1113091437 M * Bertl also I didn't rediff the q0.14.1 as it applies quite fine 1113091537 A * micah tests a compile 1113091664 M * micah does 1.2.10 not have very many config options? I only saw the virtual root as an option 1113091814 M * Bertl no, that's about it, unless you ad q0.14 1113091817 M * Bertl *add 1113091831 M * Bertl this adds the tagging styles ... 1113091837 M * DaCa http://vserver.limehouse.org/patch-2.4.27-9-vs1.2.10.diff.gz 1113091859 J * Nik ~Nik@cable-153-130.online.bg 1113091860 M * Bertl ah, and now something to compare ... but not for me, I'm off to bed now ... 1113091866 M * DaCa also untested :) 1113091880 M * Bertl but thanks to everyone! I guess it was a very productive saturday night ... 1113091895 M * DaCa n8 1113091904 M * Bertl have a good whatever everyone! and cya tomorrow ... 1113091918 N * Bertl Bertl_zZ 1113091972 M * micah DaCa: I am doing a test compile of bertl's version now 1113091997 M * maharaja what does this mean: 1113091997 M * maharaja mesg: /dev/pts/2: Operation not permitted 1113092003 M * maharaja and there is no /dev/pts/* 1113092023 M * maharaja which step during the debian setup failed? (vserver mysql build -m debootstrap ....) 1113092051 M * micah DaCa: this is his patch: http://vserver.13thfloor.at/Stuff/Debian/patch-2.4.27-9-vs1.2.10-1.diff 1113092054 M * daniel_hozac maharaja: use SSH to login to the vserver. 1113092075 P * eamacnaghten 1113092098 M * Doener maharaja: /dev/pts/2 should be visible on the host... 1113092120 M * Doener that's your current pty on the host, and not visible in the vserver, thus the error message 1113092126 M * micah DaCa: sizes are slightly different: 1113092127 M * micah -rw------- 1 micah micah 157042 2005-04-09 19:05 patch-2.4.27-9-vs1.2.10.diff 1113092130 M * micah -rw------- 1 micah micah 156280 2005-04-09 18:45 patch-2.4.27-9-vs1.2.10-1.diff 1113092171 M * micah but... those size differences could be the result of different directory names that were diffed 1113092175 M * maharaja doener: it isn't 1113092185 M * maharaja doener: i guess that grsec does something wrong 1113092205 M * Doener maharaja: maybe, but that error message is normal for linux-vserver 1113092213 M * maharaja mhm 1113092214 M * maharaja k 1113092219 M * maharaja ill try ssh 1113092239 M * Doener you don't get a pty that's owned by the vserver when you use 'vserver xxx enter', known problems are: that very error message and that screen fails to work 1113092326 M * maharaja k 1113092722 Q * Nik Ping timeout: 480 seconds 1113092945 M * micah DaCa: the differences between Bertl's version and yours are here: http://people.debian.org/~micah/Bertl-vs-Daca.diff 1113093012 M * micah I made that with interdiff -p1, not sure if thats right 1113093061 M * micah the main differences appear in the misc.S/entry.S 1113093115 M * micah but also in fs/ext2/acl.c and ext2/xattr.c ext3/acl.c ext3/xattr.c 1113093162 M * micah I'm wondering if maybe you had the q0.14.1 applied to yours and he didn't? 1113093199 M * maharaja my devpts mount somehow is not working 1113093218 M * maharaja after staring the vserver 1113093224 M * maharaja i'm not seeing the mount on the host system 1113093249 M * Doener maharaja: 2.6 with alpha tools? 1113093254 M * maharaja yes 1113093272 M * Doener you probably don't see f.e. the vserver's proc mount either, right? 1113093280 M * Doener http://linux-vserver.org/Namespaces 1113093283 M * maharaja mhm 1113093345 M * maharaja true - no proc mount 1113093353 M * maharaja and no ifcofig interface 1113093356 M * maharaja *reading* 1113093358 M * micah something seems strange with those difference 1113093441 M * Doener ifconfig is old and only knows about named interface addresses. the alpha tools use the more recent netlink interface (via iproute) to setup the ip addresses. if you don't specify a name for the interface address, it won't get one and you won't see it via ifconfig, but "ip addr show" or just "ip a" will show it 1113093467 A * Doener .oO( maybe i should finally add that to the faq ) 1113093762 J * erwan_taf ~erwan@lns-vlq-39f-81-56-133-136.adsl.proxad.net 1113093892 M * maharaja doener: strange stuff that namespaces :) 1113093997 M * Doener maharaja: is that stuff i wrote about them ok the way it is? do you miss anything there? i wrote that out of boredom back then. it proved to be useful later, but i didn't get any feedback yet... (well, i didn't ask for either ;) 1113094232 M * maharaja ill double check 1113094242 Q * erwan_taf Ping timeout: 480 seconds 1113094248 M * maharaja while doing that - is vhashify a good thing? 