1185753634 Q * jescheng Remote host closed the connection 1185753647 J * jescheng ~jescheng@proxy-sjc-2.cisco.com 1185755354 Q * yarihm Quit: Leaving 1185755686 J * FireEgl FireEgl@2001:5c0:84dc:1:4:: 1185756617 J * DoberMann_ ~james@AToulouse-156-1-104-146.w90-30.abo.wanadoo.fr 1185756723 Q * DoberMann[ZZZzzz] Ping timeout: 480 seconds 1185759259 Q * balbir Ping timeout: 480 seconds 1185761419 J * Freax ~Linus@bl7-132-73.dsl.telepac.pt 1185761811 Q * Linus Ping timeout: 480 seconds 1185764634 M * daniel_hozac derjohn: re: " Show next free context ID ", you know that util-vserver 0.30.213 handles this, right? 1185766041 M * daniel_hozac mattzerah: ping 1185766515 Q * slack101 Ping timeout: 480 seconds 1185766766 J * marcfiu ~marcfiu@c-68-39-177-97.hsd1.nj.comcast.net 1185769648 J * balbir ~balbir@122.167.1.101 1185771120 M * mattzerah pong 1185771126 M * mattzerah :) 1185771152 M * mattzerah daniel_hozac: pong 1185771163 M * mattzerah thats probably a better way to play ping pong 1185771888 P * marcfiu 1185773542 J * AStorm ~astralsto@host-81-190-179-124.gorzow.mm.pl 1185775201 Q * jescheng Remote host closed the connection 1185775211 J * jescheng ~jescheng@proxy-sjc-2.cisco.com 1185775615 Q * balbir Ping timeout: 480 seconds 1185775671 N * DoberMann_ DoberMann 1185776410 J * yarihm ~yarihm@84-74-20-183.dclient.hispeed.ch 1185776529 Q * AStorm Quit: WeeChat 0.2.5 1185776810 J * sharkjaw ~gab@158.36.44.106 1185776906 J * balbir ~balbir@122.167.6.121 1185777370 Q * sharkjaw Ping timeout: 480 seconds 1185779154 Q * balbir Ping timeout: 480 seconds 1185779423 J * sharkjaw ~gab@158.36.44.106 1185779790 J * balbir ~balbir@122.167.2.53 1185780405 J * cluk ~cluk@p5B17EFF1.dip.t-dialin.net 1185780534 J * friendly12345 ~friendly@ppp121-44-229-164.lns2.mel4.internode.on.net 1185780795 J * AStorm ~astralsto@host-81-190-179-124.gorzow.mm.pl 1185780867 J * mans ~mans@tru75-6-82-240-34-248.fbx.proxad.net 1185780897 M * mans hello 1185781670 Q * AStorm Remote host closed the connection 1185783423 M * cluk Hi 1185783476 M * cluk I have tried to turn _off_ tagxid for our vserver partition. I removed the mount option and remountet the partition. 1185783501 M * cluk but files are still tagged with the vserver's xid. 1185783562 M * cluk Does anybody know what else I could try? 1185783710 J * dna ~naucki@101-209-dsl.kielnet.net 1185783847 M * danman hi 1185783875 Q * grobie Remote host closed the connection 1185784335 M * danman Hollow: do you perhaps know what I have to change in my vserver qmail setup if I have a qmail on the host also? it sais unable to bind... ;-) 1185784345 M * danman the problem is ok, just can't find solution ;-) 1185784394 M * Hollow danman: you need to configure qmail on the host to not listen on 0.0.0.0 but only on an specific IP 1185784400 M * danman yeah 1185784405 M * Hollow otherwise no guest could bind the same potrs 1185784411 M * danman I know, but where? :-) I've been googling for an hour now ;-) 1185784419 M * danman I understand that :-) 1185784423 M * Hollow TCPSERVER_OPTS 1185784425 M * danman hmm 1185784455 M * danman ok 1185784468 M * danman I think I've got it, thx 1185784509 J * arne^ ~arne@d081226.adsl.hansenet.de 1185784672 M * Hollow danman: actually TCPSERVER_HOST 1185784675 M * danman yeah 1185784675 M * Hollow in conf-common 1185784678 M * Hollow :) 1185784678 M * danman found it 1185784684 M * danman easier than I thought ;-) 1185784689 M * danman Hollow: thx very much! 1185784698 M * Hollow yw! 1185785551 J * AStorm ~astralsto@host-81-190-179-124.gorzow.mm.pl 1185785657 Q * AStorm 1185786189 J * ema ~ema@rtfm.galliera.it 1185786210 Q * Hollow Ping timeout: 480 seconds 1185786462 J * Hollow ~hollow@proteus.croup.de 1185786787 J * AStorm ~astralsto@host-81-190-179-124.gorzow.mm.pl 1185786838 M * meandtheshell Hollow: hi there - any news on gmane? Did you reach anyone via email? 1185786874 M * Hollow meandtheshell: no, i didn't write to them yet, sorry 1185786948 M * meandtheshell I see - well, I changed the URL (made a request that is via webinterface) as well but still the old URL is there http://dir.gmane.org/gmane.linux.vserver 1185787246 Q * AStorm Remote host closed the connection 1185787266 J * AStorm ~astralsto@host-81-190-179-124.gorzow.mm.pl 1185787795 Q * FireEgl Ping timeout: 480 seconds 1185788003 Q * bzed Quit: Leaving 1185790522 N * Bertl_zZ Bertl 1185790527 M * Bertl morning folks! 1185790654 M * eyck morning Bertl 1185790768 P * AStorm ERC Version 5.2 (IRC client for Emacs) 1185790780 M * eyck ! 1185790834 J * AStorm ~astralsto@host-81-190-179-124.gorzow.mm.pl 1185791020 M * Bertl eyck: hey, how is the 'brand new' 2.4.x kernel? 1185791042 M * sid3windr :> 1185791156 M * eyck don't know yet for sure, but it looks ok, I haven't finished testing it yet 1185791300 J * kir ~kir@swsoft-mipt-nat.sw.ru 1185791581 M * derjohn hey, a question to rss.hard and rss.soft limit: ( hard-soft ) appears in the guest as swap. as the limit is called rss, it limits the guests access to physial ram. But in which case will swapping occur then? On an unlimited guest, it takes swap as well as physical ram, on a rss-limited guest not at all? Or is the "rss allocation" of a guest virtualized, so that it might be swap or physical ram on the host ? 1185791709 M * Bertl the last one is correct, although it will not show as RSS when actually swapped out 1185792210 M * derjohn that means on a 1GB RAM + 1GB swap host I can create one vserver guest, with rss.soft=1.5GB and rss.hard=1.8GB without getting in trouble (as long as the host itself need less than 200 MB) ? 1185792316 M * derjohn Bertl, BTW: I saw in the changelogm, that IO aacouting was already added to 2.2.0.x ? thx ! I wasnt able to test it, as I was not able to apply the patch cleanly. but now I'll grab the new vs version :-) 1185793626 M * Bertl excellent! 1185793954 M * Hollow Bertl: FYI, 116 subscribers so far 1185793960 M * Bertl great! 1185794585 J * bonbons ~bonbons@2001:5c0:85e2:0:20b:5dff:fec7:6b33 1185794994 M * bXi Hollow: can you see who the first was? :P 1185794994 Q * ktwilight Read error: Connection reset by peer 1185795001 J * ktwilight ~ktwilight@124.78-66-87.adsl-dyn.isp.belgacom.be 1185795162 M * phreak`` bXi: probably Hollow or Bertl, or daniel_hozac ;) 1185795190 Q * Aiken Quit: Leaving 1185795343 M * Bertl Hollow: btw, was the issue daniel_hozac had with sending/replying to the list resolved? 1185796453 J * DavidS ~david@p57A4A736.dip0.t-ipconnect.de 1185796469 N * DavidS DavidS|Juelich 1185796708 M * daniel_hozac Bertl: yeah. 1185796876 M * mattzerah daniel_hozac: ping 1185796890 M * daniel_hozac mattzerah: hey. 1185796893 M * mattzerah :) 1185796896 M * mattzerah caught you 1185796897 M * Bertl daniel_hozac: great, could you look at the O'Connor issue? 1185796910 M * daniel_hozac mattzerah: you built CentOS 5 RPMs, right? 1185796917 M * daniel_hozac Bertl: that's what i replied to ;) 1185796928 M * mattzerah yup - just needed to get around to releasing them 1185796929 M * Bertl daniel_hozac: aha, when? 1185796941 M * daniel_hozac when the mails weren't getting through. i'll catch up on the thread and resend. 1185796952 M * Bertl daniel_hozac: because it seems it didn't arrive on the list yet 1185796957 M * daniel_hozac yeah, i doubt they will. 1185796974 M * Bertl tx 1185796987 M * daniel_hozac mattzerah: just FYI, i finally got around to building some last night... 1185796992 M * mattzerah cool :) 1185797014 M * mattzerah were they from the original kernel version ? 1185797023 M * daniel_hozac no, "vanilla". 1185797029 M * mattzerah ahh, okay 1185797054 A * mattzerah had many in depth conversation with Bertl about me doing it from the original centos kernel and patches, but it was beyond me :/ 1185797079 M * daniel_hozac as it's 2.6.18-based, that would limit you to 2.0. 1185797086 M * Bertl mattzerah: do you have mainline rpms for centos now? 1185797089 M * daniel_hozac something i'm not really interested in :) 1185797101 M * daniel_hozac mattzerah: i'd really like to join forces, if you wish. 1185797140 M * mattzerah Bertl: no - i bit off more than i could chew, so i havn't done anything beyond my current efforts unfortunaltey - just a vanilla kernel - probably similar to daniel_hozac's 1185797167 M * daniel_hozac mainline == vanilla ;) 1185797175 M * Bertl mattzerah: so the best option for centos is still to take the fc6 kernels? 1185797182 M * daniel_hozac http://rpm.hozac.com/dhozac/centos/5/vserver/i386/ 1185797195 M * daniel_hozac _really_ untested at this point. 1185797251 M * daniel_hozac Bertl: btw, if you want to experiment with IPv6, the princeton machine is on an internet2-connected network. 1185797276 M * mattzerah daniel_hozac: i'm not 100% sure what i could offer - you have been building rpms for longer than me :) (only a few months) but i can contribute something to the process then i'm willing to :) 1185797329 M * daniel_hozac mattzerah: well... testing is always appreciated ;) 1185797338 M * mattzerah daniel_hozac: i probably wont get much time before the w/e but i'll d/l and try it out and let you know 1185797382 M * daniel_hozac okay, cool, thanks. 