1104282013 M * Bertl Doener: you around?
1104282738 M * mikeb but first, food
1104282746 M * mikeb heh whoops wrong window
1104284012 M * rs Bertl: where are you with the NGNET stuff?
1104284034 M * Bertl well, ng8.7 is latest ...
1104284074 M * Bertl except for the virtual 'switch' to do inter context connections almost everything should work
1104284083 M * rs did enrico started to think about how to integrate it in util-vserver ?
1104284108 M * Bertl I guess no, but I guess it's too early for that anyway
1104284115 M * rs ok
1104284161 M * Bertl it's currently dormant waiting for further testing, so feel free to give it a try ;)
1104284180 M * rs yeah would be a good idea :)
1104284260 M * Bertl btw, it seems the quad caught the first SMP issue ...
1104284280 M * rs really? =)
1104284298 J * sannes ~ace@home.skarby.no
1104284305 M * Bertl http://vserver.13thfloor.at/Experimental/OOPS/OOPS-01.txt
1104284323 M * rs was it the one we were searching for?
1104284375 M * Bertl hmm, hard to tell ...
1104284614 M * AnGeL4U` Bertl
1104284616 M * AnGeL4U` sir
1104284626 M * AnGeL4U` if i have vserv patch
1104284634 M * AnGeL4U` i can  put grsec patch ?
1104284671 M * Bertl depends on the version, and you have to slightly modify the grsec patch to apply it ontop (if that is what you are asking?)
1104284705 M * ndim Of course for vserver testing.
1104284801 M * ndim Stupid question: How does one test ngnet without util-vserver support?
1104284823 M * ndim Oh... there are command line wrappers for the syscalls, but nothing more?
1104284840 M * Bertl well, it's kind of tricky, I have to admit
1104284851 M * Bertl bascially you do the following:
1104284876 M * Bertl disable any ip related setup for util-vserver
1104284890 M * Bertl (i.e. do not assign any ip to the vserver)
1104284915 M * Bertl select just a single, non network related service to be started at vserver startup
1104284919 M * Bertl (e.g. syslog)
1104284945 M * Bertl then start the vserver, after that do the setup given in the script
1104284996 M * Bertl after that, enter the vserver, and configure your interfaces like on a normal server
1104285010 M * Bertl then startup the various network related services by hand ...
1104285015 M * Bertl that's it ;)
1104285070 M * AnGeL4U` omg
1104285072 M * AnGeL4U` :)
1104285075 M * AnGeL4U` need save this log
1104285079 M * AnGeL4U` anyone tell me
1104285088 M * AnGeL4U` vserver and grsec is conflict
1104285090 M * AnGeL4U` not good
1104285154 M * Bertl AnGeL4U`: http://linux-vserver.org/Tools+and+patches
1104285169 M * Bertl scroll down to kernel patches
1104285186 M * AnGeL4U` i see
1104285192 M * AnGeL4U` grsec and openwall
1104285266 M * AnGeL4U` i still compile my kernel
1104285267 M * AnGeL4U`  CC [M]  sound/synth/emux/emux_effect.o
1104285267 M * AnGeL4U`   CC [M]  sound/synth/emux/emux_proc.o
1104285268 M * AnGeL4U`   CC [M]  sound/synth/emux/emux_hwdep.o
1104285268 M * AnGeL4U`   CC [M]  sound/synth/emux/soundfont.o
1104285268 M * AnGeL4U` :D
1104285274 M * AnGeL4U` 3 hours for 1 kernel?
1104285281 M * Bertl well, do you use all that stuff?
1104285310 M * AnGeL4U` i type "make"
1104285318 M * Bertl (my test kernel compiles on an old k6/400 not longer than 15 minutes)
1104285328 M * AnGeL4U` hmmmmmmmmm
1104285335 M * Bertl but I just select the stuff I actually need ...
1104285362 M * AnGeL4U` u can teach me step by step?
1104285367 M * Bertl (selection can be done via the various config targets, e.g. make menuconfig or make oldconfig)
1104285397 M * Bertl there are some howtos out there how to configure and build your kernel ... basically it's figuring out what hardware you have ...