1113094257 M * maharaja because i plan to setup multiple vservers on my current box 1113094264 M * maharaja to seperate services (mysql, mail, etc) 1113094283 M * maharaja and think about security (like md5 checksums) and space consumptation 1113094291 M * Doener i have no idea ;) 1113094318 M * Doener but i guess for space consumption it is quite a good idea... 1113094361 M * Doener IIRC i've heard that vhashify is for doing unification without a reference server 1113094648 M * maharaja so you never used it 1113094649 M * maharaja hum 1113094663 M * Doener that's right... 1113095111 M * maharaja i edited your namespaces document as i understood it 1113095120 M * maharaja there are no big changes 1113095128 M * Doener yep, saw it (wiki ml ;) 1113095128 M * maharaja only spelling and stuff 1113095132 M * maharaja :P 1113095141 M * maharaja somehow the wiki does not remember my name 1113095151 M * maharaja mhm 1113095153 M * maharaja ok, it does 1113095153 M * maharaja :) 1113095166 M * maharaja i would like the recent changes to be vice versa 1113095168 M * maharaja newest on top 1113095171 M * maharaja is that possible? 1113095193 M * maharaja and what time is the wiki server set to :) 1113095238 M * Doener hm? that used to be newest on top IIRC... but you got to ask Bertl_zZ anyways ;) 1113095264 M * Doener you can set your timezone in the preferences 1113095489 M * maharaja k 1113095491 M * maharaja ill go to sleep now 1113095492 M * maharaja cu 1113095493 M * Doener "Mount is only displaying the host system's information.", that's not namespace related, mount relies on /etc/mtab ;) /proc/mounts is what's interesting here... (i'll update that) 1113095497 M * Doener good night! 1113095512 M * maharaja k 1113095528 M * maharaja mhm 1113095534 M * maharaja but when i do cat /proc/mounts 1113095537 M * maharaja ill see the vserver mounts 1113095546 M * maharaja backup:/usr/local/lib/util-vserver# cat /proc/mounts|grep mysql 1113095546 M * maharaja none /vservers/mysql/proc proc rw,nodiratime,nodev 0 0 1113095546 M * maharaja none /vservers/mysql/tmp tmpfs rw,nodev 0 0 1113095546 M * maharaja none /vservers/mysql/dev/pts devpts rw 0 0 1113095546 M * Doener on the host? 1113095567 M * maharaja yepp 1113095571 M * Doener hm... that's strange then... 1113095599 M * maharaja anyways, lets talk about that tomorrow 1113095601 M * maharaja ill go to bed now 1113095605 M * Doener k 1113095814 M * maharaja updated another page to include links to the namespaces documents 1113096008 M * maharaja nite 1113096012 M * Doener g'nite 1113096754 M * micah Bertl_zZ: just a small note -- debian patches to the kernel aren't supposed to modify EXTRAVERSION in the Makefile, so I removed that piece 1113099528 J * Nik ~Nik@cable-153-130.online.bg 1113100012 Q * Nik Ping timeout: 480 seconds 1113105230 J * Psychik Psychik@82.224.20.161 1113105255 Q * Psychik Quit: 1113106731 J * Nik ~Nik@cable-153-130.online.bg 1113107212 Q * Nik Ping timeout: 480 seconds 1113113933 J * Nik ~Nik@cable-153-130.online.bg 1113114417 Q * Nik Ping timeout: 480 seconds 1113117745 M * Doener Bertl_zZ: here's my port ... http://www.13thfloor.at/~doener/vserver/patches/patch-2.6.12-rc2-vs1.9.5.12.diff ... at the others: that doesn't work with 0.30.205 and fakeinit vserver, the init process won't get killed on 'vserver xxx stop', i'm still working on a 'nice' solution... 1113121136 J * Nik ~Nik@cable-153-130.online.bg 1113121617 Q * Nik Ping timeout: 480 seconds 1113122918 M * maharaja which is the best way to clone a vserver? 1113124877 M * DaPhreak Morning ... 