1185797653 M * Bertl daniel_hozac: ah, excellent! 1185798572 Q * AStorm Remote host closed the connection 1185799855 J * bzed ~bzed@dslb-084-059-110-179.pools.arcor-ip.net 1185800395 Q * DavidS|Juelich Quit: Leaving. 1185801319 Q * mans Read error: Connection reset by peer 1185801877 J * white- ~white-@welcome-man.man.mux.nu 1185801962 M * white- hi 1185801985 M * white- is there a way to bind to 127.0.0.1 in the host 1185802000 M * white- i have tryed nc -l 127.0.0.1 12345 1185802013 M * white- but netstat shows die first ip of the hosts 1185802029 M * daniel_hozac uh, i assume by host you actually mean guest, right? 1185802035 M * white- strace says bind(3, {sa_family=AF_INET, sin_port=htons(12345), sin_addr=inet_addr("127.0.0.1")}, 16) = 0 1185802059 M * white- # CONFIG_VSERVER_REMAP_SADDR is not set 1185802077 M * white- 2.6.20.14-vs2.2.0-i686-111 1185802086 M * white- no, the host 1185802092 M * white- not the guest 1185802096 J * mans ~mans@tru75-6-82-240-34-248.fbx.proxad.net 1185802112 M * white- i have some older hosts where it works fine 1185802114 M * daniel_hozac then you've done something silly. 1185802130 M * white- ;) 1185802131 M * daniel_hozac like, say, put sshd in a network context. 1185802134 M * white- maybe 1185802147 M * white- ohhh 1185802172 M * white- yes i do that 1185802177 M * daniel_hozac why? 1185802181 Q * mans 1185802201 M * white- because sshd should only listen on eth0s first ip 1185802205 M * white- host ip 1185802209 M * white- not the guest ips 1185802229 M * daniel_hozac right.... that's what you'd use /etc/ssh/sshd_config for.. 1185802241 M * white- hmm 1185802259 M * white- why it works with 2.6.16.20-vs2.0.2.rc22-i686-smp-78 1185802317 M * daniel_hozac probably a bug in such an old kernel. 1185802352 M * white- that means if i login on a tty it works? 1185802360 M * daniel_hozac yes. 1185802366 M * white- crazy ;) 1185802368 M * daniel_hozac unless you've put those in a network context too.... 1185802370 M * white- ok many thanks 1185802438 M * Bertl white-: wouldn't make much sense when you confine sshd into a network space (limiting it to certain IPs) and then it would be able to use all other IPs too, would it? 1185802513 J * kaner_ kaner@strace.org 1185802604 Q * kaner Ping timeout: 480 seconds 1185802697 M * white- what is a networkspace? 1185802726 M * white- i use chbind --silent --nid 2 --ip eth0 1185802739 M * white- for sshd on the host 1185802742 M * Bertl an isolation unit, which ensures that you can only use a certain number of ips for the processes 1185802759 M * white- i think chbind ist tool to do that 1185802767 M * Bertl well, funny that --ip eth0 works at all (i.e. looks like legacy to me) 1185802795 M * Bertl but basically it restricts your processes to a subset of the host ips 1185802801 M * white- util-vserver-0.30.213 1185802818 M * Bertl daniel_hozac: what is --ip eth0 supposed to do? 1185802862 M * daniel_hozac it uses the ioctl interface to get the address of that interface. 1185802879 M * Bertl ah, so the primary ip I guess? 1185802884 M * white- hmm, what should i use instead 1185802884 M * daniel_hozac yeah, i guess so. 1185802899 M * white- chbind --ip ip/mask ? 1185802899 M * daniel_hozac ListenAddress in /etc/ssh/sshd_config... 1185802902 M * Bertl white-: well, nothing wrong with that, but you have to understand what it does 1185802945 M * Bertl white-: and in your case, it does what it is supposed to do, it restricts sshd (and all children) to this single host IP 1185802963 M * white- ok, so i must include --ip lo 1185802963 M * danman do you guys use qmail in vserver? 1185802977 M * white- and it should work 1185802980 M * white- imo 1185802980 M * Bertl white-: so, whatever you start from an sshd logon, will only e able to use that single IP 1185802996 M * white- that is shit ;) 1185803002 M * Bertl white-: yes, that would be an option, the question is, why do you use chbind on sshd at all 1185803004 M * white- i look at ListenAddress 1185803028 M * white- the sshd on the host should only use the ip for the host 1185803038 M * white- so that i can start sshd in guest 1185803051 M * Bertl that is what ListenAddress is for, especially if you want to administrate the guests via host sshd :) 1185803059 M * white- but it should be automatic 1185803059 M * Guy- danman: I do 1185803074 M * white- if i change the host ip sshd should work 1185803084 M * white- i do some init script magic ;) 1185803103 M * Bertl yep, good idea, especially an if-up script would be appropriate 1185803109 M * danman Guy-: I've got an issue I just can't solve 1185803113 M * danman I have gentoo... 1185803128 M * Bertl white-: alternatively you could use NAT to accomplish the very same in a flexible way 1185803142 M * danman the mail client sais "qq trouble creating files in queue (#4.3.0)" 1185803156 M * Bertl white-: i.e. bind sshd to a private IP, and DNAT the current host IP to this specific private ip 1185803157 M * white- *argl* i think i use sshd option better... 1185803159 M * danman all I want is another SMTP server on the public address the vserver uses 1185803174 M * white- sshd is always on a private ip 1185803181 M * danman I solved the binding problems etc with the help of Hollow, but I don't know how to solve this 1185803226 M * Guy- danman: replace qmailqueue with a shellscript that invokes strace -ffF -s200 -o /tmp/qmailqueue.log qmailqueue.real "$@" or similar 1185803237 M * Guy- danman: then try again, and look at the strace log 1185803271 M * Guy- danman: or strace the inetd replacement that spawns your qmail-smtpd 1185803284 M * Guy- danman: that will include the strace log of the qmail-queue process 1185803302 M * Guy- danman: my guess is you just botched the permissions of the queue directory, but strace will tell you for sure 1185803318 M * danman hmm 1185803350 M * danman qmailqueue replace you mean in /etc/tcprules.d ? 1185803354 M * danman well I guess so 1185803354 M * danman ;-) 1185803365 M * danman never used strace, I'll try what you wrote 1185803387 M * Guy- no, I don't mean tcprules 1185803398 M * Guy- I mean move the qmail-queue binary to qmail-queue.real 1185803407 M * Guy- and place a shell script called qmail-queue there 1185803412 M * Guy- which does what I said 1185803425 M * danman aaa ok 1185803426 M * Guy- or strace tcpserver/tcpsvd/whatever starts your qmail-smtpd 1185803437 M * danman 1 sec 1185803456 M * Guy- the latter is easier, but you'll have to sift through more logs 1185803479 M * Guy- strace -ffF -s200 -p { pid of tcpserver or equivalent } 1185803489 M * Guy- -o /tmp/qmail.log 1185803503 M * danman hmm 1185803655 N * BobR_oO BobR 1185803687 M * danman compiling strace ;-) 1185803801 M * Guy- strace is your best friend, you should install it by default :) 1185803895 M * danman Guy-: http://pastebin.com/d4e90baf3 1185803905 M * danman I have no idea what to tell from this :-) 1185803927 M * Guy- [pid 26364] chdir("queue") = -1 EACCES (Permission denied) 1185803931 M * Guy- this is your problem 1185803940 M * Guy- qmail-queue can't change into the queue directory 1185803957 M * danman hmm 1185803966 M * danman interesting 1185803992 M * danman Guy-: do you have any idea why? it's a default installation ;-) 1185804002 M * Guy- did you replace qmail-queue with a wrapper or did you strace tcpserver? 1185804011 M * danman strace tcpserver 1185804032 M * danman drwxr-x--- 11 qmailq qmail 4096 Jul 30 12:40 queue 1185804033 M * Guy- is the qmail-queue binary setuid qmailq and setgid qmail? 1185804043 M * danman -rws--x--x 1 qmailq qmail 20656 Jul 30 12:40 qmail-queue 1185804047 M * danman yup 1185804055 M * Guy- it's not setgid qmail 1185804061 M * danman hm? 1185804065 M * Guy- (but maybe that's not strictly necessary) 1185804070 M * Guy- it's just g+x, not g+s 1185804073 M * danman aham 1185804090 M * danman well I'll check 1185804098 M * danman what it does then :-) 1185804128 M * danman same thing 1185804149 M * danman strace is a fun game :-) 1185804178 M * daniel_hozac in general, you can't strace setuid/setgid apps. 1185804190 M * danman oh great :-) 1185804191 M * daniel_hozac i mean, you can, but they won't setuid/setgid. 1185804207 M * danman :-) 1185804209 M * Guy- daniel_hozac: I thought this was only the case when you straced them directly 1185804231 M * Guy- not when they were spawned by a process you were stracing 1185804234 M * Guy- but you could be right 1185804253 M * danman Guy-: do you have any other idea? 1185804260 M * danman log does not show anything 1185804261 M * danman etvc 1185804267 M * danman I am absolutely clueless ;-) 1185804269 M * daniel_hozac ptrace gives you sufficient capabilities to use the process as a puppet. 1185804271 M * Guy- in that case, a more complex wrapper is called for :) 1185804285 M * danman Guy-: hmm, like qmail-scanner etc? 1185804292 M * Guy- danman: no 1185804313 M * Guy- danman: you'll need sudo 1185804319 M * danman I have root 1185804326 M * Guy- I mean the sudo package 1185804329 M * danman ok ;-) 1185804332 M * danman 1 sec 1185804335 M * Guy- give qmaild the right to invoke strace as root 1185804360 M * danman okay :-) 1185804385 M * Guy- then replace qmail-queue with a wrapper that does this: 1185804387 M * Guy- #!