1104285409 M * AnGeL4U` yes
1104285412 M * AnGeL4U` and next step ?
1104285449 M * Bertl well, if you know what hardware you have, you 'just' select the drivers/features you need, and disable the others completely
1104285476 M * Bertl http://www.compute-aid.com/linux-kernel/
1104286580 M * Bertl okay, folks, have a good night everyone!
1104286600 N * Bertl Bertl_zZ
1104286614 M * rs have a good night bertl
1104291094 Q * rs Quit: Lost terminal
1104295288 J * mikeb_ ~baptiste@64-40-83-12.dsl.mebtel.net
1104295686 N * lilo_ lilo
1104295725 Q * mikeb Ping timeout: 480 seconds
1104298160 N * mikeb_ mikeb
1104308804 Q * virtuoso Quit: leaving
1104308811 M * sannes morning :)
1104308960 J * virtuoso ~s0t0na@tranq.dorms.spbu.ru
1104310967 J * monrad ~monrad@213083190130.sonofon.dk
1104312378 J * rs rs@ice.aspic.com
1104312749 J * Hollow ~bene@home.xnull.de
1104312919 J * _are_ ~are@mail.foehl.de
1104313170 M * _are_ hi
1104313356 M * Loki|muh moin
1104313393 M * bill Bertl: sory i went to bed. It is solved and work well thx.
1104313441 M * bill Bertl: my postgresql work fine
1104314382 M * _are_ anyone tried samba with vserver? my smbd complains about packets not coming from 127.0.0.1 :-/
1104314385 M * _are_ Dec 29 10:59:27 samba smbd[28617]:   receive_local_message: invalid 'from' address (was 10a010a should be 127.0.0.1)
1104314440 M * _are_ due to this it claims not to reach some client for breaking an oplock and just breaks it after some time
1104314483 M * _are_ this again leads to a) a delay and b) a client that assumes to have correct data, but might have not (e.g. filebased databases like access)
1104314919 J * comdata ~mertins@I3e6e.i.pppool.de
1104317036 M * pusling I am trying to use some gnupg in a vserver for verifying downloads, but it fails, because it cannot lock memory pages. Any good advice? this faq question describes the problem 
1104317039 M * pusling 09:29 -!- virtuoso [~s0t0na@tranq.dorms.spbu.ru] has joined #vserver
1104317042 M * pusling ups
1104317051 M * pusling http://www.gnupg.org/documentation/faqs.html#q6.1
1104323214 Q * sannes Read error: Connection reset by peer
1104324236 J * sannes ~ace@home.skarby.no
1104324289 M * sannes Bertl_zZ : you don't have a fresh BME patch lying around? :)
1104324862 Q * grecea Remote host closed the connection
1104325132 J * grecea ~grecea@h-195-22-237-74.mdl.net
1104325525 J * t___ ~tnichols@CPE-139-168-208-9.sa.bigpond.net.au
1104325685 Q * t__ Ping timeout: 480 seconds
1104325815 J * t____ ~tnichols@CPE-139-168-209-206.sa.bigpond.net.au
1104326010 Q * t___ Ping timeout: 480 seconds
1104326446 M * sannes Bertl_zZ : hm, going to try do some changes on that BME patch .. it'll actually be smaller.. :)
1104326643 Q * t____ Ping timeout: 480 seconds
1104328468 Q * comdata Ping timeout: 480 seconds
1104329050 J * t____ ~tnichols@CPE-139-168-209-162.sa.bigpond.net.au
1104332995 J * mboman ~michael@cm48.sigma230.maxonline.com.sg
1104334083 N * Bertl_zZ Bertl
1104334121 M * Bertl morning folks!
1104334160 M * mboman Hi Bertl
1104334165 M * Bertl pusling: you can give the corresponding cap (but unfortunately it also allows other things)
1104334177 M * Bertl hey michael! how are you?
1104334236 M * Bertl _are_: hmm, seems to be the result of the ip remapping, maybe you can tell samba that those ips are equivalent?
