1159056756 J * Johnnie ~jdlewis@jdlewis.org
1159057920 M * Bertl ok, I'm off to bed now .. have a good one everyone! cya tomorrow!
1159057927 N * Bertl Bertl_zZ
1159060761 J * _node node@c-69-143-148-254.hsd1.md.comcast.net
1159063999 Q * ensc Killed (NickServ (GHOST command used by ensc_))
1159064006 J * ensc_ ~irc-ensc@p54B4E61E.dip.t-dialin.net
1159064062 N * ensc_ Guest1227
1159067815 Q * comfrey Quit: Lost terminal
1159068010 Q * ebiederm Quit: Leaving
1159068936 N * DreamerC_ DreamerC
1159075202 Q * _node Ping timeout: 480 seconds
1159078760 M * VxJasonxV Could anybody in here give me a hand with a problem occuring inside a vserver? I have a vserver which cannot start most init.d processes, because a cat command blocks it.
1159078797 M * VxJasonxV For example, (this I can do consistently), I just restarted sshd due to an update, the terminal in which I run the /etc/init.d/sshd restart command goes to "Stopped sshd \t\t\t [ok]" and then just stops.  Then in another already active terminal (or vnc if I'm on it) I have to open up another xterm and kill the cat process, something like: cat /var/lib/init.d/exclusive/net.lo
1159078838 M * VxJasonxV when I kill it, the terminal that was restarting sshd finally says that it's started, there's a process error, but not on for sshd: /lib/rcscripts/sh/rc-services.sh: line 375: /var/lib/init.d/exitcodes/net.lo: No such file or directory
1159078850 M * VxJasonxV oh, and also: /lib/rcscripts/sh/rc-services.sh: line 377: return: : numeric argument required
1159078888 M * VxJasonxV it's bothersome, because this vserver "didn't come back up" (it did, but no process started because of that cat command blocking them) after the last host server reboot.  I had to get the host to go it and (literally) just run: killall cat [enter] [up] [enter] [up] [enter], tons of times
1159079171 J * s0undt3ch_ ~s0undt3ch@bl7-245-88.dsl.telepac.pt
1159079458 Q * s0undt3ch Killed (NickServ (GHOST command used by s0undt3ch_))
1159079458 N * s0undt3ch_ s0undt3ch
1159080258 J * derjohn ~aj@dslb-084-058-195-087.pools.arcor-ip.net
1159080792 J * dna_ ~naucki@53-238-dsl.kielnet.net
1159081196 Q * coocoon Ping timeout: 480 seconds
1159081993 J * coocoon ~coocoon@p54A063C7.dip.t-dialin.net
1159082548 Q * click Ping timeout: 480 seconds
1159083278 Q * ruskie Remote host closed the connection
1159083824 J * ruskie ~ruskie@ruskie.user.oftc.net
1159084917 Q * ruskie Ping timeout: 480 seconds
1159085997 J * bonbons ~bonbons@83.222.36.111
1159086423 J * ruskie ~ruskie@ruskie.user.oftc.net
1159087138 A * waldi just disabled legacy interfaces in the debian packages, lets check if this works now
1159090354 M * daniel_hozac CONFIG_VSERVER_LEGACY? you'll need a very recent (read: 0.30.211-rc1) util-vserver for reboot -f to work inside guests.
1159090553 M * waldi and LEGACY_NETWORK
1159090669 M * daniel_hozac that should work with 0.30.210.
1159091727 M * waldi okay
1159091843 M * daniel_hozac hmm, i'm seeing some strange behaviour with 2.6.18-vs2.0.2.1-t7.
1159091882 M * daniel_hozac chcontext --xid <some guest> cat /proc/1/status; vps auxc will list the host's init as the guest's init.
1159092173 M * Guest1227 daniel_hozac: afais, that's also the case with 2.6.17.11 2.02  (with non-initfake guests)
1159092287 M * daniel_hozac you really should use 2.0.2.1.
1159092339 N * Guest1227 ensc
1159092499 M * VxJasonxV could anybody go just a smidge back into their scrollback and offer any insight if at all possible?