1113125131 M * maharaja hi phreak 1113125335 M * DaPhreak maharaja: well the simplest way would if you would do a 'rsync -av /path/of/your/old/vserver/ /path/of/your/new/vserver/' and the same to your config. The only thing you have to change is some stuff in your config (e.g. utsname, ips, contextnumber) 1113125453 M * maharaja ill give it a try 1113125454 M * maharaja thnx 1113125473 M * DaPhreak you're welcome 1113126701 Q * DaCa Ping timeout: 480 seconds 1113128461 J * duckx ~Duck@dyn-83-157-154-185.ppp.tiscali.fr 1113128622 Q * duckx Quit: 1113129924 J * erwan_ho ~erwan@lns-vlq-39f-81-56-133-136.adsl.proxad.net 1113133787 Q * alexx Quit: Bye 1113133919 J * DaCa ~danny@mail.limehouse.org 1113134107 Q * DaPhreak Quit: leaving 1113134430 J * DaPhreak ~phreak@lms.rz.uni-greifswald.de 1113140053 N * Bertl_zZ Bertl_oO 1113142513 M * SiD3WiNDR any known problems with proftpd? 1113142703 M * Doener bot that i'd know 1113142707 M * Doener s/bot/not/ 1113142726 M * SiD3WiNDR okay, then it's just proftpd acting up 1113142726 M * SiD3WiNDR hmm 1113142735 M * SiD3WiNDR indeed I do have proftpd on 2 vservers without problems 1113142737 A * SiD3WiNDR slaps himself 1113142745 M * Doener heh 1113142806 M * SiD3WiNDR problem solved :) 1113143996 J * Pazzo ~thomas@host130-250.pool8172.interbusiness.it 1113148479 J * muadda ~muadda@gar31-1-82-234-50-167.fbx.proxad.net 1113148870 M * muadda On Debian, when I install util-vserver, I get /etc/init.d/vservers-legacy 1113148900 J * kjo ~krischan@pD9E2DF14.dip.t-dialin.net 1113148915 M * muadda Is it Debian specific or is /etc/init.d/vservers-legacy written by the vserver team? 1113148919 M * Doener that's to start vservers that use a legacy configuration 1113148934 M * Doener that's upstream stuff, but debian folks may have modified it 1113148977 M * muadda And /etc/init.d/vservers-default is for the new directory based configuration ? 1113148995 M * Doener yes, for those having 'default' set as mark 1113149137 M * muadda echo "default" > /etc/vservers/vserver04/apps/init/mark 1113149200 M * muadda Doener, it works ;) thanks 1113149207 M * Doener you're welcome 1113150033 Q * kjo Quit: Verlassend 1113151375 N * Bertl_oO Bertl 1113151382 M * Bertl evening folks! 1113151444 M * Pazzo Hi Bertl! 1113151450 M * muadda hello ! 1113151466 M * Bertl hey Pazzo! muadda! 1113151536 M * Doener Bertl: just found that one http://weather.ou.edu/~apw/projects/stress/ 1113151539 M * Doener evening Bertl° 1113151548 A * Bertl is looking ... 1113151554 M * Bertl evening Doener! 1113151579 M * Bertl Doener: excellent ... 1113151581 M * Doener system stress tool... might be a nice replacement/addition for the cpuhog stuff we have 1113151604 M * Bertl yeah, best would be if we could integrate/combine it with my stress tool .. 1113151652 M * Bertl reminds me that I have to release it ;) 1113151696 Q * pusling Ping timeout: 480 seconds 1113151698 M * Bertl ah, it's already there, excellent ;) 1113151700 M * Bertl http://vserver.13thfloor.at/Experimental/TOOLS/vstress-0.03.tar.bz2 1113151701 M * Doener regarding my port to 2.6.12-rc2... that includes the clrsid stuff and thus needs modified tools... did you and rs find a solution for that? adding a call to "vkill --xid -s 9 1" to the stop script doesn't seem to work 1113151735 M * Bertl hmm, it doesn't? 1113151797 M * Bertl what happens with that? IIRC rs did exactly this ... plus a while loop which waits for the context to go away 1113151805 M * Doener no... init survives... but if i do the same manually it works... 1113151857 M * Bertl hmm, could you enable syscmd debugging and see if it calls the same cmd? 1113151886 M * Doener syscmd debugging? kernel option? 1113151923 M * Bertl echo 255 >/proc/sys/vserver/debug_switch 1113151941 J * sebd ~sebd@lesdeveloppementsdurables.org 1113151995 M * Bertl welcome sebd! 1113152003 M * sebd hi :) 1113152125 M * Doener building a qemu kernel now to test that... 1113152144 M * Bertl btw, it's interesting how similar challenges lead to similar solutions (looking at the stress output) 1113152221 M * Doener porting to 2.6.12-rc2 was pretty annoying btw ;) they introduced macros for rss accounting, so all those hunks had to be adapted... 1113152300 M * Bertl sounds interesting, did you enable the full monty (debuging wise) and run the -L tests yet? 1113152310 M * Doener hm, and my qemu build fails... checking... 