1237334842 Q * dowdle Remote host closed the connection 1237335797 J * saulus_ ~saulus@c150011.adsl.hansenet.de 1237336175 Q * saulus Ping timeout: 480 seconds 1237336309 J * juscelio ~muntro@189.70.215.181 1237336639 Q * harobed Ping timeout: 480 seconds 1237336812 J * block ~ove@c-e677e455.19-26-64736c14.cust.bredbandsbolaget.se 1237336831 M * block is a dummy0 device really needed for nated vservers? 1237336848 M * block or interface rather 1237337057 M * daniel_hozac no. 1237337087 M * daniel_hozac interfaces don't really matter to Linux. 1237337146 M * daniel_hozac you can put the addresses on eth0 or lo if you don't want dummy0. 1237337212 M * daniel_hozac or any other interface that's up. 1237337649 M * block ok 1237337665 M * block i got this problem tho.. http://paste.linux-vserver.org/12787 1237337695 M * block and if i would try to ssh to 192.168.254.21:40999 (even tho sshd is not running in guest) im getting connected to the host's sshd 1237337718 M * daniel_hozac did you restrict the host's sshd to the host's IP addresses? 1237337749 M * daniel_hozac http://linux-vserver.org/Frequently_Asked_Questions#When_I_try_to_ssh_to_the_guest.2C_I_log_into_the_host.2C_even_if_I_installed_sshd_on_the_guest._What.27s_wrong_here.3F 1237337795 M * block no, but i have never ever needed to before 1237337815 M * block always used 2.6.22.19 before tho 1237337835 M * block 2.6.28.7-grsec2.1.13-vs2.3.0.36.7 now 1237337939 M * daniel_hozac you've always needed that. 1237337949 M * block no 1237337952 M * daniel_hozac yes. 1237337960 M * block no really no :P 1237337968 M * daniel_hozac it's the way Linux-VServer has always worked. 1237337992 M * daniel_hozac it's IP isolation. if the host binds the port, guests can't. 1237338068 M * block so you're saying that for example two guests with different "nated" ip's cant listen at :22 at the same time? 1237338078 M * daniel_hozac sure they can. 1237338085 M * block on the other hand 1237338090 M * daniel_hozac it's only the host that's unlimited. 1237338095 M * daniel_hozac and as such needs to limited manually. 1237338100 M * block the host's sshd listened at another port than 22 before 1237338135 M * daniel_hozac which would be why you haven't seen the problem before. 1237338141 M * daniel_hozac guests would have been unable to use that port. 1237338181 M * block mhm, okey. i did as in the link, but im getting 1237338184 M * block Mar 18 01:58:36 [sshd] error: Bind to port 22 on 192.168.254.21 failed: Cannot assign requested address. 1237338184 M * block Mar 18 01:58:36 [sshd] fatal: Cannot bind any address. 1237338192 M * daniel_hozac on the host? 1237338200 M * block oh 1237338201 M * block lol 1237338210 M * block shouldve read 1237338468 M * block okay now its working, thank you for the help daniel_hozac :) 1237338873 M * daniel_hozac you're welcome. 1237339360 Q * jp- Ping timeout: 480 seconds 1237339614 J * jp- ~jp@itdev.judelawfirm.com 1237341487 Q * juscelio 1237343857 P * glen 1237344153 J * snooze ~o@1-1-4-40a.gkp.gbg.bostream.se 1237344225 Q * block 1237349403 J * ghislainocfs21 ~Ghislain@LPuteaux-151-41-11-129.w217-128.abo.wanadoo.fr 1237349733 Q * ghislainocfs2 Ping timeout: 480 seconds 1237350767 Q * ex Ping timeout: 480 seconds 1237351089 J * ex ex@valis.net.pl 1237351666 Q * ex Ping timeout: 480 seconds 1237352008 J * ex ex@valis.net.pl 1237356727 J * sharkjaw ~gab@149-240-82.oke2-bras6.adsl.tele2.no 1237359953 J * duckx ~Duck@81.57.39.234 1237360590 N * pmenier_off pmenier 1237362132 J * doener ~doener@i59F5482F.versanet.de 1237362521 J * dna ~dna@53-215-103-86.dynamic.dsl.tng.de 1237363198 J * yarihm ~yarihm@hg-public-dock-155-dhcp.ethz.ch 