1159092557 M * daniel_hozac i can't reproduce it as easily with 2.6.17.13-vs2.0.2.1, (i.e. i haven't been able to yet).
1159092722 M * daniel_hozac VxJasonxV: i don't know Gentoo, but it sounds like something is pretty broken.
1159092745 M * VxJasonxV yeah... it does
1159092760 M * VxJasonxV but I'm totally at a loss as to where to start
1159092789 M * daniel_hozac IMHO it looks like a space got replaced by a linefeed.
1159092796 M * VxJasonxV O_o
1159092804 M * daniel_hozac so it's cat\n<filename>, rather than cat filename.
1159092853 M * daniel_hozac ensc: did you have to do anything more to reproduce it on that kernel?
1159092876 M * VxJasonxV I'm confused daniel.  If it were a newline, it would have been on a new message.
1159092886 M * VxJasonxV a new message line... not sure how else to describe it
1159092892 M * daniel_hozac what?
1159092917 M * ensc daniel_hozac: sorry, I do not have 2.0.2.1
1159092941 M * daniel_hozac well, 2.0.2 -> 2.0.2.1 is pretty minimal changes.
1159092998 M * daniel_hozac reproducing it on 2.6.18-vs2.0.2.1-t7 is almost too easy...
1159093145 M * ensc afair, host's init was always (from early kernel 2.4 days) shown in vserver
1159093159 M * daniel_hozac ah, well, i meant the other way around.
1159093174 M * daniel_hozac i.e. the host's init is replaced by the guest's init, in the vps output.
1159093236 M * daniel_hozac http://paste.linux-vserver.org/404
1159093270 M * ensc lunch-time; will be back later...
1159093299 M * daniel_hozac ok, cya.
1159093876 M * ensc daniel_hozac: ah ok; seems to be new in -t7
1159093894 M * ensc -t6 does not show any init
1159093951 M * daniel_hozac t6 to t7 was just the find_task_by_real_pid fix though.
1159093986 M * daniel_hozac (i.e. making it work in the host context too)
1159094050 M * ensc mmh... strange, now I am seeing init again
1159094155 M * ensc I am pretty sure, that I executed the same command when it was hidden; perhaps the 'cat /proc/1/status' between both commands triggered something