1113152359 M * Doener hmm... #201 suddenly fails... 1113152368 M * Doener ah ok, xid 100 is in use... 1113152380 M * Doener so -L is working fine 1113152416 M * Doener debugging is turned off mostly, i just abused my desktop box to see if it works at all ;) 1113152431 M * Bertl well, -L doesn't make sense without debugging ;) 1113152467 M * Doener ah right... limit tests... 1113152498 M * Doener ok, bug in the nfs code was an easy one... forgot to rename a variable... 1113152792 J * pusling ~pusling@195.215.29.124 1113152830 M * Doener updated patch: http://www.13thfloor.at/~doener/vserver/patches/patch-2.6.12-rc2-vs1.9.5.12-1.diff 1113152867 M * DaPhreak oh .. bleeding edge :) 1113152902 M * Bertl hey DaPhreak! pusling! 1113152987 M * pusling hey 1113153025 M * DaPhreak morning Bertl :) or is it nearly good night ?! :-P your day/night rhythm is really strange :) 1113153063 M * Bertl yeah, well, I can't get everything right, can I? 1113153087 M * DaPhreak yeah, as nearly everyone on the world ;) 1113153172 M * Doener working: vxD: vc: VCMD_12_1[0], 10 1113153180 M * Doener now modifying the tools inside qemu... 1113153267 M * Doener not working: vxD: vc: VCMD_12_1[0], 10 1113153336 M * Doener hmm... the third run was succesful this time... 1113153643 M * Doener doing the vkill call twice seems to work as expected all the time... 1113153678 M * Bertl Doener: maybe it would be better to send -15 first then -9 1113153690 M * Bertl with half a second pause or so ... 1113153738 M * Doener hm, might be worth a try... i expected init not to care about SIGTERM 1113153746 M * Bertl and I think until the tools use the wait syscmd I had to implement instantly but which was never tested (AFAIK) you should do a loop checking for the /proc/virtual/ dir 1113153806 M * Bertl okay, SO is back, so I'm off for now .. back later (around 2200) 1113153817 M * SiD3WiNDR hmm, 195 years to wait for Bertl :p 1113153828 M * Doener heh 1113153831 M * Doener cya Bertl 1113153833 N * Bertl Bertl_oO 1113153863 M * Bertl_oO SiD3WiNDR: LOL it took some time .. ;) 1113155393 Q * ndim Ping timeout: 480 seconds 1113156036 M * trulux Bertl_oO: heya 1113156048 A * trulux was writing on physics, last section of the school paper! 1113156060 A * trulux desperately looking for someone skillful with povray 1113156082 A * trulux is being raytraced by a group of mad heffalumps in need of crack 1113156637 J * ndim hun@helena.bawue.de 1113156866 M * albeiro trulux here ? 1113156872 M * albeiro no, it can't be ;p 1113156892 M * albeiro i just eat too much 1113157580 J * Nik ~Nik@cable-153-130.online.bg 1113157583 M * Nik hi all 1113157955 M * trulux albeiro: haha, hey 1113158067 Q * Nik Ping timeout: 480 seconds 1113158105 M * albeiro nice to see you 1113158235 M * trulux albeiro: nice to see you too, but I would prefer to see a povray freak, I'm fscked up with this sun eclipse model 1113158361 M * albeiro hehe, not me 1113158798 M * trulux albeiro: If you were not albeiro, I could have kicked your ass so hard ;D 1113158813 A * trulux feels sad on the fscking raytracing work 1113158845 M * trulux what the fuck, they didn't asked me about it, but I have done all the work in LaTeX so, I can't do what others might be doing (drawing it by hand) 1113158849 M * trulux I need some solar powah 1113158852 M * trulux :D 1113159722 M * muadda I did something wrong on my vserver and my vserver don't have access to internet anymore from my vserver 1113159748 M * muadda are my capabilities enough? -- http://alban.blop.info/~alban/vserver/vserver04/ 1113159796 M * muadda I have "iptables -t nat -A POSTROUTING -o eth0 -s 10.0.0.0/24 -j MASQUERADE" 1113160207 N * Bertl_oO Bertl 1113160217 M * Doener wb Bertl 1113160221 M * Bertl thanks! 1113160235 M * Doener and good night folks... got to get up early tomorrow... 1113160235 M * Bertl muadda: sure you want to use MASQUERADE? 1113160240 N * Doener Doener_zZz 1113160243 M * Bertl night Doener_zZz! 1113160274 M * muadda Bertl, no, I'm not sure 1113160278 J * Nik ~Nik@cable-153-130.online.bg 1113160313 M * Bertl muadda: I'd a) make sure that you do not have 'old' rules lying around in that chain, then use -SNAT --to-source 1113160388 M * muadda Bertl, perfect ;) 1113160401 M * muadda it works 1113160449 M * Bertl the pleasure is mine! 1113160475 M * micah howdy Bertl! 