1237363921 J * cga ~weechat@62.196.2.6 1237363984 J * grosgis ~admin@abeille.lifl.fr 1237364016 M * grosgis hello everyone 1237364184 M * grosgis Have a problem automounting nfs share (netapp appliance) in vserver guest. Mounting take a long time to be up. 1237364283 Q * hparker Quit: Read error: 104 (Peer reset by connection) 1237364380 J * ard ~ard@shell2.kwaak.net 1237364933 M * cehteh grosgis: first guess would be name lookup, next some portmapper issues 1237365114 J * davidkarban ~david@193.85.217.71 1237365623 M * grosgis thank for reply, name lookup is ok . looking for portmaper issue. 1237365646 J * esa ~esa@ip-87-238-2-45.static.adsl.cheapnet.it 1237366093 Q * yarihm Ping timeout: 480 seconds 1237366129 Q * esa Ping timeout: 480 seconds 1237366172 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d 1237367364 J * harobed ~harobed@arl57-1-82-231-110-14.fbx.proxad.net 1237367390 Q * xdr Quit: Lost terminal 1237367395 Q * davidkarban Ping timeout: 480 seconds 1237367420 J * davidkarban ~david@88.86.104.103 1237368511 J * yarihm ~yarihm@guest-docking-nat-3-023.ethz.ch 1237369767 Q * larsivi Ping timeout: 480 seconds 1237370233 J * Punkie ~Punkie@2a01:5f0:1:80:216:d4ff:fe05:5987 1237370384 J * gnuk ~F404ror@pla93-3-82-240-11-251.fbx.proxad.net 1237370780 J * cluk ~cluk@p5B17F866.dip.t-dialin.net 1237371096 J * joseph ~jop@213.151.173.198 1237371499 N * DoberMann[ZZZzzz] DoberMann 1237372170 Q * davidkarban Ping timeout: 480 seconds 1237372285 N * Bertl_zZ Bertl 1237372289 M * Bertl morning folks! 1237372474 J * Pazzo ~ugelt@reserved-225136.rol.raiffeisen.net 1237372672 Q * zbyniu_ Quit: leaving 1237373646 M * ghislainocfs21 hello ! :) 1237373751 J * davidkarban ~david@199.123.broadband11.iol.cz 1237374599 M * Guy- hi 1237374624 M * Guy- I'm running 2.6.28.7-vs2.3.0.36.7 and have noticed something strange in one of the guests 1237374628 M * Guy- ip addr shows this: 1237374659 M * Guy- inet 172.18.0.56/32 scope global {dummy0,vbox2,vbox1,vbox0} 1237374662 M * Guy- (actually on four lines of course) 1237374668 M * Guy- however, ip ro sh shows this: 1237374676 M * Guy- 172.18.0.0/16 dev if8 proto kernel scope link src 172.18.1.180 1237374701 M * Guy- what's "if8"? and how does 172.18.1.180 creep in, which is the host's address? 1237374747 P * joseph 1237374827 M * _Shiva_ Bertl: patch-2.6.28.*7*-vs2.3.0.36.8 applies perfectly (from patch's PoV) to 2.6.28.*8* - i propose it simply needs a re-diff for /Experimental/ ..? 1237374844 M * Bertl yep, I presume so 1237374867 M * Bertl Guy-: I presume you are running some emulation software, yes? 1237374877 M * Guy- Bertl: in the vserver, I'm running virtualbox, yes 1237374915 M * Bertl maybe that does some weird routing or so? 1237374936 M * Bertl does 'ip link ls' show some if8? 1237374975 M * Guy- nope 1237374991 M * Bertl then this route should not be able to exist at all 1237375016 M * Guy- I also don't see how virtualbox should be able to create it; I didn't give the vserver any special privileges 1237375036 M * Bertl not talking about the vserver guest, talking about the host here 1237375083 M * Guy- the host also doesn't have 'if8', and the same route looks like this: 172.18.0.0/16 dev br0 proto kernel scope link src 172.18.1.180 1237375123 M * Bertl hmm, so you have a bridge setup? 