1104334293 M * Bertl sannes: sounds good ... keep me up to date ;)
1104334318 M * mboman I'm ok. Installing FC2 on one of my work laptops..
1104334331 M * Bertl isn't fc3 already out/there?
1104334358 M * mboman yeah.. looked at it and ran away..
1104334374 M * mboman by distro of choice is RHEL3/CentOS3 at the moment
1104334378 M * Bertl ah, i.c. similar with recent mandrake ...
1104334395 M * mboman but I need a 2.6.x kernel for the laptop to get the 802.11b card to work..
1104335066 M * rs hey Bertl!
1104335096 M * _are_ Bertl: found nothing to tell samba lo and eth0:1 is equivalent/valid for localhost. Ofc I could take the samba source and just get rid of that part, but I'd hate to do this with every new samba release again
1104335132 M * Bertl hey rs!
1104335146 M * Bertl _are_: well, it will probably be solved with ngnet anyway ...
1104335161 M * rs Bertl: how are you?
1104335209 M * Bertl hmm .. don't know yet, but should be fine ... ;)
1104335231 M * rs just out from bed ? :)
1104335409 M * Bertl yep, obviously I had to catch up a little ... (sleep wise)
1104335529 M * _are_ Bertl: well, I consider myself somewhat daring to use devel-vserver with devel patch on an mostly untested architecture on a production server. Guess it would turn me into a fool if I added ngnet. :->
1104335712 M * Bertl probably ;) but hey, the fools rule the world ...
1104335726 M * _are_ yes, but I try and be a somewhat reasonable fool.
1104335761 M * _are_ somewhere end of January/February the 2nd box should be there, then I can test the things I have not been daring enough this time
1104335778 M * Bertl btw, what about ignoring 127.0.0.1 as far as samba is concerned?
1104335778 M * rs _are_: the devel version is running on some production servers btw
1104335807 M * mikeb _are_: Heh you and me both - I'm doing the same thing - hadn't actually started up Samba yet though - bummer
1104335813 M * Bertl yes, actually _most_ provider already swithed to 2.6/1.9.x
1104335818 M * _are_ especially I consider gfs on lvm over drbd, anyone ever tried this?
1104335829 M * _are_ ofc with vserver patch
1104335838 M * Bertl drbd is reported working ...
1104335848 M * _are_ rs: but probably not on amd64
1104335850 M * rs why drbd with gfs ?
1104335858 M * rs _are_: good point :)
1104335883 M * Bertl gfs is interesting IMHO, and I'm willing to support that (linux-vserver wise)
1104335901 M * _are_ rs: 2 machines with 2 raids each, 1 box holds 1 raid primary for drbd, the other one the othe raid, gfs on both to enable service migration to the other box, but mostly they should stay as local as possible
1104335921 M * rs we plan to try gfs as a replacement of our current nfs
1104335925 M * Bertl sounds like kind of raid10 over network to me ;)
1104335932 M * _are_ Bertl: it is.
1104335953 M * _are_ basically a share nothing cluster. take out 1 machne and just continue working with the rest.
1104335970 M * rs I guess you can use sw raid with iSCSI with a gfs on top (it's my dream)
1104336007 M * _are_ erm, the machine owner didn't pay for dual-opteron (quad-opteron in the next machine) to let it calc sw raid :->