/bin/sh 1185804405 M * Guy- exec sudo strace -u qmailq -ffF -s200 -o /tmp/qmailqueue.log qmail-queue.real "$@" 1185804423 M * Guy- this should do the trick 1185804439 M * danman 1 sec (thx for helping me btw) 1185804464 M * Guy- that's what we're here for 1185804546 M * danman qmaild ALL=(ALL) NOPASSWD: /usr/bin/strace 1185804548 M * danman ? :-) 1185804553 M * danman does that look ok? 1185804580 J * Dominik user@sub87-230-126-84.he-dsl.de 1185804583 M * Guy- yes 1185804706 M * danman here goes nothing ;-) 1185804732 M * danman no log created 1185804737 M * danman and qq temporary problem ;-) 1185804754 M * Guy- aren't you setting QMAILQUEUE to something funny? 1185804759 M * danman no, default 1185804778 M * Guy- and you're saying /tmp/qmailqueue.log wasn't created? 1185804781 M * danman yes 1185804805 M * Guy- did sudo log anything into syslog? 1185804814 M * danman no 1185804816 M * danman it's created 1185804817 M * danman sorry 1185804821 M * danman I was stupid ;-) 1185804832 M * danman 1 sec, I'll put it on pastebin 1185804870 J * FireEgl FireEgl@Sebastian.Atlantica.IRCNut.Com 1185804919 M * danman http://pastebin.com/d2abbc36f 1185804979 M * Guy- it exits with an exitcode of 54, which means "54 Unable to read the message or envelope." 1185805004 M * danman sudo: must be setuid root (this is in qmail log) 1185805014 M * danman aham 1185805032 M * danman well that's odd :-) 1185805042 M * Hollow chmod 4711 /var/qmail/bin/qmail-queue 1185805057 M * Hollow and owned by qmailq:qmail 1185805060 M * Bertl that needs to be suid root? 1185805063 M * Hollow no 1185805065 M * Guy- Hollow: been there, done that 1185805069 M * Hollow ok :) 1185805071 M * danman :-) 1185805079 M * Guy- danman: is your sudo installed setuid root? 1185805090 M * danman how can I tell that? :-) 1185805096 M * Guy- danman: ls -la $(which sudo) 1185805098 M * Bertl ls -la? 1185805104 M * danman 1 sec 1185805122 M * danman ---s--x--x 2 root root 107128 Jul 30 16:06 sudo 1185805143 M * daniel_hozac again, that's probably strace ruining things for you. 1185805144 M * Guy- that's somewhat stupid (no read access), but otherwise OK 1185805150 M * Guy- daniel_hozac: how? 1185805155 M * Guy- daniel_hozac: we're doing sudo strace 1185805161 M * Guy- (strace -u in fact) 1185805184 M * Guy- danman: what's the uid of qmailq? 1185805191 M * danman I do not need vpopmail for just sending, right? 1185805202 M * Guy- right 1185805206 M * daniel_hozac so i assume all previous strace's have been killed? 1185805227 M * Guy- daniel_hozac: sure, they exited along with the processes they were tracing 1185805247 M * danman 1 thing I remembeered now is when I was emerge-ing it it said that I have to create a group with gid 200... So I looked at another installation and created it with "nofiles" 1185805251 M * Guy- danman: please confirm you killed the strace that was tracing tcpserver? 1185805253 M * daniel_hozac Hollow: i really like how responsive the list is now, it's pretty much instantenous, whereas before it... well, could take quite some time. 1185805264 M * danman Guy-: 204:201 is qmailq 1185805271 M * danman Guy-: confirmed 1185805274 M * Hollow well, my mailservers don't use greylisting for good reason ;) 1185805287 M * Guy- danman: then the setuid thing worked, getuid() returns 204 as it should 1185805297 M * danman aham 1185805303 M * daniel_hozac greylisting shouldn't delay confirmed addresses though, should it? 1185805305 M * Bertl Hollow: hmm, and that reason would be? 1185805314 M * Dominik Is there an automatic way to update old vserver-configurations-files to the new "directory with configstuff in it"-configuration? I set-up a new Server and put old vservers on it, they dont run well, newly builded with the new config-dir, did. 1185805315 M * Hollow you have to wait for mails 1185805323 M * daniel_hozac why? 1185805328 M * danman Guy-: one more thing, I've installed qmail many times without vserver :-) 1185805339 M * danman with no problem 1185805342 M * Hollow because greylisting denies delivery for xx seconds/minutes 1185805352 M * Bertl danman: on whatever distro you are using right now? 1185805359 M * danman gentoo 1185805363 M * Bertl Hollow: only for _new_ addresses 1185805369 M * danman http://www.gentoo.org/doc/en/qmail-howto.xml 1185805371 M * danman installation method 1185805374 M * Guy- danman: I'm somewhat puzzled by your problem, I've never seen this happen before 1185805386 M * danman :-( 1185805390 M * Bertl danman: is that a test machine? 1185805391 M * daniel_hozac Dominik: not really. 1185805400 M * danman Bertl: the guest is a test, the host isn't ;-) 1185805408 M * danman the host is running qmail with the other public IP 1185805408 M * Hollow Bertl: i just don't think it is needed, RBL filters almost all spam, i only get about 10 spam messages per day which are filtered by dspam then .. 1185805431 M * daniel_hozac Dominik: the old configuration file is sufficiently minimal that converting them manually shouldn't be much of a problem... 1185805432 M * Bertl danman: hmm, you could try to start qmail via chroot, that would eliminate Linux-VServer from the equation 1185805433 M * Guy- danman: does the qmail-inject test described in the gentoo howto work? 1185805442 M * Hollow without RBL it would be like 1000 messages a day 1185805448 M * Bertl Hollow: no problem with that, as long as it works :) 1185805450 M * danman Guy-: didn't try, but now I will 1185805451 M * Hollow :) 1185805468 M * danman Bertl: aham, well maybe I'll try that if there's not another solution 1185805475 M * Dominik daniel_hozac: I tried that, when i type "vserver vserver start" nothing happens. 1185805491 M * Hollow danman: i did not follow entirely, what is your problem? 1185805499 M * Bertl Dominik: well, maybe it simply started? 1185805514 M * Bertl Dominik: what does vserver-stat show? 1185805523 M * daniel_hozac Dominik: so what did you do? vserver .. build -m skeleton --context ? 1185805527 M * Guy- Hollow: 'qq trouble creating files in queue (#4.3.0)' is his problem 1185805558 M * Dominik Bertl: "vps aux | grep vserver" shows nothing :p 1185805594 M * Dominik daniel_hozac: I build just a debootstraped vserver, then copy the config and edit it to fit the old vserver. 1185805600 M * Hollow danman: did you try to repair the queue? 1185805619 M * daniel_hozac Dominik: from vserver ... build? 1185805632 M * Guy- Hollow: http://pastebin.com/d2abbc36f 1185805647 M * Hollow yeah, i saw that 1185805651 M * daniel_hozac Dominik: so you updated all of the symlinks, changed the context, all the names? 1185805658 M * Dominik daniel_hozac: yes 1185805663 M * danman Hollow: I had a queue-repair 1185805667 M * danman but I'll try it again 1185805668 M * Hollow ok 1185805698 M * daniel_hozac Dominik: so what does vserver --debug ... start say? (use paste.linux-vserver.org) 1185805711 M * danman Guy-: qmail-inject: fatal: qq trouble creating files in queue (#4.3.0) 1185805753 M * danman Hollow: doesn't helpp 1185805825 M * Guy- danman: did you get a new strace log in /tmp? does it look any different? 1185805830 M * danman Hollow: well, my problem is that I install everthing here: http://www.gentoo.org/doc/en/qmail-howto.xml until vpopmail (1,2 chapter) and I think I should be able to smtp now (modifying /etc/tcprules.d to let me in with my IP), but it sais "qq trouble creating files in queue (#4.3.0)" 1185805836 M * danman Guy-: oh 1185805839 M * danman 1 sec 1185805840 M * danman ;-) 1185805862 M * danman actually I didn't get a new log 1185805885 M * danman but I'll check that again 1185805885 M * Guy- wasn't the old one replaced? 1185805900 M * danman no 1185805917 M * Bertl hmm, interesting what a google search for this one gives :) http://marc.info/?l=qmail&m=112656545311133&w=2 1185805929 M * Hollow heh 1185805932 M * Hollow i just read the same 1185805945 M * Hollow actually the qmail ebuilds were horrible, and still are to some degree 1185805951 M * Hollow but i'm working to fix the situation ;) 1185805960 M * danman :-)))) 1185805967 M * Hollow see http://planet.gentoo.org/developers/hollow/2007/07/15/experimental_qmail_ebuilds 1185805974 M * danman should I try this? :-)http://pyropus.ca/software/queue-repair/ 1185806003 M * danman Guy-: it writes 1185806004 M * danman sudo: must be setuid root 1185806004 M * danman qmail-inject: fatal: qq temporary problem (#4.3.0) 1185806013 M * danman now 1 sec and I'll see the log 1185806023 M * danman because before I put back qmail-queue ;-) 1185806031 M * danman no log created... 1185806084 M * Guy- I'm beginning to feel that this is a gentoo specific issue, btw :) 1185806088 M * danman ;-))))))))) 1185806104 M * Guy- I use qmail on Debian 1185806106 M * danman I'll check this: http://pyropus.ca/software/queue-repair/ 1185806108 M * Bertl ah, no, can't be, as danman has it working on another gentoo host, yes? 1185806116 M * danman Bertl: yes :-) on many hosts 1185806123 M * Hollow i have plenty of qmail installs too 1185806125 M * danman I've been using qmail for years 1185806144 M * danman but this is in vserver ;-) 1185806158 M * danman maybe I should try an earlier ebuild version 1185806166 M * Hollow no, please not 1185806170 M * Hollow :) 1185806173 M * Dominik daniel_hozac: paste.linux-vserver.org -> No code specified. 