1128729610 M * gndmstr i have a definite need to do it so i can free up an entire machine that sits on 24/7 to get used a few times a month
1128729619 M * Bertl remote X should be fine, just use gdm without hw X?
1128729642 M * gndmstr it only starts a grey x window with no programs and kdm doesnt load
1128729681 M * gndmstr but if x is running via starting xdm init script, then it works just fine, but its also waiting on a tty
1128729701 M * Bertl Hollow: still around?
1128729878 J * antagonist ~antagonis@ip68-13-182-17.om.om.cox.net
1128729894 M * Bertl welcome antagonist!
1128729922 M * antagonist thanks!
1128729972 M * antagonist i have a question for you folks... most of the documentation i refer to mentions that there'll be a vhost specific configuration file under /etc/vservers/ ... but all i have is a directory tree for each instead of a .conf
1128729987 M * antagonist is this just because i'm using gentoo and a gentoo package?
1128730007 M * Aiken gndmstr this the line from my kdmrc that issues the login to the clients with no local X -> StaticServers=fred.bedrock:11,bambam.bedrock:0
1128730013 M * gndmstr the directory tree and various files inside it are the configs
1128730018 M * mnemoc antagonist: .conf is the legacy format
1128730024 M * gndmstr ahh
1128730025 M * gndmstr ok thanks
1128730033 M * antagonist ah, thank you
1128730057 M * mnemoc antagonist: take a look into the weed page
1128730072 M * antagonist my other question... how do i properly setup a gateway within a vhost?  i've tried 'route add default gw <ip>' to no avail
1128730078 M * antagonist mnemoc, aye, i saw that
1128730085 M * antagonist i just didn't read it closely enough i suppose
1128730087 J * Nicoli ask@28.burningd.enterconnect.net
1128730090 M * Bertl antagonist: util-vserver _also_ supports the legacy .conf but it won't create them ...
1128730094 M * antagonist ah
1128730110 M * antagonist i'm fine with the current way of doing things
1128730118 M * antagonist i was just wondering what i'm missing
1128730119 M * Bertl antagonist: networking is on the host, you do not setup anything regarding networking on the guest (for now)
1128730122 M * antagonist ah
1128730123 M * antagonist hmm
1128730123 M * gndmstr are you putting 2 cards in or 2 networks on one card
1128730128 M * antagonist i have 2 cards
1128730134 M * antagonist and eth1 is the outside world
1128730136 M * Bertl antagonist: you need a second gateway setup?
1128730139 M * antagonist and eth0 is my internal net
1128730142 M * antagonist yes
1128730144 M * antagonist something like that
1128730149 M * gndmstr use iproute2 module in your net file
1128730177 M * gndmstr i have a complete setup for adding route2 gateways and rules for almost any number of nics and networks
1128730180 M * Bertl antagonist: do you have a real second network, or do you actually want to send packages to the outside via the first interface?
1128730182 M * gndmstr took days to find all the info
1128730184 M * antagonist so i just need to route my vhosts alias on the host end
1128730208 M * antagonist i want to masquerade traffic from the vhosts to my router
1128730223 M * Bertl ah, I see .. well actually you do not want to masquerade anything
1128730226 M * antagonist ah
1128730235 M * Bertl you want to SNAT (source nat) packets
1128730237 M * antagonist aye
1128730244 M * Bertl something like:
1128730249 M * antagonist and i want to use DNAT to route incoming traffic to each individual server
1128730279 M * Bertl iptables -t nat -I POSTROUTING -s <guest ip> -j SNAT --to <public ip>
1128730313 M * Bertl with incoming traffic it's a little trickier, if you use a single public ip
1128730330 M * Bertl you have to 'assign' certain ports for certain guests
1128730351 M * Bertl (but the principle is the same)
1128730358 M * antagonist aye
1128730371 M * antagonist what i've done so far, though
1128730394 M * antagonist is iptables -t nat -A POSTROUTING -s 192.168.1.0/24 -j MASQUERADE
1128730399 M * antagonist that i will remove
1128730411 M * antagonist and i'll change the end to SNAT --to 192.168.0.1
1128730416 M * antagonist which is my external interface
1128730421 M * antagonist that goes to yet another router
1128730426 M * antagonist then to my cable modem
1128730430 M * antagonist one sec, lemme try
1128730431 M * Bertl ah, ic. yeah, that should work
1128730496 M * Bertl the thing is, masquerading applies only to forwarded/routed packets, but the packets are created 'on' the host so you need the SNAT
1128730537 M * antagonist my problem is name resoluting isn't working
1128730549 M * antagonist but the resolv.conf in my vhost is exactly the same as my resolv.conf on the host
1128730567 M * Bertl once the SNAT is active, it should work ...
1128730569 M * antagonist my vhost can't reach anything outside of the 192.168.1.0 subnet
1128730569 M * antagonist ah
1128730571 M * antagonist hmm
1128730579 M * antagonist maybe i didn't issue the right iptables command
1128730625 M * antagonist if the packets are originating on one interface (eth0) and i want them to go out another (eth1), what should i do?
1128730633 M * antagonist i issued 'iptables -t nat -L'
1128730646 M * antagonist and it returned one rule: SNAT       all  --  192.168.1.0/24       anywhere            to:192.168.0.1
1128730652 M * antagonist under the postrouting chain
1128730661 M * antagonist all policies are currently ACCEPT
1128730664 M * daniel_hozac as the first rule?
1128730671 M * Bertl that looks good ... let's try a few things 
1128730685 M * antagonist ip_forward is set to 1 in sysctl
1128730689 M * antagonist okay
1128730699 M * Bertl antagonist: ping -c 1 www.google.com
1128730707 M * Bertl (on the host)
1128730725 M * antagonist that works fine
1128730748 M * Bertl what's your guest ip?
1128730758 M * antagonist 192.168.1.10
1128730763 M * Bertl ping -I 192.168.1.10 -c 1 www.google.com
1128730767 M * Bertl (also on the host)
1128730790 M * antagonist okay, that's hanging
1128730796 M * antagonist 0 received
1128730811 M * antagonist it's my iptables rules somehow then?
1128730811 M * Bertl ping -I 192.168.0.1 -c 1 www.google.com
1128730814 M * antagonist okay
1128730847 M * antagonist well, that's my external router's address
1128730851 M * antagonist i substituted .1.1
1128730854 M * antagonist and that hung too
1128730859 M * antagonist that's the actual address of eth0
1128730863 M * Bertl ah, what is _your_ address?
1128730869 M * antagonist eth0 is 192.168.1.1
1128730894 M * Bertl okay, so you probably have 192.168.0.0/16 then with the router?
1128730896 M * antagonist eth1 (home network) 192.168.0.3, dynamically assigned from my home router, 192.168.0.1
1128730902 M * antagonist it's /24
1128730919 M * Bertl so what is your router's ip to the world?
1128730947 M * antagonist 68.13.182.17
1128730950 M * antagonist sekt0r.info
1128730952 M * Bertl or better, what is your address which is routed via that router?
1128730977 M * Bertl (hint: it's the address your eth0 has)
1128730986 M * antagonist 192.168.0.1
1128730989 M * antagonist sorry
1128730990 M * antagonist .3
1128730995 M * antagonist that's the one that goes to that router
1128730999 M * antagonist hrm
1128731001 M * daniel_hozac you just said that was eth1.
1128731009 M * antagonist eth1 is to the router
1128731010 M * antagonist yes
1128731015 M * Bertl ah, so let's try: ping -I 192.168.0.3 -c 1 www.google.com
1128731021 M * antagonist that succeeds
1128731025 M * Bertl good!
1128731031 M * Bertl then your SNAT rule is:
1128731047 M * antagonist doh
1128731050 M * antagonist i just realized what i did
1128731057 M * antagonist i put .0.1 in the SNAT rule
1128731058 M * antagonist sorry
1128731066 M * Bertl :)
1128731070 M * Bertl so I save the paste :)
1128731082 M * antagonist indeed
1128731093 M * antagonist and i have connectivity now
1128731097 M * antagonist thank you much
1128731107 M * antagonist i guess i need to read an iptables primer
1128731111 M * Bertl you're welcome!
1128731112 M * antagonist and i need to label my interfaces
1128731113 M * antagonist heh
1128731122 M * antagonist i've got a convoluted configuration right now
1128731152 M * antagonist awesome.
1128731183 M * antagonist well i'll have to come back and visit more often... i'm just now starting to use vserver, and i'm getting a real kick out of it.