1104336016 M * _are_ there are cheaper chips for that. ;)
1104336039 M * _are_ though I'd definitly used sw-raid if the cpu usage was not that much of an issue
1104336061 M * rs sw raid is better than hw raid most of the time
1104336072 M * Bertl I guess you are overestimating the impact of sw raid on recent machines
1104336077 M * sannes Bertl : It compiles atleast (with all filesystems enabled) .. want the patch? 
1104336085 M * rs anyway your drbd+lvm solution looks nice
1104336092 M * _are_ but this machine should replace 2 HP9000 HP-UX machines with oracle, 2 linux server, 1 NT server (vmware or some other emulation, not yet fixed)
1104336097 M * Bertl sannes: sure, send it over (or make it available via some url=
1104336150 M * _are_ well, the raid5 code is definitly faster in hardware, 190MB/s <-> 150MB/s (U320 bus, 8*U320 disk)
1104336168 M * sannes two secs, just going to remove some tmp files.. heh
1104336181 M * _are_ for raid1/0 it should not really make a difference if you use hardware or software
1104336208 M * Bertl _are_ raid 5 heavily benefits from reading the disks independantly ... does your hw raid allow for that?
1104336226 M * Bertl (the sw raid code does, btw)
1104336254 M * _are_ I have not asked it. however, measurement on this box with these disks say hw is fatser than sw here.
1104336259 M * _are_ ofc might be different on next box
1104336282 M * rs _are_: which hw raid equipment are you using ?
1104336313 M * _are_ 0000:02:00.0 RAID bus controller: LSI Logic / Symbios Logic MegaRAID (rev 02)
1104336324 M * _are_ it is an U320 X2
1104336345 M * Bertl _are_ what measurements? bonnie?
1104336356 M * _are_ yes
1104336375 M * Bertl do you ahve the results somewhere?
1104336384 M * Bertl (I'm just curious)
1104336397 M * _are_ only the hw-version, I think
1104336424 M * rs _are_: we did some tests with a similar card and got really bad results
1104336428 M * rs comparing to sw raid
1104336430 M * _are_ yes, found the hw-version results
1104336431 M * rs 0000:04:00.0 RAID bus controller: American Megatrends Inc. MegaRAID (rev 20)
1104336479 M * Bertl well, probably depends on the disks too ...
1104336485 M * rs yeah sure
1104336492 M * _are_ oh, amazing: I have 16 disks, so built 2 hw raid5, when I striped over both raidsets with an appropriate chunk size, it had been 10-20% slower than on a single hw raid
1104336516 M * rs hmm you did your tests with raid 5 ? we only tested raid 1+0
1104336525 M * mikeb _are_ That kinda makes sense
1104336532 M * Bertl ah, wait, raid5 over 8 disks?
1104336537 M * _are_ and another surprise (to me at least): ext3 is fatser in reading than reiserfs
1104336544 M * _are_ yes, raid5 with 8 disks
1104336551 M * Bertl okay, that explains it for me ...
1104336553 M * mikeb since in a striped setup you went from being able to access 8 disks in parallel to accessing 2 'vitural' drives in parallel - limited by the bandwidth of the RAID card
1104336569 M * rs reiserfs is most of the time only faster as corrupting :o)
1104336581 M * _are_ well, can't aford raid1 for 2TB in SCSI :->
1104336597 A * Bertl .oO( ah, good that hans isn't here ;)
1104336611 M * rs oops :P
1104336631 M * _are_ well, using reiserfs sucessfully here. can be grown online.
1104336638 M * Bertl _are_: i.c. would be interesting to test a setup with, hmm, 4 disks raid5, all 4 raids striped
1104336644 M * mikeb _are_ You realyl want XFS :)
1104336650 M * _are_ I had corruptions with jfs, xfs, reiserfs and ext3 so far, so I don't care which one anymore
1104336657 M * Bertl _are_: 128k chunk size on both
1104336658 M * mikeb _are_ LOL true
1104336669 M * _are_ mikeb: check google for xfs and amd64 and welcome to Elm Street
1104336688 M * rs _are_: yeah was just kidding, we are also mainly using reiserfs here
1104336689 M * mikeb _are_: Yeah well, cutting edge = beleding edge :0
1104336696 M * mikeb bleeding even
1104336702 M * _are_ jfs had been the worst so far
1104336713 M * _are_ ext3 stabilized very very much since my first encounter
1104336809 M * _are_ xfs crashes now and then, xfs_repairfs fixes this but claims to have no errors found. :->
1104336827 M * sannes there are two patches .. one to make nameidata available in generic_permissions and one for bme .. http://www.sannes.org/bme/
1104336839 M * sannes (bme depends on the first one)
1104336889 M * sannes it is quite straightforward .. the things I worry about is in ncpfs, it should do the same thing.. also I had to update update_inode_time() usage.. but, no biggies..