1185806174 M * Hollow rather try the new ones 1185806180 M * danman :-)))) 1185806188 M * Bertl danman: give the chroot a try, maybe with a ncontext (just to save your soul) 1185806197 M * Hollow there is also a draft of the qmail-howto for the new ebuilds 1185806198 M * daniel_hozac Dominik: what? 1185806207 M * Hollow http://dev.gentoo.org/~hollow/qmail-howto.xml 1185806212 M * danman Bertl: I don't know how to do that :-) 1185806219 M * danman Hollow: wow 1185806293 M * danman Hollow: maybe I'll try your package: -) 1185806313 M * Dominik daniel_hozac: http://rafb.net/p/idx4nK98.html , paste.linux-vserver.org didnt work. 1185806314 M * danman I've got nothing to lose I guess ;-) 1185806315 M * Hollow ok :) 1185806323 M * Hollow report any problems so i can fix them in my overlay 1185806327 M * danman ok ;-) 1185806335 M * daniel_hozac Dominik: WORKSFORME... 1185806338 M * Bertl Dominik: interesting, works here ... 1185806364 M * danman Guy-: well thanks very much!!! I will try Hollow's stuff for the next step 1185806367 M * Hollow Bertl: btw, a friend of mine said the spam filter is too restrictive, but i don't know what he tried to paste ;) 1185806388 M * Hollow danman: probably remove the qmail stuff before (also the queue etc) 1185806400 M * Bertl Hollow: well, would be interesting ... maybe you can ask him? 1185806406 M * Hollow will do 1185806408 M * danman Bertl: do you have by any chance any howto on how to do this chroot? are you only say I should "chroot /vserver/... /bin/bash" and start it somehow? :-) 1185806416 M * daniel_hozac Dominik: it's starting the guest, so you'll have to check inside why the processes aren't sticking around. 1185806417 M * danman Hollow: yeah, I will 1185806478 M * Bertl danman: I would use chbind --nid 42 --ip -- chroot /path/to/guest/ /bin/bash 1185806509 M * danman hmm 1185806511 M * Bertl danman: then (re)start the service 1185806521 M * danman well ok 1185806547 M * Bertl if the error is the same, it is not Linux-VServer process context related 1185806549 M * Guy- danman: szivesen :) 1185806564 M * danman Guy-: :-)))))) 1185806573 M * danman lol 1185806663 M * danman Bertl: 1185806663 M * danman * nothing in /proc [ !! ] 1185806666 M * danman * ERROR: svscan failed to stop 1185806669 M * danman :-) 1185806689 M * Bertl simply mount proc 1185806719 M * danman ok 1185806771 M * danman samething 1185806775 M * danman Bertl: thx btw 1185806807 M * Bertl danman: so it _looks_ very much like a non Linux-VServer issue after all 1185806820 M * danman maybe 1185806820 M * danman ;-) 1185806843 M * Bertl well, if you leave out the chbind, you can test that for sure 1185806849 M * danman ah well, I'll play around on another server without vserver and do the same steps 1185806860 M * Bertl danman: but be careful, it could mess with your host IPs 1185806864 M * Guy- danman: you could just try qmail-inject in the chroot 1185806870 M * danman Bertl: then maybe at midnight :-) 1185806874 M * Guy- danman: that doesn't need tcpserver or svscan 1185806888 M * danman Guy-: well ok ;-) 1185806889 M * Bertl danman: once you get a working installation, you can simply move that into a Linux-VServer guest ... 1185806902 M * danman Bertl: I have working installation ;-) 1185806912 M * danman Bertl: that seems like a good idea 1185806914 M * danman ;-) 1185806918 M * Bertl danman: so, maybe 'virtualize' that for a test? 1185806922 M * danman ok 1185806932 M * Bertl danman: you can use the rsync build method for that 1185806933 P * white- bye 1185806963 M * danman hmm 1185806975 M * danman I understand 1185806975 M * danman ok 1185807017 M * Dominik is it possible that there is a problem when the vservers are in image-files which are mounted to the needed location? 1185807050 M * Bertl Dominik: no, that just adds unnecessary overhead but it otherwise fine 1185807078 M * danman Guy-: doesn't work 1185807111 M * Guy- danman: then it's not a vserver issue, but I was almost certain of that 1185807119 M * danman Guy-: ok, could be ;-) 1185807120 M * Guy- danman: it's some obscure problem with your particular qmail installation... 1185807135 M * danman Guy-: ok, I understand... time to try Hollow's stuff :-) 1185807143 M * Guy- definitely 1185807158 M * Hollow i also fixed tons of vpopmail bugs yesterday 1185807163 M * Hollow but you do not want vpopmail right? 1185807173 M * danman yup, just simple smtp 1185807212 M * danman (I got on cbl.abuseat.org all the time, and since I seperated to 2 IPs I don't get back) 1185807215 M * danman ;-) 1185807226 M * danman so this is why I need the other smtp too ;-) 1185807468 Q * cluk Quit: Ex-Chat 1185807828 M * derjohn daniel_hozac, re: " Show next free context ID " -> no, are user-space context id implemented? I.e. I can leave the context file out in the config ? 1185807864 M * Bertl derjohn: if you leave it out, it won't work in the future :) 1185807889 M * Bertl derjohn: after all, dynamic context have been deprecated a _long_ time ago (so nobody should use them anymore) 1185807910 M * derjohn aren't they replaced by userspace-generated conext ids ? 1185807915 M * derjohn *context 1185807923 M * Bertl yes, that is likely what you are referring to 1185807972 M * derjohn i mean, daniel_hozac wrote that to me, as he obsiously revwied the page i created: http://linux-vserver.org/Somehandyscripts 1185808008 M * daniel_hozac derjohn: if you don't specify --context when you build it, one will be generated for you. 1185808033 M * Bertl s/generated/allocated/ 1185808053 M * derjohn ah, so only at build-time. 1185808064 M * daniel_hozac well, there's not really an allocation algorithm. it's just context.next + 1. 1185808077 M * daniel_hozac i.e. currently no check if another guest already has it. 1185808088 M * Bertl hmm, maybe you should add that :) 1185808105 M * Dominik someone know something that would be needed inside the config-directory, when the old configfile looks like this http://rafb.net/p/TVaSAm90.html ? Or something that must be changed in the image of the vserver. 1185808110 M * Bertl daniel_hozac: also would make sense to wrap around once you run out of contexts 1185808128 M * daniel_hozac sure, there are a lot of ways to improve it. 1185808220 M * daniel_hozac Dominik: i'm really not familiar with the legacy utils, but maybe you should try switching to the plain initstyle. 1185808275 M * Bertl Dominik: vserver new build -m skeleton --context 42 --interface eth0:123.123.241.229/?? --interface eth0:123.123.240.18/?? --hostname nointernet.com --initstyle sysv 1185808298 M * Bertl Dominik: that should give you a config quite similar to the old one 1185808317 M * Bertl Dominik: replace the ?? with a proper prefix 1185808370 M * Bertl Dominik: although I doubt that nointernet.com is the proper hostname :) 1185808475 M * Dominik yeah i changed it due to privacy interest 1185808483 M * danman :-) 1185808508 M * Dominik i tried that idea.. exactly the same as further above 1185808538 J * DavidS ~david@p57A48416.dip0.t-ipconnect.de 1185808548 M * Dominik nothing is starting, i think there ist something crappy with the rc script from the SuSE-Linux-Guest 1185808557 M * daniel_hozac most likely. 1185808558 M * Bertl Dominik: was the guest moved? 1185808575 N * DavidS DavidS|Juelich 1185808575 Q * FireEgl Read error: Connection reset by peer 1185808585 M * Bertl Dominik: what util-vserver version? 1185808597 M * Bertl daniel_hozac: in what version is the environment fixed? 1185808608 M * Dominik util-vserver-0.30.213 1185808627 M * Dominik i don't like SuSE-Linux.. 1185808649 M * Bertl fortunately it is gone for some time now, IIRC :) 1185808650 M * daniel_hozac trunk only, i think. 1185808665 M * Bertl Dominik: it probably fails on the missing ENV entries then 1185808681 M * Bertl Dominik: you might want to try the CVS/SVN version 1185808710 M * daniel_hozac yeah, it's post-0.30.213. 1185808712 M * Bertl Dominik: or simply switch to plain init style with a proper guest init 1185808733 M * daniel_hozac http://people.linux-vserver.org/~dhozac/t/uv-testing/util-vserver-0.30.214-pre2571.tar.bz2 is a fairly recent snapshot. 1185808752 M * Hollow do we expect a relase soon? 1185808762 M * Bertl we do, we do :) 1185808770 M * Dominik you think the svn version will solve that problem with the enviroment? 1185808771 M * Hollow otherwise i'll add the snapshot to portage 1185808888 M * Dominik although.. the work is over now.. i will try the svn version tomorrow, thanks for your help. 1185808901 M * Bertl you're welcome! 1185808957 M * daniel_hozac Bertl: have you heard anything from blino? i haven't seen any patches yet. 1185808971 P * friendly12345 1185808971 M * Bertl sec, let me ask him ... 