[00:05] ahs3: well http://people.canonical.com/~serge/netcf-sid.debdiff and http://people.canonical.com/~serge/netcf-wheezy.debdiff are fine. [00:08] hallyn_: okey dokey. i'll give the sid one a spin later [00:08] ahs3: thx. i'm going to to saucy before i look into what is going boinkers with jessie [00:10] nod. i'll see if i can do that tonight or tomorrow [00:13] so. hooking a server directly via a static ip to the internet is bad? or i just need to setup a sw firewall first? [00:13] ahs3: thx. pls let me know when you look at sid, as i suppose ideally i should sync that into saucy rather than pushing separately [00:14] hallyn_: will do [00:16] i3luefire: a firewall is a useful backup to correct configuration of your intended services, but it isn't strictly necessary for running a host directly on the internet. [00:18] because i have been running my server behind a router for years... but now i am transitioning to a business internet connection with 5 static IPs and would like to put them to good use [00:18] i need a friend in IT [00:19] I'd take a good long look at existing systems before putting them directly online :) brand-new systems ought to have nearly nothing open by default and update during the install to avoid the worst of the potential issues [00:20] check sudo netstat -nlp output and make sure the services listening on external ports by default are intentional and properly configured. [00:22] wow [00:25] lots of stuff there [00:27] so the stuff with :::someport and 0.0.0.0:* are open to any and all basically? [00:27] yes [00:27] jeez [00:28] i have been sloppy [00:28] and the 192.168.... might be as well, depending upon how you configure things [00:28] no. i have a non-default ip scheme on the internal network [00:29] you didn't use rfc1918? [00:30] no [00:30] i just googled it because i didnt know what it was [00:31] this is a good time to fix that too :) hehe [00:31] yea [00:32] for some reason in the past i thought that was helpful in obfuscating my internal network for intruders. [00:33] but i would not even need to worry about the IPs on my internal network anymore except for wifi which will still use my router [00:33] since the hardwired ones will have static external IPs [00:34] you might still wish to have internal addresses and external addresses both on your machines; it might be convenient to have a CIFS share for internal-only use, for example [00:35] oh. how can i have 2 addresses without 2 nics? or did you mean with separate nics? [00:36] Vlan? [00:38] i3luefire: you can assign a few thousand IPs per NIC without any trouble.. I'm not confident of the best way to make it persistent via the interfaces(5) file, but "ip addr add" can add addresses all day long to existing NICs :) [00:38] wow. i need to take a class or read a book === Ursinha is now known as Ursinha-afk [01:03] roaksoax: hey are you around? === justizin_ is now known as justizin [01:45] Hi, Does anybody know how can I have application server in ubuntu? [01:47] is there a replacement for ethtool? [01:47] WARNING: The following packages cannot be authenticated! [01:47] ethtool [01:50] replacement? [01:50] maybe fixed your package repo [01:50] mojtaba, what is *application server* [01:50] I updated and did the --fix-missing [01:51] Beatstreet, well, those two have NOTHING to do with the problem [01:51] fix the gpg keys [01:55] Patrickdk: I want to run my applications from PI. [01:55] remotely* [01:55] mojtaba, talk english? [01:55] Patrickdk: English [01:55] what is *applications* [01:56] if you fail to explain, I can't help you [01:57] Patrickdk: I have lots of applications installed in my computer at home in ubuntu, and I want to run them from PI remotely. [01:57] you mean you want remote X sessions? [01:57] I mean, you keep saying applications [01:58] until you name atleast ONE, I cannot help you [01:58] but your welcome to read the manual [01:58] Patrickdk: for example I have installed Matlab in my computer [01:58] I want to run it remotely [01:58] Patrickdk - I re-downloaded the keys using the hexidecimal numbers but ethtool still fails to install [01:59] mojtaba, well, you either want to run X forwarding, or a remote X session using like vnc [02:00] Beatstreet, what version ubuntu? [02:00] Patrickdk: Does X forwarding works for Matlab? [02:00] question: anyone using multipath with a FC storage system? [02:00] Description: Ubuntu 11.04 [02:00] Release: 11.04 [02:00] Codename: natty [02:00] Beatstreet, that would be an issue [02:01] 11.04 and 11.10 aren't supported [02:01] ok, so I'm SOL - ok, thanks [02:01] I just deleted a bunch of volumes and now my multipath daemon is showing me all sorts of failed paths. is there a good way to clean them out, or a different procedure I should be following? [02:01] your not SOL, but your unsupported :) [02:01] look into adjusting apt to use the archive server [02:02] and upgrade as soon as possible :) [02:04] other question; is there something like snapshot.debian.org for ubuntu package repos? if not, does anyone know of tools which can make setting up something locally like this possible? [02:06] Thanks Patrickdk [02:10] jkitchen: looks like weasel published the code: http://anonscm.debian.org/gitweb/?p=mirror/snapshot.debian.org.git [02:12] sarnold: sweet, looking at that now, thanks [02:13] I'd love to set something like that up locally. in my environment I'm all about absolute reproducibility of machines, right down to the packages they're running, even if that means running older versions of packages. if I can auto snapshot like that then I can just promote things along later [02:13] I was just gonna do an apt-mirror and leave it at that :( [02:14] making it more flexible will be a good thing :) [02:14] ahs3: cjwatson saves us. he has a patch for grub2 which fixes teh gnulib failure in netcf in jessie. i'll post a full debdiff in a bit [02:41] ahs3: but infinity points out that pushing to jessie is weird and to ignore that. (the fix at people.canonical.com/~serge/netcf-jessie.debdiff did work though,easing my mind :) [02:59] hallyn_: here [03:06] roaksoax: nm, thanks :) [03:07] hallyn_: lol k ;) [03:42] I currently have a sFTP server up and running. I chrooted it to a folder on my root directory. I really like how I have it all setup but none of my clients etc want to use sFTP. Is there a reasonably easy way to have ftp go there as well? [03:42] almost just have sFTP allow for ftp connections? [03:47] I guess I would need to have ssh allow a FTP connection as well as sFTP, is that possible? [06:34] utlemming: sheesh, new raring images again? :-P [07:31] zanzacar: just setup some ftp server? [07:32] zanzacar: vsftpd or something [09:58] Hi. i'm new to set up a minecraft server on ubuntu straight from terminal. How do I make it so i can enter my server? do i put in my ipv4 or ipv6? nad if so.. how do I check it? [10:07] ? [10:17] Cyrax: in your client, put the ip of the server you are running minecraft on? [11:37] jamespage: FYI, in case you didn't notice, the autopkgtest for squid3 failed [11:41] mdeslaur, I did notice and I can't reproduce [11:41] they all pass when I run locally [11:41] jamespage: :( [11:41] mdeslaur, I think its actually something in vsftpd borking but I can't figure out what [11:48] What is the command on terminal to find my ip like ipv4 and stuff? [11:51] ip a [11:52] or 'ip address list' if you want the long version [11:58] sgran: i'm trying to run minecraft server and i start it with a command.. can't remember all,but it starts with java. Do i need to put my real ip address or ipv4 or ipv6 address into server.properties? [11:59] Cyrax: I don't know anything about running a minecraft server, sorry [11:59] I'm sure there are docs about this online? [12:15] Cyrax: you don't need to put any ip into that file, afaik [12:16] Cyrax: check the various minecraft server documentation you can find on the web, or maybe ask in #minecraft. [12:58] jamespage: the sqlalchemy stuff got fixed in heat so we can drop out patch [12:58] zul, great [13:08] zul, http://people.canonical.com/~jamespage/ca/havana/ [13:09] jamespage: +1 i should have some for you today as well === blitzkrieg3 is now known as jmleddy [14:24] jamespage/roaksoax: https://code.launchpad.net/~zulcss/heat/refresh-aug14/+merge/180150 [14:46] adam_g, when you start today - the nova-cloud-controller redux is missing its templates dir in the branch [14:51] jamespage: i think i have a fix for the glance testsuite failures (its sqlalchemy related) [14:52] zul, the one where is completely crashed the machine its running on? [14:52] jamespage: yeah [14:53] * zul is in the zone [14:53] rbasak, OK - so my panda crashed 3.5 hrs into testing building mongodb [14:54] rbasak, as it can't get any worse in proposed right now I'm going to upload my fix for the armhf build failure [14:55] is it normal for quotacheck to take a long time? [15:00] jamespage: I've been building on a highbank node [15:01] rbasak, me too! [15:01] https://launchpad.net/ubuntu/+source/mongodb/1:2.4.5-1ubuntu2/+build/4877049 [15:01] jamespage: :) [15:01] lol [15:01] I have a parallel build that's been going for 45 minutes [15:01] I guess we'll see. [15:02] rbasak, tbh I'm not 100% sure why the DM started using the internal libv8 [15:02] I'd not seen any issues [15:28] jamespage: wait...nevermind [15:39] roaksoax: hey! when you get a few minutes, would you mind pushing: http://people.canonical.com/~serge/netcf-precise.debdiff http://people.canonical.com/~serge/netcf-quantal.debdiff and http://people.canonical.com/~serge/netcf-raring.debdiff ? [15:40] hallyn_: sure! give me ~15 mins or so [15:41] roaksoax: thanks! [15:47] Hi, anyone using vmbuilder with a preseeed file? I'm trying to make the pkgsel/include directive actually work [15:47] *preseed [15:51] !anyone [15:51] A high percentage of the first questions asked in this channel start with "Does anyone/anybody..." Why not ask your next question (the real one) and find out? See also !details, !gq, and !poll. [15:53] hallyn_: btw... changelogs versions should preferably be ubuntu3.1, ubuntu3.2 and so on, instead of ubuntu4, ubuntu4 (for precise case) when it comes to SRU [15:56] jamespage, oh, right. i have a bunch of stuff to push there [15:56] adam_g, please do [15:56] jamespage, on its way [15:56] roaksoax: iirc the last time that was discussed itw as decided that was only needed if the next release shared the version. but it does make it clearer what's going on,so i i like your suggestion. do you mind doing it inline? [15:56] adam_g, can I suggest that we land the redux first and then do the neutron rename afterwards [15:57] adam_g, we should probably do a re-sync with charm-helpers as well [15:57] quite a few things have landed [15:57] jamespage, renaming of quantum-gateway and associated interfaces to neutron-server? tahts fine by me. ive tried to make the nova charms agnostic to the name, at least in most places [15:57] adam_g, yes [15:57] jamespage, ya. i need to push some stuff to charm helpers and resync around [15:57] adam_g, OK [15:58] jamespage, switching from helper local configs to using the main core helpers for relation_get() may have introduced some bugs [15:58] adam_g, most things seem OK - I hit one problem with an older helper in cinder which was using socket.get_hostname [15:58] hallyn_: yeah I guess that i might have missed that discussion, but I guess my thinking might be old school :) [15:58] but I already fixed that in charm-helpers :-) [15:59] jamespage, https://bugs.launchpad.net/charm-helpers/+bug/1203241 [15:59] roaksoax: it makes it clearer whether it's beenupdated in an sru so i like that [15:59] Launchpad bug 1203241 in charm-helpers "relation_get() on a non-set relation setting does not return None" [Undecided,New] [15:59] hallyn_: but you are right, It doens' really matter cause the versions are not shared between releases :) [16:00] adam_g, well it did do the right thing once upon a time... [16:00] jamespage, the relation_get() in core.hookenv? [16:00] i thought so too [16:00] adam_g, yes - because I made it do that [16:01] hah [16:01] unit/config/relation [16:01] thats what i thought. i wonder if juju's interface changed (as mentioned in the bug) [16:03] jamespage, pushed [16:04] jamespage, the main nova.conf template is getting ugly. we may want to look at using template inheritance to