1159094228 M * ensc yes; chcontext --xid 152 cat /proc/1/status ; vps axfu
1159094234 M * ensc --> init not visible
1159094252 M * ensc cat /proc/1/status ; vps axfu
1159094256 M * ensc --> init visible
1159094287 M * daniel_hozac indeed.
1159094322 M * daniel_hozac i'd say it was introduced in t6, the pid_revalidate VX_WATCH change.
1159095177 J * vorbote ~vorbote@201.244.231.115
1159095490 P * vorbote Parting is such sorrow.
1159096717 Q * cehteh Read error: Connection reset by peer
1159096741 J * cehteh ~ct@pipapo.org
1159098688 Q * fosco charon.oftc.net venus.oftc.net
1159098826 J * fosco fosco@konoha.devnullteam.org
1159099253 Q * fosco charon.oftc.net venus.oftc.net
1159099424 J * fosco fosco@konoha.devnullteam.org
1159100987 N * Bertl_zZ Bertl
1159100990 M * Bertl morning folks!
1159101201 M * Bertl had a feeling that the VX_WATCH was wrong there, but I couldn't say why .. now we know ... :)
1159101220 M * Bertl ensc, daniel_hozac: try -t8 or the delta-vpid-fix01.diff
1159101318 M * Bertl okay, off for breakfast .. back later!
1159101322 N * Bertl Bertl_oO
1159101543 A * waldi testbuilds xen-vserver ...
1159104472 J * root__ ~root@ACaen-152-1-46-86.w83-199.abo.wanadoo.fr
1159104566 M * root__ hello all
1159104574 Q * root__ 
1159104619 J * pmenier ~pmenier@ACaen-152-1-46-86.w83-199.abo.wanadoo.fr
1159104654 M * pmenier hello all
1159105055 M * ensc mmh... is it really expected that /dev/hw_random outputs only 0xff bytes?
1159105263 M * Loki|muh its /dev/hwrng here
1159105279 M * Loki|muh crw-rw---- 1 root root 10, 183 2006-09-08 17:33 /dev/hwrng
1159105555 M * ensc nevertheless, a stream of 0xff has very low entropy
1159106135 M * daniel_hozac do you have a hardware random number generator?
1159107414 M * ensc daniel_hozac: the intel_rng module thinks this at least
1159107495 M * ensc but probe() function looks very weak to me (checking for PCI device and whether a bit is set at some memory location)
1159107996 J * phedny_ ~mark@volcano.p-bierman.nl
1159107997 Q * phedny Read error: Connection reset by peer
1159108420 M * ensc that's a really clever design... intel datasheet recommends to do some FIPS checks with the read data to check whether there is really an rnd
1159109459 Q * shedi Quit: Leaving
1159112354 M * waldi FIPS?
1159112800 Q * Johnnie Read error: Connection reset by peer
1159113058 J * Johnnie ~jdlewis@jdlewis.org
1159114066 M * ensc Federal Information Processing Standard
1159114677 M * yang hey Bertl_oO !
1159114690 Q * coocoon Quit: KVIrc 3.2.0 'Realia'
1159114866 J * coocoon ~coocoon@p54A073B1.dip.t-dialin.net
1159118216 P * bone_idol 
1159119874 J * glut glut@no.suid.pl
1159120875 Q * gerrit Ping timeout: 480 seconds
1159120877 M * starlein anyone needs newest patch for kernel 2.6.17.13 with grsec2.1.9?
1159121002 M * daniel_hozac better just post it so when someone needs it, it's in the logs ;)
1159121018 M * starlein and vs2.0.2.1 patches I forgot to say
1159121032 M * starlein hmmm posting where?
1159121043 M * starlein mailinglist?
1159121043 M * daniel_hozac if anyone feels like living on the edge: http://people.linux-vserver.org/~dhozac/p/uv/experimental/util-vserver-0.30.211-rc2.tar.bz2
1159121053 M * daniel_hozac just post the URL here.
1159121097 M * daniel_hozac harry will probably take a look at it and upload it to his people account if he agrees ;)
1159121100 M * starlein ahh :) okay publishing is in progress :)
1159121202 M * starlein just took me an hour because of that binfmt_elf changes
1159121334 J * click click@ti511110a080-4180.bb.online.no
1159121591 Q * cehteh Quit: I shouldn't really be here - dircproxy 1.0.5
1159121593 J * cehteh ~ct@pipapo.org
1159121685 Q * pmenier Quit: leaving
1159122210 M * derjohn daniel_hozac, do you utils contain v6 foo ?
1159122216 M * daniel_hozac derjohn: no.
1159122228 M * daniel_hozac derjohn: that requires an IPv6 patched set of kernel headers.
1159122244 M * daniel_hozac util-vserver is using 2.6.17.13-vs2.0.2.1 kernel headers, right now.
1159122251 M * derjohn why dont you include them?
1159122286 M * daniel_hozac because IPv6 isn't even in devel yet? :)
1159122307 M * derjohn well, but would it harm to have chbind6 in?
1159122323 M * daniel_hozac as i said, it requires compiling against IPv6 kernel headers.
1159122330 M * daniel_hozac so you'd still have to manually compile it.
1159122338 M * daniel_hozac and chbind6 is really just a hack tool.
1159122370 M * derjohn daniel_hozac, you will probably have ;) besides that, could that part of the code simply check, if the headers are there and if not compile without?
1159122409 M * daniel_hozac for chbind6? its purpose is to setup IPv6 addresses, so that really doesn't make much sense, IMHO.