1128731195 M * antagonist you folks have a good night/day... i'm out.
1128731200 M * Bertl feel free to stay/visit here as often as you like
1128731207 M * antagonist thank you.
1128731213 Q * antagonist Quit: Leaving
1128731811 M * Bertl gndmstr: okay, so could we try the find and start/stop qmail guest again?
1128731888 M * gndmstr ok i did it at 5:30 so its been 3 hrs
1128731935 M * Bertl and could you check the /proc/virtual/status info for me?
1128731975 M * gndmstr both are at 3 right now with everything running
1128731980 M * Bertl excellent!
1128732013 M * gndmstr i cant find where i put that find command
1128732025 M * Bertl np, sec
1128732093 M * gndmstr hmm after 3 hrs its not much.. meminfo shows  Cached:          75864 kB
1128732131 M * Bertl find / -xdev -type f -exec ls {} \; >/dev/null
1128732168 M * gndmstr ok running
1128732245 M * Bertl but I'm optimistic that the restructuring avoids the issue
1128732272 M * gndmstr the first 3 hrs looked like it. didnt grumble at all over anything
1128733349 M * gndmstr find is still running
1128733378 M * gndmstr still close to what it was
1128733381 M * gndmstr Cached:          76236 kB
1128733490 J * Johnsie ~john@acs-24-154-53-217.zoominternet.net
1128733727 M * Bertl gndmstr: hmm, maybe the ls is a little extreme ...
1128733734 M * Bertl let's change that to:
1128733761 M * Bertl find / -xdev -type f -print0 | xargs -0 ls
1128733769 M * Bertl oops
1128733775 M * Bertl with >/dev/null
1128733954 M * gndmstr find / -xdev -type f -print0 | xargs -0 ls >/dev/null  right?
1128733974 M * Bertl yep
1128734069 Q * yarihm Quit: Leaving
1128734188 M * gndmstr ok its done
1128734202 M * gndmstr shut individuals down or just reboot the whole mess
1128734218 M * Bertl you#re the crash expert :)
1128734253 M * gndmstr ok individuals
1128734347 M * gndmstr this time the first one, name server timed out but it says it stopped and it left no pid
1128734402 M * gndmstr it took the radius server down with it
1128734416 M * gndmstr but there is no pid left for the radius either and there were no messages it was going down
1128734460 M * gndmstr qmail guest shut down cleanly
1128734482 M * gndmstr rebooting
1128734499 M * Bertl okay, how are the name/radius connected?
1128734526 M * Bertl (because you say: it took the radius server down with it)
1128734532 M * gndmstr they arent connected
1128734539 M * gndmstr the radius server just stopped silently
1128734572 M * Bertl hmm ... I'd inspect that, as the guests do not share anything with eachother (at least not in a typical setup)
1128734588 M * gndmstr i get that a lot when a guest times out
1128734601 M * gndmstr i got this error when it shut down to reboot bu tit rebooted cleanly anyway
1128734602 M * gndmstr * Remounting remaining filesystems readonly ...vxW: !!! limit: f541e050[NPROC,6] = 3 on exit.
1128734602 M * gndmstr vxW: !!! cvirt: f541e27c[nr_threads] = 3 on exit.
1128734636 M * Bertl hmm ... yeah, that's possible, I forgot about that ...
1128734690 M * Bertl well, let me put it this way, if a guest shutdown results in 'killing' other guests then they are either somehow connected or something is broken
1128734716 M * gndmstr could be broken with the crashes.. 
1128734844 M * Bertl I'd suspect some kind of interconnection, maybe same xid or liks pointing to the wrong guest?
1128734862 M * Bertl in your histor dump, I only see xid=3405 and 3905
1128734879 M * gndmstr 3904 is ns
1128734888 M * gndmstr i can see that in vps
1128734895 M * Bertl okay ..
1128734902 M * gndmstr and it shows in vservers.rev
1128734928 M * gndmstr been getting one other thing odd
1128734938 M * gndmstr when i see the prometheus init msgs after this one
1128734941 M * gndmstr INIT: no more processes left in this runlevel
1128734948 M * gndmstr i have to press enter to get the login prompt
1128734960 M * gndmstr then when i try to log in i get a timeout then i can log in
1128734975 M * gndmstr apollo login: root
1128734975 M * gndmstr root
1128734975 M * gndmstr     Password: Login timed out
1128734980 M * gndmstr apollo login: root
1128734980 M * gndmstr Password:
1128734980 M * gndmstr Last login: Fri Oct  7 17:30:15 on tts/0
1128734981 M * gndmstr apollo ~ #
1128735022 M * gndmstr this is on the serial console
1128735026 M * gndmstr ssh is fine
1128735066 M * Bertl hmm, that's probably a minor issue with the mgetty
1128735068 M * gndmstr ok so this  time the nameserver timed out, neither of the others did they went fine reboot went fine with that one msg
1128735085 M * gndmstr yeah i think once i figure out how to shut that thing up it will all be ok
1128735106 M * gndmstr and really that doesnt matter since ill be building a totally new guest for each anyway
1128735123 M * gndmstr just would be nice to figure out why its doing it someday
1128735145 M * gndmstr something must have corrupted since it just started all by itself after a particularly nasty crash
1128735160 M * Bertl okay, I'll update the patch once again, to handle the warning
1128735188 M * gndmstr i tried reinstalling the base layout but that didnt help. thinking of reinstalling the host tools in case its in there
1128735190 M * gndmstr ok
1128735196 M * Aiken Bertl I have a few hours today if you still want me to try stuff
1128735228 M * Bertl Aiken: great! I'd appreciate a test with some different guests (on alpha or x86)
1128735263 M * Aiken this x86 is 2.1.0-pre5 and the alpha is 2.1.0-rc3 so either
1128735269 M * Bertl Aiken: also, I have a patch which adds a config option for xid propagation
1128735288 M * Bertl which could need some testing ...
1128735335 M * Bertl s/need/use/?
1128735424 M * Aiken ok
1128735522 M * Bertl @all native speakers, please correct me whenever I'm talking nonsense ...
1128735553 M * Bertl (and yes, I consider australian folks native speaker :)
1128735579 M * mnemoc who is a native speaker here?
1128735607 M * Aiken at last austalians know how to spell words like  colour
1128735719 M * gndmstr btw Bertl that phrase was correct
1128735747 A * Aiken does not have a problem with it as well
1128735788 M * gndmstr hehe but for spelling, i go 'round and 'round with my bud in britain about spellings and phraseology
1128735809 M * gndmstr literally both are correct depending on the dialect spoken
1128735815 M * gndmstr american or queen's english
1128735834 M * Aiken the proper one, the 2nd one :)
1128735850 M * gndmstr LOL im leaving that one alone
1128735879 M * gndmstr basically it is proper for that half of the world
1128735880 M * gndmstr :)
1128735923 M * Bertl okay, basically what I would like to get is some feedback if I 'create' new grammar and such (or leave out endings, plurals, whatever)
1128735936 M * gndmstr ok
1128735947 M * gndmstr i do it all the time just because i've gotten lazy in my old age
1128735953 M * Bertl that's something you do not imemdiately realize as non-native speaker ...
1128735959 M * gndmstr true
1128735966 M * Aiken so do I, I blame the keys moving around on my keyboard
1128735974 M * gndmstr hehe yep me too
1128735977 M * Bertl and if you get it wrong a few times, you stick to it sooner or later ...
1128735987 M * gndmstr exactly
1128736007 M * gndmstr i have started using cause instead of because without taking the trouble to type any '
1128736007 M * Bertl so I definitely prefer to be corrected :)
1128736069 Q * Johnsie Quit: G'bye!
1128736094 M * Bertl (I also have absolutely no problem with 'suggestions' how to say something better, but those should come via private /msg to avoid cluttering the logs)
1128736147 M * Aiken Bertl I thought you were ok with english, certainly no worse that a lot of native english speaking people on irc
1128736173 M * gndmstr absolutely. i have not noticed anything at all since i started here
1128736198 M * gndmstr and being an english major in high school i just seem to catch those things even when i dont want to