1113160501 M * Bertl hey micah! hows' going? 1113160542 M * micah Bertl: pretty good, getting ready to leave tomorrow! 1113160558 M * Bertl what's the destination? 1113160560 M * micah Bertl: and its going well over there I assume? 1113160568 M * micah Bertl: i'm headed to seattle for work for two weeks 1113160578 M * micah west coast of the US 1113160588 M * Bertl cool area? 1113160623 M * micah its really nice there, very beautiful 1113160637 M * Bertl so you've been there before, right? 1113160652 M * micah yeah, my parents live out there, and I lived there for a while, so I know it some :) 1113160661 M * eyck Bertl is moving to US? 1113160676 M * Bertl evening eyck! and no ;) 1113160684 M * micah eyck: no ;) we were just talking about where I am going tomorrow! 1113160690 M * eyck evening, 1113160692 M * eyck wassup? 1113160694 M * micah Bertl: I compiled the 2.4.27-9-vs.1.2.10 patch you made, it compiled fine, I didn't get a chance to run any tests on it 1113160717 M * micah Bertl: DaCa finished his version and there are some differences that might be because one of you had the quota diff applied or maybe not 1113160750 M * micah I did an interdiff -p1 between your two patches which shows me these as the differences between your two patches: http://people.debian.org/~micah/Bertl-vs-Daca.diff 1113160779 M * micah i thought if you looked at it you would instantly know what caused the differences :) 1113160917 M * micah oops, I have to go, be right back 1113160973 M * SiD3WiNDR Bertl: well, I figured 2200 was the year, not the military time ;) 1113160975 M * SiD3WiNDR so 195 years =) 1113160989 M * SiD3WiNDR I guess there is no way of getting ipv6 inside a 1.2 :( 1113161297 M * Bertl SiD3WiNDR: not really ... 1113161315 M * Bertl micah: don't do interdiffs ;) 1113161338 M * Bertl just apply both patches to a clean source and diff the result ;) 1113161371 M * Bertl SiD3WiNDR: you want it for the vserver, not the host, right? 1113161418 M * SiD3WiNDR both.. :) 1113161421 M * SiD3WiNDR hehe 1113161443 M * SiD3WiNDR but it's not that urgent 1113161547 M * Bertl well, no chance with 1.2 inside a guest ... 1113161574 J * kalou ~kalou@AToulon-201-1-3-28.w81-53.abo.wanadoo.fr 1113161582 M * kalou good evening 1113161651 M * Bertl evening kalou! 1113161663 M * Nik Bertl: 1113161671 M * Bertl hey Nik! 1113161677 M * Nik Bertl: heya :-) 1113161712 M * Nik Bertl: I'm trying to shape the CPU time of some vservers and somehow I don't really get what i need 1113161752 M * Nik Bertl: currently trying with 2.6.11-vs1.9.5-rc2 1113161795 M * Bertl please elaborate ... 1113161810 M * Nik Bertl: it makes no difference how I set the schedulers 1113161826 M * Nik Bertl: e.g. vsched --xid 10005 --tokens-max 10 --tokens-min 2 --prio-bias 5 --interval 40 --fill-rate 10 1113161832 M * Bertl did you at least set one of the scheduler flags? 1113161841 M * Bertl sched_prio or sched_hard? 1113161854 M * Nik Bertl: and the default settings make no differende 1113161893 M * Bertl btw max=10 with a fill rate of 10 is extremely low 1113161911 M * Nik Bertl: CONFIG_VSERVER_PROC_SECURE=y 1113161939 Q * erwan_ho Remote host closed the connection 1113161955 M * Nik Bertl: default is 1. I want to see some difference, not 4 cpuhog's consuming 25%CPU each 1113161993 M * Nik Bertl: CONFIG_VSERVER_PROC_SECURE=y CONFIG_VSERVER_HARDCPU=y CONFIG_VSERVER_HARDCPU_IDLE=y 1113162070 M * Bertl please show me the output of /proc/virtual//status 1113162078 M * Bertl for one of those contexts 1113162115 M * Nik Bertl: # cat /proc/virtual/10005/status 1113162115 M * Nik UseCnt: 51 1113162115 M * Nik Tasks: 25 1113162115 M * Nik Flags: 0000000202000010 1113162115 M * Nik BCaps: ffffffffd44c04ff 1113162115 M * Nik CCaps: 0000000000000101 1113162115 M * Nik Ticks: 0 1113162128 M * Nik cat /proc/virtual/10001/status 1113162128 M * Nik UseCnt: 48 1113162128 M * Nik Tasks: 24 1113162128 M * Nik Flags: 0000000202000010 1113162128 