1104336907 M * _are_ reiserfs crashed a few times on me, reiserfsck usually fixed it and actually admitted to have found errors
1104336963 M * _are_ it is like with these TV mechanics: you bring in a broken one, they replace the fuse, hand it back and charge you extra for bringing in a working TV.
1104336991 M * Bertl sannes: know what, I have to start a larger recompile (will take about 10min to get it running) after that I'll compile your BME and put it through my BME test scripts, okay?
1104337090 M * sannes Bertl : superb! :) I'm on way for some family christmas thing, so I'll be gone, but aslong as I don't time out I see what you have commented on.. :) and I am in no hurry for the test, because I it doesn't touch upon reiserfs which is the fs I'm using.. heh
1104337146 M * Bertl okay, have fun!
1104337290 M * Loki|muh Bertl: does vserver run on sparcs, too?
1104337404 M * Bertl I hope so! at least it did some time ago ...
1104337796 M * Bertl _are_: did I miss the benchmark results somehow? (sorry for being a nuisance)
1104337846 M * _are_ Bertl: no, as I found only parts of them, they are un labeled and I don't know exactly anymore what had been the difference, I didn't post them
1104337870 M * Bertl what a pity ... well, shit happens ...
1104337880 M * _are_ it had all 3 been hardware, this is what i know, the software-results are gone anyway
1104338050 M * Bertl rs: everything working fine at your firends' ?
1104338150 M * _are_ btw, how comes the clock is out of sync on the various vservers?
1104338168 M * _are_ from log-vserver:
1104338170 M * _are_ Dec 29 17:34:15 source -- MARK --
1104338170 M * _are_ Dec 29 17:35:32 web -- MARK --
1104338214 M * _are_ or is this not happening every n minutes in the hour, but every n minutes from syslogd-start?
1104338214 M * Bertl well, it's not a mark at **:*5 is it, it's a mark _every_ 5 minutes (for example)
1104338229 M * _are_ ok, then I am happy again
1104338258 M * Bertl the time base itself should be synced (e.g. via ntpd) on the host
1104338265 M * _are_ yes
1104338372 M * Bertl btw, I was (some time ago) thinking about allowing the various vservers to run with 'unsynchronized' clocks (i.e. arbitrary time settings) but I abandoned that idea because it seemed a very useless feature to me ... any comments on that?
1104338401 M * Bertl (this was a question @ all)
1104338412 M * _are_ it is useful for people that want to fake system time for various applications.
1104338420 M * _are_ legal and non-legal applications for that
1104338481 M * _are_ btw, I'd like to have an interface-renaming / fake mac feature
1104338499 M * Bertl ngnet!
1104338518 M * Bertl (it's already there)
1104338532 M * _are_ have some software here that will only run on 1 mac/ip, as the servers are meant to be redundant and the softwrae has no way to give it 2 licenses, it is quite uselss.
1104338599 M * _are_ vendor doesn't even understand the problem, i just can't fake a mac for a virtual intrface :->
1104338662 M * _are_ it is beyond their brains to understand there might be an eth0 with *no* IP (e.g. dsl-modem linked there)
1104338705 M * _are_ fortunately not the case in the current project, but definitly an issue in the next, unless i want to restructure scripts/network :-(
1104338801 M * _are_ ridiculous anyway, 1 company licenses the software to the mac, then shows you tools to fake the mac. argh. the other licenses it to the ip at least.
