[00:08] anyone have installed ubuntu-server on thin client? [00:09] somebody here? [00:24] * eboyjr doesn't understand why you would install a server on a thin client [00:24] unless its an x server :) [01:49] I am trying to set up my server with static IP. I have configured the interfaces file, and the resolv file. I can ping any computer on my local net, but not on the internet. [01:49] you also need a default route [01:50] like ip route add default via 192.168.0.1 dev eth0 [01:50] where 192.168.0.1 is your router [01:50] if you have something like that, you may also need to setup your dns [01:50] route -n does show a correct gateway to my router (192.168.1.10) [01:50] or at least check your resolv settings [01:51] can you ping 192.168.1.190 [01:51] err [01:51] 192.168.1.10 [01:51] oops sorry the router is at 192.168.1.1 the server is at 192.168.1.10 [01:52] can you ping 192.168.1.1 [01:52] ? [01:52] I have pinged the routher [01:52] yes [01:52] can you ping 209.85.225.106 ? [01:54] at the moment yes. because I had to use dhcp. one moment.... [01:56] Ok, yes I have pinged 209.85.225.106 [01:57] NoobFukaire, thanks for your help. for some reason it is now working. === Pilif12p|afk is now known as Pilif12p [02:46] New bug: #658036 in samba (main) "package samba 2:3.4.0-3ubuntu5.7 failed to install/upgrade:" [Undecided,New] https://launchpad.net/bugs/658036 [03:02] is there a xen optimized kernel and virtualzation utilties set and ready to go? [03:03] or does this require a good deal of configuration to get running? [03:03] ehnde: Ubuntu can be a DomU out of the box. It can't be a Dom0 -- that isn't supported. [03:03] that was what i was wanting, a Dom0 [03:03] You can't have it [03:04] any other type I hypervisor solutions? [03:04] The virtualization technology Ubuntu recommends is KVM. [03:04] i've always preferred and been most comfortable on ubuntu, would rather use it [03:09] I need to access an ubuntu computer behind a firewall via SSH without forwarding ports. I have tried autossh but some of my locations don't maintain the connection and I'm left without access..any ideas or solutions? [03:10] Sivarts: you have ssh access to the firewall? [03:11] ssh -oProxyCommand='ssh firewall -W %h:%p' foo [03:12] twb nope, basically I have servers that run kiosks at different 3rd party stores, I don't maintain their networks and they're all different so I don't bother trying to forward ports [03:13] Sivarts: then there is NO WAY to guarantee a connection initiated externally will get through. [03:13] As you saw, ssh -w, even wrapped in autossh, is pretty flaky. [03:14] twb Well if theres a way to for it to keep trying that would work, these stores just use basic routers no security or anything like that [03:14] I suggest you try openvpn or some other UDP-flavoured VPN. === Pilif12p is now known as Pilif12p|afk [03:15] twb ok yeah and autossh worked great it just seems like over time (having the computers on for a day or two) I could not longer connect on my computer [03:15] does openvpn make it through most routers easily? my goal is to not have to configure their routers or anything like that [03:16] Sivarts: as easily as any other UDP-flavoured tunnel initated from behind the NAT. [03:16] I'm assuming when you say "firewall" you actually mean a masquerade. [03:17] so if I use openvpn that will maintain a connection to my main server and keep it alive at all times? [03:18] Well, it operates over a connectionless transport layer. [03:18] That should mean that it only sends packets when it has something to say. [03:18] Although come to think of it, the OpenVPN layer probably sends a keepalive packet occasionally to ensure the quintuple association stays in the router's masquerade list. [03:19] gotcha I'll give it a shot. Thanks! === harrisonk_away is now known as harrisonk === harrisonk is now known as harrisonk_away [04:15] <_Techie_> will a dist-upgrade to 10.04 LTS possibly break any of my configurations, proccesses in question are as follows [04:15] <_Techie_> postfix, dovecot, SASL, lighthttpd, apache2 [04:15] <_Techie_> oh yeah, also scgi in lighthttpd [04:22] <|rt|> _Techie_: I think the appropriate course of action is to try to do that in your test environment before rolling it out to the production environment. That's really the only way to insure that it's not going to break anything. [04:23] <_Techie_> this isnt a production environment [04:23] <_Techie_> its a