1237375125 M * Guy- but I just noticed that same 'ifX' style interface in other vservers too 1237375149 M * Guy- yes, on this particular box I have the eth0 of the host bridged with the vbox* interfaces) 1237375170 M * Guy- but on another box I have no bridge, and ip ro sh inside a guest says: 172.18.0.0/16 dev if2 proto kernel scope link src 172.18.8.1 1237375187 M * Guy- this is "172.18.0.0/16 dev eth0 proto kernel scope link src 172.18.8.1 " on the corresponding host 1237375228 M * Guy- but the guest's only ip is 172.18.9.78/32, and it's on dummy0 1237375246 M * Bertl please describe your setup once again, as I'm confused by the part where the virtualbox is involved 1237375269 M * Guy- just forget the virtualbox, I have the same symptom on the other box with the much simpler setup :) 1237375292 M * Bertl okay, then let's go there for further investigation 1237375306 M * Bertl and please describe the setup 1237375338 M * Guy- the host has two interfaces: eth0 and dummy0; eth0 is configured thus: inet 172.18.8.1/16 brd 172.18.255.255 scope global eth0 1237375365 M * Guy- dummy0 thus: inet 172.18.9.254/32 brd 172.18.255.255 scope global dummy0 1237375391 M * Guy- I have a guest with "inet 172.18.9.78/32 scope global dummy0" 1237375408 M * Guy- and in this guest, ip ro sh gives me "172.18.0.0/16 dev if2 proto kernel scope link src 172.18.8.1" 1237375434 M * Bertl okay, that's kind of expected 1237375449 M * Bertl and the direct result of your 'funny' dummy setup 1237375463 M * Guy- ah? :) 1237375482 M * Bertl 172.18.9.78/32 is in the 172.18.0.0/16 network 1237375507 M * Bertl putting it on dummy0 doesn't make much sense, but results in eth0 not being visible to the guest 1237375531 M * Bertl but, OTOH, the only route to other IPs in that network is through eth0 1237375550 M * Bertl so, the route through eth0 shows up in the guest, but without interface 1237375569 M * Bertl and as the interface is only referred to by number/entry you see 'if2' 1237375580 M * Bertl or 'if#' in general 1237375606 M * Guy- OK, that's pretty clear so far; I'm just somewhat confused by the src 172.18.8.1 part 1237375628 M * Guy- because actually, when this guest sends packets, they have a source address of 172.18.9.78 1237375654 M * Bertl that's te route from the host, and it has 172.18.8.1 as source 1237375684 M * Bertl the guest doesn't have a network per se, which is the second config error 1237375707 M * Guy- it doesn't need one 1237375707 M * Bertl it would be much better to put the ip with the correct netmask on eth0 1237375735 M * Guy- I'm not sure I agree; if I ifdown eth0 on the host, won't that cause the IP of the guest to disappear from it? 1237375746 M * Guy- (even if I ifup again afterwards) 1237375779 M * Bertl yes, but it also would destroy any connectivity for the guest as is 1237375793 M * Guy- as far as I remember, this was why I came up with this dummy setup in the first place: to make the guests' IPs independent of the status of physical interfaces 1237375827 M * Guy- this was I can freely ifconfig eth0 down, and then bring it up later, and everything will continue working as before 1237375830 M * Bertl no problem with that, still doesn't explain the /32 mask? 1237375882 M * Guy- there's really no point in pretending to have my physical /16 network on the dummy0 interface, is there? 1237375912 M * Bertl well, your guest won't be able to send or receive broadcasts for example 1237375929 M * Guy- I know, and that's intended 1237375943 M * Bertl and that other IP addresses in that range are reachable is only due to the 'generic' route on the host 1237375988 M * Bertl so what the guest sees is basically correct, a route, which doesn't really belong to the guest, resides on an unknown interface, but is the only connection to the outside 1237375988 M * Guy- yes, but how is that bad? 1237376000 M * Bertl nothing bad in that, it works, doesn't it? 1237376012 M * Guy- you said "configuration error" :) 1237376037 M * Bertl well, let me