1104338845 M * mikeb _are_ Yeah well it coudl be felxlm *shudder*
1104338895 A * mikeb has had nothign but grief every time we've gotten a new flexlm based app - each one has a customized daemon that may or may not work with the current release of linux, etc - silly
1104338943 M * Bertl hmm, what is flexlm? never heard of that?
1104338957 M * mikeb Bertl: flexlm is from Macrovision - its for concurrent use licenses
1104338962 M * _are_ well, I usually refuse to use software that has stupid license blockers, personally I only accept 'commercial' sotware for viru scanners and backup anyway
1104338974 M * _are_ and backup only as I am unable to find a good free one.
1104338982 M * mikeb great concept - even supports license checkouts for laptop users to borrow for offline use
1104338984 M * _are_ always hassle with closed source and no way to fix it.
1104338999 M * mikeb but each vendor has to write their own daemon that runs on top of flexlm
1104339004 M * Bertl _are_: interesting ... you mean commercial software with available source, I hope ...
1104339013 M * mikeb and we've had a number that 'only supported on RH 7.3' etc 
1104339048 M * mikeb Once you get it going - it works great - give you central view of how many license you have, how many are out - the whole 9 yards.  But each new app approaches things ever so differently
1104339088 M * _are_ Bertl: unfortunately not. that is the problem here. If you know a good backup solution, free or commercial with source, running on various unix flavours and windows, supporting libraries and so on, give me a shout.
1104339115 M * _are_ and virus scanners I accept to be comercial as someone has to sit there the whole day and look for new virii / add new patterns
1104339146 M * Bertl what about BackupPC?
1104339153 M * mikeb _are_ Have you checked out Bacula?
1104339154 M * _are_ some link?
1104339170 M * _are_ bacula looked quite bad from the description
1104339175 M * Bertl http://backuppc.sourceforge.net/
1104339191 M * mikeb http://www.bacula.org/
1104339212 M * mikeb _are_ Well its a client server setup which seemed to be what you described above (various client flavors, etc)
1104339235 M * _are_ oh, and I am old fashioned, I insist on backing up to tape.
1104339246 M * _are_ and I want to be told when the tape is full
1104339254 M * Bertl the thing is, both areas (virus protection and backups) are somewhat sensible ... and exactly this stuff you hand over to a blackbox, in the hope it will do the right thing?
1104339260 M * _are_ amanda is a good solution here unless the data doesn't fit on one tape
1104339282 M * _are_ Bertl: i run clam afterwards
1104339286 M * mikeb _are_ yeah thats why we've moved to Bacula - amanda got to be a real PIA given we've got 10+TB to backup
1104339341 M * _are_ and as said, the free backup solutions I know all had some major drawbacks so far.
1104339343 A * mikeb is still amazed the extor^H^H^H^money most commercial backup firms are charging these days - Even with academic pricing - $10-15K was the minimum for a couple Novell boxes, 15 linux boxes, and one Window sbox
1104339349 M * _are_ my own backups are still done via amanda
1104339405 M * Bertl ah, well, it's not my data ... ;)
1104339415 M * _are_ erm, the last one i had to install was for 10k, backups 10 linux boxes, 2 hp-ux, 1 windows server, 2 oracle online (na dyes, I think it is to expensive)
1104339484 M * _are_ na dyes = and yes. I have to get smaller fingers.
1104339560 Q * AnGeL4U` Quit: 
1104339643 M * _are_ backup-pc can't handle windows, it uses smb. you may not access everything on a windows box cia smb :-/
1104339805 M * _are_ seems it is mainly some frontend to collect the data ifrom clients in multiple instances, then use some programm to get it to tape. So it is basically an rsnapshot with restore option and a user interface, not a backup system ...
1104339805 M * _are_ as it won't make sure the data is on taoe after wards.
1104339839 M * _are_ have seen enough disks dieing not to trust any diskbased backup anymore.
1104339840 M * Bertl yep, but those jobs can be done on the backup/tape server ...
1104339867 M * Bertl usually that's what you want ... doing lvl1 backup to a server/disk, and lvl2 backups to tape
1104339902 M * Bertl but YMMV and as I said, it's not my data ;)
1104339922 M * _are_ well, yes, this is the problem. for a comfortable backup you need to know what tape to insert now, and this is what i look for.