1159122427 M * derjohn well, the the v6 patch for the utils?
1159122438 M * daniel_hozac what?
1159122517 M * derjohn well, the package should check at compile time, if the v6-patched kernel headers are there. if so, they should complie v6 in. I *hope* Bertl_oO will include the v6 patch in devel some day
1159122695 M * daniel_hozac some day, probably.
1159122702 M * derjohn :) k
1159122719 M * daniel_hozac but how would it check for those headers?
1159122759 M * daniel_hozac right now util-vserver is using a private set of headers.
1159122853 M * derjohn hm, yes I notices that (by patching > 16 per guest in). So it would be easy to pre-patch them for v6? and how does your util-v6 patch work? does that use the 'real' kernel headers instead?
1159122893 M * daniel_hozac the util patch for v6 just changes the scripts to use chbind6 instead of chbind.
1159122910 M * daniel_hozac and alters the configuration reading/option generating functions.
1159122936 M * derjohn so i need to build chbind6 in any case (additionaly to utils)
1159122940 M * daniel_hozac exactly.
1159122953 M * daniel_hozac but you could quite easily include chbind6 in your package.
1159123060 M * derjohn well, that what I just think about, as my newest kernel deb package include the v6 patch. I need a util package that has 256 IP, so I could add chbind6, too and alltogether it would be the "Bleeding edge vserver packerls" on debian (in contrast to Debian which ships stable)
1159123076 M * daniel_hozac yep.
1159123083 M * daniel_hozac so you're using 0.30.211-rc2 then as well? :)
1159123110 M * derjohn not yet., but thats the next step !
1159123616 J * shedi ~siggi@inferno.lhi.is
1159123767 J * id23 ~id@p5081271B.dip0.t-ipconnect.de
1159124268 M * daniel_hozac derjohn: http://people.linux-vserver.org/~dhozac/p/uv/experimental/util-vserver-0.30.211-ipv6.patch is a far less intrusive patch that applies to 0.30.211-rc2 as well (but it does require recompiling the utils, or changing util-vserver-vars manually).
1159125039 M * harry i will indeed... butg not now
1159125039 M * harry ;)
1159125932 Q * shedi Quit: Leaving
1159128553 N * Bertl_oO Bertl
1159128559 M * Bertl evening folks!
1159128728 M * _are_ hi Bertl
1159128747 M * Bertl hey _are_! LTNS!
1159128819 M * _are_ oh, since I pretend and have vacation I don't need to carry around my laptop all the time -> logged in Friday and I am still here.
1159129117 M * Bertl ah, so you sneaked in :)
1159129288 M * _are_ well, reconnecting to 3+ irc servers and jhoining all channels again every few hours sort of becomes work. shoukd patch my irc client to finally do this automatically.
1159129382 Q * cirrus Remote host closed the connection
1159129645 M * Bertl what about having an irc proxy or a dedicated vserver to irc from?
1159129906 J * duckx ~Duck@tox.dyndns.org
1159129931 M * _are_ well, i started irc with ircII and it had been fine, ran it within screen. But nowadays it is so many channels and so much traffic i much prefer having it in seperate windows. So unless I run VNC-sessions on some host and ...
1159129932 M * _are_ reattach them, i won't gain anything.
1159129966 M * Bertl i.c. well, still using irssi here :)
1159129973 M * Bertl welcome duckx!
1159129974 M * _are_ and i tried VNC, unfortrunately I tghen havea single-window irc, while atm a new PM/channel is in a seperate window opening wherever i am.