1185809033 M * daniel_hozac as for a release, i guess we could do that... 1185809035 Q * Dominik Quit: leaving 1185809044 M * daniel_hozac it needs some testing first though, especially the debootstrap stuff. 1185809131 M * Bertl can you put it on the princeton machine? 1185809137 M * daniel_hozac sure. 1185809183 M * Bertl I'll do the complete guest isntall thing for all 'known working' distros including the untested rpm/yum cases 1185809186 J * stefani ~stefani@tsipoor.banerian.org 1185809192 M * Bertl welcome stefani! 1185809203 M * stefani h. 1185809242 M * harry short on words, long on name 1185809254 M * harry is stevani a girl/woman? 1185809263 M * Bertl harry: indeed :) 1185809275 M * harry then maybe: short on words, long on legs ;) 1185809284 M * harry Bertl: is she your wife? ;) 1185809293 M * harry Bertl: how old are you? ;) 1185809293 M * Bertl harry: nope :) 1185809308 M * danman lol 1185809325 M * Bertl harry: (are you) long on words and short on legs? :) 1185809376 M * daniel_hozac Bertl: note that "support" for various Ubuntu versions has been added. 1185809395 M * harry Bertl: /me is long on words, meaning: i talk a shitload of crap 1185809399 M * harry and long on legs 1185809404 M * harry since i'm 1,97m tall 1185809409 M * Bertl daniel_hozac: ah, great, I had them tested before, so they will be tested again 1185809455 M * daniel_hozac so how do you build the utils on the princeton machine? packages or manually? 1185809472 P * stefani 1185809563 J * FireEgl FireEgl@4.0.0.0.1.0.0.0.c.d.4.8.0.c.5.0.1.0.0.2.ip6.arpa 1185809596 M * danman scared her off :-) 1185809638 N * BobR BobR_oO 1185810008 M * Bertl daniel_hozac: package with ta should be fine 1185810024 M * Bertl daniel_hozac: just override the missing dependancies (everything should be there) 1185810045 M * daniel_hozac okay. 1185810397 M * daniel_hozac they're installed. 1185810407 M * Bertl excellent, tx! 1185810836 J * fluor- ~fluor@tanneries.squat.net 1185810840 M * fluor- hello 1185810901 M * fluor- working on a vserver-based setup at the moment, and wondering about Mailman 1185810923 M * fluor- if apache and postfix are to run within separate vservers, 1185810932 M * fluor- where should Mailman be? 1185811103 M * Bertl you mean, because it uses apache and postfix? 1185811210 M * Bertl I think you will have several options here, the most obvious ones are: 1) in the apache guest, with access to the ML dirs, 2) on a separate guest, with access to the ML dirs ... 1185811239 M * fluor- yes 1185811250 M * fluor- well, it has to sit on the same vserver as apache 1185811258 M * fluor- and communicating with postfix is not so hard, 1185811268 M * fluor- but communication from postfix to mailman in that case is more tricky 1185811322 M * Bertl I don't know the mailman details, but what kind of postfix-mailman communication is going on there? 1185811475 Q * ensc Ping timeout: 480 seconds 1185811664 J * ensc ~irc-ensc@p54B4D9BE.dip.t-dialin.net 1185811983 Q * meandtheshell Remote host closed the connection 1185812039 J * meandtheshell ~markus@85.127.114.20 1185812047 M * Bertl daniel_hozac: http://paste.linux-vserver.org/4664 1185812216 Q * fs Read error: Connection reset by peer 1185812222 J * fs fs@213.178.77.98 1185812464 Q * fluor- Ping timeout: 480 seconds 1185812854 Q * ema Quit: leaving 1185812904 A * Freax boas sir's :) 1185813040 M * Bertl Freax: aha? 1185813073 J * onox ~onox@kalfjeslab.demon.nl 1185813130 M * Freax sorry Bertl i'm portuguese and a do a /ame for my people :P 1185813133 M * Freax lol 1185813169 M * Freax i translate 1185813179 A * Freax hello sir's :þ 1185813321 M * Bertl ah, i.c. ... 1185813530 M * Bertl daniel_hozac: note: that is issued even after a 'service util-vserver restart' 1185813530 Q * DavidS|Juelich Read error: Connection reset by peer 1185813654 J * DavidS ~david@p57A48416.dip0.t-ipconnect.de 1185814028 N * DoberMann DoberMann[PullA] 1185815553 M * daniel_hozac Bertl: looks like there are two installs, one in /lib, and one in /usr/lib64. 1185815573 M * daniel_hozac the latter being the initscript and the first being the vserver you execute. 1185815579 M * Bertl how comes? 1185815598 M * daniel_hozac how come there are two installs? i have no idea :) 1185815621 M * Bertl well, I would expect there was only one there last time I checked :) 1185815645 M * Bertl I assume you updated the existing rpm/install now, so that should have been replaced, no? 1185815652 M * daniel_hozac yep, definitely. 1185815661 M * daniel_hozac but /lib isn't used for RPM-installs at all. 1185815670 M * Bertl indeed, as it is 64bit 1185815708 M * Bertl okay, maybe we can fix that once and for all in the next release? 1185815729 M * Bertl i.e. the fact that the tools use different pathes they get from I do not know where? 1185815751 M * Bertl I mean, is that a problem or why do different 'lib' dirs pop up every now and then? 1185815759 M * daniel_hozac because you have two installs. 1185815792 M * Bertl okay, fine, but obviously, vserver start checks for /usr/lib64/util-vserver/vshelper, which I assume is the correct one 1185815795 M * danman hi! I emerged layman, how do I have to use Hollow's special overlay? does anyone know/ :-) 1185815798 M * danman ? 1185815807 M * danman it doesn't get shown with layman -L 1185815815 M * Hollow danman: layman -f -a hollow 1185815817 M * daniel_hozac /usr/lib64/util-vserver/vshelper is what the initscript sets. 1185815826 M * Bertl daniel_hozac: I would also assume that 'servive util-vserver restart' does the right one too, no? 1185815832 M * daniel_hozac exactly. 1185815833 M * danman ok 1185815842 M * daniel_hozac but you're running the vserver script from the other installation. 1185815856 M * Hollow danman: -f tells layman to look at the remote overlay list 1185815858 M * danman wow, that was easy :-) 1185815861 M * daniel_hozac (i.e. the one in /sbin rather than the one in /usr/sbin) 1185815864 M * Hollow somewhere on the gentoo websapce 1185815865 M * danman Hollow: cool, thx 1185815946 M * Bertl daniel_hozac: hmm, why does it end up in /usr anyway? 1185815955 M * daniel_hozac that's where the RPMs install them. 1185815971 M * Bertl interesting ... 1185816030 M * Bertl okay, so there must have been a manual install before 1185816040 M * daniel_hozac indeed. 1185816048 M * Bertl I removed that now (with uninstall) but I guess we have to reinstall the rpms 1185816067 M * daniel_hozac i don't think so. 1185816100 M * Bertl oh yes, I think so :) 1185816102 M * daniel_hozac with the different prefixes, they should've only shared files in /etc and /var... both of which i hope aren't removed by uninstall :) 1185816104 M * Bertl rpm -V util-vserver-build-0.30.214-0.pre2571 1185816127 M * Bertl from uninstall: 1185816128 M * Bertl rm -f '/etc/vservers/.distributions/rh9/apt/sources.list' 1185816159 M * daniel_hozac right.... well, i guess it wouldn't hurt, but those files aren't used anymore. 1185816169 M * Bertl and what about those: 1185816170 M * Bertl rpm -V util-vserver-sysv-0.30.214-0.pre2571 1185816187 M * daniel_hozac it trashed the initscripts too? that's not cool. 1185816225 M * daniel_hozac though i guess it's kind of expected... 1185816237 M * daniel_hozac so i guess, just a re-install should fix those problems. 1185816285 M * Bertl that's what I did :) 1185816385 M * Bertl hmm, a build now gives: Can not find file for 'file'; aborting 1185816413 J * duckx ~Duck@tox.dyndns.org 1185816438 M * daniel_hozac d'oh. 1185816495 M * daniel_hozac which dist? 1185816503 M * Bertl currently potato :) 1185816514 M * Bertl vserver potato32 build -m debootstrap --context 10101 --hostname potato01.debian.org --interface eth1:10.1.1.32/24 -- -d potato -m http://archive.debian.org -- --arch i386 1185816567 M * daniel_hozac ah, yeah, i didn't add potato/woody/sarge to util-vserver. i need to relax that check somehow though... 1185816580 M * Bertl note: it worked before 1185816609 M * daniel_hozac yeah. 1185816727 M * daniel_hozac okay, try it again now. 1185816832 M * Bertl /usr/lib64/util-vserver/vserver-build.debootstrap: line 159: syntax error near unexpected token `"$__DISTRIBDIR/$DISTRIBUTION/debootstrap.script 1185816839 N * DoberMann[PullA] DoberMann 1185816956 M * daniel_hozac wow, well, i told you it was raw ;) 1185816970 M * Bertl np, that is why we are testing :) 1185816980 M * daniel_hozac okay, try now? 1185816992 M * Bertl looks much better ... 1185817180 M * Bertl W: Failure trying to run: chroot /etc/vservers/.defaults/vdirbase/potato32 dpkg --force-depends --install var/cache/apt/archives/base-files_2.2.0_i386.deb var/cache/apt/archives/base-passwd_3.1.10_i386.deb var/cache/apt/archives/ldso_1.9.11-9_i386.deb 1185817205 M * Bertl so it installs, but leaves nothing working ... 1185817249 M * daniel_hozac is potato still supposed to be installable? 