1104339936 M * _are_ it is not me sitting in front of the tape drive and restoring peoples data, fortunately :-)
1104340044 M * _are_ well, atm all backups run smoothly (and restore works, too, btw ;) - but sooner or later some systems will be replaced and I will have to search again.
1104340497 J * Rusty` ~rusty@rev.193.226.233.94.euroweb.hu
1104340500 M * Rusty` hy guys
1104340508 M * _are_ Hi Rusty`
1104340577 M * Rusty` i have a quastion: can i use bind9 inside a vserver? :)
1104340640 A * mikeb recalls reading some mailing list posts that said yes, but you had to do some fancy footwoork
1104340684 M * _are_ some days ago, someone did a debian-package with a vserver-enabled bind9, if I recall correctly
1104340705 M * mikeb Yeah I think this was one post talking about it http://www.paul.sladen.org/vserver/archives/200211/0172.html
1104340734 M * mikeb I think you just had to compile bind9 with --disable-linux-caps
1104340844 M * Rusty` mikeb & _are_: thanks for the help
1104341151 M * Bertl btw, it's in the FAQ IIRC
1104341187 Q * rs Quit: home
1104341202 M * Rusty` what is "IIRC"? :)
1104341214 M * Bertl if I remember correctly
1104341240 M * Rusty` oh i see
1104341309 M * Rusty` yes, i read it :)
1104341366 M * Rusty` G. Software compatibility
1104341366 M * Rusty` Bind does not work in a vserver 
1104341420 M * Bertl k
1104341492 M * Rusty` yes, the solution is in the FAQ, sorry...
1104341768 M * Bertl hey, no problem, just checking that it is there ...
1104341778 M * Bertl (otherwise we would have to add it)
1104342091 M * Doener evening!
1104342111 M * Bertl hey Doener! everything fine?
1104342141 M * Doener yeah, feeling better... got a major head ache attack yesterday...
1104342197 M * Bertl that's ugly, you want to do something useful, but you can't because it is keeping you from thinking ...
1104342204 M * _are_ uhm, anyone ever did ssh with X11 Forwarding into a vserver? it claims the authorization cookie is wrong for localhost:36 (played with display offsets as I thought it messed up), however, replacing 'localhost' by the vserver ...
1104342204 M * _are_ ip won't help
1104342221 M * Bertl ssh config, change use localhost to no
1104342260 Q * mboman Quit: One day I'll get that peer and reset HIS connection!
1104342278 M * Bertl X11UseLocalhost no
1104342329 M * _are_ :-)
1104342368 M * ndim Is there a "database" of the issues the different services and programs have with vservers?
1104342395 M * _are_ I'd say the FAQ, but this ssh-issue is not there (yet)
1104342453 M * Bertl then please add it (similar to the other issues)
1104342610 M * Doener dinner time... back later
1104342614 N * Doener Doener|gone
1104342627 M * Bertl enjoy your meal!
1104342772 M * ndim It already is in http://www.paul.sladen.org/vserver/fix0r.txt
1104342882 M * _are_ Bertl: done.
1104342938 M * Bertl thanks!
1104343021 M * Bertl sannes: looks good, at least it passes all my BME tests ... will look into the patch a little later ...
1104343455 M * _are_ uhm, I am lost in doc-space, where can I find a list of the capabilities? especially needed the one to allow a vserver to bind to tcp sockets?