1159129982 M * duckx Hy Bertl :)
1159129997 M * _are_ well, i user ircII or irssi atm, but wtrapped into a slightly patched tkircII
1159129999 M * duckx I just went to the new looking webserver ...
1159130008 M * duckx Nice to see your face ;)
1159130035 M * Bertl hehe, now you have to upload yours somewhere too, no? :)
1159130046 M * duckx lol :)
1159130058 M * duckx Give me second
1159130191 M * _are_ http://linux-vserver.org/Special:Recentchanges <-- topmost entry, someone created a spam page.
1159130206 M * duckx Bertl: http://tox.Dyndns.org/00013.jpg
1159130260 J * soatola ~soatola42@82.153.18.114
1159130263 M * duckx I just recompile a 2.6.17 kernel for my vserver box ...
1159130271 M * _are_ hmm, I am sure I've seen that face already, if I just had a better brain for names....
1159130285 M * duckx Is there any special requirement for 2.0.2 (from 2.0.1) ?
1159130302 M * _are_ How can I delete a page on the wiki? if the spam stays we end up as a linkfarm :-/
1159130373 M * cehteh moinmoin has spam protection .. lala :P
1159130487 A * _are_ has not been involved on any wiki advocacy. But if in doubnt I would use moinmoin for the simple reason I know one of the maintainers and have physical access to him ;)
1159130504 M * Bertl no idea how to delete a page, I would like to know that myself :)
1159130511 M * Bertl welcome soatola!
1159130542 M * _are_ ok, who runs the wiki/knows how to do this? I'd just move it, but then it is an alias and in the end nothing changes
1159130554 M * soatola ..lo Bertl
1159130589 M * soatola So, I'm trying to move a vserver to another box, tar the /etc/vserver/`vservername` and /var/lib/vservers/`vservername` and copy across. However the group ids on the newbox don't match. Whats the proper method?
1159130604 M * Bertl _are_: send email to hollow, or maybe derjohn
1159130639 M * Bertl soatola: dump/restore or rsync usually
1159130662 M * Bertl soatola: tar has no option to preserve numeric ids by default (at least not that I know of)
1159130665 M * soatola I'll do that next, chears.
1159130671 M * _are_ rsync -n should work
1159130690 M * _are_ nope
1159130693 M * _are_ --numeric-ids
1159130696 M * _are_ -n is dry-run
1159130701 M * Bertl soatola: yeah, if you decide to use rsync, check the man page thoroughly, you want quite some options, like:
1159130709 M * Bertl -azxH for example
1159130762 M * soatola I've used rsync in the past but was under the impression it was more efficient to transfer the bulk first and rsync differences
1159130787 M * _are_ rsync is efficient as it is
1159130817 M * _are_ well, ok, tar | nc is more efficient
1159130872 M * soatola It was probable in relation to real work like distro images held centrally.
1159130917 J * BeLu ~B.Lukas@88.134.54.56
1159130924 M * BeLu hello
1159130931 M * _are_ Hi BeLu
1159130943 Q * bonbons Quit: Leaving
1159132223 P * BeLu 
1159132465 J * Aiken ~james@tooax6-169.dialup.optusnet.com.au
1159134476 J * shedi ~siggi@inferno.lhi.is
1159134619 M * derjohn Bertl, I do not have any rights to admin the wiki. sry.
1159134849 M * harry starlein: you there?
1159134872 M * Bertl derjohn: well, sent an email to Hollow, hope he'll respond soon
1159134882 M * derjohn Bertl: tar --help |grep numer -> --numeric-owner        always use numbers for user/group names
1159134908 M * Bertl ah, and that works?
1159134912 M * derjohn Bertl, yes, I mailed him, too
1159134960 M * derjohn Bertl, I hope so ;) I am doing  some backups that way. In cases where the users dont have root access to a rsync machine, because r00t is needed to set uid/gid on files
1159135055 M * Bertl good to know! tx for the info!
1159135070 M * Bertl soatola: got the part with the --numeric-owner?
1159135141 M * soatola I just got to the part where I'm told that --owner --group will associate matching names which sounds good.
1159135169 M * Bertl nah, take the --numeric-owner for tar
1159135190 M * Bertl the thing is, the host often has a different view of id<->name relations
1159135204 M * Bertl and you actually want to keep exactly the same numeric ids
1159135220 M * _are_ soatola: do you want to sync a vserver elsewhere from the host or from within the vserver? If from the host: names might have different UIDs/GIDs, so using names *will* fail as you sync your vserver-/etc/passwd with it
1159135223 M * yang heya Bertl
1159135230 M * derjohn Bertl, so that backup is _inside_ a guest? 
1159135246 M * Bertl derjohn: nope on the host
1159135270 M * derjohn h, but alles files with a guest belong to root, nor?
1159135282 M * soatola I was thinks on the host not in the vserver.
1159135303 M * soatola My issue is that when I start postgresql it says
1159135306 M * Bertl derjohn: no, the files of a guest belong to whatever users are valid inside the guest
1159135346 M * Bertl derjohn, soatola: whatever kind of backup tool you use, you want the files to be recreated with _exactly_ the same uid.gid. period.