1128736217 M * gndmstr correction.. having been an english major
1128736219 M * gndmstr :_)
1128736237 M * Bertl well, there is always space to improve ...
1128736244 M * gndmstr absolutely
1128736336 M * Bertl my pronunciation is still horrible (well, that's nothing we can improve here :)
1128736388 M * gndmstr it is difficult as a second language when the first is so widely different
1128736419 M * gndmstr if the rules were similar it would not be so hard but since they are not... 
1128736437 M * Bertl I'm not good at languages at all .. and unfortunately english wasn't _that_ interesting when I was in school :/
1128736465 M * gndmstr be thankful you don't have to learn afrikans
1128736488 M * gndmstr thats a mixture of 11 totally disconnected languages that have similar roots but have varied widely over the years
1128736504 J * Johnsie ~john@acs-24-154-53-217.zoominternet.net
1128736508 M * Bertl similar to the indian language desaster?
1128736529 M * gndmstr probably
1128736558 M * gndmstr i only know a few it is made from, dutch, german, zulu and some others
1128736595 M * gndmstr oh .. for correction
1128736598 M * gndmstr its disaster
1128736608 M * Bertl k, tx
1128736640 M * gndmstr many pronounce it desaster but.. it is also incorrect 
1128736671 M * gndmstr first 3 letters sound like disease
1128736789 M * Bertl it's desaster in german/austrian ... a funny thing in german is that package and packet both are translated to 'Paket'
1128736812 M * Bertl this always causes confusion when I mean packet and say package :)
1128736836 M * gndmstr heh i took one month of german in school and switched quickly to latin instead
1128736839 M * gndmstr hehe yeah
1128736918 Q * tchan Quit: leaving
1128736973 J * tchan ~tchan@c-67-174-18-204.hsd1.il.comcast.net
1128737718 M * Bertl okay, test compiling here, patch should be available shortly
1128737772 M * gndmstr ok
1128737800 M * Aiken the delay suited me fine, just got back inside from dealing with the tv antenna, 5 hours of car racing and the tv antenna has to play up
1128737927 M * gndmstr gonna walk dog be back in 10
1128738121 M * Bertl Aiken: I switched from terrestrial reception to satellite some years ago ... never had good reception before ...
1128738198 M * Aiken all free to air here from about 50 - 60 miles away
1128738254 M * Aiken far enough away to show how badly digital handles fading signals
1128738539 M * gndmstr i wont use satellite here.. with our weather in the summer it would be out as much or more than in :)
1128738557 M * gndmstr if i lived farther north yes
1128738592 M * gndmstr but in northern tropics, it does not pay since our winter is only a few weeks long and the rest of the time we have almost daily tstorms
1128738622 M * Aiken I know someone only a few km from me that watches satellite but I can not be bothered
1128738638 M * Aiken except for the racing I prefer our dvd collection
1128738648 Q * tchan Quit: WeeChat 0.1.6-cvs
1128738662 M * Bertl it's pretty optimal in europe ... excellent picture quality, maybe one or two incidents a year (right before heavy thunderstorms) absolutely not impressed by snow :)
1128738671 M * gndmstr for what little i watch tv, the 3 channels i can pick up with my indoor antenna is fine.. mostly only watch the news and local emergency weather channel
1128738699 J * tchan ~tchan@c-67-174-18-204.hsd1.il.comcast.net
1128738771 Q * tchan Quit: 
1128738818 J * tchan ~tchan@c-67-174-18-204.hsd1.il.comcast.net
1128738840 M * Bertl wb tchan! reconfiguring your work station?
1128738895 M * tchan sorry about all the quits and joins, testing an irc client (weechat) was using irssi
1128738940 M * Bertl np, wasn't irssi good enough?
1128738984 M * tchan irssi works great for me for a long time.  one of my irc buddies just wanted me to try weechat and it seems nice as well
1128739132 J * RoT ~bob@203-59-118-101.dyn.iinet.net.au
1128739141 M * Bertl gndmstr: patch against vanilla kernel is fine?
1128739160 M * RoT Hi, I am having a problem if anyone can help please?
1128739169 M * Bertl welcome RoT!
1128739178 M * RoT hi :)
1128739190 M * gndmstr ok
1128739206 M * RoT I am getting an error Unknown symbol vx_rmap_pid when trying to load some modules, this will be a show stopper for me :(
1128739215 M * Bertl ah, openAFS?
1128739223 M * RoT no bestcrypt
1128739237 M * Bertl ah, okay ... addon module?
1128739253 M * RoT yes, but also in gentoo portage
1128739311 M * Bertl okay, could you upload the relevant code section to pastebin.com (or similar)?
1128739328 M * Bertl i.e. the file which complains on kernel build :)
1128739348 M * RoT well actually, there are no errors or warnings during the build
1128739355 M * RoT I checked that already
1128739362 M * Bertl there are, trust me ...
1128739371 M * Bertl but the final link section tells the file too
1128739389 M * RoT well, nothing that mentions vx_rmap_pid
1128739391 M * Bertl Aiken, gndmstr: http://vserver.13thfloor.at/Experimental/patch-2.6.13.3-vs2.1.0-rc4.diff
1128739471 M * gndmstr ok
1128739526 M * Aiken still downloading
1128739600 A * Bertl .O( hmm, maybe I should make .bz2s by default ) 
1128739666 A * Aiken could really do with something faster then 56k
1128739736 M * Bertl RoT: I assume the 'warning' was related to find_task_by_pid()
1128739781 M * gndmstr Bertl:   is it ok to have this enabled for vservers? i have always been enabling it just because it seemed the right thing to do when using more than 1gb ram
1128739786 M * gndmstr [*] Allocate 3rd-level pagetables from highmem
1128739817 M * Bertl depending on the code, you want either to replace it with find_task_by_real_pid() and/or include include/linux/sched.h in that file :)
1128740013 J * Aiken_ ~james@tooax6-073.dialup.optusnet.com.au
1128740065 M * Bertl gndmstr: you should disable highmem for performance reasons with less than or equal to 3 GB memory
1128740095 M * Bertl (above 1GB the different split settings are useful)
1128740120 M * RoT Bertl, actually there are no warnings or errors during the build
1128740132 M * gndmstr for the big one what is the recommended setting then. there are 3g in it now and a 4th will go in before it goes to production
1128740170 M * gndmstr jon wants to update it to 8g but i cautioned him to check the mobo specs i dont think it will take it
1128740189 M * Bertl RoT: *sigh* okay, could you upload a patch between vanilla 2.6.13.x and the kernel source you have right now, including a .config file?
1128740221 M * Bertl RoT: (i.e. a single patch against a vanilla (kernel.org) kernel)
1128740262 M * RoT Bertl, umm, Ill have to look into that :)
1128740268 J * menomc ~amery@200.75.27.15
1128740333 M * RoT Bertl, give me an hour or so mate
1128740335 M * Bertl RoT: you can do the following (given that you ahve your 'build tree' at hand
1128740350 M * RoT ok
1128740357 M * Bertl cp -la /path/to/build/kernel /some/other/path
1128740360 Q * Aiken Ping timeout: 480 seconds
1128740364 M * Bertl cd /some/other/path
1128740371 M * Bertl make mrproper
1128740376 Q * mnemoc Ping timeout: 480 seconds
1128740376 N * menomc mnemoc
1128740389 M * Bertl then get the 'mainline' kernel source, unpack it
1128740393 M * Bertl then do:
1128740415 M * Bertl diff -NurpP /path/to/mainline/kernel /some/other/path >delta-rot01.diff
1128740472 M * gndmstr Bertl: on memory splits what is the optimum then for 3gb and for 4gb ram
1128740488 M * gndmstr with i assume high mem page tables off for both
1128740496 M * Bertl depends, on x86, everything above 3GB means overhead
1128740515 M * Bertl you have to use highmem, or cut the memory to 3GB
1128740543 M * Bertl the help to the various split options lists the available memory
1128740557 M * gndmstr ok so dont use it for 3gb but use it for 4gb, and the splits.. i dont know.. its a 440ex mobo with 4 p3