personal server [04:23] <_Techie_> i just dont want to have to reconfigure everything if it goes belly up [04:24] <|rt|> well either test using virtualization or backup your configurations so you can easily rebuild it back on 10.04 if for some reason there are issues that you can't work around [04:25] <|rt|> in general Ubuntu is pretty good with leaving any configuration files that you've modified alone through the upgrades [04:34] hmmmm, my Ubuntu 10.04 LTS server has a mind of its own with regards to accepting any kind of connection, most of the time it works fine but sometimes it just kind of... dies and stops accepting any connection from any part of my network, and I either have to wait for it to fix itself or reboot (and even that sometimes doesn't work...) [04:34] for example, reading/writing to the thing or SSHing into it [04:35] the server itself is still responsive, it doesn't crash or anything [04:35] any ideas? [04:39] Callum__: is it under load? [04:39] Callum__: is NetworkManager installed? [04:39] twb: no, and no [04:39] Dunno then [04:39] Check your logs, dmesg, etc. [04:40] dmesg says nothing out of the ordinary, I should probably check some of my system logs.. [04:43] well, there is an error or two in my dmesg but none of it sounds like it would cause any problems... [05:03] hey, just wondering if anyone can give me a hand? i just install package dovecot-postfix, and allowed that to configure my mail server for me, and I can log in to receive mail, but whenever i try to send mail to user@host, it replies saying "RCPT TO failed: No such user here" [05:03] could anybody shed some light on this? [05:07] You're trying to send mail to a user that doesn't exist on the local system. [05:07] Because postfix doesn't recognize that you're coming from a trusted/local network, it won't relay such mail. [05:08] In postfix's main.cf there's something like "local_network = 192.168/16" [05:10] twb: thanks for the response - the user does exist on the system I'm logging into, if that's what you mean, but basically I want to set this box up as a mail server, so will I need to add more networks to the "mynetworks" parameter? [05:10] i've currently got the hosts IP address, as well as localhsot in there [05:10] (mail hosts IP address that is) [05:11] if the system is the final destination for the email address you're sending to, mynetworks should not make any difference. [05:11] rcsheetsKRC: yes, the final destination is a local system user [05:11] at least in typical configurations, mynetworks only comes into play when you want the postfix machine to relay mail elsewhere, for a certain set of other systems. [05:12] ah ok, well basically, i want this box to be able to send mail to any other email address, and receive mail for a local user (which can be retrieved via pop/imap/etc) [05:12] so is dovecot and postfix a good choice for this? [05:12] or would you suggest something else [05:12] dovecot and postfix are an excellent choice for that [05:13] but what do you mean by "this box" and "send mail to any other email address"? which box... the postfix server or some other one? [05:13] and which email address? like someone@gmail.com? [05:14] by "this box", i do mean the postfix server, and by "send mail to any other email address", i mean to act as any other email provider would - be able to send to user@gmail.com, amazingstart@finish.com, whateveryouchoose@wherever.com [05:14] i just want to use it as my email address, to email anybody [05:14] if that makes sense? [05:14] MrPPS: the other thing it could be is that those angle brackets look weird [05:14] Where did you get that line? Did you manually create an envelope by talking SMTP to server, or are you looking at a log? [05:14] yes, it makes sense, though i should point out that if you are running this server at your home, for example, it may not be practical for it to do direct-to-MX sending of mail. [05:15] twb: my apologies, those angled brackets are from Evolution mail, as i send from a working address to my new mail host [05:15] Oh: "the user does exist on the system I'm logging into" [05:15] rcsheetsKRC: no, im not running this server at home, I'm running it on a dedicated hosting box [05:15] Then make sure that the @host hostname corresponds to what postfix believes the hostname to be (/etc/mailname and something like accept_hosts in main.cf) [05:16] MrPPS: ok, just making sure. [05:16] rcsheetsKRC: i understand :) [05:17] twb: /etc/mailname does match up [05:17] MrPPS: can you pastebin the output of 'postconf'? [05:17] twb: i cannot find "accept_hosts" in main.cf, but "myhostname" is set to the host [05:17] and myorigin = /etc/mailname [05:17] oh, i should mention this is running on ubuntu 10.04 [05:19] should i take the "hostname out of "mydestination", and just have localhost values? [05:20] mydestination should be set to the list of domains for which your server is the final destination, *and* which should be delivered via the $local_transport. [05:20] MrPPS: yeah, that one [05:20] normally that would be the case if local users on that box actually use that server to read their mail. [05:20] MrPPS: I'm going from memory because ICBF getting out a postfix host [05:20] You could also try #postfix [05:21] so if your server is fred.example.com and you only want to have it accept mail for user@fred.example.com, then fred.example.com would be a fine $mydestination value [05:21] but if it's to be the MX for all of example.com, and all the example.com users will have an account on that server, then example.com should be included in $mydestination as well. [05:21] if i wanted to accept mail for user@fred.example.com, user2@fred.example.com, etc., that would still be acceptable i take it? [05:21] ok, cool [05:22] thanks [05:22] yes, provided user and user2 are both users on your system. [05:22] correct [05:22] do i need to manually created the mailbox's for each user? [05:22] perhaps that's where this error is stemming from [05:22] no, the actual mailbox files should be created automatically once a message is received [05:23] hmm, then i shouldn't really be getting that error if it's an existing user, should I? [05:23] no. the error seems to suggest that postfix is unclear about which users actually exist. [05:24] again, the output of 'postconf' would be helpful. [05:24] shall do - wont be a sec [05:26] http://pastie.caboo.se/8763 [05:27] that... seems unrelated [05:27] sorry, incorrect url [05:27] wont be a moment (i'll just found out where the real one went) [05:27] if that's your postconf output, please reinstall ubuntu from scratch :) [05:29] pastebin.com/35gFap4J [05:29] that should be right this time :P [05:30] that does seem better :) [05:30] q [05:30] sorry, wrong window [05:31] is the system hostname really the bare name tuxcon.com? [05:31] the local system's hostname is "rusty", the domain is "tuxcon.com" [05:32] myhostname should then be rusty.tuxcon.com [05:32] there is currently no cname pointing rusty.tuxcon.com to 178.63.80.110 - should i set up the cname and myhostname as such, even if i want the email address to be user@tuxcon.com? [05:33] an A record would seem more appropriate [05:33] sorry, momentary mixup [05:33] if 178.63.80.110 is the address of rusty.tuxcon.com [05:33] it is, but it is also the address of tuxcon.com [05:33] that's fine too [05:34] okey, won't be a moment then [05:34] well, i doubt the absence of an A record is causing any of the problems you are currently facing [05:35] actually DNS says that tuxcon.com has address 178.63.80.87 [05:35] not 178.63.80.110 [05:35] sorry, i incorrectly typed that - it is 87, for both [05:35] ok [05:35] 178.63.80.110 is canyouget.in [05:36] my current host, yes [05:36] hence why i mistyped ;) [05:36] ahh i see [05:36] alright, records created, postfix reloaded, still the same issues though [05:37] yeah those were more general system/dns configuration things. probably not relevant to your error message. what exactly was your error message again? [05:37] i shall just retrieve it for you [05:37] oh sorry i see it [05:37] "RCPT TO failed: No such user here" [05:37] and no worries, it's fine :) [05:37] is it really and not like ? [05:38] rcsheetsKRC: no, sorry, that was a direct copy from before - it's not user@host, it's steve@tuxcon.com [05:38] ok [05:39] and steve is a user in /etc/passwd? [05:39] correct, logged in at this moment [05:40] just a minute. i don't understand what your $local_recipient_maps means [05:41] that was automatically configured, i have not set that value [05:41] im unsure of its usage also [05:41] yeah, i just don't actually run any systems that deliver mail to local users ;) [05:41] and i never have thus far :P [05:41] hence the new venture for me :) [05:42] is there anything showing up in /var/log/mail.log when you're attempting to send mail? [05:43] i will just check [05:43] no [05:43] none [05:45] nothing there whatsoever? [05:45] there should be something. at least postfix telling us it has started. [05:45] well yes, there's the usual of me logging in, and disconnecting [05:45] but that's it [05:46] nothing directly related to receiving or failing to receive mail [05:47] hmm [05:48] well [05:48] I just connected and successfully sent you mail. [05:48] 250 2.0.0 Ok: queued as 6F75B583C4F [05:49] anything in the log now? [05:50] yes, i have received a message from you, according to those logs [05:50] do you see it in your mailbox? [05:50] i do [05:50] well that's good then [05:50] how exactly did you try to send the mail when it didn't work? [05:50] fired up evolution [05:50] typed the mail [05:50] clicked send :P [05:51] where was evolution running? [05:51] on my local machine, unrelated and unconnected to the mail server [05:52] and how is that evolution configured to send mail? [05:52] and i just tried sending from a separate mail address on a separate server - and it all worked: just looks like there's something wrong with the mail server i'm using [05:53] yes, your smarthost may be broken [05:53] or your local mail configuration, if you have evolution configured to use sendmail [05:53] but your server seems fine :) [05:54] well, thanks for all the help [05:54] it's very much appreciated [05:54] happy to [05:54] sorry to put you through all that when the problem wasn't even related to the server :P [05:54] that's alright. and your configuration is a bit better now. [05:55] yes, thank you :) [05:55] you're welcome [06:21] I've changed a hostname on my EC2 instance. I get "sudo: unable to resolve host samfordwebsites" come up when I do an update (the update works fine). If I change the resolv.conf so that there is a nameserver appropriate for samfordwebsites the message disappears but then I get signature verification errors. I can switch back and forth between the two problems by changing the nameserver line in resolv.conf. Can someone point me to a way to better un [06:27] paul_whipp: i don't know how to help you, but your message was cut off at "way to better un" [06:28] rcsheetsKRC: thx, it shows ok in my window :( [06:28] I am trying to get my head around /etc/resolv.conf [06:28] paul_whipp: could just try /etc/hosts [06:28] add samfordwebsites to there [06:28] and keep /etc/resolv.conf pointed to the proper nameservers [06:28] All done there and its generally working [06:29] Is the resolv.conf supposed to be a *local* nameserver? [06:29] no supposed to be, but can be, if you run one [06:29] i usually use opendns [06:29] 208.67.222.222 and 208.67.220.220 [06:30] so you would have those entries in resolv.conf [06:30] ? [06:30] correct [06:30] simply like: [06:30] nameserver 208.67.222.222 [06:30] and [06:30] nameserver 208.67.220.220 [06:31] without the "and" in between [06:31] obviously [06:31] :) [06:31] That works thx! [06:32] * rcsheetsKRC hits the deck [06:33] no worries :) [06:34] Seriously MrPPS, quick answers like that are a great help. I'm still not sure I understand exactly what it is doing but at least the warning messages are gone and the signatures are being verified properly. [06:36] without being there and knowing what you used before, it's hard to tell exactly what the cause was, but it's possible that the nameservers you specified were pointing you towards servers with unverified/modified packages [06:36] but again, hard to tell from here [06:36] i've found opendns to be trustworthy, and hence why i personally use them [06:37] They were the 'default' ec2 ones or my dnsmadeeasy ones depending upon which problem I was having. I'm happy using the opendns ones. It looks like there is a speed benefit too. === sailerboy is now known as zz_sailerboy [08:23] how do i change my time pdt to pst? or are they the same? command line [08:26] jmazaredo: sudo dpkg-reconfigure tzdata [08:27] tried that but cannot find the pst it only gives me pdt [08:28] isn't the difference daylight savings? [08:28] * kinygos doesn't know cos he's from the UK [08:28] ah hehe [08:28] oks [08:49] trying to install kvm guest. any idea what goes wrong here: http://pastebin.org/134413 [09:41] New bug: #658176 in samba (main) "package samba-common 2:3.3.2-1ubuntu3.6 failed to install/upgrade: el subproceso post-installation script devolvió el código de salida de error 10" [Undecided,New] https://launchpad.net/bugs/658176 === cs278|| is now known as cs278^kinslayer [10:11] hi...i'm