1104343550 M * Bertl the simples way is in /usr/include/linux/capability.h
1104343554 M * Bertl +t
1104343567 M * Bertl but there is a description in the paper ...
1104343655 M * _are_ found it now, http://www.lids.org/lids-howto/node49.html (has distriptions, too)
1104343862 M * Bertl okay, descriptions _are_ in the includefile, btw ;)
1104343901 M * _are_ hmm, ok, last time I had quite some difficulties with these descriptions. it least if I remember that one right
1104343958 M * Bertl not saying that those link doesn't contain _better_ descriptions ;)
1104345246 J * Sirius ~sirius@p548543D5.dip.t-dialin.net
1104345247 M * Sirius ^^
1104345250 M * Sirius +g*
1104345296 M * Bertl welcome Sirius!
1104345322 M * Sirius Bertl: is there any patch for the 2.4.29-pre3 ^^ ?
1104345373 M * Bertl does the 2.4.28 patch fail?
1104345380 M * Sirius mh
1104345381 M * Sirius lol
1104345385 M * Sirius <-- testing
1104345386 M * Sirius *g*
1104345539 J * nox ~nox@c150087.adsl.hansenet.de
1104345570 M * Bertl evening nox! 
1104345576 M * Bertl okay, folks, back later ...
1104345580 M * nox hi Bertl !!
1104345584 M * nox and bye
1104345592 N * Bertl Bertl_oO
1104345601 Q * serving Ping timeout: 480 seconds
1104346037 M * _are_ would one need special capabilities to use a call like 'setsockopt(8, SOL_SOCKET, SO_RCVBUF, [32768], 4)'?
1104346180 Q * Sirius Quit: 
1104349093 Q * _are_ Ping timeout: 480 seconds
1104350847 Q * Rusty` Quit: using sirc version 2.211+KSIRC/1.3.10
1104352530 J * serving ~serving@213.186.172.38
1104354067 J * DuckMaster ~Duck@dyn-83-157-153-59.ppp.tiscali.fr
1104354240 J * _are_ ~are@dsl-213-023-038-196.arcor-ip.net
1104354488 Q * DuckKing Ping timeout: 480 seconds
1104355105 Q * ndim Ping timeout: 480 seconds
1104355277 J * mikeb_ ~baptiste@64-40-83-12.dsl.mebtel.net
1104355368 J * ndim U2FsdGVkX1@helena.bawue.de
1104355430 M * _are_ hi
1104355536 N * mikeb_ miekb
1104355718 Q * mikeb Ping timeout: 480 seconds
1104357014 Q * bill Quit: Leaving
1104357649 N * Doener|gone Doener
1104359210 Q * albeiro Ping timeout: 480 seconds
1104359296 J * albeiro albeiro@linux.gentoo.pl
1104361401 N * Bertl_oO Bertl
1104361404 M * Bertl evening folks!
1104361530 M * Doener wb Bertl 
1104361553 M * Bertl hey Doener! got a few minutes?
1104361599 M * Doener in about 15 minutes, was just about to move the computer to another location... got annoyed of my room-layout ;)
1104361613 M * Bertl okay, let me know when you are there ;)
1104361690 N * Doener Doener|gone
1104361739 M * Loki|muh hehe
1104362905 N * Doener|gone Doener
1104362976 M * Doener i hope my connection won't break... wlan card decided to drop datarate to 1Mbit/s...
1104363018 M * Loki|muh uh
1104363115 M * Doener guess i've chosen the wrong corner of the room as location for the box... ap stands on the opposite side of the building (and a floor below)
1104363152 M * Loki|muh ouch
1104363161 M * Bertl well, get a booster ;)
1104363207 M * Doener hm, datarate went up to 5.5Mbit/s ... i can live with that... it's just dsl that comes over it after all
1104363465 M * Doener Bertl: i'm ready now btw ;)
1104363497 M * Bertl the quad PIII (from lycos) caught the first SMP bug ...
1104363503 M * Bertl http://vserver.13thfloor.at/Experimental/OOPS/OOPS-01.txt
1104363642 M * Bertl btw, the 0x00100100 is a list poison 
1104363857 M * Doener i guess you got the line numbers somewhere? ;)
1104363886 M * Bertl sure, sec
1104364236 M * Bertl longer sec ... brb
1104364254 M * Doener hehe, no problem