1128740576 M * Bertl typically you want the 1/3 split for up to 3GB
1128740595 M * gndmstr ok
1128740603 M * Bertl with 4GB you are forced to highmem on x86
1128740615 M * Bertl probably a 2/2 split is then the best option
1128740627 M * gndmstr ok
1128740629 M * gndmstr thanks
1128740638 M * Bertl you're welcome!
1128741026 M * RoT Bertl, do you want the diff against a 2.6.12 kernel since I am using linux-2.6.12-vserver-2.0-r1
1128741041 M * Bertl yes, in this case, 2.6.12
1128741493 M * RoT Bertl, ok I have it
1128741536 M * Bertl excellent!
1128741608 M * RoT the diff is 2.9 M
1128741628 M * Bertl sounds reasonable, bzip2?
1128741661 M * RoT 184K
1128741670 M * Bertl excellent, dcc or url?
1128741682 M * RoT no actually 574K :)
1128741685 J * stefani ~stefani@c-24-19-46-211.hsd1.wa.comcast.net
1128741698 M * RoT ok url .. sec
1128741703 M * Bertl welcome stefani!
1128741819 M * RoT Bertl, http://members.iinet.net.au/~xite/delta-rot01.diff.bz2
1128741852 M * Bertl tx
1128741858 M * RoT cheers mate
1128741866 M * Bertl what is the .config option to enable that module/stuff?
1128742087 M * RoT nothing
1128742096 M * Bertl RoT: what's the module called?
1128742188 M * RoT well the modules are not in the kernel
1128742201 M * Bertl ah, well, so I need the module code too :)
1128742207 M * RoT ok
1128742224 M * Bertl and in this case, the warning/error is with the module compile
1128742274 M * RoT no, the error is with modprobe
1128742287 M * Bertl well, yes, there again ...
1128742292 M * RoT ok
1128742415 M * RoT http://members.iinet.net.au/~xite/BestCrypt-1.5-10.tar.gz
1128742417 M * RoT 165K
1128742475 M * Bertl how do you compile it?
1128742483 M * RoT just make
1128742510 M * Bertl well, you need to specify the kernel source, no?
1128742520 M * RoT oh hang on thats the gentoo source 1 sec
1128742572 M * Bertl btw, what is it used for?
1128742589 M * Bertl I mean, all the fancy ciphers are in the kernel IIRC :)
1128742603 M * Aiken_ Bertl should backing out the rc3 patch then applying the rc4 patch trigger a full kernel rebuild instead of just building the changed files?
1128742604 M * RoT http://members.iinet.net.au/~xite/BestCrypt-1.6-2.tar.gz
1128742611 M * RoT try that one
1128742623 M * Bertl Aiken_: well, almost ...
1128742659 M * Bertl RoT: so what's the advantage of bcrypt?
1128742671 M * Aiken_ after 45 min it is only abt 1/2 way though the build, cross compiling a kernel instead
1128742687 M * Bertl probably faster, yes ...
1128742710 M * RoT well nothing really its not open source, truecrypt is better but not compatible
1128742758 M * Bertl hmm, I emant over the kernel ciphers ...
1128742804 M * RoT stronger encryption alograthyms I believe, like rijn
1128742829 M * RoT and it does tricks like hidden containers
1128742843 M * RoT and stuff for the paranoid :)
1128742869 M * Aiken_ 14 min compared to a projected 90 min
1128743032 M * Bertl RoT: well, it's badly coded ... so much I can tell
1128743041 M * RoT hehe
1128743087 M * Bertl I have to change it to compile it ... because you can not specify the kernel source, instead it does some guesswork from the running kernel
1128743158 M * RoT i see
1128743188 M * Bertl well, it's even worse ...
1128743200 M * RoT thats just the generic stable source
1128743207 M * Bertl okay, I'm not going to test compile it ... but I guess I know what you have to change ...
1128743219 M * Bertl grep -nr find_task_by_pid *
1128743219 M * Bertl mod/bc_dev22.c:287:		if (NULL == find_task_by_pid(bc_pid_table[i])) {
1128743219 M * Bertl mod/bc_dev24.c:273:		if (NULL == find_task_by_pid(bc_pid_table[i])) {
1128743219 M * Bertl mod/bc_dev26.c:164:		if (NULL == find_task_by_pid(bc_pid_table[i])) {
1128743235 M * Bertl you#re only interested in the bc_dev26.c
1128743252 M * RoT and bc_rijn
1128743261 M * RoT ok
1128743283 M * Bertl hum, not here ... at least it doesn't contain the find_task_by_pid() ;)
1128743294 M * Bertl change the line to:
1128743297 J * sebi ~sebi@Fd576.f.strato-dslnet.de
1128743331 M * Bertl if (NULL == find_task_by_pid_type(PIDTYPE_PID, bc_pid_table[i])) {
1128743362 M * Bertl and recompile the bcrypt
1128743380 M * Bertl you might need to define:
1128743386 M * Bertl extern struct task_struct *find_task_by_pid_type(int type, int pid);
1128743404 Q * sebi_ Ping timeout: 480 seconds
1128743502 M * Aiken_ rc4 booted and the one auto start guest also started as it should
1128743545 M * Bertl excellent .. with or without XID_PROPAGATE enabled?
1128743589 M * Aiken_ :( without
1128743608 M * Bertl np, simple recompile, not much changed ... :)
1128743675 M * RoT /lib/modules/2.6.12-vs2.0-gentoo-r1/source/include/asm/elf.h:139: error: storage size of `__kernel_vsyscall' isn't known
1128743680 M * RoT make[1]: *** [bc_dev26.o] Error 1
1128743743 M * Bertl hmm, unrelated to the cahnges, I'd say
1128743760 M * Bertl sure that the unmodified version doesn't give you the same?
1128743764 M * RoT actually the topmost error is:
1128743767 M * Aiken_ 1568 bytes below that 4 meg vmlinux limit I get to deal with
1128743791 M * Bertl Aiken_: not too shabby! :)
1128743797 M * RoT /lib/modules/2.6.12-vs2.0-gentoo-r1/source/include/linux/sched.h:879: error: storage size of `default_exec_domain' isn't known
1128743844 M * gndmstr 2.6.13.3-vs2.1.0-rc4 up and running
1128743899 M * Bertl RoT: also unrelated ...
1128743924 M * Bertl RoT: please try with the version you successfully compiled before
1128743932 M * Bertl (of bcrypt, that is)
1128744197 M * RoT ohh mate :) :)
1128744204 M * RoT left out the defines
1128744208 M * RoT working
1128744257 M * RoT thanks so much for that, what is the verdict of the problem?
1128744277 M * Bertl so the module is working for you now?
1128744284 M * RoT perfectly
1128744288 M * Bertl excellent!
1128744320 M * Bertl well, the 'problem' is that the modules use kernel internal functions without really including the kernel stuff
1128744351 M * Bertl as the linux-vserver code does some virtualization in that area, some things changes
1128744354 M * Bertl *changed
1128744361 M * RoT I see 
1128744385 M * Bertl without including the proper defines, the bcrypt doesn't realize the change and ompiles wrongly
1128744407 M * RoT well thats a good result, now I can migrate my server off vmware
1128744416 M * RoT should I file a gentoo bug against it?
1128744418 M * Bertl okay, please could you also do a diff of your changes (to the source of the bcrypt stuff) and submit that to the mailing list?
1128744439 M * RoT gmane?
1128744441 M * Bertl so that other folks could search (and find) the patch later
1128744464 M * Bertl http://linux-vserver.org/ (see mailing lsit)
1128744481 M * Bertl (you don#t need a subscription to post there)
1128744499 M * RoT ok will do, thanks again
1128744506 M * Bertl hmm, not sure if that is still correct 
1128744523 M * Bertl anyway, should be easy ...
1128744524 M * RoT ill find it
1128744534 M * Bertl you're welcome! have fun!
1128744550 M * Bertl (and feel free to linger around)
1128744584 M * RoT I have a feeling I might :)
1128744638 M * Bertl gndmstr: okay, can we test some qmail start/stop/restart cycles?
1128744659 M * Aiken_ it is running with CONFIG_XID_PROPAGATE=y
1128744665 M * Bertl (and watch for the warning)
1128744676 M * gndmstr ok in a sec
1128744690 M * gndmstr just getting my head around these multiple scsi controllers the dell has
1128744718 M * Bertl Aiken_: excellent, you 'should' be able to do a mount --bind -o xid=42 ... and it should apply the xid tagging to all files inside that bind mount (if newly accessed)
