the init dir or the lib 1116805909 M * Bertl # 1116805910 M * Bertl ++ UNHIDEFILE=/etc/vservers/.defaults/apps/vprocunhide/files 1116805914 M * Bertl # 1116805915 M * Bertl + read filename 1116805918 M * Bertl # 1116805919 M * Bertl + test -n 1 1116805923 M * Bertl 157. 1116805924 M * Bertl + exit 0 1116806091 M * badiane I see so it failed reading that file I had placed there 1116806144 M * Bertl I'd say it was simply empty ... 1116806165 M * badiane hmmm 1116806171 M * badiane I actually opened it 1116806178 M * badiane and read it after copying it 1116806194 M * Bertl well, you can try to copy a single line or two there ... 1116806201 M * Bertl then run the script in debug again ... 1116806211 M * badiane I'll try it later 1116806227 M * badiane following what you said in a irc transcript 1116806238 M * badiane I created a 00 directory 1116806244 M * badiane in the interfaces directory 1116806259 M * badiane actually it's a 01 dir 1116806268 M * Bertl try 'ip addr ls' inside the guest 1116806275 M * badiane bcast ip mask nodev prefix 1116806287 M * Bertl nodev? 1116806307 M * badiane yep 1116806311 M * Bertl you're assigning the ip on the host before starting the vserver? 1116806337 M * badiane you said something about the possibility of bringing down the interface 1116806348 M * badiane if the vserver was brought down ??? 1116806362 M * badiane no 1116806366 M * Bertl with certain configurations, yep, that's possible ;) 1116806369 M * badiane no ip is assigned 1116806391 M * badiane no ip is assigned in the hosts interfaces file 1116806393 M * Bertl then I would replace the nodev by a dev entry with the interface name 1116806395 M * badiane but in the vserver config 1116806399 M * badiane I place the ip file