setting up a web application (apache2) and a mail server (dovecot+postfix) on a dedicated ubuntu 10.04 dedicated server...it seems i need a SSL certificate in 3 places, https in my web application, authentication for mailbox access, and authentication for sending mail by smtp...naive question, can i use the same SSL certificate bound to the root domain (eg. mydomain.com) or do i have to buy 3 certificat [10:18] kinygos: the certificate must match the domain name presented by the services or it is useless. If the names are different you either need a wildcard certificate (e.g. *.mydomain.com) or a separate one for each service. [10:19] maedox: thank you :) i didn't know you could get a wildcard certificate...i'll investigate that...would it be strange to use a generic qualifier in the 3 places i suggests...such as secure.mydomain.com? === cs278|| is now known as cs278 === cs278|| is now known as cs278^kinslayer [10:46] * RoyK just upgraded his first box to 10.10 === cs278|| is now known as cs278 [11:26] New bug: #658227 in openldap (main) "won't start after Maverick upgrade; bdb "Program version 4.8 doesn't match environment version 4.7"" [Undecided,New] https://launchpad.net/bugs/658227 [11:33] hi...i've set up a mail server on my dedicated server using dovecot+postfix...as i was told last week, the online docs are very good....i have one quick question...for sending mail, i couldn't get TLS to work...i don't have a valid SSL certificate yet...could that be the reason? in Evolution, i have no encryption and login authentication for sending mail [11:36] https://help.ubuntu.com/10.04/serverguide/C/jeos-and-vmbuilder.html I think the bridging guide doesn't apply 100% at the moment [11:46] kinygos: you need a cert for tls, yes [11:47] zoopster: i thought that was the case...but i've never done this before so needed a bit of reassuring :) i appreciate your time [11:48] kinygos: my pleasure === cs278|| is now known as cs278^kinslayer [12:05] New bug: #658193 in eucalyptus (main) "Cluster Controller Forgets Routing at Reboot" [Undecided,New] https://launchpad.net/bugs/658193 === _Techie_ is now known as _TechAway_ [12:19] Hi, can anyone help me with an HP BL685 G7 problem please? [12:20] I keep getting "Disk quota exceeded" where as: http://pastebin.com/vFun6rb7 [12:22] asantos: What issue are you having with it? [12:23] Hello again :) I'm unable to get it to recognise the onboard NICs. [12:26] asantos: During or after install? [12:28] During. === WinstonSmith_ is now known as WinstonSmith [12:33] It's got to be something to do with the BL685 G7 being recently release as we've not had any problems with the previous generations. We're speaking to HP at the moment but I thought it might be worth asking here to see whether anyone else has had any issues... [12:39] <\sh> asantos: which ubuntu release? [12:40] <\sh> asantos: lucid (aka 10.04 LTS?) [12:40] 10.04 [12:40] <\sh> asantos: ok, and you are trying to use the cd image of 10.04 right? [12:40] yes correct [12:40] <\sh> it's missing the bnx2x drivers inside Debian Installer UDEB [12:40] <\sh> asantos: http://www.shermann.name/2010/10/server-hardware-of-hewlett-packard.html <- read the bnx2x bits on G7 [12:41] <\sh> asantos: it has nothing to do with HP or G7 it's a bug in the installer kernel, which has been recently fixed [12:41] <\sh> but not landed on any iso image [12:41] <\sh> asantos: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/360966 [12:41] Launchpad bug 360966 in linux "bnx2x missing in initrd for install media" [Medium,In progress] [12:42] <\sh> asantos: when you install without network and reboot into production system, there you have your bnx2x drivers and firmware, and then you can configure your network === cs278|| is now known as cs278 [12:43] I see. ok guys, I'll give that a go. [12:43] <\sh> asantos: I had a testdrive with new HP G7 blade server series a couple of weeks ago here @HP germany [12:44] <\sh> it's really just the installer which has problems...the installed system works like a charm with those flex10 devices ;) [12:46] <\sh> asantos: https://bugs.launchpad.net/ubuntu/lucid/+source/linux/+bug/494052 [12:46] Launchpad bug 494052 in linux "bnx2 driver cannot find firmware" [Medium,Fix released] [12:48] I'll check out the links etc... Thanks for all your help in advance. === cs278|| is now known as cs278^kinslayer [13:16] Hello people. I have a problem with an upgraded 10.10 system: VLANs are not working anymore. I see the tagged traffic going out and in on the physical interface but nothing comes in in the VLAN interface (but the outgoing traffic is working well). Any idea ? === ivoks-afk is now known as ivoks === cs278|| is now known as cs278 [13:27] Good afternoon. Would somebody possibly be able to assist me with an installation problem i have on an HP BL685 G7 Server? Any version of Ubuntu i try and install, even the supported 10.04 LTS release, fails to recognise the NIC of the Server. This seems to be a known problem but the ISO's have yet to be updated. The NIC in this Server is a HP NC551m Dual Port FlexFabric 10Gb Network Adapter. WOuld someone please be able to advise how i can buil [13:27] Any help would be greatly appreciatted [13:27] \sh: Is that issue fixed in Lucid? [13:27] Afriad not [13:28] I have doenloaded all ISO's from Ubuntu and they all have the same issue [13:28] I assume the ISO simply doesn't have the correct drivers within? [13:28] \sh: Sorry, confused that bug with bug #360966. [13:28] Launchpad bug 360966 in linux "bnx2x missing in initrd for install media" [Medium,In progress] https://launchpad.net/bugs/360966 [13:30] I've seen that link. Sorry to be a layman here, but are you able to advise specifically what i need to do? I'm not really an advanced Ubuntu user and the link doesn't entirely make sense [13:31] tgwoollard: Yes; it's an issue in the installer udeb, see: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/360966 [13:31] Launchpad bug 360966 in linux "bnx2x missing in initrd for install media" [Medium,In progress] [13:31] is there anyone that can help me with software raid? [13:31] tgwoollard: The installer will not pick up the card; however once the system is installed, it'll work like a charm. [13:32] Thanks jpds. How can i include the driver in the image in order to sucesfully run the installer? [13:32] THe instructions in the link aren't clear to me i'm afraid [13:33] jpds i will give that a go and report back. Incidentally the actual version of Ubuntu i want to run is 9.10 but the installer won't progress past the network stage that it fails on so i can't get the os loaded. Do you know why this might be? [13:34] tgwoollard: I think you should be able to skip the networking stage and allow the installation to continue without networking? [13:35] tgwoollard: As for respinning the image; I'm afraid I do not know what to do about that. [13:37] I will give it a try and report back. So if i basically just install the OS, and ignore the network setup, are you saying that the NIC should automatically just work on system startup? [13:38] tgwoollard: Yes. [13:56] <\sh> jpds: AFAIK "yes" but I don't know if there is already a new iso for lucid with this fix...I had the the std iso of 10.04 and it wasn't fix at this time [13:58] <\sh> jpds: 10.04.1 iso didn't have the fix for the d-i udebs [13:59] \sh: Hmm, strange. [14:00] Hello again jpds. i have installed Ubuntu 10.04 and my interfaces file is empty. It just contains the loopback info on auto lo. At this stage are you advising that if i populate the interface file with my eth0 details and then bring up eth0 using ifconfig i should have networking? [14:00] tgwoollard: Yes. [14:00] <\sh> jpds: oh wait..sorry... [14:00] Okay i'll let you know [14:00] <\sh> jpds: I had the problem with bug #628776 [14:00] Launchpad bug 628776 in linux "HP NC511i Driver (be2net and be2scsi) is missing in kernel module udebs" [Low,In progress] https://launchpad.net/bugs/628776 [14:01] Thanks uvirtbot. how can i check within the OS if the be2net and be2scsi modules have loaded? [14:01] <\sh> jpds: that was fixed in maverick and is waiting for a 10.04.2 updated iso [14:02] <\sh> and that is the problem with the G7 series...not the bnx2x ;) *damnineedholiday* [14:02] <_ruben> dont use ifconfig, if you use /etc/network/interfaces you should use ifup [14:02] <_ruben> and even if you wouldnt use that file, you shouldnt use ifconfig either [14:03] Many Thanks all. For my own education here, how can i check within the OS if the be2net and be2scsi modules have loaded? [14:03] <\sh> tgwoollard: lsmod on the second tty during d-i === cs278|| is now known as cs278^kinslayer [14:03] <\sh> tgwoollard: but for lucid you won't succeed...after installation you have the modules though and a working network [14:04] <\sh> tgwoollard: and it only affects the d-i kernel packages ... [14:06] tgwoollard: You could also check 'dmesg' for anything related to network modules. [14:09] hello. i have 10.4.1 server, and i installed ebox - a web configuration manager [14:09] it installed apache2, and now i have 2 copies, one running as root and one as the ebox user. do you know how i can disable the apache running as root? [14:11] lightstep, that's normal, one process runs as root, the rest run as www-data [14:13] but this is a machine with low memory, and i don't want it to be a web server. so i want to stop the normal copy [14:14] ebox requires apache to be running [14:15] the root-owned process is what allows it to listen on low ports [14:15] the rest actually handle the web requests [14:15] you only have one instance of the apache service running, apache is a multithreaded application [14:16] so it doesn't take extra memory, it only shows so in "top"? the figures are different for the two processes [14:17] 8 apache threads is a pretty typical setup, you can tune those settings in apache2.conf [14:17] no, each process uses a certain amount of memory [14:17] but apache cannot be run with a single process [14:17] that's how it's designed [14:18] ok, thanks [14:20] if you're running out of resources, you could use a lighter weight httpd, like lighttpd or nginx, but I don't know if ebox supports them or how you'd go about switching it [14:22] you can also reduce the values in apache2.conf, StartServers and MinSpareServers to 1 [14:22] it seems to only use apache, so i'll try to work out the memory issues some other ways [14:22] apache is tuned for full feature web serving by default, but for ebox you're probably the only user [14:22] so that would be safe to do [14:32] New bug: #658341 in clamav "libclamav6: Some PDFs cause "Unknown error code ERROR"" [High,In progress] https://launchpad.net/bugs/658341 [14:36] New bug: #658346 in php5 (main) "Please enable process control support (--enable-pcntl) for cgi/cli" [Wishlist,Triaged] https://launchpad.net/bugs/658346 === cs278|| is now known as cs278 [14:50] hello again. jdps, i'm still failing to get network activity even after populating my interfaces file and bringing up eth0. [14:50] Am i missing something here? Apologies for all the q's === zoopster1 is now known as zoopster [14:54] Forgive me. It came up shortly after typing. Thanks to all who replied [14:56] tgwoollard: Ah, brilliant to hear. [14:56] tgwoollard: let me guess dhcp not static [14:56] Thanks for your help jpds. I would never have thought to simply just build it and assumed it would work. Bloody ISO!! === cs278|| is now known as cs278^kinslayer [15:11] <_ruben> bah, needa get the office line upgraded, only getting 60Mbps downloading the maverick repo === harrisonk_away is now known as harrisonk === harrisonk is now known as harrisonk_away [15:38] What is removed from the server OS to make it a virtual machine installation? [15:42] drivers, most likely === cs278|| is now known as cs278 [15:42] only need to support a given set of hardware for each VM software [15:50] <_ruben> and other "useful" stuff (like tab completion) is removed as well to reduce disk/memory footprint [16:01] Hello again all. You were kind enough to assist with a network issue earlier. I now have a slightly different question. I am installing Ubuntu 9.10 from USB on a HP BL685 G7 Server. I am installing from USB for two reasons. The CDROM and the NIC are not detected during the install. Running from USB gets around the CDROM issue, but after failing to discover my NIC the 9.10 installer wants me to choose a mirror for my installation. Obivousley beca [16:03] tgwoollard, just mount the cdrom using the oa [16:04] CDROM install doesn't work. THe install boots but then fails to detect the CDROM during actual installation. It's a known bug apparently with G7 servers. THis is why i need to install via USB. Can you advise how to get around the mirror issue and not the cd issue? [16:04] what mirror issue? [16:04] as you type too much on one line, irc cuts you off [16:04] tgwoollard, you're getting cut off [16:05] When i run the install from USB, it fails to detect NIC during installation [16:05] your client isn't smart enough to line break at the limit [16:05] hmm, actually, g7 isn't supported on 9.10 at all [16:05] it therfore can't install at the choose a mirror section and fails [16:05] I know, it looks like 10.04 LTS is the only official supported version [16:06] Not that, that really helps given how