1128745114 M * Aiken_ this is confusing
1128745270 M * Bertl hmm, in what way?
1128745332 M * Aiken_ I do the bind mount with a directory from the host to a directory in a guest
1128745345 M * Aiken_ the guest does not see any files created from the host
1128745355 M * Aiken_ and the host does not see anything created by the guest
1128745376 M * Aiken_ it looks like the guest is ignoring the mount and using the mount point as a normal directory
1128745385 M * Bertl yes
1128745393 M * Bertl that's private namespaces :=
1128745421 M * Bertl (i.e. you have to enter the guest namespace to make the mount visible for the guest)
1128745442 M * Aiken_ but if I umount the bind the files created by the guest in the directory
1128745451 M * Aiken_ from that point of view I may as well not have done the mount
1128745473 M * Bertl which is true, for the guest perspective
1128745488 M * Aiken_ and from the hosts perspective
1128745500 M * Bertl there you did the --bind mount
1128745513 M * Bertl you can verify that by writing to it _on the host_
1128745533 M * Bertl (data will end up in the bind mounted dir)
1128745567 M * Aiken_  /vservers/p is the directory on the host
1128745578 M * Aiken_  /vservers/hoppy/var/tmp/p is the directory in the guest
1128745604 M * Aiken_ mount --bin -o xid 42 /vservers/hoppy/var/tmp/p /vservers/p
1128745628 M * Aiken_ echo blah > /vservers/p/test 
1128745633 M * Aiken_ does not appear in the guest
1128745643 M * Bertl expected
1128745667 M * Bertl (first, for the record, it's mount --bind -o xid=42 /vservers/hoppy/var/tmp/p /vservers/p )
1128745686 M * Bertl now, what is the host 'view':
1128745688 A * Aiken_ make a typo in irc :)
1128745701 M * Bertl /vservers/hoppy/var/tmp is a dir
1128745703 M * Aiken_ did a simillar echo in the guest 
1128745709 M * Aiken_ yes
1128745714 M * Bertl you probably created the p dir there
1128745726 M * Aiken_ I keep /tmp and /var/tmp separate
1128745729 M * Bertl for the guest view:
1128745747 M * Bertl tmpfs is probably mounted on /vservers/hoppy/var/tmp 
1128745762 M * Aiken_ the host never can see the guest's file and the guest can not see the file created by the guest
1128745782 M * Aiken_ no
1128745789 M * Aiken_ I keep /var/tmp real filesystem on disk
1128745805 M * Bertl okay, so it's 'just' a dir
1128745821 M * Aiken_ so /var/tmp/p in the guest and /vservers/p are on the same disk and same filesystem
1128745857 M * Aiken_ if I umount to undo the mount the host can then see the guests file in /vservers/hoppy/var/tmp/p/
1128745868 M * Aiken_ which leads me to think why bother with the mount
1128745902 M * Aiken_ on the host if I look in /vservers/hoppy/var/tmp/p/ I never see the file created by the guest
1128745908 M * Bertl anyway, I just realized that the xid propagation is not working correctly yet :/
1128745930 M * Bertl Aiken_: check the /pro/mounts inside the guest please
1128745936 M * Bertl */proc/mounts
1128745971 M * Aiken_ (root@hoppy) cat /proc/mounts 
1128745971 M * Aiken_ rootfs / rootfs rw 0 0
1128745971 M * Aiken_ /dev/root / ext3 rw 0 0
1128745971 M * Aiken_ none /proc proc rw,nodiratime,nodev 0 0
1128745971 M * Aiken_ none /dev/pts devpts rw 0 0
1128746016 M * Aiken_ the way it is behaving now is the guest is completely ignoring the mount
1128746031 M * Bertl and you write to yes, that's expected, but
1128746037 M * Bertl *hum*
1128746046 M * Bertl yes, that's expected, but ...
1128746070 M * Bertl writing to the guest /tmp from the host, should make files show up on the guest
1128746116 M * Aiken_ no
1128746122 M * Bertl and having a --bind mount for that tmp (or a subdirectory) somewhere else on the host, is also supposed to allow you to write a file to the guest
1128746132 M * Aiken_ checking both direcotories on the host shows the mount is working
1128746175 M * Aiken_ but the guest only sees it's own copy of /var/tmp/p/ not what it should see considering it is a mount point
1128746196 M * Bertl did you mount anything there?
1128746267 M * Aiken_ the guests /var/tmp/p/ in reality is /vservers/hoppy/var/tmp/p/ which is --bind mounted to /vservers/p/
1128746277 M * Aiken_ (root@pebbles) ls vservers/p/ vservers/hoppy/var/tmp/p/
1128746277 M * Aiken_ vservers/hoppy/var/tmp/p/:
1128746277 M * Aiken_ gh  test
1128746277 M * Aiken_ vservers/p/:
1128746277 M * Aiken_ gh  test
1128746283 M * Aiken_ (root@hoppy) ls /var/tmp/p/
1128746283 M * Aiken_ bye
1128746311 M * Aiken_ pebbles is the host, hoppy is the guest
1128746337 M * Bertl okay, let's do  ls -di /var/tmp/p (guest)
1128746349 M * Bertl and ls -di vservers/hoppy/var/tmp/p (host)
1128746378 M * Aiken_ (root@hoppy) ls -di /var/tmp/p
1128746378 M * Aiken_ 122552 /var/tmp/p
1128746386 M * Aiken_ (root@pebbles) ls -di vservers/hoppy/var/tmp/p
1128746386 M * Aiken_ 98309 vservers/hoppy/var/tmp/p
1128746397 M * Bertl so those are different inodes ...
1128746408 M * Bertl i.e. your directories are different
1128746426 M * Bertl (which again suggests a tmpfs on the guest :)
1128746427 M * Aiken_ and on the host they are the same inode
1128746433 M * Aiken_ the guest is not using tmpfs
1128746455 M * Aiken_ (root@pebbles) cat /etc/vservers/hoppy/fstab 
1128746455 M * Aiken_ none    /proc           proc    defaults                0 0
1128746455 M * Aiken_ #none   /tmp            tmpfs   size=16m,mode=1777      0 0
1128746455 M * Aiken_ none    /dev/pts        devpts  gid=5,mode=620          0 0
1128746457 M * Bertl well, --bind mounts do not affect inodes ...
1128746482 M * Bertl let's try the same thing with a dir in /, okay?
1128746494 M * Bertl i.e. vservers/hoppy/p
1128746513 M * gndmstr Bertl: it is noticably faster. is this something you did or a result of shutting off hi mem paging? anyway individual multiple stop start and restart passed flawlessly. using the init script for stop start and restart for all servers was flawless. system reboot with servers still running was flawless. will try this again in the morning after it runs for 6 hrs or so.
1128746542 M * Bertl yeah, highmem has significant overhead ...
1128746547 M * Aiken_ (root@pebbles) ls -di vservers/hoppy/var/tmp/p
1128746548 M * Aiken_ 122552 vservers/hoppy/var/tmp/p
1128746548 M * Aiken_ (root@pebbles) ls -di vservers/p              
1128746548 M * Aiken_ 98309 vservers/p
1128746553 M * Aiken_ and that is after a umount
1128746557 M * gndmstr dang. im shutting it off on all of them then :D
1128746560 M * gndmstr my systems too
1128746574 M * Bertl gndmstr: regarding testing: excellent!
1128746595 M * gndmstr does this mean this is could be a finished product i can use on the dell?
1128746629 M * Bertl well, you probably want to wait until Hollow updates the gentoo sources .. but in general ...
1128746653 M * gndmstr i dont use gentoo sources.. too much overhead with all the patching they do that i dont use anyway
1128746659 M * gndmstr i use vanilla kernels
1128746666 M * Bertl you should not consider the devel tree a 'finished product' at any time ... it will/should work ... but it might also break
1128746684 M * gndmstr understood
1128746694 M * Bertl (dosn't mean that you should not use them if the 'work for you' :)
1128746706 M * gndmstr however if i dont use any of the experimental stuff, and they seem to work they should be fine :)
1128746750 M * Bertl yes, but have you tried the stable branch yet? it should work fine for you too (if you do not use/nneed the devel stuff)
1128746776 M * gndmstr that rush job on the one monitoring guest is running the avalonnet percival monitoring system and its running perfectly
1128746788 M * Bertl anyway, I'm now going to add you to the Hall'o'Fame for identifying the issue and helping with debug/testing ...