rephrase it as 'unfortunate and partially unnatural' :) 1237376079 M * Guy- OK, thanks for the explanation :) 1237376085 M * Bertl np 1237376803 J * ktwilight__ ~ktwilight@91.178.152.79 1237376868 Q * cga Quit: WeeChat 0.2.6 1237377105 Q * ktwilight_ Ping timeout: 480 seconds 1237377163 J * ktwilight ~ktwilight@91.178.150.165 1237377540 Q * ktwilight__ Ping timeout: 480 seconds 1237377588 J * cga ~weechat@62.196.2.6 1237378158 Q * Piet Quit: Piet 1237378191 Q * DreamerC Quit: leaving 1237378207 J * DreamerC ~DreamerC@122-116-181-118.HINET-IP.hinet.net 1237379521 J * jsullivan ~jsullivan@cpe-74-75-231-235.maine.res.rr.com 1237379563 M * jsullivan Hello, all. 1237379581 M * jsullivan Sorry to trouble you but I've got an immortal runaway process. 1237379607 M * jsullivan In testing ossec, I have an ossec-syscheck process consuming 100% of one of the processors. 1237379616 M * jsullivan kill does not work 1237379621 M * jsullivan kill -9 does not work. 1237379636 M * jsullivan chcontext --ctx 40005 bash and then kill from there doesn't work. 1237379654 M * jsullivan How does one slay the immortal beast? 1237379679 M * jsullivan Without slaying the mortal vserver host underneath? 1237379740 M * jsullivan kill from top says there is no such process. 1237379801 M * jsullivan The console is filled with "soft lockup CPU#3 stuck for 61s [ossec-syscheckd:4270]" 1237380662 M * Bertl what about vkill? 1237380698 M * Bertl but the fact that you get soft lockups reported sounds like a kernel module or something like that going awry 1237380849 M * jsullivan I'll try it right now although if I understand vps correctly, the process is in the host! 1237380887 M * Bertl then a normal kill -9 will suffice, as long as it isn't in D state (i.e. doing something in the kernel) 1237380984 M * jsullivan Argh!! No such luck 1237380993 M * jsullivan I assume this means it is running on the host: 1237380996 M * jsullivan 4270 0 MAIN ? 01:02:28 ossec-syscheckd 1237381005 M * jsullivan yet kill -9 4270 does nothing. 1237381025 M * Bertl what does 'ps auxwww' say about that process? 1237381042 M * jsullivan Thankfully, it is not yet in production but I'd sure like to know how to address such a problemwithout rebooting BEFORE we put this into production. 1237381090 Q * FireEgl Remote host closed the connection 1237381091 M * jsullivan root 4270 73.2 0.0 4132 836 ? R 07:30 64:01 /usr/local/ossec/bin/ossec-syscheckd 1237381108 M * Bertl well, first thing I would make sure is not to run such processes on the host :) 1237381159 M * jsullivan 1237381162 M * Bertl doesn't look like 100% cpu usage to me 1237381180 M * jsullivan eight processors - it's hogging 100% of one. 1237381189 M * Bertl most likely it will die after it logged a million more kernel log lines 1237381196 M * jsullivan Thanks :( 1237381208 M * jsullivan :) 1237381267 M * jsullivan Am I flat out of luck and, were this production, need to schedule a reboot? 1237381338 M * Bertl well, I have no idea what the ossec-syscheckd is supposed to do, but running it on the host allows it to do almost anything, including trashing your system until you cannot even logon anymore 1237381454 M * Bertl what kernel version are we talking about, btw? 1237381564 M * jsullivan 2.6.28.7 1237381590 M * jsullivan It's an integrity checker - HIDS 1237381594 J * BenG ~bengreen@94-169-110-10.cable.ubr22.aztw.blueyonder.co.uk 1237381635 M * jsullivan The idea was to see if anyone has compromised the host - not to compromise the host :) 1237381658 M * Bertl well, it seems it is doing a very poor job then :) 1237381734 M * jsullivan Well, thanks anyway. 