handle that [16:08] jamespage, also a bunch of nova-compute changes pushed [16:18] stgraber: so fwiw i pushed lxc-user-nic (bc it doesn't change anything, just adds a so far unused binary) and the cgroup rework sets (bc without it nested containers are broken) ) to staging. [16:19] if the cgroup set breaks anything i'll git-revert it poste haste, but there was no comment and "it works here" :) [16:33] hallyn_: done! [16:35] roaksoax: thx [17:04] adam_g, I might know what the relation_Get issue is [17:05] there was a change in behaviour from pyjuju -> juju-core [17:05] in that 'null' was being returned for a missing relation key rather than empty string [17:05] I suspect that has been fixed! [17:05] jamespage, hmm [17:07] jamespage, so juju-core returns empty string for config-get/relation-get when nothing is set? [17:07] adam_g, not sure - I would need to re-test [17:07] it certainly did not at some point in time [17:50] hallyn_, stgraber utlemming i just hit 'publish' on http://ubuntu-smoser.blogspot.com/2013/08/lxc-with-fast-cloning-via-overlayfs-and.html [17:51] thanks for your great work. [17:51] anyone else using lxc, espeicailly with the lxc ubuntu-cloud template might find that interesting. [17:55] smoser: cool :) [18:04] sarnold: awesome, will these new changes be availible in openstack too? [18:04] I havn't really played with lxcs in openstack yet. [18:04] blkperl: sorry, no idea [18:13] blkperl, no, unfortunately nova's current lxc support uses a different implementation (libvirt+lxc) [18:16] jamespage, all grizzly branches migrated to lp:~ubuntu-server-dev, build scripts updated and 2013.1.3 update retargetted [18:17] adam_g: how would you test a section of a charm that looks like this: http://pastebin.ubuntu.com/5985960/ [18:18] adam_g: (the test in it fails because it seems that it only catches the last relation_set function being executed [18:18] so I can only assert against the last one [18:19] roaksoax, which assertion is failing, the very last? [18:20] adam_g: yep [18:20] roaksoax, i think you want to look at self.relation_set.call_args_list [18:20] adam_g: because it compares against the relation_set for image-service [18:20] adam_g: http://pastebin.ubuntu.com/5985970/ [18:21] roaksoax, something like this http://paste.ubuntu.com/5985971/ [18:22] adam_g: so the actual assert comparison is against "Actual call: relation_set(glance-api-server='https://10.10.10.10:9292', relation_id='identity-service:0')" whcih is the second if statement, when I want to compare against the first statement [18:22] * roaksoax looks [18:24] adam_g: :( [18:40] adam_g: thanks [18:59] Daviey: ping [18:59] Daviey: can you review oslo.messaging tomorrow its in the source new queue [19:01] What version of ubuntu is best paired with postgres? [19:07] smoser: via-via :) [19:07] adam_g: glance didn't require to save an rc script with the environment variables right? [19:08] smoser: your blog requires js [19:09] hallyn_, do you need me to give you a recommendation for operating system that has a browser capable of rendering it ? [19:09] roaksoax, if its not in the bas charm dont bother, im not even sure anyone is actually using those rc files naymore [19:09] hallyn_, https://gist.github.com/smoser/6199772 [19:09] smoser: you're asking me if i need my computer to run your code for you? [19:10] this should be good :) [19:10] hallyn_, you already run my code. your machine last checked in at [19:10] Wed Aug 14 19:11:04 UTC 2013 [19:10] zul: Yes [19:10] adam_g: so yueah I had this in config_changed: http://pastebin.ubuntu.com/5986146/ i guess I can simply drop it [19:10] smoser: that's not my machine, just one of my decoys [19:10] Daviey: merci buckets [19:11] no, thats the real one. one of your decoys failed checkin last night. [19:11] hallyn_, ifyou'd rather git checkout https://gist.github.com/smoser/6199772 is basically it. [19:11] i just really didn't like having to deal with blog entires, so i started putting stuff in gists [19:11] smoser: no no i trusted you implicitly so enabled js just for your blog post [19:12] smoser: drat. the nice thing about regular blog posts