1185817266 M * Bertl no idea, it was last time, otherwise I would not have it in my list 1185817283 M * daniel_hozac that's strange then. 1185817284 M * Bertl I doubt they changed the repository though 1185817293 M * Bertl moving on to woody now 1185817310 M * danman Hollow: I'm lame at this overlay thing ;-) how can I emerge it after I've done what you did, and it downloaded it? :-) I put some PORTDIR_OVERLAYs in make.conf, but nothing happens :-) I've used this howto: http://gentoo-wiki.com/TIP_Overlays 1185817332 M * danman sorry :-) 1185817339 M * Hollow you need to include the layman make.conf: 1185817347 M * Hollow source /usr/portage/local/layman/make.conf 1185817354 M * danman that's in my /etc/make.conf 1185817357 M * Hollow in /etc/make.conf 1185817357 M * danman still no change ;-) 1185817367 M * danman emerge netqmail -p ,right? 1185817370 M * Hollow well, did you keyword the ebuilds? 1185817371 M * danman I should see yours? :-) 1185817375 M * danman no ;-) 1185817382 M * Hollow they are all ~arch of course 1185817392 M * danman eix should show it then, no? 1185817400 M * danman eix netqmail 1185817407 M * danman it only shows the normal ones... 1185817418 M * Hollow if you updated eix after fetching the overlay, yes 1185817430 M * danman that's what happened ;-) 1185817433 M * danman hmm 1185817458 M * danman and simply mail-mta/netqmail etc? 1185817463 M * Hollow source /etc/make.conf; echo $PORTDIR_OVERLAY 1185817486 M * danman /usr/portage/local/layman/hollow 1185817504 M * Hollow emerge '>=mail-mta/netqmail-1.05-r9' 1185817506 M * danman ok, it's in 1185817517 M * danman great, thx!!! 1185817533 M * Hollow this forces the new version andw ill show you why it cannot use it 1185817570 M * danman hmm 1185817793 Q * matti Remote host closed the connection 1185817799 J * mans ~bellinuxi@86.68.254.93 1185817815 M * tam Is there a trick to get traceroute to work in a guest? 1185817817 M * mans hello 1185817839 M * daniel_hozac tam: alias traceroute=tracepath 1185817870 M * tam thanks. We should add that to the FAQ on the wiki. 1185817876 M * daniel_hozac feel free. 1185817957 M * mans i've build vserver with vserver, i got some network trouble, then i'm trying to create with newvserver, but i'd like to use --force because my vserver directory is a partition named as the name of the vserver 1185817994 M * mans but i can't use --force, is there a solution 1185818019 Q * balbir Ping timeout: 480 seconds 1185818040 M * daniel_hozac what is the problem, exactly? 1185818063 M * mans the error 1185818066 M * mans /etc/vservers/sirius1' 1185818066 M * mans exist already; please try to use '--force', or remove them manually. 1185818084 M * daniel_hozac that's the configuration directory. 1185818098 M * mans but i don't know how tu sue --force with newvserver 1185818119 M * daniel_hozac you probably can't. you shouldn't use newvserver anyway. 1185818136 M * Bertl mans: newvserver is deprecated 1185818168 M * mans okay, then i better to solve my network trouble than use newvserver 1185818225 M * mans thx for the info 1185818333 M * Bertl maybe you can elaborate on 'your network trouble' 1185818401 Q * jescheng Remote host closed the connection 1185818412 J * jescheng ~jescheng@proxy-sjc-2.cisco.com 1185818799 Q * duckx Remote host closed the connection 1185818822 J * duckx ~Duck@tox.dyndns.org 1185818918 M * Bertl daniel_hozac: http://svn.mandriva.com/cgi-bin/viewvc.cgi/packages/cooker/util-vserver/current/SOURCES/ 1185818955 M * Bertl daniel_hozac: according to blino "far from perfect, but it works (somewhat)" 1185819217 M * daniel_hozac there are a couple changes that might break other build methods, and will definitely make it less secure. 1185819277 M * Bertl maybe you can adapt/integrate/fix it somewhat? 1185819293 M * Bertl I could probably push it into cooker (via blino) 1185819322 M * Bertl (which means that it will be in the upcoming stable mandriva) 1185819343 J * matti matti@acrux.romke.net 1185819351 M * daniel_hozac i'm just curious if there's a reason for moving the rpmdb mountpoint... 1185819424 M * daniel_hozac also, that patch can't possibly build. 1185819473 M * daniel_hozac oh, it can, unless you rerun auto*. 1185819717 M * Bertl let me put it this way: if we can fix/integrate it, there is no reason whatsoever to add this patch to the mandrake rpm :) 1185819748 M * daniel_hozac well, the reason i didn't implement urpmi support was because i couldn't find a way to configure it. 1185819750 M * Bertl which probably saves us a lot of: 'broken/insecure in mandriva, please use mainline' 1185819780 M * daniel_hozac and unless this is setting some magic environmental variables, i don't see this doing it either. 1185819792 M * Bertl soo, how does it work then? 1185819806 A * Bertl assumes it does at least work for blino :) 1185819833 M * daniel_hozac probably uses whatever configuration is on the host for the initial build, and then whatever's inside the guest after that. 1185820019 M * Bertl so we need a patch to urpmi? 1185820045 M * daniel_hozac that was my conclusion the last time i looked at it, i'd love to proven wrong though... 1185820133 M * daniel_hozac i.e. if one of these variables is the magic knob, let me know and we can work out the other things, i guess. 1185821831 Q * trapni Ping timeout: 480 seconds 1185821860 M * Supaplex humm 1185822494 Q * meandtheshell Quit: Leaving. 1185822874 J * meandtheshell ~markus@85.127.112.109 1185823045 M * Bertl daniel_hozac: debian guests are still not cleaned up, right? 1185823059 M * daniel_hozac etch, lenny and sid should be 1185823075 M * Bertl what about sarge? 1185823082 M * daniel_hozac nope. 1185823098 M * daniel_hozac not right now, at least. 1185823098 M * Bertl okay, let's see how it goes for etch :) 1185823162 M * Bertl daniel_hozac: lenny is already there? 1185823176 M * daniel_hozac how do you mean? 1185823182 M * Bertl I thought it would not be released before 2008 or so 1185823193 M * daniel_hozac it's not released, but that's "testing" for now. 1185823207 M * daniel_hozac as far as i understand Debian terminology, at elast :) 1185823226 M * Bertl aha, hmm ... so what do I specify for that then? 1185823235 M * Bertl lenny or testing ? 1185823236 M * daniel_hozac -d lenny? 1185823249 M * Bertl okay 1185823262 M * daniel_hozac i guess technically either should work... but only lenny is handled by the scripts. 1185823409 J * Radiance 823c898cd6@halt.1984world.eu 1185823867 Q * arne^ Quit: WTF 1185824055 Q * hardwire Quit: Coyote finally caught me 1185824131 J * hardwire ~bip@radius1.mtaonline.net 1185824656 Q * hardwire Quit: Coyote finally caught me 1185824670 J * hardwire ~bip@216.152.176.9 1185824768 J * fatgoose ~samuel@76-10-147-136.dsl.teksavvy.com 1185824774 N * fatgoose samuel 1185824794 M * samuel hi all 1185824857 M * samuel I'm playing with zabbix ( www.zabbix.com ) 1185824932 M * samuel anyone has developped snmp module to monitor vservers activity? 1185825085 J * Pazzo ~ugelt@195.254.225.136 1185825136 Q * DavidS Quit: Leaving. 1185825246 M * Bertl samuel: I'm pretty sure somebody did :) 1185825252 M * samuel yeah 1185825262 M * Pazzo sorry for disturbing, I have a quick qestion: how do I get an overall netstat output? e.g. "netstat -antupo" in context 1? 1185825272 M * daniel_hozac Pazzo: yep. 1185825281 M * daniel_hozac just make sure it's _network_ context 1. 1185825299 M * daniel_hozac (for vs2.2+) 1185825332 M * Bertl daniel_hozac: seems the post install script/whatever for etch brings in the locale setting (into the guest) 1185825351 M * daniel_hozac yeah, i'd guess so 1185825353 M * Bertl daniel_hozac: making perl complain quite a lot 1185825380 M * Bertl can't we get a clean environment for all those scripts (being run inside the guest)? 1185825398 M * Pazzo daniel_hozac: it's 2.2.x - and how do I set a network context? (I didn't really follow latest changes) 1185825401 M * Pazzo Hi Bertl! 1185825417 M * daniel_hozac well, i'd prefer not to remove the ability to pass in environment variables. 1185825423 M * Bertl Pazzo: check out ncontext --help 1185825430 M * daniel_hozac i.e. without using env on the inside. 1185825438 M * Pazzo Bertl: regarding etch and perl complaints: just apt-get install locales && dpkg-reconfigure locales 1185825450 M * Bertl Pazzo: yeah, I know, tx 1185825477 M * Pazzo Bertl: I have never noticed "ncontext" - is it quite new? 1185825483 M * Bertl daniel_hozac: what about having a global 'env' file? 1185825501 M * Bertl daniel_hozac: or what kind of env vars are you talking about? 1185825582 M * Bertl daniel_hozac: or to put it the other way: how can I get a 'clean' guest install without 'polluting' that guest with ENV info from the host? 1185825651 M * daniel_hozac env -i PATH=$PATH vserver ... i guess. 1185825682 M * daniel_hozac Pazzo: added in 0.30.212 or 0.30.213... can't quite remember. 1185825693 M * Pazzo daniel_hozac: thnx! 1185825810 M * Pazzo Bertl: still regarding the etch thingy: I used to create my own debootstrap scripts (slightly modifying the original ones), letting them call a cleanup script at the end of the "second_stage_install()" 1185825845 M * Bertl Pazzo: nice, but that should not be required 1185825945 M * Bertl daniel_hozac: hmm, is it intentional, that the post install script starts syslog and cron? 1185825986 M * daniel_hozac i think so... IIRC, the Debian post-install script will start the guest as normal. 