M * Nik BCaps: ffffffffd44c04ff 1113162129 M * Nik CCaps: 0000000000000101 1113162129 M * Nik Ticks: 0 1113162155 M * Nik 10005 is modified, iirc 10001 - not 1113162219 M * Bertl #define VXF_SCHED_HARD 0x00000100 1113162227 M * Bertl #define VXF_SCHED_PRIO 0x00000200 1113162245 M * Bertl as I suspected and already mentioned, you have not set any of the scheduler flags 1113162285 M * Nik Bertl: ok. where should I set this? 1113162305 M * Bertl you can get the hard cpu scheduling if you do 1113162348 M * Bertl vattribute --set --xid --flag sched_hard 1113162371 M * Bertl or just by adding the sched_hard to the vserver config (see flower page) 1113162412 M * Bertl as a rule of the thumb, the max should be at least 2 times the fill rate (better 5-10 times) 1113162484 M * Nik Bertl: ok, 10x. Just one more thing... how can I make the vservers show the virtual load average, their own uptime etc? 1113162517 M * Bertl with the 'other' flags ;) 1113162527 M * Bertl http://savannah.nongnu.org/cgi-bin/viewcvs/util-vserver/util-vserver/lib/cflags-v13.c?rev=HEAD 1113162536 M * Bertl (link from the flower page) 1113162547 M * Nik Bertl: ok :-) 10x a lot 1113162556 M * Bertl you're welcome! 1113162717 Q * monrad Ping timeout: 480 seconds 1113162961 M * kalou Bertl: I'd love to have /proc/virtual//status giving VC_VXF_* instead of 0000000202000010 ... 1113163015 M * kalou Bertl: I guess it would be better for all, is it ? 1113163027 Q * Nik Ping timeout: 480 seconds 1113163038 M * Bertl kalou: that's a job for userspace ... 1113163050 M * kalou Bertl: hmm. oki 1113163063 M * Bertl doesn't really make sense to bloat the kernel with what a simple script can do 1113163122 M * kalou Bertl: I was looking for some useful but simple coding to do in kernel space ... 1113163149 M * kalou Bertl: (did not even write a line in kernel space yet, and no more than 100 lines of C in my whole life) 1113163190 M * kalou Bertl: If you have ideas for such jobs, I'd be pleased to help 1113163227 M * Bertl well, we have the kernel todo page 1113163242 M * Bertl and there is the linux kernel janitors stuff ... 1113163281 J * monrad ~monrad@213083190130.sonofon.dk 1113163318 M * Bertl kalou: http://linux-vserver.org/ToDo+List+Kernel 1113163366 M * Bertl kalou: what about the stack usage part for example ... 1113163402 M * Bertl evening monrad! 1113163410 M * kalou Bertl: sorry to ask, but what are the janitor patches for ? a projet held by kernel mentors ? 1113163465 M * Bertl http://janitor.kernelnewbies.org/TODO 1113163473 M * kalou Bertl: ok .. I'll take a look at stack usage .. Will remind me of my old HP48 days :) 1113163562 M * kalou Bertl: thank you for the janitor project.. I understand that these are easy jobs that have to be done on the kernel, but no one has time for them, probably 1113163596 M * Bertl yes, precisely and they are a good way to start hacking the kernel 1113163630 M * kalou WooowwwwwWW next days are gonna be ... sunny ;-) 1113163645 M * Bertl but as I said, feel free to work on linux-vserver it's just a little more challenging ... 1113163859 M * albeiro kernel hacking is interesting - once you start you are in it for live ;] 1113164977 J * Nik ~Nik@cable-153-130.online.bg 1113165462 Q * Nik Ping timeout: 480 seconds 1113165701 J * sOCiETY ~society@p549AF3FB.dip.t-dialin.net 1113165703 M * sOCiETY hi 1113165722 M * sOCiETY i wrote to the mailing list that i have a problem with the kernel 2.6.11.5 :=) 1113165740 M * sOCiETY can any1 help me out? 1113165768 M * sOCiETY I use the latest patch for 2.6x kernel 1113165779 M * sOCiETY and also the latest util-vserver 1113165948 M * monrad evening 1113166598 M * TheSeer sOCiETY: shoot 1113166626 Q * trulux Read error: Connection reset by peer 1113166757 M * Bertl sOCiETY: you're the one with x86 and amd64? 1113166941 M * sOCiETY yep 1113166941 M * sOCiETY gi 1113166943 M * sOCiETY hi 1113166980 M * Bertl did you try with a different compiler? 1113166997 M * sOCiETY i used this compiler on the 2.6.10 too 1113167002 M * sOCiETY with this it work 1113167015 M * sOCiETY i compare the syscall 1113167018 M * sOCiETY they are right 1113167048 M * Bertl so did you try to boot the 2.6.10 kernel? 