is i can read them from readitlater on my ereader [19:12] yeah, that does suck. i agree. [19:12] but getting reaasonable syntax highlighting in blogger sucks [19:12] and then updating it also sucks [19:13] heck just getting code to render properly in blog sucks [19:13] no arguments there [19:13] yeah, that is basically what drove me to the "just use github" [19:14] smoser: you switched to github pages? [19:17] Daviey: that was awhiel ago, but now he has switched to github scripts which you download which use sed to noninteractively write the blog post on your system. [19:18] he finds that more authentic [19:18] no. i use blogger and just include a gist in the file [19:21] smoser: and is that pushed to planet.u.c? [19:22] should be. [19:22] yeah, syndicated there. [19:22] oh. [19:22] shoot. i see. [19:22] that sucks. [19:22] you have a better solution for that? [19:23] nope, not even sure what the problem is === dosaboy_1 is now known as dosaboy [19:24] getting syntax highlighting / formatted code into a blog post. [19:24] no, last time i looked into it ppl were suggesting some wordpress pluging - bleh [19:24] right. [19:24] so that sucks. [19:25] so i just reference gists via javascript [19:25] but clearly that does not syndicate well [19:25] hallyn_: haha.. is this how smoser browses ? http://article.gmane.org/gmane.os.openbsd.misc/134979 [19:25] I'm using Pelican (Python) now as a static blog generator, so now I can write my posts in markdown. It supports syntax highlighting apparently, but I haven't tried it. Also I have little experience, having now written a grand total of one blog post (with Pelican that is). [19:26] rbasak: I also have been experimenting with Pelican, then pushing it up to Heroku [19:26] I'm pushing to a server running apache (I know. Old school!) [19:26] rbasak: hm, i may try that [19:27] rbasak: I was more looking for a reason to learn Heroku.. But I agree, i prefer simply pushing it up to a server [19:27] rbasak: what about comments? (bc we all want commnets :) [19:28] Scaling blogs down is what the cool kids do nowadays. I'm sure my cheap VPS will be able to do plenty of traffic when it's serving only static pages. [19:28] hallyn_: I added google plus to the footer for comments [19:28] hallyn_: I haven't done comments, but Pelican supports Disqus out of the box I think. [19:28] Daviey: i dn't do g+ [19:28] rbasak: ok, i'll have to look into it. thanks for the tip [19:29] I've stopped believing in comments. People can email me, blog a rebuttall, or tweet, or something I figure. [19:29] I'm using Octopress for static blog [19:29] agreed, i dno't think i actually want comments [19:29] Moderating comments is too much of a pain when in reality I'll get a comment once a year or something. [19:30] https://twitter.com/AvoidComments is relevant here [19:32] jcastro: there's another timesink. :) [19:34] Is there any USB 3.0 issued or issues with the USB 3.0 cont rollets? [19:35] hallyn_, i donr tknow if you consider it a bug or not [19:36] but: lxc-destroy -n hallyn_loves_lxc_but_this_container_doesnt_exist [19:36] exits with '0' [19:39] and that is a change from previous behavior also. [19:39] lxc-stop also behaves like that [19:39] (in Raring at least) [19:39] hm.. [19:40] adam_g: ok I added the tests for all fo the glance_relations, you want a MP against the branch in ~openstack-charmers ? [19:40] roaksoax, sure [19:41] oh wait. i'm wrong. it does exit failure (1) it just doesn't complain anywhere. [19:41] adam_g: done! https://code.launchpad.net/~andreserl/charms/precise/glance/port/+merge/180224 [19:41] roaksoax, word [19:42] adam_g, hey - I just synced the reprepro archive for havana in the CI lab from havana-staging [19:42] smoser: yeah just pulled up the source ... lxc_destroy could spit out an error [19:42] adam_g, we suck at backporting there first... [19:42] (lxcapi_destroy() should not) [19:42] jamespage, ah cool [19:43] http://paste.ubuntu.com/5986246/ [19:43] hallyn_, ^ [19:44] http://paste.ubuntu.com/5986259/ [19:44] better [19:45] i think you make ad money off pastebin somehow [19:45] well, i get a referrer fee