1237381738 M * jsullivan I appreciate the help. 1237381751 M * jsullivan I knew I was in pretty deep trouble when kill -9 didn't work :( 1237381764 M * jsullivan Take care. 1237381774 M * Bertl u2 1237381891 J * FireEgl ~FireEgl@Sebastian.Atlantica.US.TO 1237381977 Q * BenG Quit: I Leave 1237382910 Q * jsullivan Quit: using sirc version 2.211+KSIRC/1.3.12 1237384240 Q * pmenier Quit: Konversation terminated! 1237384753 Q * sharkjaw Quit: Leaving 1237385229 Q * davidkarban Ping timeout: 480 seconds 1237385239 J * davidkarban ~david@88.86.104.103 1237385308 J * pmenier ~pme@LNeuilly-152-22-72-5.w193-251.abo.wanadoo.fr 1237386063 Q * FireEgl Ping timeout: 480 seconds 1237386348 Q * scientes Ping timeout: 480 seconds 1237386459 M * Bertl k, off for now .. bbl 1237386467 N * Bertl Bertl_oO 1237386668 J * FireEgl ~FireEgl@173-16-9-10.client.mchsi.com 1237387287 J * critch ~critch@c-71-228-230-70.hsd1.tn.comcast.net 1237388377 J * xdr ~xdr@gote2.244.cust.blixtvik.net 1237389895 Q * Punkie Quit: Leaving 1237390549 J * hparker ~hparker@2001:470:1f0f:32c:290:96ff:fe50:40fa 1237391231 Q * cga Quit: WeeChat 0.2.6 1237392455 J * dowdle ~dowdle@scott.coe.montana.edu 1237394046 Q * gnuk Remote host closed the connection 1237394160 J * gnuk ~F404ror@pla93-3-82-240-11-251.fbx.proxad.net 1237396094 Q * daniel_hozac Remote host closed the connection 1237397079 N * DoberMann DoberMann[PullA] 1237397376 J * daniel_hozac ~daniel@2002:5043:693::102 1237398005 J * cga ~weechat@94.36.110.25 1237398979 Q * pmenier Quit: Konversation terminated! 1237399101 Q * harobed Ping timeout: 480 seconds 1237399249 Q * davidkarban Remote host closed the connection 1237400947 Q * Pazzo Quit: Ex-Chat 1237401188 Q * hparker Quit: Read error: 104 (Peer reset by connection) 1237401304 N * Bertl_oO Bertl 1237401309 M * Bertl back now ... 1237401503 J * hparker ~hparker@2001:470:1f0f:32c:215:f2ff:fe60:79d4 1237402434 Q * hparker Quit: Quit 1237402442 J * hparker ~hparker@2001:470:1f0f:32c:215:f2ff:fe60:79d4 1237402808 M * critch so everyone primarily idles through the US workday here? 1237402827 M * Bertl well, as long as nobody has questions ... yes 1237402855 M * critch I may have a question, going to bang head on it one more time to see if it works. 1237402860 Q * hparker Quit: Quit 1237402881 M * Bertl good, when you have one ... just ask 1237402928 M * critch have much experience converting from vmware to vserver? 1237403012 J * hparker ~hparker@2001:470:1f0f:32c:290:96ff:fe50:40fa 1237403028 M * Bertl not really, but it should be simple, you just drop everything hardware related and forget about bridging and vmware networking :) 1237403053 M * pmjdebruijn it's a pretty drastic change 1237403056 M * critch hehe, for the simple network mode, it has been fairly straight forward 1237403080 M * critch I just finished converting my my mail server 1237403131 M * critch for my slightly more complicated networking instances, I think I am going to have to do a refresher course in iproute2 1237403176 M * Bertl you probably use several public IPs right now (or used to use them with vmware) right? 1237403250 M * Bertl daniel_hozac: are you around? 1237403253 M * critch no, My vmware servers had 2 NICs, one on a public IP, and one on a private network. I have a firewall as a guest doing NAT and reverse proxy on the private network. 1237403285 M * critch So many of the machines I want to convert have to access the private network. 1237403318 M * Bertl okay, and the 'Guests' (as we calle the virtual machines) have access to both networks, private and public? 1237403321 M * critch Problem being that I had the hosts on 10.254.0.* addresses, and the guests on 192.168.*.