shit HP support is these days [16:06] Was curious as to how i could get 9.10 up and running as that is our preffered falvour [16:06] your USB install must be using a netboot image [16:06] would have to find backported drivers if they exist [16:06] probably download those backport packages manually and load them up, via your usb stick [16:06] you need to actually install from USB [16:07] I think it might be quickler and safer to just go with 10.04 as i've tested and confirmed all okay with that release. [16:07] Thanks for the quick response [16:07] qman, he would still have an issue, for network drivers even if he did fix that :) [16:07] yes [16:07] backporting network drivers is no simple task [16:14] anybody know offhand how I would enable the partner repository from a preseed file? [16:14] I see how to add a local repo, but not how to enable partner [16:14] ah! found it [16:14] d-i apt-setup/partner boolean true === luis__lopez is now known as luis_lopez [16:35] wishful thinking... [16:45] huh? [16:46] pmatulis: I agree. [16:55] ttx: ping === felimwhiteley_ is now known as felimwhiteley [17:23] anyone faced this vmbuilder problem before: http://ubuntuforums.org/showthread.php?p=9951621#post9951621 [17:24] JamesPage: pong [17:26] hi [17:26] trying to install 10.04 server [17:26] on my armada e500 [17:27] but the installation "stops" at detecting hardware [17:27] the screen starts flashing, it does that while detecting network and some other times as well but at this stage it doesn't move forward [17:28] i've tried installing debian and that works like a charm and i've been able to install previous versions of ubuntu on the same hardware [17:28] i believe the installation fails to detect my hard drive [17:29] since when i go into the expert mode detecting the hard drive fails... [17:30] please help me, i've search on google but i'm unable to find a solution for this and i really don't want to use another distro for this pc [17:32] ttx: wanted a second opinion on the severity of bug 658227 if you have time [17:32] Launchpad bug 658227 in openldap "won't start after Maverick upgrade; bdb "Program version 4.8 doesn't match environment version 4.7"" [Undecided,Confirmed] https://launchpad.net/bugs/658227 [17:33] JamesPage: I'm on a swap day, will look at it tomorrow. Sounds bad :) [17:34] ttx: thats fine (wondered why it was so quiet today); workaround documented in ticket so not that bad.... [17:35] ok [17:39] any ideas, anyone? [17:40] after a while of flickering in the screen i see the message, "killed" [17:40] and the hdd shows activity and the cd boots up [17:41] and after about 5 minutes or so i get a message about kernel panic [17:41] not syncing: out of memory and no killable processes... [17:48] Perhaps 10.10 doesn't have this problem but I doubt it, when I buy a new set of CD's I'll give it a try but until then I'll install 8.04 and perhaps try to upgrade it to 10.04. [17:49] X-Sleepy-X, I'd suggest installing 9.10 and upgrading instead [17:50] since it shares more in common with 10.04 [17:50] there will be less junk left behind in the upgrade [17:51] qman__, I would if I could but I tossed those CD's a while back... [17:51] qman__, I'll try to drag my ass to the shop today and buy some.... [17:51] :) [17:52] Perhaps I'll buy a pizza while I'm at it... :D [17:52] upgrading from hardy would work fine, but you would have some leftovers from stuff that lucid doesn't use anymore [17:53] pretty sure hardy uses a different syslog, it uses grub1, etc [17:53] ok [17:53] so you wouldn't have exactly the same package set as a fresh lucid install [17:53] For how much longer will 8.04 be supported? [17:54] I might just skip the upgrade... [17:54] https://wiki.ubuntu.com/LTS [17:54] Thanks! [17:54] you've got time. :) [17:54] haha, no worries there... :) [17:57] bbl === xfaf is now known as zul [18:03] How do I safely eject a SATA disk from the linux CLI? [18:03] jiboumans_: or jiboumans: how are the blueprints coming along? [18:03] umount all drives, then pull it :) [18:03] jcastro: they're being worked on this week [18:04] seriously, patdk-wk? [18:04] Hmmmm [18:04] sjr, well all drives mounted from that drive [18:04] no [18:04] jcastro: and next probably [18:04] heh? [18:04] Well I mean nothing is mounted, I'm trying to fix a mount problem, and thought perhaps it was caused by just turning the power off on the bay [18:05] jiboumans_: ok if you have any problems lmk [18:05]