1113167056 M * sOCiETY yep 1113167066 M * Bertl and it works quite fine, with the same tools? 1113167071 M * sOCiETY do not work with the new util-vserver but with the old 1113167079 M * Bertl ehe? 1113167087 M * sOCiETY because of the compiling 1113167096 M * Bertl I'm listening? 1113167104 M * sOCiETY do you speak german? 1113167125 M * Bertl yes, but the channel ist english .. 1113167132 M * sOCiETY i know :) 1113167137 M * sOCiETY but for me is easer in german *G* 1113167141 M * Bertl (if you really have to, please msg me in private) 1113167148 M * sOCiETY ok :) 1113168983 M * micah Bertl: ok, going to apply both patches as you recommend 1113169031 M * Bertl yeah, and just make a diff -NurpP between the results 1113169034 M * Bertl wb, btw 1113169566 M * micah thanks :) 1113169569 M * micah its been a long day 1113169583 M * micah ok... http://people.debian.org/~micah/bertl_vs_daca.diff 1113169656 M * Bertl hmm, okay, could you redo it with -NurpbBP ? 1113169683 M * Bertl (that will remove the whitespace changes) 1113169751 M * micah sure 1113169800 M * sOCiETY good night all 1113169801 M * sOCiETY cya 1113169809 Q * sOCiETY Remote host closed the connection 1113169983 M * DaCa evening all 1113170005 M * micah ok, new version up 1113170006 M * Bertl evening DaCa! 1113170008 M * DaCa I hope I didnt mess up too much in that patch :) 1113170012 M * micah http://people.debian.org/~micah/bertl_vs_daca.diff 1113170020 M * micah DaCa: we're just looking at the differences now 1113170024 M * micah between yours and bertl's 1113170112 M * Bertl micah, DaCa: did you compile test any of those? 1113170177 M * DaCa for the xattr syscalls in ppc64, I thought I saw Bertl already removed them in his patch (they where conditioned with #if 0 initially) 1113170230 M * Bertl well, no problem with the ppc syscalls ... 1113170250 M * DaCa the other significant difference I see is using IS_IMMUTABLE vs IS_IMMUTABLE_FILE, which I honestly don't understan where it comes from 1113170260 M * micah Bertl: I compile tested yours, I hadn't had a chance to with DaCa's 1113170280 M * DaCa I didnt have the chance to compile/test either 1113170366 M * Bertl np 1113170402 M * Bertl I'd go for my patch with two or three changes from DaCa's version 1113170409 M * Bertl explicitely: 1113170421 M * Bertl kernel-source-2.4.27_daca/fs/xfs/xfs_dinode.h 1113170442 M * Bertl kernel-source-2.4.27_daca/Makefile 1113170520 M * Bertl yeah, that's it ... 1113170574 M * Bertl micah: I'll upload a revised patch ... 1113170602 M * Bertl DaCa: thanks for your help, it shows that the patches should be fine ... 1113170633 M * Bertl the udp is the only one I have to check ... 1113170670 M * DaCa the udp one I didnt do manually, its done with wiggle 1113170717 M * Bertl it looks a little broken to me ... 1113170748 M * micah Bertl: are these with quota support? 1113170762 M * Bertl nope, but the quota patch is fine ontop of that 1113170762 M * micah I want to make sure they also have the quota pieces because debian users expect them 1113170796 M * Bertl there should be two patches one without and another with quota IMHO (s/patches/kernels) 1113170834 Q * kalou Read error: Connection reset by peer 1113170868 M * micah I need to learn how to do this so you can get back to the coding ;) 1113170911 M * Bertl sure, help is always appreciated ... 1113170976 M * micah Bertl: ok, thats a good idea, I'll take the one that you are going to put up as the one without quota, then I'll add the quota patch to that and provide that in addition 1113170997 M * Bertl that's the idea ... 1113171023 M * Bertl quota is still very intrusive, so if somebody doesn't need it, why have it in the kernel ... 1113171034 M * micah good point 1113171098 M * Bertl btw, the quota patch applies perfectly, not even a changed line number 1113171214 M * micah excellent 1113171218 M * Bertl http://vserver.13thfloor.at/Stuff/Debian/patch-2.4.27-9-vs1.2.10-2.diff 1113171225 M * Bertl http://vserver.13thfloor.at/Stuff/Debian/patch-2.4.27-8-vs1.2.10-q0.14.1.diff 1113171283 M * Bertl btw, no need to depend on the 'stable' tools, the alpha tools work quite fine with 1.2.10 ... 