* addresses. 1237403367 M * critch very few machines stradled the public/private networks. Most stayed on the private network. 1237403398 M * Bertl so it should be rather simple then .. all go on one host or do you have several Linux-VServer hosts? 1237403422 M * critch I'm converting probably 4 out of my 5 vmware servers to vservers 1237403432 M * Bertl and do the host machines have two network interfaces? 1237403443 M * critch Yes, all of them do. 1237403469 M * Bertl so then you simply setup the public IP plus default route, and a private network/i on the other interface for the host 1237403499 M * Bertl for the guests, you decide whether you want to give them an IP on the public network or on the private network only 1237403519 M * daniel_hozac Bertl: pong 1237403552 M * Bertl hey, I wonder where I would put a script which does a --bind mount on a file (with 0.30.215 and with recent util-vserver) 1237403565 M * daniel_hozac any of the scripts should do. 1237403570 M * Bertl i.e. which binds something from outside the guest into the guest 1237403583 M * daniel_hozac secure-mount doesn't yet do the right thing. 1237403585 M * critch I understand that so far. My issue is with default routing. the hosts do not default route to the firewall on the private network, and therefore my guests aren't getting outbound access. 1237403602 M * Bertl daniel_hozac: I didn't have much success with my testing there, was putting it into post-start to no avail 1237403631 M * Bertl daniel_hozac: i.e. the mount succeeds, but the --bind doesn't show up in the guest 1237403642 M * daniel_hozac that's weird. it's all in the guest's namespace. 1237403657 M * daniel_hozac or, hmm... dual namespaces might change that. 1237403662 M * daniel_hozac you probably want something pre-start. 1237403664 M * Bertl entering the guest namespace with vnamespace and doing the mount there doesn't work either 1237403699 M * Bertl critch: if the hosts already have public access, you do not need to route over the private network 1237403701 M * daniel_hozac really? 1237403711 M * daniel_hozac that is weird... 1237403716 M * Bertl yep, it is visible in the vnamespace, but not inside the guest 1237403728 M * Bertl i.e. it doesn't propagate to the sub-namespace 1237403766 M * Bertl tool version here is 0.30.216-pre2824 (maybe too old?) 1237403779 M * critch Bertl: actually I do if I am running webservers routing via a reverse proxy that is accessable only from behind the firewall. The idea is to make migration of the guests to any other host painless and not require reconfiguration. 1237403811 M * Bertl one doesn't require the other, let me give you an example 1237403832 M * Bertl let's assume, you have 3 hosts, with public IPs 1.1.1.1-1.1.1.3 1237403834 M * daniel_hozac i guess MS_REC|MS_BIND|MS_SHARED doesn't work as expected... 1237403865 M * Bertl you then put a number of guests on them, with private IPs 10.0.0.1-10.0.0.10 1237403918 M * Bertl to allow them access to the internet, a simple DNAT rule on each host is required, mapping 10.0.0.0/24 to the public IP 1237403960 M * Bertl i.e. that rule doesn't need to know what guest is running on it 1237403997 M * critch So your option right now is that every host do the NATing to the public network. Not as nice as I would have it. 1237404021 M * Bertl it is one option, of course you can do any other setup as well 1237404022 M * critch But it is a workable option 1237404037 M * daniel_hozac Bertl: ah, okay. i need to change it to rbind + shared. 