1128746809 M * gndmstr never tried the stable branch. i want to stay out of the .12 series of kernels.. the .13 series is so much better
1128746832 M * Bertl http://vserver.13thfloor.at/Experimental/patch-2.6.13.2-vs2.0.1-pre2.diff.bz2
1128746876 M * Aiken_ the guest is not using the inode from the bind mount, it is using the original inode
1128746926 M * gndmstr harry makes a huge argument why his production machines are still running 2.6.11 vanilla ... and i can agree with him somewhat but i have always been a 'bleeding edge' type person... i moved our servers to gentoo long before anyone had ever heard of it and they have been rock stable :)
1128746988 M * gndmstr hmm... ok made note of that... will think it over.. problem is, all my configs are for 2.1 im gonna have to relearn the older config styles 
1128747010 M * Bertl naah, 2.0 uses same config ... even 1.x does
1128747030 M * Bertl the legacy config is soemthing we try to get rid of since ever ...
1128747087 M * gndmstr oh ok. mark files, the pre/post scripts etc... will the stable branch get these fixes or are they only needed in 2.1
1128747135 M * Bertl IMHO the fixes we did should not be required for stable (2.0) but it won#t hurt to verify that ...
1128747241 M * Aiken_ the guest has to be restarted
1128747307 M * Aiken_ the guest uses the old inode until it is restarted, then it uses the correct inode for the bind mount
1128747457 M * Bertl hum, you did not bind mount anything _on_ the guest's dir, did you?
1128747476 M * Aiken_ the bind mount was done on the host
1128747497 M * Aiken_ but until the guest was restarted it was still using the original directory
1128747545 M * Bertl gndmstr: okay, added you to the Hall'o'Fame ...
1128748114 M * gndmstr wow thank you sir
1128748125 M * gndmstr im honored:)
1128748151 M * gndmstr and im off to bed and let this thing run for 6 or more hours then ill try the tests again
1128748152 M * Bertl the pleasure is all mine :)
1128748170 M * Bertl okay, have a good sleep, and cya!
1128748188 M * gndmstr yeah i will ... seeing double already :)
1128748192 M * gndmstr good night!
1128748200 P * gndmstr 
1128748246 M * Bertl Aiken_: the bind you mentioned before was binding the guest's tmp/p to some other dir, not the other way round?
1128748331 M * Aiken_ mount --bin -o xid=42 p/ hoppy/var/tmp/p/
1128748350 M * Bertl ah, so the other way round?
1128748356 M * Bertl well, that explains it ...
1128748379 M * Bertl of course, in this case the guest will not see it, unless you restart it
1128748416 M * Bertl (or do the bind mount inside the guest namespace)
1128748880 M * Aiken_ the way I did it treats the directory inside the guest as the mount point
1128748903 M * Aiken_ doing it the other would make the directory on the host as the mount point
1128748949 P * stefani parting (is such sweet sorrow)
1128749199 M * Bertl yes, that's what I thought you did (judging from the command you pasted)
1128749233 M * Bertl anyway, guess I'm off to bed now too ... thanks for testing the rc4!
1128749243 M * Aiken_ bye
1128749270 M * Bertl night everyone! have a good whatever ... cya later
1128749273 N * Bertl Bertl_zZ
1128752126 J * eugenesan eugenesan@bzq-219-189-203.cablep.bezeqint.net
1128752200 P * eugenesan 
1128757980 J * mrec_ ~revenger@p54B02FC8.dip0.t-ipconnect.de
1128758286 Q * mrec Ping timeout: 480 seconds
1128758734 J * liquid3649 ~inet@p54975195.dip.t-dialin.net
1128763943 Q * RoT Ping timeout: 480 seconds
1128768864 Q * case Ping timeout: 480 seconds
1128768925 J * case ~case@donpanic.faveve.uni-stuttgart.de
1128771703 J * Blissex pcg@82-69-39-138.dsl.in-addr.zen.co.uk
1128774870 Q * Aiken_ Quit: Leaving
1128776220 Q * yungyuc Remote host closed the connection
1128776923 J * yungyuc ~yungyuc@220-135-53-220.HINET-IP.hinet.net
1128779081 M * mnemoc how can i get vserver name (to call vserver stop) from /proc/virtual/ ?
1128779944 N * Bertl_zZ Bertl
1128779949 M * Bertl morning folks!
1128779963 M * Bertl mnemoc: not directly, but via syscall
1128779992 M * Bertl (provided it was set, which is not mandatory)
1128780009 M * mnemoc ic
1128780051 M * Bertl vuname --xid <xid> context
1128780067 M * Bertl it seems enrico stores the path there ...
1128780111 M * mnemoc great
1128780168 M * Bertl but per se, there is no relation between the 'guest name' and the xid ... so you/somebody can easily modify/change this info
1128780210 M * mnemoc i just want to stop all running vservers (not only marked as default) on shutdown/reboot
1128780309 M * Bertl then you should do what the runlevel scripts do, but without checking the mark, no?
1128780328 M * Bertl (not invent a 'new' way to figure which guests are active :)
1128780415 M * mnemoc isn't simpler to loop over /proc/virtual/*/, calling vserver stop for each vuname --xid <xid> context ?
1128780450 M * Bertl no, because doing this will fail in several cases
1128780462 M * mnemoc ok
1128780496 M * Bertl if you just want to 'kill' those processes off, you can use the proc info and send vkills to the xid
1128780943 Q * case helium.oftc.net oxygen.oftc.net
1128780943 Q * sebi helium.oftc.net oxygen.oftc.net
1128780943 Q * BWare helium.oftc.net oxygen.oftc.net
1128780943 Q * virtuoso helium.oftc.net oxygen.oftc.net
1128780943 Q * SiD3WiNDR helium.oftc.net oxygen.oftc.net
1128780943 Q * Millox helium.oftc.net oxygen.oftc.net
1128780943 Q * nox helium.oftc.net oxygen.oftc.net
1128780943 Q * ag- helium.oftc.net oxygen.oftc.net
1128780943 Q * eyck helium.oftc.net oxygen.oftc.net
1128780943 Q * click helium.oftc.net oxygen.oftc.net
1128780943 Q * Loki|muh_ helium.oftc.net oxygen.oftc.net
1128780943 Q * SNy helium.oftc.net oxygen.oftc.net
1128780943 Q * michal helium.oftc.net oxygen.oftc.net
1128780943 Q * liquid3649 helium.oftc.net oxygen.oftc.net
1128780943 Q * neofutur helium.oftc.net oxygen.oftc.net
1128780943 Q * mountie helium.oftc.net oxygen.oftc.net
1128780943 Q * alexx helium.oftc.net oxygen.oftc.net
1128780943 Q * Hunger helium.oftc.net oxygen.oftc.net
1128780943 Q * derbien helium.oftc.net oxygen.oftc.net
1128780945 Q * DaCa helium.oftc.net oxygen.oftc.net
1128780945 Q * AndrewLee helium.oftc.net oxygen.oftc.net
1128780945 Q * sladen helium.oftc.net oxygen.oftc.net
1128780945 Q * Nicoli helium.oftc.net oxygen.oftc.net
1128780945 Q * litage helium.oftc.net oxygen.oftc.net
1128780945 Q * jkl helium.oftc.net oxygen.oftc.net
1128780945 Q * Vudumen_ helium.oftc.net oxygen.oftc.net
1128780945 Q * Bertl helium.oftc.net oxygen.oftc.net
1128780945 Q * yungyuc helium.oftc.net oxygen.oftc.net
1128780945 Q * Blissex helium.oftc.net oxygen.oftc.net
1128780945 Q * mrec_ helium.oftc.net oxygen.oftc.net
1128780945 Q * mnemoc helium.oftc.net oxygen.oftc.net
1128780945 Q * tchan helium.oftc.net oxygen.oftc.net
1128780945 Q * Johnsie helium.oftc.net oxygen.oftc.net
1128780945 Q * Sonarman helium.oftc.net oxygen.oftc.net
1128780945 Q * samuel helium.oftc.net oxygen.oftc.net
1128780945 Q * Hollow helium.oftc.net oxygen.oftc.net
1128780945 Q * lonewolff helium.oftc.net oxygen.oftc.net
1128780945 Q * Greek0 helium.oftc.net oxygen.oftc.net
1128780945 Q * flock helium.oftc.net oxygen.oftc.net
1128780945 Q * ag-2 helium.oftc.net oxygen.oftc.net
1128780945 Q * monrad helium.oftc.net oxygen.oftc.net
1128780945 Q * cryo helium.oftc.net oxygen.oftc.net
1128780945 Q * hippo helium.oftc.net oxygen.oftc.net
1128780945 Q * kevinp helium.oftc.net oxygen.oftc.net