1185825995 M * Pazzo Bertl: hmmm... just gave a look at it, it does mostly things for my personal need (bashrc, shadowconfig, vimrc) but also some useful things: working sources.list, fill apt cache, remove unused / evil start scripts in an upgrade-safe manner, set correct mailname... 1185826021 M * Bertl daniel_hozac: okay, just asking 1185826023 M * daniel_hozac i guess that should be changed to the way the redhat post-install script works. 1185826050 M * Bertl btw, the env trick seems to work fine 1185826211 M * daniel_hozac okay, good. 1185826471 M * Bertl maybe some comment on the 'post install script' wouldn't hurt 1185826485 M * daniel_hozac ? 1185826494 M * Bertl i.e. something like: >>> Executing post install script (this may take a while) ... 1185826527 M * daniel_hozac well, it shouldn't take a while... 1185826533 M * Bertl could maybe come from the post install script itself, which here seems to have troubles resolving debian/kernel.org? 1185826544 M * Bertl Err http://mirrors.kernel.org etch Release.gpg Temporary failure resolving 'mirrors.kernel.org' 1185826548 M * Bertl Err http://security.debian.org etch/updates Release.gpg Temporary failure resolving 'security.debian.org' 1185826575 M * Bertl host side works fine here ... 1185826583 M * daniel_hozac mkdir -p /etc/vservers/.defaults/files; cp /etc/resolv.conf /etc/vservers/.defaults/files 1185826584 M * matti Eh. 1185826588 A * matti is sad :< 1185826599 M * daniel_hozac how come? 1185826743 M * matti My aunt died. 1185826744 M * matti ;/ 1185826883 M * daniel_hozac oh, i'm so sorry. 1185826962 M * eyck we're sorry. 1185827207 M * matti Heh. 1185827228 M * matti Now I am completely alone. I hate my life. 1185827230 M * matti ;-/ 1185827272 M * danman Hollow: same problem :-(, but now I have some logs about it ;-) 1185827281 M * danman @4000000046ae49d205dce09c alert: unable to opendir todo/0, sleeping... 1185827296 M * danman Hollow: anyways, nice package! 1185827298 M * Hollow do you have enabled highvolume use flag? 1185827301 M * danman yup 1185827306 M * danman I shouldn't? :-) 1185827322 M * Hollow well, i never actually tested this one, and i doubt it is necessary 1185827356 M * danman ok, I'll take it out 1185827381 M * Hollow from qmail.org: 1185827383 M * Hollow Big-todo is only useful if you need to inject email into the queue when qmail-send is not running. This is not the usual case. 1185827417 M * danman hmm, then it's is definitely not important ;-) 1185827441 M * danman 1 sec 1185827490 M * Hollow but it's probably a good idea to test this patch, otherwise it should be removed ;) 1185827508 M * danman :-) 1185827523 M * danman we'll find out 1185827529 M * danman if that's the problem :-) 1185827531 M * Hollow well, i'm going to setup a qmail test vserver anyways 1185827539 M * Hollow before the new eclass goes live. 1185827552 M * danman sounds great 1185827552 M * Bertl matti: make some friends, go bowling or play billards ... 1185827609 M * danman Hollow: still have the same log message ;-) 1185827658 M * danman ah well 1185827666 M * danman I'll google a bit 1185827708 M * Hollow danman: you need to recreate your queue 1185827721 M * Hollow the highvolume patch assumes a different layout of todo 1185827775 M * danman I deleted /var/qmail before reemerge-ing 1185827775 M * danman ;-) 1185827779 M * Hollow danman: looks like this here: http://paste.linux-vserver.org/4667 1185827780 M * Hollow hmm 1185827821 M * danman yup 1185827847 M * danman hmm 1185827855 M * danman on my host todo has directories 1185827862 M * danman maybe all I need is to create them... ;-) 1185827867 M * danman will try 1185827878 M * Hollow maybe the host has highvolume but works? 1185827879 M * Hollow :) 1185827887 M * danman oh yeah, you're right :) sorry 1185827888 M * danman hehe 1185827983 M * danman this is probably not the main problem 1185827997 M * danman because otherwise everything is the same 1185827999 M * danman ah well 1185828094 J * gerrit ~gerrit@166.129.20.54 1185828136 M * Hollow danman: please try the following: 1185828180 M * Hollow strace -ff -o qmail-send.trace qmail-start "`cat /var/qmail/control/defaultdelivery`" 1185828190 M * Hollow then qmail-inject a message 1185828198 M * Hollow and try to find the right trace log :) 1185828234 M * danman http://pastebin.com/d4d7275e2 1185828243 M * danman I didn't qmail-inject 1185828245 M * danman w8 1185828297 M * Hollow looks odd. it should search for mess/x instead of mess/x/x 1185828316 M * danman hmm 1185828323 M * danman now that's true! 1185828334 M * Pazzo I'm off, thnx & cya! 1185828335 M * danman injecting didn't produce new messages in strace 1185828348 M * danman qmail-inject: fatal: qq trouble creating files in queue (#4.3.0) 1185828350 Q * Pazzo Quit: ... 1185828371 M * danman what if I create does dirs? :-) 1185828372 M * danman lol 1185828400 M * danman new msg 0 1185828400 M * danman warning: trouble reading todo/0 1185828400 M * danman new msg 1 1185828400 M * danman warning: trouble reading todo/1 1185828403 M * danman :-) 1185828408 M * danman (I created those 2) 1185828416 M * danman hmm 1185828438 M * danman something is definitely wrong ;-) 1185828462 M * Hollow grep 'chdir.*queue' qmail-send.trace* 1185828464 M * Hollow :) 1185828511 M * danman aaaa /me stupid :-) 1185828513 M * danman sry ;-) 1185828520 M * danman that was lame ;-) 1185828533 M * Hollow so? :) 1185828576 M * danman qmail-send.trace:chdir("queue") = 0 1185828576 M * danman qmail-send.trace.18755:chdir("queue/mess") = 0 1185828576 M * danman qmail-send.trace.18756:chdir("queue/mess") = 0 1185828576 M * danman qmail-send.trace.18757:chdir("queue") = 0 1185828589 M * danman but I created mess/1/1 and so on 1185828590 M * danman ;-) 1185828600 M * samuel forgot about zabbix 1185828604 M * Hollow and it works now? 1185828611 M * danman no ;-) 1185828659 M * Hollow "trouble createing files in queue" is only returned if qmail-queue cannot chdir to the queue directory 1185828664 M * danman yes 1185828671 M * danman I'll give you the strace, 1 sec 1185828766 M * danman I think we already found out that chdir is the problema 1185828813 M * danman http://pastebin.com/d349e800b 1185828913 M * danman qmHandle sais "Qmail isn't running, can't send messages! 1185828944 M * danman and it isn't runnign 1185828945 M * danman great 1185828951 M * danman 1 sec 1185828961 M * danman I forgot the bind :-(((( 1185828987 Q * gerrit Quit: Client exiting 1185828988 Q * onox Quit: leaving 1185829042 M * danman /var/lib/supervise... :-) 1185829047 M * danman didn't see that one ;-) 1185829063 M * danman you changed that :-) 1185829069 M * Hollow yeah, maybe it is not a good idea, as it will break a lot of stuff ;) 1185829082 M * Hollow but maybe we find a neat way to convert automagically 1185829090 M * danman :-) 1185829112 M * Hollow but doesn't really matter for our test 1185829185 M * danman yeah, just as I see, qmail doesn't even start ;-) 1185829195 M * danman conf-common is fixed to bind to the right IP 1185829198 M * danman ln -s is ok 1185829201 M * danman in /service 1185829212 M * Hollow well, this is all just for qmail-smtpd 1185829213 M * danman and now I don't see why it doesn't start ;-) 1185829218 M * Hollow but the error has nothing to do with it 1185829221 M * danman yes 1185829228 M * danman :-( 1185829237 M * Hollow the error is in qmail-send 1185829251 M * danman true 1185829273 M * danman and it is there constantly 1185829274 M * danman :-) 1185829297 M * Hollow ok, please stop all qmail processes and we'll try the trace again 1185829316 M * Hollow and recreate the queue before that 1185829328 M * danman ok 1185829342 M * danman is there a nice way of creating the queue? 1185829348 M * danman or just emerge --config netqmail? :-) 1185829356 M * Hollow /usr/bin/queue-repair.py --create --no-bigtodo --split 23 /var/qmail 1185829357 M * danman after deleting /var/qmail/queue? 1185829363 M * danman ok ;-) 1185829490 M * mnemoc hi, anyone with experience with netconsole? 1185829522 M * Hollow danman: qmail-start "`cat /var/qmail/control/defaultdelivery`" 1185829522 M * danman strace -ff -o qmail-send.trace qmail-start "`cat /var/qmail/control/defaultdelivery`" 1185829525 M * danman alert: cannot start: qmail-send is already running 1185829528 M * Hollow this will show log messages on the console 1185829531 M * danman ok 1185829543 M * danman it sais it's running, but it isn't running ;-) 1185829548 M * Hollow well, is it running? :) 1185829551 M * Hollow hm 1185829562 M * danman ok 1185829570 M * danman status: local 0/10 remote 0/20 1185829573 M * danman now I'll inject 1185829791 M * daniel_hozac mnemoc: some. why? 1185829814 J * Aiken ~james@ppp121-45-205-95.lns1.bne1.internode.on.net 1185829826 M * danman http://pastebin.com/d10784a79 1185829861 M * mnemoc daniel_hozac: i can't get it working :\ 1185829882 M * daniel_hozac what's the problem? 