1237404046 M * Bertl critch: for example, dedicate a firewall, and route to that as default gateway 1237404078 M * critch right now the problem I am trying to tackle is a machine that wont start, vcontext: execvp("/etc/init.d/rc"): No such file or directory 1237404120 M * Bertl and that file exists inside the guest? 1237404130 M * Bertl (you can run the start with --debug) 1237404138 M * daniel_hozac as does the interpreter specified in it and any libraries it depends on? 1237404143 M * critch yes, and so does the shell. Didn't know about --debug yet. 1237404189 M * Bertl daniel_hozac: okay, so please let me know when there is a test version (newer pre) to try 1237404455 M * daniel_hozac Bertl: curl 'http://svn.linux-vserver.org/projects/util-vserver/changeset/2831?format=diff&new=2831' | patch /usr/lib*/util-vserver/vserver.functions 1237404752 J * yarihm_ ~yarihm@guest-docking-nat-1-003.ethz.ch 1237404758 Q * yarihm_ 1237405003 Q * yarihm Ping timeout: 480 seconds 1237405220 M * critch hmm, seems vserver start believes my guest is a RH based machine. How do I fix it's misconception? 1237405248 M * critch :q 1237405321 M * daniel_hozac what kind of OS is it? 1237405328 M * critch debian 1237405366 M * daniel_hozac /etc/init.d/rc is correct then. 1237405386 M * critch right, but it is trying to go to runlevel 3, and it should be 2 1237405421 M * daniel_hozac i've never seen the two differ on Debian. 1237405431 M * critch that may be true 1237405435 M * daniel_hozac but if you really care, set apps/init/runlevel.start to 2. 1237405564 J * BenG ~bengreen@94-169-110-10.cable.ubr22.aztw.blueyonder.co.uk 1237405823 M * hparker Is this normal output for a skeleton build? http://rafb.net/p/03HuSt60.html 1237405907 M * daniel_hozac specifying a distribution doesn't really make sense for a skeleton. why are you? 1237405927 Q * BenG Quit: I Leave 1237405928 M * hparker Cuz I know no better? :P 1237405949 M * daniel_hozac you can't really do distribution-specific cleanup on an empty tree. 1237405952 M * Bertl hparker: maybe you think that a skeleton is actually something else .. while in fact a skeleton is just an empty dir with devices and mount entry for proc 1237405974 M * critch hmm, wtf, test -x /etc/vservers/webfs/vdir//etc/init.d/rc returns 0 while ls -l /etc/vservers/webfs/vdir//etc/init.d/rc shows -rwxr-xr-x 1 root root 8045 2009-03-18 14:45 /etc/vservers/webfs/vdir//etc/init.d/rc 1237405999 M * hparker Makes sense.. I'm needing to rsync a server over to it.. So, skeleton with no distro would be proper? 1237406002 M * daniel_hozac sounds right to me. 1237406030 M * Bertl hparker: you could use the rsync build method? 1237406080 M * hparker Hrrmm.. Will it resume properly? I foresee connection issues with this one 1237406370 Q * gnuk Quit: NoFeature 1237406380 M * critch any clue why test and ls dissagree over the same file? 1237406430 M * Bertl in what way? 1237406449 M * critch nm, not I understand what I am doing wrong 1237406470 M * critch at least in trying to debug this. 1237406739 M * critch --debug doesn't seem to help me in trying to figure out why vserver start claims to fail at running /etc/init.d/rc when it is visible. 1237406766 M * Bertl check two things to narrow it down: 1237406800 M * Bertl - chroot into the guests data dir, look around and see if that script is there and the interpreter is working 1237406854 M * Bertl - change into a context, with 'chcontext --xid 42 --secure -- /bin/bash' 1237406884 M * Bertl and see if you can stat the script (make sure to start from within the guest data) 1237407185 M * critch can't seem to run /bin/bash from within the chroot 1237407228 M * Bertl so, there is your problem then 1237407252 M * Bertl possible causes are: -wrong architecture 1237407261 M * Bertl - missing dependancies/libraries 1237407273 M * Bertl - missing execute permission on libraries/binaries 1237407332 M * critch yep, seems a library is missing. 