everytime i use pastebinit. [19:45] most of hte money goes to stgraber [19:45] lol [19:46] smoser: yeah lxc_stop is doing exit(-1) instead of exit(1) [19:46] i'll fix both of those right now in upstream [19:46] fix to what ? [19:46] what is right ? [19:46] 0 ? [19:46] lxc_destroy will spit out a msg, [19:46] and lxc-stop will exit 1 [19:47] i consider it an error to ask to stop c when ci s not running [19:47] disagree? [19:47] (also, note there's a big cgroup rewrite so next time i build ppa it *may* cause troubles... but shouldn't, i've obviously tested) [19:47] its argubable, but i think its bikeshed, so i'd recommend staying consistent with previously published behavior. [19:48] hallyn_, do you want a bug for "keep track of child ephemeral mounts" [19:49] ie, clones that depend on a master [19:49] smoser: yeah sure [19:51] smoser: wait. lxc_destroy *does* print an error if container is not found. [19:52] smoser: oh. right. this is actually more complicated than it seems [19:52] the problem is that there is the concept of containers withotu a config file [19:52] ok, nm, should be ok [19:53] i didn't make that pastebin up. [19:55] smoser: what's to stop you? [19:55] yeah nm, fixed in git head nwo [19:57] https://bugs.launchpad.net/ubuntu/+source/lxc/+bug/1212414 [19:57] Launchpad bug 1212414 in lxc "lxc-destroy allows unsafe destruction of overlayfs sources" [Undecided,New] [19:57] hallyn_, ^ [19:58] smoser: great, thanks [20:01] Hi [20:03] Evening [20:04] Im curious, trying to setup Ubuntu as a NAS, running iperf to test network througput but.. well for a GigE network with Intel Pro1000's on each side its only reporting about 650Mbit/sec [20:04] Shoudlnt a Gige network be faster? [20:08] Slyboots_: it should, clocking gige at >950Mbit/s on the LAN hasn't been hard for, hm, 8-10 years now. [20:09] Im at a total loss as to what the hell is wrong with this server, I really am [20:09] Had a samba/nfs setup that heh, could WRITE to the shares at about 80MB/sec which is about what iperf is reporting [20:09] READING from those exact same shares, in either networking format (smb/nfs) 30MB/sec [20:11] So, I've got it totally stripped down to the nuts and bolts and testing each componant, one at a time lol [20:11] I've tried increasing the TPC window size to 1Mbyte, and thats got the speed up to 920Mbits/sec which.. is nice [20:47] Anyone any idea why my NIC's would be detected.. but not work (no dhcp responce, and they are marked as DOWN in ip addr) [20:48] But they work fine in a liveCD enviroment [20:49] Slyboots_: wierd - does it work with a static IP? [20:50] Just about to try that now [20:53] jamespage: my mongodb build test succeeded from your branch. As did the buildd. Looks like it's migrated from proposed too. [20:53] \o/ [20:54] rbasak, indeed! [20:54] 2.5 hrs - not bad [20:54] only x2 amd64 [20:54] I got 02:48:34 on my highbank node [20:54] I might try -j5 next time. That was -j4. [20:59] Right.. So I put two entries in /etc/network/interfaces for static.. [20:59] and.. they arnt "taking" [20:59] if I run ifconifg eth0 down/up the interface comes "up" but it doesnt have an IP assigned [21:03] zul, please can you fixup the versioning on the oslo.messaging upload [21:03] zul, 1.2.0~a4-0ubuntu1 [21:03] hello all. I get this warning after an installed Ubuntu13.04-64-mini, advanced mode: "" Please check that your locale settings: LANGUAGE = (unset), LC_ALL = (unset), [after that all as] "pt_PT@euro", except LANG = "en_US.UTF-8" [as i want main language as english] . Could you please help me solve this issue... [21:03] otherwise you will end up epoching when it releases to 1.2.0 [21:05] btw, during certain operations i get: "locale.Error: unsupported locale setting" [21:06] zul, can you please start merging your approved changes into packaging branches when you push them instead of just pushing your local branch directly? i use tarmac locally to do it, i just run it whenever something is ready for merging [21:07] zul, my tarmac.conf http://paste.ubuntu.com/5986517/ [21:07] jamespage: ack [21:07] adam_g: ack [21:07] morning folks [21:07] hey thumper [21:10] zul, you might want to fixup your bzr whoami from wherever you are working right now as well [21:10] (typo in your email address) [21:13] someone please help me with that... [21:15] MraMaria: I'm sorry, I thought that was just background information before the question.. [21:15] MraMaria: can you restate your question? [21:15] sarnold: :) [21:16] I get this warning after an installed Ubuntu13.04-64-mini, advanced mode: "" Please check that your locale settings: LANGUAGE = (unset), LC_ALL = (unset), [after that all as] "pt_PT@euro", except LANG = "en_US.UTF-8" [as i want main language as english] . Could you please help me solve this issue... [21:16] ah, see, I don't see an issue to solve. [21:17] sarnold: i'll pastebin the output of all warning... [21:17] MraMaria: perhaps pastebin the current output of 'locale' and maybe add to it what you'd -like- the output to be.. [21:17] that's a good idea :) [21:18] sarnold: http://paste.ubuntu.com/5986558/ and i'll put also for 'locale' [21:20] hi there, super newbie question, I'm trying to test a saucy package in 12.04, can someone point to some info on how to do it? thnx [21:21] MraMaria: do you have both language-pack-pt language-pack-pt-base packages installed? [21:22] sarnold: http://paste.ubuntu.com/5986563/ : output for 'locale' - the main idea is to have the lang as english but all other settings as PT€euro [21:22] sarnold: i've selected basic ubuntu-server during the installation... [21:23] MraMaria: you may need all four packages, language-pack-pt language-pack-pt-base language-pack-en language-pack-en-base installed [21:23] sarnold: i'll do it now.. :) [21:26] sarnold: it seems like is solved. thanks so much :) [21:26] MraMaria: nice! :) have fun [21:26] sarnold: can i post anothet matter, pls :) [21:27] MraMaria: sure [21:28] i know we should not have any graphic interface on server. altough this is a new (rebuilded) machive and i would like to explore it before going into "production". due to that i've installed lubuntu minimal. however, although i get a lubuntu logo i don't get the login dialog. startx didn't work... [21:29] sarnold: wait.. i've installed basic-server + lubuntu minimal... [21:29] MraMaria: try "serice start lightdm" ? [21:30] sarnold: okay [21:31] sarnold: i have 1 phone call.. brb. sorry [21:40] I really cant figure this out.. Oh great Tux! Please help.. I've got fish! tasty.. tasty fish! [21:41] Slyboots_: pastebin your interfaces file? [21:41] I've setup Samba (and NFS) on my Ubuntu server, I can write to the server at about 110MB/sec which is fine [21:41] sarnold, Oh I fixed that :) .. typo in the interfaces file -.- [21:41] Slyboots_: oh yay :) [21:41] I can only read files OFF the server.. at about 30MB [21:42] if I use DD to write zeros to the disk, I get about 650MB/sec. Same thing to read files from disk to /dev/zero I get 564MB/sec [21:42] Which, is a lil weird in itself but.. [21:42] Slyboots_: how large of files are you reading and writing? [21:42] 2-4GB in size [21:42] I've tried larger (up to 14GB) but it doesnt seem to maky any difference [21:42] nice ssd then. :) [21:42] I can write to the disks at blazing speeds, but reads are terrible [21:43] .. what? [21:43] no.. metal SATA2 disks [21:43] really? [21:43] WEll I have a SSD in my desktop.. [21:43] the best metal disks I've seen go around 80-100 MBps. [21:43] The metal disks _I_ owned went around 40-60 MBps. [21:44] and reading cold-cache files from disk at 30 MBps would make sense if they were 'smallish', say half a meg or smaller.. [21:45] I dunno, these files should really be cached anyway [21:46] Im copying right to the disk, then right back off [21:46] But the reads are terrible no matter what I seem to do [21:46] Im going to try pulling all the RAM, see what that does :P === NomadJim_ is now known as NomadJim [21:50] sarnold: 'service start lightdm' didn't work. something might be wrong with the graphics adapter... when is the time for the login dialog the monitor displays it... i did dpkg-reconfigure lightdm and didn't help also. i think i'm going to purge lubuntu-core or.. any other suggestion? [21:50]