1159135348 M * soatola main error Cluster is owned by group id 106 which does not exit
1159135403 M * Bertl yeah, that's the result of tar 'remapping' uid->name->uid
1159135627 A * derjohn rethinks ... makes sense. I always used rsync -azv from the host, so I had luck ;)
1159135627 M * _are_ soatola: the answer stays the same, user --numeric-owner for tar / --numerix-ids for backups of vservers from outside
1159135627 M * _are_ you only don't use this if you e.g. want to move your vservera postgres installation to vserverb that has different UID/GIDs
1159135627 M * soatola OK thanks for that, I'll do that next.
1159135675 M * Bertl derjohn: well, actually you want to add -H and --numeric-ids too
1159135884 M * derjohn Bertl, so i copied hard links als own files? well, I rarely use hardlinks, but: agreed. the -numeric-ids stuff: If the uid onthe remote system doesnt exist, it's replaced the the uid anyway. or did I misunderstood something?
1159135952 M * Bertl no, it's replaced with the best match in your /etc/passwd
1159136049 M * harry haha... for the funneh part: http://www.enexgee.net/pics/gif/What%20Really%20Happend.gif
1159136200 M * soatola Bertl: Got the postgresql working on the new host, many thanks for the pointers.
1159136339 M * Bertl you're welcome!
1159136420 M * daniel_hozac Bertl: t8 seems to work fine here.
1159136464 M * Bertl yeah, t6 was a step in the wrong direction
1159136928 M * daniel_hozac Bertl: btw, i played around a bit and http://people.linux-vserver.org/~dhozac/p/k/delta-pidtype-feat01.diff seems to work as well.
1159137276 M * Bertl yeah, that's roughly what I had in mind
1159137305 M * Bertl would prefer to have it kind of maskable though
1159137325 M * Bertl i.e. REAL = 0x10 or so (over whatever type is there)
1159137338 M * id23 Hi Bertl - do you know how many people are using patch-2.6.18-vs2.0.2.1-t8.diff ? try8 ;)
1159137348 M * daniel_hozac yeah, i thought about that, but figured it didn't really buy us anything.
1159137367 M * Bertl guess that is correct ...
1159137369 M * daniel_hozac id23: works fine for me, but i'm only running it on my test box right now.
1159137381 M * Bertl id23: countless :)
1159137408 M * id23 :) ok i will use them in heavy load - upgrade from 2.6.16.16
1159137432 M * daniel_hozac so that should really help ironing out any potential bugs it has left ;)
1159137464 M * Bertl I hope so, please keep us posted
1159137619 M * Bertl id23: and do not hesitate to report any issue you encounter
1159137663 M * id23 sure i will
1159137761 Q * dhansen Ping timeout: 480 seconds
1159137761 Q * zobel Quit: leaving
1159138202 M * doener evening!
1159138298 M * Bertl evening!
1159138320 Q * dna_ Quit: Verlassend
1159138537 M * vasko Bertl: hi, ad my yesterday's problem -- vanilla .18 boots up and runs ok with the same config
1159138700 M * doener may I get a short summary of the problem? :)
1159138807 M * vasko sure - amd64 happily running with 2.6.17.11-vs2.0.2-rc31, some sata some pata disks, nothing special, with the same config anf .18...t6 it does not boot up
1159138833 M * vasko well, actually, boot process is in progress, but always hangs
1159138837 M * vasko on different points
1159138887 M * doener any backtrace? oops? bug?
1159138898 M * vasko no oopses
1159138908 M * vasko it seems to boot slower
1159138956 M * vasko latest place where it got after one of reboots was a note about fscking file systems
1159138959 M * doener hm, my amd64 is not near me and no way to remote reboot it, so I'd rather not mess with it right now :(
1159138983 M * Bertl vasko: might be the same issue akpm encountered
1159139009 M * Bertl give me a second to find the patch, can you test a vserver kernel with one patch applied/removed easily?