1237407608 M * critch seems I missed the /lib64 dir when archiving out the source machine. 1237407618 M * critch Now that I know what is up, I'll fix it tomorrow. 1237407621 M * critch later gents. 1237407630 M * Bertl have fun! 1237408967 J * zbyniu ~zbyniu@ip-62.181.188.13.static.crowley.pl 1237409103 Q * cluk Quit: Ex-Chat 1237409646 M * hparker Bertl: How do I pass options to rsync using the rsync build method? 1237409713 M * Bertl vserver - build --help 1237409883 M * hparker I see, thanks! 1237409893 M * Bertl np 1237409948 Q * cga Quit: WeeChat 0.2.6 1237410634 J * thierryp ~thierry@home.parmentelat.net 1237412463 Q * duckx Read error: No route to host 1237412571 J * duckx ~Duck@81.57.39.234 1237414409 J * Piet ~piet@asteria.debian.or.at 1237414790 J * harobed ~harobed@arl57-1-82-231-110-14.fbx.proxad.net 1237415152 Q * Piet Ping timeout: 480 seconds 1237416950 J * Piet ~piet@asteria.debian.or.at 1237417066 M * hparker Finally getting back to this, sometimes telephones suck.. Does this look right? vserver hosting2.pcsrvc.com build --context 112 --hostname hosting2.pcsrvc.com --interface eth0:10.0.0.112 -m rsync --force -o --bwlimit=45 --source user@example.com 1237417088 M * hparker --force just so I don't have to clean up the skeleton 1237417146 M * Bertl the prefix is missing and you might want some additional arguments to rsync, check with daniel_hozac because IIRC, that depends on the util-vserver version 1237417167 Q * bonbons Quit: Leaving 1237417247 M * hparker Prefix for the netmask? 1237417253 J * zugowaq ~zugowaq@193.43.249.169 1237417261 M * Bertl yup 1237417286 M * Bertl and personally I would choose a shorter guest name 1237417314 M * hparker heh.. I usually make them the machine name to keep my old brain less confused 1237417330 M * hparker util-vserver is 0.30.215-r2 1237417348 M * hparker Latest in Gentoo portage :P 1237417410 M * Bertl maybe bug Hollow a little to update that then :) 1237417615 A * hparker pokes Hollow 1237417692 M * hparker Does vserver have --dry-run or anything? 1237417701 M * hparker Not seeing it in --help 1237417757 M * Bertl what would that do? (or not do?) 1237417795 M * hparker Just show what it would do without any changes 1237417814 M * hparker So old farts like me don't get so antsy when hitting enter 1237417816 Q * thierryp Remote host closed the connection 1237417835 J * thierryp ~thierry@home.parmentelat.net 1237417851 M * Bertl hmm, mkdir doesn't have a --dry-run either .. I wonder why :) 1237417879 M * hparker It's not logging into a remote server as root either 1237417885 M * hparker Just a thought 1237417945 M * hparker Are you speaking of the rsync options as in this page? http://linux-vserver.org/util-vserver:Howto_virtualize_an_exisiting_Linux_server 1237417973 M * hparker (Which is what I was using as a guide before to setup the guest) 1237418010 Q * thierryp 1237418035 M * Bertl yes, but IIRC, most or maybe all of them should be in recent util-vserver (as defaults) 1237418061 M * hparker I see.. Wonder if I can find it in the tarball/source 1237418064 A * hparker pokes 1237419341 M * hparker test -z "${OPTS[*]}" && OPTS=( -Hazx --numeric-ids ) .. Looks like it does 1237419346 Q * pflanze Remote host closed the connection 1237419356 M * Bertl yep, looks good 1237419365 M * hparker And needed to move --source before -o 1237419482 M * hparker Or not :P 1237420277 Q * dna Quit: Verlassend 1237420504 M * Bertl off to bed now .. have a good one everyone! 1237420509 N * Bertl Bertl_zZ