1113171319 M * micah i notice the one with the quota added on is -8, not -9? 1113171346 M * Bertl yep, that's an older one, but as I said, it applies without slightest deviation 1113171355 M * micah ah, I misunderstood 1113171362 M * Bertl feel free to apply and rediff it ... 1113171364 M * micah when you say no need to depend on the stable tools --- what are you referring to? 1113171378 M * Bertl util-vserver-0.30 1113171398 M * Bertl 0.30.20x works fine with 1.2.10 (and older kernels) 1113171412 M * micah ah, thats not my package to mess around with yet :) 1113171421 M * Bertl of course you can also use the older tools with newer kernels 1113171437 M * Bertl yeah, just thought to mention it, don#t know the dependancy system of debian 1113171441 M * DaCa its ola's package iirc 1113171454 M * micah although, I dont see the dependency you are talking about 1113171468 M * Bertl that's the point, there is none ;) 1113171507 M * micah oh! you mean I should add a dependency on the kernel-patch for util-vserver? 1113171511 M * DaCa kernel-patch-vserver only suggest util-vserver 1113171537 M * Bertl micah: well, that depends on what you want to get ... 1113171557 M * Bertl a) 2.4/1.2 with 0.30 1113171565 M * Bertl b) 2.4/1.2 with 0.30.20x 1113171571 M * Bertl c) 2.6/1.9 with 0.30 1113171579 M * Bertl d) 2.6/1.9 with 0.30.20x 1113171603 M * DaCa there is only 0.30.20x in sarge and sid atm 1113171604 M * Bertl a) and c) works like it did before 1.3/1.9 1113171621 M * Bertl b) supports legacy and works quite fine 1113171644 M * Bertl d) supports legacy (if configured) and allows for the new features 1113171665 M * Bertl so I could very well live with 0.30.20x being the one and only tools ;) 1113171679 M * micah i believe that is what we have no 1113171680 M * micah now 1113171689 M * Bertl good, then! 1113171698 M * micah because 0.30.20x is the version of util-vserver that will be shipped 1113171710 M * micah and we made patches for 2.6/1.9 and 2.4/1.2 1113171746 M * micah DaCa: right? 1113171817 M * DaCa micah: indeed 1113173602 Q * muadda Quit: Leaving 1113173656 M * micah Bertl: maybe you didn't see when I said before because you were gone, but Debian doesn't allow for modification of EXTRAVERSION in the Makefile 1113173678 M * Bertl yes, you have to remove it ;) 1113173680 M * micah so I am removing that from http://vserver.13thfloor.at/Stuff/Debian/patch-2.4.27-9-vs1.2.10-2.diff and did so for the 2.6.8-15 patch 1113173695 M * micah err, well, I am not removing it from your URL, but I thought you might wish to 1113173709 M * Bertl will confuse everybody .. but if that is what debian wants ;) 1113173763 M * micah Bertl: I think its because when the debian kernels are built, they add their own extraversion so they conflict 1113173790 M * micah thats my guess anyways :) 1113173791 M * Bertl well, that's what the kernel config option is for ;) 1113173927 J * ciphernaut ~a@61.88.18.130 1113173972 M * Bertl evening ciphernaut 1113173977 M * Bertl +! 1113174015 M * ciphernaut hi Bertl, all 1113174049 M * ciphernaut do people mess up the wiki reguglarly? 1113174061 M * ciphernaut 1113174243 M * Bertl well, no actually not that often ... 1113174600 M * Bertl ciphernaut: any reason for asking? 1113174677 M * ciphernaut just wondering 1113174691 M * ciphernaut I removed the casino links 1113174714 M * Bertl hmm, where? 1113175176 Q * ciphernaut Ping timeout: 480 seconds 1113175313 J * ciphernaut ~a@61.88.18.130 1113175325 M * Bertl wb ciphernaut! where did you remove them? 1113175464 M * ciphernaut they were tacked onto the bottom of the page 1113175491 M * Bertl do you have an url? 1113175516 M * ciphernaut one of them was http://online-roulette.myplace.com/, the others also started with online- 1113175527 M * Bertl ahem, of the wiki page ;) 1113175558 M * ciphernaut http://vserver.strahlungsfrei.de/tiki-index.php?page=FrequentlyAskedQuestions 1113175573 M * Bertl ah, different wiki, that explains it ;) 1113175582 M * ciphernaut hmm 1113175589 M * ciphernaut another wiki 1113175604 M * ciphernaut which one are you refering too? 1113175615 M * Bertl the linux-vserver.org wiki