1159139030 M * doener akpm boots vserver kernels? :)
1159139044 M * Bertl nope, but amd64 kernels :)
1159139053 M * vasko yes, i can
1159139106 M * doener hm, same config runs on vanilla, so...
1159139274 M * Bertl doener: it is a patch I included (to avoid having issues with the soft lockup) which akpm rejected, because he had bootup issues
1159139294 M * doener ah! that makes sense
1159139599 M * Bertl vasko: could you also upload some details regarding your config and system (cpu/memory/etc) to paste.linux-vserver.org please?
1159139644 M * Bertl vasko: here is the patch, you want to revert (-R) http://vserver.13thfloor.at/Experimental/patch-resched-01.diff
1159140247 M * vasko ok, i reverted this patch and it boots up ok
1159140369 M * Bertl okay, so we will remove it again, but it would be really interesting what config you have
1159140410 M * vasko i will paste it, sec plz
1159140421 M * Bertl doener: when you get to it, it would be interesting to test if it is an amd issue
1159140440 M * Bertl doener: personally I do not see how this could harm the bootup at all
1159140466 M * doener vasko: SMP box?
1159140485 M * Bertl I'd assume so
1159140505 M * doener so do I, but asking is for free ;)
1159140508 J * ekc2 ~EKC@netblock-66-245-252-180.dslextreme.com
1159140513 M * Bertl welcome ekc2!
1159140520 M * ekc2 hi
1159140540 M * vasko pasted on paste...
1159140626 M * vasko it is not a smp box, but if you wish i could try on morning in office on amd64 X2
1159140647 M * vasko (CET)
1159140766 M * Bertl hmm, hmm .. so UP kernel shows that .. interesting
1159140861 M * vasko i do have preemption
1159140868 M * Bertl yeah, I saw that
1159140869 M * vasko could it be related?
1159140886 M * Bertl yes could be, if you want to try, it would be the first I'd check
1159140913 M * vasko need to go to work on morning ;)
1159140938 M * Bertl np, no need to check it _right_now_ :)
1159140954 M * vasko :)
1159140954 M * vasko ok, see you later
1159140962 M * Bertl have a good night!
1159140975 Q * ekc2 
1159140990 J * ekc2 ~EKC@netblock-66-245-252-180.dslextreme.com
1159141057 M * ekc2 If I want to create a copy of a vserver that is busy running a database app, is it possible to pause the vserver, or am I better off stopping the vserver first?
1159141069 M * ekc2 I'm thinking of something like Vmware's "snapshot" feature
1159141121 M * Bertl you can pause it with the hard cpu scheduler's pause feature
1159141148 M * Bertl but as you are interested in a snapshot of the disk, I'd go for the dm snapshot functionality
1159141351 M * ekc2 trying to find out what dm-snapshot is... Is that a vserver-specific feature?
1159141413 M * Bertl nope, dm is device mapper
1159141422 M * Bertl part of basically any 2.6 kernel
1159141509 M * ekc2 ah. modprobe dm-snapshot. very neat. hot backups of mysql
1159141966 Q * ekc2 Remote host closed the connection
1159141995 J * ekc2 ~EKC@netblock-66-245-252-180.dslextreme.com
1159142046 M * ekc2 is online dm-snapshot the best practice for creating regular backups of vservers? or do most people do offline rsync?
1159142114 A * cehteh uses rdiff-backup
1159142164 M * cehteh and i try to avoid software which cant be hot-backuped
1159142270 M * ekc2 yes, but unfortunately i wasn't that carefuly in configuring my vserver
1159142295 M * ekc2 rdiff-backup looks interesting
1159142297 Q * mire Quit: Leaving
1159142328 M * cehteh what software is there that cant be hot-backupped?
1159142347 M * cehteh .o(mysql as usual?)
1159142367 M * ekc2 yup. at least not without a lot of configuration
1159142398 M * cehteh locka tables, do a compressed dump, unlock