1128780945 Q * entroposcope helium.oftc.net oxygen.oftc.net
1128780945 Q * micah helium.oftc.net oxygen.oftc.net
1128780952 J * case ~case@donpanic.faveve.uni-stuttgart.de
1128780952 J * liquid3649 ~inet@p54975195.dip.t-dialin.net
1128780952 J * sebi ~sebi@Fd576.f.strato-dslnet.de
1128780952 J * Nicoli ask@28.burningd.enterconnect.net
1128780952 J * litage ~nick@203.220.55.70
1128780952 J * BWare ~bware@office.intouch.net
1128780952 J * jkl eric@c-67-173-249-8.hsd1.co.comcast.net
1128780952 J * virtuoso ~s0t0na@shisha.spb.ru
1128780952 J * neofutur ~neofutur@neofutur.net
1128780952 J * SiD3WiNDR luser@bastard-operator.from-hell.be
1128780952 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com
1128780952 J * alexx ~alexx@proxy.ikse.net
1128780952 J * Hunger Hunger.hu@Hunger.hu
1128780952 J * Millox ~mattias@shienar.csbnet.se
1128780952 J * nox ~nox@nox.user.oftc.net
1128780952 J * ag- ag@caladan.roxor.cx
1128780952 J * eyck eyck@81.219.64.71
1128780952 J * SNy 713549e1e4@bmx-chemnitz.de
1128780952 J * click click@ti511110a080-1724.bb.online.no
1128780952 J * michal ~michal@michal.usercloak.oftc.net
1128780952 J * Loki|muh_ loki@satanix.de
1128780952 J * derbien ~derbien@whiterabbit.nbmc.de
1128780952 J * DaCa ~danny@mail.limehouse.org
1128780952 J * sladen paul@starsky.19inch.net
1128780952 J * AndrewLee ~andrew@tnlug.linux.org.tw
1128780952 J * Vudumen_ vudumen@perverz.hu
1128780952 J * Bertl herbert@212.16.62.52
1128780982 J * yungyuc ~yungyuc@220-135-53-220.HINET-IP.hinet.net
1128780982 J * Blissex pcg@82-69-39-138.dsl.in-addr.zen.co.uk
1128780982 J * mrec_ ~revenger@p54B02FC8.dip0.t-ipconnect.de
1128780982 J * mnemoc ~amery@200.75.27.15
1128780982 J * tchan ~tchan@c-67-174-18-204.hsd1.il.comcast.net
1128780982 J * Johnsie ~john@acs-24-154-53-217.zoominternet.net
1128780982 J * Sonarman ~cleetus@adsl-67-113-235-31.dsl.snfc21.pacbell.net
1128780982 J * samuel ~samuel@Quebec-HSE-ppp242189.qc.sympatico.ca
1128780982 J * Hollow ~hollow@home.xnull.de
1128780982 J * lonewolff ~lonewolff@host86-128-128-38.range86-128.btcentralplus.com
1128780982 J * Greek0 ~greek0@85.255.145.201
1128780982 J * flock ~restless@l192-117-111-12.broadband.actcom.net.il
1128780982 J * ag-2 ag@muaddib.roxor.cx
1128780982 J * monrad ~monrad@213083190134.sonofon.dk
1128780982 J * cryo ~say@gw.psoft.od.ua
1128780982 J * hippo ~trh45gwer@57.66.77.129
1128780982 J * kevinp ~kevinp@ny.webpipe.net
1128780982 J * entroposcope ~entroposc@user-0c992og.cable.mindspring.com
1128780982 J * micah micah@micha.hampshire.edu
1128781807 M * mnemoc Bertl: something like http://mnemoc.pastebin.com/387093 ?
1128781861 M * mnemoc vkill all looks too agressive :)
1128781874 M * Bertl could work ... looks good
1128781907 M * mnemoc Enrico's sh scripting is arcane :p
1128781950 M * mnemoc pure bourne shell
1128782148 M * mnemoc too much time coding on ksh and now bash3 i guess 
1128782357 M * Bertl well, feel free to 'improve' it (whatever that may mean)
1128782503 M * mnemoc didn't work
1128782515 M * mnemoc i will ;)
1128782695 M * Hollow mnemoc: lol.. i just rewrote some functions for looking up the name per vuname
1128782717 M * Hollow http://dev.croup.de/proj/util-vserver/file/trunk/tools/gentoo-functions.sh?rev=15&format=txt
1128782744 M * Hollow morning Bertl btw
1128782793 M * Bertl hi Hollow! but I 'guess' you missed my comment on that :)
1128782813 M * Hollow no, there's no relation per se
1128782815 M * Hollow i got it ;)
1128782851 M * Bertl okay, guess the $2 is the criterion :)
1128782873 M * Bertl .. so be it .. btw, you should update to rc4 if not already done so
1128782884 M * mnemoc :)
1128782889 M * Hollow rc4?
1128782893 M * Hollow 2.1.0-rc4?
1128782896 T * Bertl http://linux-vserver.org/ | latest stable 2.0, 2.0.1-pre2, 1.2.10, 1.2.11-rc1, devel 2.1.0-rc4 | He who asks a question is a fool for a minute; he who doesn't ask is a fool for a lifetime -- share the gained knowledge on the wiki, and we'll forget about the minute ;)
1128782898 M * Hollow k
1128782918 M * Bertl it seems it fixes the issues chuck experienced ...
1128782927 M * mnemoc Hollow: ls -1 is your friend :)
1128782949 M * Hollow does ls not work?
1128782955 M * Hollow Bertl: which issues?
1128782970 M * Hollow seems like i missed last 2 irc days
1128782973 M * mnemoc Hollow: yes, but you have to strip */ later
1128782983 M * Bertl hitting some sanity checks (because of bad kernel structure)
1128782987 M * Hollow mnemoc: ah, k.. thx
1128783005 M * Bertl http://vserver.13thfloor.at/Stuff/BUGHUNT/gndmstr-0001/
1128783038 M * mnemoc Bertl: do you dream with gndmstr these days?
1128783137 M * Hollow Bertl: will there be a 2.0.1 release?
1128783144 M * Hollow before 2.1.0 is stable
1128783157 M * Bertl definitely!
1128783171 M * Bertl (as 2.1.0 will never become stable :)
1128783172 A * Hollow slaps DaPhreak
1128783176 M * Hollow lol
1128783180 M * Hollow why?
1128783185 Q * litage Ping timeout: 480 seconds
1128783187 M * Bertl because it's the devel branch :)
1128783194 M * mnemoc :)
1128783194 M * Hollow ah
1128783196 M * Hollow ;)
1128783198 M * Bertl stable branch <-> devel branch :)
1128783215 M * Hollow i missed the versioning change too
1128783223 M * Hollow so it'll be 2.2?
1128783226 M * Bertl hmm, no real change there ...
1128783238 M * Bertl 2.0.1 will be the next stable release
1128783245 M * Bertl 2.1.0 the first devel release ...
1128783257 M * Bertl after some time, the 2.1.x will become 2.2.0 (stable)
1128783267 M * Hollow ok, got it ;)
1128783305 M * Bertl but 2.0.1 will definitely add some stuff from 2.1.0
1128783323 M * Bertl okay, have to leave now .. will be back tonight ...
1128783341 N * Bertl Bertl_oO
1128785000 M * micah morning Bertl_oO 
1128785013 M * micah err
1128785704 Q * lonewolff Quit: leaving
1128786141 J * lonewolff ~lonewolff@host86-128-128-38.range86-128.btcentralplus.com
1128786609 J * lilo_ ~lilo@lilo.usercloak.oftc.net
1128786701 Q * lilo Ping timeout: 480 seconds
1128786948 J * yarihm ~yarihm@84-74-18-28.dclient.hispeed.ch
1128787694 Q * yungyuc Read error: Connection reset by peer
1128788276 J * yungyuc ~yungyuc@220-135-53-220.HINET-IP.hinet.net
1128788692 Q * yungyuc Remote host closed the connection
1128789588 J * yungyuc ~yungyuc@220-135-53-220.HINET-IP.hinet.net
1128789598 Q * liquid3649 Ping timeout: 480 seconds
1128791228 J * dddd44 dhb55@60.49.78.240
1128791999 Q * dddd44 Read error: Connection reset by peer
1128792201 J * dddd44 dhb55@60.49.78.240
1128792295 Q * Blissex Remote host closed the connection
1128792544 J * antagonist ~antagonis@ip68-13-182-17.om.om.cox.net
1128792560 M * antagonist hullo!
1128792773 J * litage ~nick@203.220.55.70
1128793324 Q * cryo Ping timeout: 480 seconds
1128793504 Q * dddd44 Ping timeout: 480 seconds
1128793805 J * cryo ~say@gw.psoft.od.ua
1128793992 J * dddd44 dhb55@60.49.78.240
1128794034 J * liquid3649 ~inet@p54973FB9.dip.t-dialin.net
1128795497 Q * lonewolff Read error: Connection reset by peer
1128795551 J * lonewolff ~lonewolff@86.128.128.38
1128797860 J * gndmstr ~gndmstr@ip1.pathworx.sbbsnet.net
1128798714 Q * liquid3649 Ping timeout: 480 seconds
1128799370 M * micah looks like ola built a new util-vserver adding fix03 and depending on libbeecrypt so vhashify will be built
1128800309 M * gndmstr interesting
1128800334 M * gndmstr i have another friggin new bug on the big machine im setting up. putting a vanilla kernel in to see if i still get it..