1185829908 M * mnemoc daniel_hozac: i have a netcat listening port udp/514, which receives ls > /dev/udp/$ip/514 from the host with troubles 1185829922 M * danman Hollow: do you think this is a vserver issue btw? 1185829927 M * Hollow danman: no 1185829944 M * Hollow looks ok to me.. 1185829948 M * Hollow but no inject 1185829950 M * mnemoc daniel_hozac: but netconsole (loaded with modprobe) doesn't send anything there 1185829958 M * danman Hollow: maybe I'll rebuild a vserver 1185829965 M * danman and try with fresh install 1185829974 M * daniel_hozac mnemoc: how did you load netconsole? 1185829982 M * danman I have no idea why it does what it does 1185830001 M * mnemoc daniel_hozac: modprobe netconsole netconsole=@/eth0,514@$ip/ 1185830026 M * danman Hollow: there was inject, I don't know if I should paste qmail-send.trace.????? also :-) 1185830034 M * danman there are many 1185830037 M * mnemoc daniel_hozac: i tested with an umoun/mount, dmesg shows the printk stuff but the remote never saw it 1185830047 M * Hollow yeah, did you delete the old ones before? :) 1185830053 M * danman yup :-) 1185830088 M * mnemoc daniel_hozac: that damn machine is crashing twice per day, and i need "logs" to probe the problem :\ 1185830104 M * daniel_hozac mnemoc: is that IP local to the box in question? 1185830110 M * daniel_hozac i.e. on the same network? 1185830133 M * daniel_hozac otherwise i'm pretty sure you're gonna need your gateway's MAC-address after the /. 1185830142 M * Hollow danman: no output on the console beside status? 1185830148 M * mnemoc daniel_hozac: remote, on a diffenret DC 1185830186 M * danman Hollow: yes, only that 1 line 1185830187 M * danman http://pastebin.com/d48e46772 1185830212 M * danman http://pastebin.com/d31e764e9 1185830228 M * danman http://pastebin.com/d77b5c27e 1185830243 M * danman http://pastebin.com/d12a6d38e 1185830259 M * danman http://pastebin.com/d6b3697f 1185830262 M * danman :-) 1185830274 J * onox ~onox@kalfjeslab.demon.nl 1185830289 M * danman but Guy- said that it is definitely an issue with chdir 1185830308 M * Hollow well, none of the chdir in the traces failed :) 1185830318 M * danman :-) 1185830333 M * Guy- it was only an issue with chdir when we tried to strace without sudo 1185830348 M * danman ah ok 1185830348 M * danman ;-) 1185830385 M * danman I have a feeling that something really sucks in my system ;-) maybe I should rebuild it ;-) 1185830466 M * Hollow well, all the traces you posted look ok 1185830472 M * Hollow nothing failed 1185830482 M * Guy- I don't think stracing qmail-send will tell us anything 1185830492 Q * meandtheshell Quit: Leaving. 1185830493 M * Guy- the problem is with qmail-queue, and that's not started by qmail-send 1185830516 M * danman yes 1185830542 M * danman and what if I would install qmail-scanner for queue? :-) 1185830550 M * danman I have that on the host 1185830554 M * danman or that uses qmail-queue too? 1185830563 J * thei0s ~G0D@BSN-61-50-203.dial-up.dsl.siol.net 1185830573 M * Hollow qmail-queue is started by qmail-inject 1185830583 M * Guy- and by qmail-smtpd 1185830600 M * danman aham 1185830645 M * Guy- did you already try the sudo strace qmail-queue wrapper with qmail-inject? 1185830665 M * danman well :-) 1185830671 M * danman I think we tried it, yes 1185830685 M * Guy- and what was in the strace log? 1185830694 M * danman :-) 1185830695 M * Bertl welcome thei0s! 1185830700 M * Hollow oh. i ate to much gingerbread 1185830704 M * danman Guy-: goodquestion :-) 1185830704 M * thei0s Bertl: re 1185830711 M * danman ah well, thx guys! I think I will try it on a freshly built system (Hollow's stage4) with no upgrades 1185830726 M * danman it's probably the best idea now ;-) 1185830739 M * Hollow you never know .. :) 1185830741 M * mnemoc daniel_hozac: may i paste you something (8 lines) on query? 1185830743 M * danman ;-) 1185830751 M * daniel_hozac mnemoc: sure. 1185830753 M * Hollow debugging qmail is not a simple task unfortunately 1185830764 M * danman that's for sure :-) 1185830766 M * Guy- danman: be sure to tell us when you find out what it was :) 1185830774 M * danman Guy-: you can count on me ;-) 1185830800 M * danman good night! 1185830815 M * danman and thx for everything, you guys are great! no other channel like this! 1185830816 M * Hollow danman: if it still does not work for you with a fresh install ping me and i'll tar you a working qmail guest for testing 1185830834 M * danman Hollow: great, ok, thx very much! 1185830876 M * Hollow you're welcome 1185831032 M * Guy- yeah 1185831121 M * Bertl danman: did the 'rsync working system' fail too? 1185831182 M * danman Bertl: haven't tried that 1185831220 M * danman it's not that simple ;-) 1185831227 M * danman but I will try maybe 1185831228 M * danman now 1185831240 M * Bertl hmm, why not? (that simple :) 1185831316 M * danman :-) 1185831624 M * thei0s I was wondering if there is a vserver documentation rewrite planned and a filename cleanup/reordering of configuration files and programs? imho it would be great to somehow start all over and this time the documentation and detailed specifications should be written as first thing, later the kernel patches and user programs -- this way everything would have documentation and be elegantly designed and possibly more intuitive to use 1185831677 M * daniel_hozac no :) 1185831703 M * thei0s how many people are actually involved in the project? 1185831719 M * daniel_hozac depends on your definition of involvement, i guess. 1185831751 M * thei0s yeah... the most active ones.. 1185831760 M * daniel_hozac active ones doing what? 1185831789 M * thei0s heh :) anything? 1185831844 M * danman Hollow: 1185831875 M * danman http://pastebin.com/d7ef1e1ef 1185831877 M * daniel_hozac no idea... 10-20 people? 1185831891 M * danman this is something I've never met when not in vserver... 1185831891 M * danman ;-) 1185831914 M * danman this might be the source of the problem perhaps? 1185831921 M * Hollow danman: yeah, this is a bug in the current qmail ebuild.. workaround by: groupadd -g 200 nofiles 1185831926 M * danman ok 1185831933 M * danman then no ;-) 1185831937 M * Hollow well, the ebuild in my overlay does The Right Thing(tm) 1185831939 M * daniel_hozac though technically, i guess most active ones would limit it to 3 people (most active coder, most active tester, most active documenter). 1185831940 M * danman (fresh system is soon done) 1185831948 M * danman Hollow: lol , yeah ;-) 1185831990 M * Hollow daniel_hozac: i'm not actively involved then :) 1185831998 M * daniel_hozac neither am i ;) 1185832014 M * thei0s heh 1185832083 M * Hollow i think the new ml subscriptions will give an idea how many people are actively following the development/discussions 1185832099 M * Hollow "actively" 1185832101 M * Hollow :) 1185832103 M * daniel_hozac right 1185832122 M * Hollow 124 subscribers currently 1185832217 N * DoberMann DoberMann[ZZZzzz] 1185832994 M * danman Hollow, Guy-: after building Hollow's stage4 I ran emerge netqmail, and the problem is the same!!! :-) 1185833013 M * danman totally fresh stuff 1185833080 M * danman so I am back to saying that this must be a vserver issue 1185833089 M * danman :) 1185833150 M * daniel_hozac so if you do the exact same thing on a regular machine, it doesn't happen? 1185833174 M * danman daniel_hozac: that I don't know ;-) 1185833183 M * danman that's a bit harder to test 1185833195 M * danman but I will 1185833427 Q * dna Quit: Verlassend 1185833894 Q * thei0s Quit: Leaving. 1185834377 M * danman daniel_hozac: it works like a charm 1185834383 M * danman on a clean non-vserver system 1185834397 M * danman Hollow, Guy-: so this must be a vserver issue ;-) 1185834412 M * danman (I use stage4 from Hollow) 1185834419 M * danman (in the vserver) 1185834427 M * Hollow no idea.. i have a lot of qmail vservers that run like a charm :) 1185834435 M * Hollow but you never know 1185834447 M * danman well yeah 1185834466 M * danman ah well, I'll get postfix I guess 1185834471 M * Hollow hehe 1185834473 M * danman all I need is an smtp :-) 1185834486 M * danman maybe later on it will just start working :-) 1185834500 M * danman thx for every effort! 1185834513 M * danman this is really a channel with the best! thx 1185834518 M * danman good night! 1185834536 M * Hollow i have heard a lot of things about qmail, but it has worked for me until now :) 1185834545 M * Hollow i never used another MTA :) 1185834550 M * danman :-) 1185834552 M * Hollow night 1185834570 M * danman well I'll see, but for now I give up :-) 1185834604 M * Hollow sorry for that, maybe postfix will work better for you :) 1185834704 M * danman :-))) well we'll see and really thx, byebye 1185836065 Q * bonbons Quit: Leaving 1185836428 J * lilalinux__ ~plasma@dslb-084-058-220-073.pools.arcor-ip.net 1185836861 Q * lilalinux_ Ping timeout: 480 seconds 1185837363 M * baldy Bertl: ping? 1185837430 M * Bertl baldy: pong! 1185837673 Q * Vudumen Quit: leaving 1185837708 J * Vudumen ~vudumen@perverz.hu 1185837836 Q * Vudumen 1185837850 J * Vudumen ~vudumen@perverz.hu 1185838203 Q * mans Quit: Ex-Chat 1185838686 M * Bertl okay, off to bed now ... have a good one everyone! 1185838692 N * Bertl Bertl_zZ 1185838791 J * thei0s ~G0D@BSN-61-50-203.dial-up.dsl.siol.net 1185839159 Q * Aiken Ping timeout: 480 seconds 1185839750 Q * samuel Quit: samuel