1128801831 M * antagonist hmm
1128801845 M * antagonist i'm trying to mount an already mounted filesystem in a second location
1128801854 M * antagonist it's mounted in the host and i want to mount it in each of my vservers
1128801868 M * antagonist it works fine when i mount it read-write in both locations
1128801878 M * antagonist and /etc/mtab reports that it is in fact mounted in multiple locations
1128801904 M * gndmstr but a running guest doesnt see it
1128801906 M * antagonist but if i try to make it read-write on the host, and read-only in the directory tree of the vservers, it won't let me mount it again
1128801914 M * antagonist well, the guests see it fine
1128801915 M * gndmstr hmm
1128801922 M * antagonist everything is fine if i go ahead and mount read-write
1128801933 M * antagonist my only problem is that mounting seems to be inconsistent on the host
1128801956 M * antagonist it gives me the 'filesystem is already mounted or mountpoint is busy' error
1128801959 M * gndmstr are you using this syntax?
1128801987 M * antagonist on the host: mount /dev/vg/portage /usr/portage
1128802006 M * gndmstr mount -t bind -o ro /dev/vg/portage /usr/portage
1128802019 M * antagonist on the host, in the guest's filesys: mount -o ro /dev/vg/portage /vservers/blah/usr/portage
1128802019 M * antagonist hmm
1128802024 M * antagonist i didn't try to bind it
1128802027 M * antagonist one moment
1128802047 M * gndmstr from what ive seen all the mounts to the host that the guests see need bind 
1128802077 M * antagonist okay
1128802091 M * antagonist this works: mount --bind -o ro /usr/portage /vservers/blah/usr/portage
1128802103 M * antagonist but is there a way to reflect that in my /etc/fstab on my host?
1128802113 M * antagonist or do i need to add that to a local startup script to automate it?
1128802113 M * gndmstr thats the way they have it in the fstab inside the guest definition
1128802124 M * antagonist ah
1128802131 M * gndmstr there is an fstab inside /etc/vservers/guestname
1128802133 M * antagonist i haven't even looked at the fstab in the guest definition
1128802135 M * antagonist whoops
1128802135 M * antagonist ;)
1128802143 M * gndmstr hehe thats ok i found it by accident
1128802170 M * antagonist what would a line look like in that file?
1128802182 M * antagonist would i use bind as the filesystem type?
1128802194 M * antagonist or actually put the same line i'd put in the host?
1128802205 M * antagonist it can't actually mount the device from /dev, can it?
1128802218 M * gndmstr i dont think you can bind to dev .. wasnt thinking
1128802224 M * gndmstr i think you bind to a premounted point
1128802227 M * antagonist right
1128802241 M * antagonist yeah, the line i mentioned above had the correct syntax
1128802250 M * antagonist i just wasn't sure how to specify a bind in fstab
1128802276 M * gndmstr goes in the options column
1128802281 M * antagonist hmm hmm!
1128802284 M * antagonist okay, i'll try it
1128802645 M * gndmstr hmm it appears that the vprocunhide init script was forgotten in the latest util-vserver for gentoo
1128803225 M * antagonist gndmstr, do you run gentoo in your guests?
1128803236 M * antagonist is it necessary to have mount available in order to accomplish a bind mount?
1128803246 M * antagonist i just emerged util-linux to be on the safe side
1128803274 M * antagonist and to clear up the ambiguity, i meant is it necessary to have mount available in the guest to accomplish a bind mount through its fstab
1128803882 M * antagonist cool
1128803892 M * antagonist i figured it out now, thanks very much for your help
1128803914 M * antagonist that's a lot less cluttered looking
1128803916 M * gndmstr no they are mounted before the guest starts
1128803921 M * antagonist now my host doesn't see all of those mounts
1128803926 M * gndmstr hehe
1128803935 M * antagonist thanks a bunch
1128803941 M * gndmstr im not sure how they accomplish it but it seems to mount inside
1128803946 M * antagonist aye
1128803953 M * antagonist it's done in the context somehow
1128803967 M * antagonist neat
1128803994 M * antagonist now i just need to learn to use conditionals in my shell scripting
1128804023 M * gndmstr fun time:) me too
1128804045 M * antagonist i wrote a long convoluted script to automate making a new vserver from my template, with a custom hostname and ip, context, and setup a unique port range for it in my iptables nat table
1128804062 M * gndmstr cool
1128804139 Q * cryo Ping timeout: 480 seconds
1128804608 J * Aiken ~james@tooax6-106.dialup.optusnet.com.au
1128805036 J * cryo ~say@gw.psoft.od.ua
1128805344 Q * mrec_ Ping timeout: 480 seconds
1128807626 Q * lonewolff Read error: Connection reset by peer
1128807680 J * lonewolff ~lonewolff@host86-128-128-38.range86-128.btcentralplus.com
1128809022 J * mrec ~revenger@p54B0413D.dip0.t-ipconnect.de
1128810019 N * Bertl_oO Bertl
1128810023 M * Bertl evening folks!
1128810038 M * Bertl gndmstr: you are trying to bind mount devices?
1128810361 M * Bertl Hollow: from ML: Missing Gentoo init script?
1128811202 M * antagonist Bertl, no, that was me
1128811206 M * antagonist the bind mounting
1128811209 M * antagonist i was the slowpoke
1128811293 M * Bertl ah, okay, np, the --bind or -o bind  (never -t bind) does only affect vfs ...
1128811806 M * gndmstr Bertl if i enabled acpi in the kernel would that cure that oops?
1128811881 M * gndmstr we try not to have any power management capabilities in the kernels at all.. will have to have the boss check the cmos since it doesnt support serial redirection:(
1128812238 M * Bertl gndmstr: how often are you going to power down the server?
1128812258 M * Bertl how much do you care about this message (and the fact that it doesn't power down)
1128812279 M * gndmstr nothing at all but it had to be power cycled. that was a reboot command i issued
1128812283 M * gndmstr not a halt
1128812325 M * gndmstr the machines should never turn themselves off since we have remote power managament boxes for that
1128812339 M * gndmstr but i dont understand why it said power down when i said reboot
1128812435 M * Bertl your mail said 'power down' IIRC
1128812466 M * Bertl you should try with 'shutdown -r ..' vs 'shutdown -h ..'
1128812469 M * gndmstr oops. my mistake. boss told me to shut it down
1128812473 M * gndmstr i remember now
1128812477 M * gndmstr ok normally i dont care
1128812485 M * gndmstr we set machines to always on conditions
1128812519 M * gndmstr after 12 hrs running rc4 is rebooting perfectly with init shutting the guests down
1128812520 M * Bertl so then better ignore it
1128812530 M * gndmstr ok cool
1128812534 M * Bertl excellent! so we finally nailed it :)
1128812538 M * gndmstr i think so
1128812570 Q * lonewolff Read error: Connection reset by peer
1128812661 M * gndmstr rebooted perfectly :)
1128812694 M * gndmstr shall i try again after 24 hrs or can we call it 'case closed'
1128812776 M * Bertl I'd say it's done ... but I'd actually appreciate a test run with the stable branch, if that would be possible?
1128813058 M * gndmstr ok.. lets see i need to change kernel, util-vserver tools, what about baselayout-vserver
1128813103 M * gndmstr brb about 30 min
1128813288 Q * Nicoli Read error: Connection reset by peer
1128813315 J * Nicoli ask@28.burningd.enterconnect.net
1128813497 M * Bertl gndmstr: no changes required, just use the 2.0.1 patch instead of the 2.1.0 one ... vanilla kernel as usual ...
1128813586 J * lonewolff ~lonewolff@host86-128-128-38.range86-128.btcentralplus.com
1128814557 Q * yarihm Quit: Leaving
1128815061 Q * Nicoli Ping timeout: 480 seconds
1128815190 M * gndmstr ok ill find it.. is there one for 13.3 kernels
1128815696 M * Bertl yep
1128815718 M * Bertl http://vserver.13thfloor.at/Experimental/patch-2.6.13.2-vs2.0.1-pre2.diff.bz2