[00:18] Yesterday I sent an email to the list about postfix and ethernet adapters. The two replies I got appear to refer to mapping postfix to an IP address, which in turn "connects" it to a physical adapter. Only, I cannot pre-determine what the IP address will be. What I really want is postfix to use whatever Internet connection there is. At the moment it just picks one and tells me that it cannot resolve an address. [00:19] owh: you're talking about a postfix bug? [00:19] giovani: I don't think it's a bug, I think it's a "feature". [00:20] owh: ok, what do you want #ubuntu-server to do? [00:20] is there a question? [00:20] Yes, I'm asking if there are thoughts on how postfix might be coerced into doing what I need it to do. [00:21] Or if there are alternative approaches. [00:22] so you have two ethernet adapters, and both are getting dynamically assigned ips? [00:22] Some background: There are two network adapters. They get their IP address from a DHCP server on the network. I'm running postfix to send out mail to an external smtp server. [00:22] Yup [00:23] why do they both have to be dhcp? [00:23] owh just a thought, postfix should be behind a firewall and nat'd. why don't you set them up with static addresses? [00:25] I'm not sure I'd advise NAT for any server, if avoidable [00:25] Static IP is also not really possible. [00:25] you can't realistically run a mail server on the internet today on dynamic addresses [00:25] you'll get spam-filtered at most ISPs ... totally prevented from sending mail [00:26] Well, if I connect via SSL to an external smtp server, all is just fine and dandy. I don't send mail out to the end-user, it goes via an extra hop. [00:26] giovani, You have to forward the ports, so it kinda defeats the purpose of NAT. I totally agree. [00:26] Incoming mail is via fetchmail. [00:26] bytor4232: and NAT is just messy and never advisable unless required [00:26] owh: ok, so you're not running an internet mail server [00:26] just a local one [00:27] Yup [00:27] so I'm still unclear why you can't just issue inet_interfaces = all [00:27] and then have postfix use the default route [00:27] like it will by default [00:27] and listen on both interfaces [00:28] In the firefox profile directory, which file contains the stored password? [00:28] oops [00:28] sorry, wrong channel, sorry [00:28] giovani: Well, inet_interfaces = all is already configured. It seems to just grab "the first interface" and then decide that it cannot resolve the smtp server. [00:29] owh: it "grabs" your default route [00:29] and if it can't resolve dns -- it sounds like your network settings aren't properly configured [00:29] Right, so if the connected interface gets its IP address after postfix has started, it uses a bogus route and dies. [00:30] uh [00:30] your dhclient process should be running much earlier than postfix in startup [00:30] besides, how often is this server being rebooted that that's a problem? [00:31] maybe you need to rearrange your startup scripts [00:31] giovani: The server is a laptop. [00:31] ... [00:31] sigh [00:31] rearrange your startup scripts -- something is wrong [00:31] no server processes should be starting before dhclient executes and finishes [00:31] otherwise they'd all have this problem [00:32] Well, the "server" goes to sleep. When it wakes up, it takes a little while for the network to be available. The adapters are managed by Network Manager, postfix is always running. [00:32] haha [00:33] this isn't a postfix problem [00:33] Excellent. [00:33] this is a ... don't run mail servers on laptops with dual dhcp nics problem [00:33] ROTFL [00:33] Thanks, that was helpful :) [00:33] well, be serious [00:33] this setup of yours creates tons of problems [00:33] I'm open to suggestions. [00:34] don't use a laptop [00:34] for your mail server [00:34] and why, if this is just an internal mail server, can you not use static ips? [00:34] Because it's a laptop and it moves around - onto other networks :) [00:34] well don't do that [00:34] sounds like you need an internet mail server [00:35] either sitting at home on stationary hardware, or out on the internet on a $5/mo VPS [00:36] owh: restart postfix in /etc/network/if-up.d/ once everything is up? [00:36] There is a long history behind this configuration. I am mobile. I have a satellite dish. I setup impromptu internet cafes alongside the road. People connect to my wifi hotspot and can send email. [00:36] cjwatson: Will the id-up.d scripts run if Network Manager is in control? [00:37] yep, that's actually a decent idea [00:37] let's see if it works [00:38] Hmm, there is already a postfix script there. It reloads if it's running, but I'm guessing that a reload won't re-check the route. [00:39] There is also one in if-down.d - also reloads. [00:40] What if I change reload to restart - any nasty side effects anyone can think of? [00:41] You know, like mail vanishing that is currently being sent, "little" things like that? [00:47] BRB [00:50] Hmm, seems no scripts are running at all. There is bug #336736 pointing at pre and post scripts not running, but up and down are. I'm not seeing any run. [00:50] Launchpad bug 336736 in network-manager "NetworkManager does not call /etc/network/if-pre-up.d scripts" [Medium,In progress] https://launchpad.net/bugs/336736 [01:52] hello people [01:52] i am new to ubuntu server and i got couple questions [01:53] i am about to to fresh install of os and how should i do my hard drive?? i mean boot, / , swap and more [01:54] wizardslovak: how many users? One, or more than one? [01:55] this will be my personal server for test and study purposes and LAN for now [01:55] wizardslovak: in addition to number of users, it really depends on what kind of applications it will be running [01:55] admin only me users only me [01:55] wizardslovak: are you going to use software RAID? Will there be a large amount of user data in /home (e.g. a file server)? [01:55] i want to do web server (apache , mysql,php) [01:56] OK, so it's a LAMP server? [01:56] yes [01:56] wizardslovak: are you using software RAID? [01:56] this is my old machine celeron 2.4 , 512mb 80gb hda [01:56] no [01:58] its for study purposes , i always wanted to learn servers and command line [01:58] well, not many web projects are going to use anything approaching 80 gigs, so partitioning isn't going to be a terribly big concern for this machine [01:58] I recommend you use LVM, first of all. [01:58] LVM allows you to change your mind later. [01:59] logical volume manager? [01:59] Secondly, I recommend you allocate 4GiB to the root partition (/), and then (possibly after installation) allocate say 10GiB to /var/www or wherever you expect to put your data. On a file server it would be /home. [01:59] why? [01:59] wizardslovak: yes. [01:59] 11:02 LVM allows you to change your mind later. [01:59] LVM allows you to resize partitions or move them between disks *while the system is running*. [01:59] yeah, for experimental stuff, it's hard to say in advance; better to put yourself in a position where you can change things as needed [01:59] well thats what i wanted how many Gbs to what [02:00] It's good to use LVM and only allocate what you need short-term, then lattr allocate more space when you need it [02:00] e.g. 4GiB for root and 100GiB for /home (on a file server), and leave the other 376GiB unallocated. [02:01] lol i got only 80gbs for now [02:01] on an 80 gb hard drive??? [02:01] yes [02:01] wizardslovak: so allocate 4Gib and 10GiB. [02:02] ok 4gb root, 1.5gb swap and 10gb /home [02:02] It doesn't matter if you screw up with LVM, because you can allocate more later. [02:02] that system's unlikely to swap unless you go live with a high-traffic sight, so minimal swap should be okay [02:02] like 256MB [02:02] wizardslovak: well in your case you probably won't have anything in /home, right? [02:02] again, you can increase as necessary [02:02] Right. [02:02] well no [02:03] it might actually not make any sense to use a separate /home partition then [02:04] separate /var would be more useful, if anything [02:04] where then i will install lamp and www docs? [02:04] the default httpdocs directory is /var/www [02:05] you can change that to anywhere you like, but might as well leave it as is until you have a specific reason to change it [02:05] the elements of the LAMP stack go in /usr, /etc, and /lib [02:08] so 4gb root lest say 1gb swap and ?? [02:17] what's a sane way to get hardy to get glibc 2.8 [02:17] I need a patched pthread in hardy that just doesn't seem to exist in Ubuntu that exists in Debian [02:17] :| [02:18] Damm: why 2.8? [02:19] Damm: Debian, at least, never even shipped that (AFAIK). [02:20] If you enjoy pain and instability, you can always (very carefully!) cherry-pick bits from Debian. But if you're not a DD, you probably don't have experience to do so safely. [02:23] twb, I'd rather upgrade to Jaunty [02:23] if i have to do that [02:23] tbh [02:23] Fair enough. [02:24] is Jaunty going to be LTS? [02:24] Cherry-picking from ubuntu+1 is safer than cherry-picking from Debian. [02:24] or Karmic Koala? [02:24] Koala? Bah. I would have chosen Kookaburra. [02:24] Not if the world is watching a Koala being watered during a bush fire :) [02:25] there was a two year interval between the last LTSs, so I'd guess the next one won't be at least until 10.04 [02:25] I concur with p_quarles [02:25] naah [02:25] yep p_quarles [02:25] i saw that [02:26] welp i might as well go Jaunty and see if it fixes it, if not I can reinstall [02:26] definetly easier then trying to apt-get dist-downgrade [02:27] getting Ulrich to admit pthread bugs is like [02:27] trying to scale the chrysler tower with 2 fingers [02:27] or is that too polite? [02:27] should i install lamp in os install or wait and install it later? [02:27] Damm: If you have spare space in your LVM, I strongly recommend making an LVM snapshot or otherwise keeping your old root filesystem around as-is while you do the test upgrade. [02:28] twb, reinstalling takes 15minutes [02:28] and it's automated [02:28] Damm: well, if you have preseeding and know exactly what you want [02:28] twb, no it's fai [02:28] and i didn't set it up [02:28] wizardslovak: I would install it later with tasksel; I recall that giving you more options [02:28] Yeah, fai is preseeding [02:28] but by god it's easy enough to get done [02:28] when I have time I'll do the preseeding [02:28] I'm not really a fan of FAI, I'd prefer a combination of manual preseeding and puppet. [02:29] * Damm stabs puppet with a fork. [02:29] chef 4ever [02:29] Not that I like puppet... [02:29] tasksel??? can i use apt-get? [02:30] wizardslovak: tasksel is another front-end for dpkg, like apt-get; it's useful for deplying common setups, like lamp [02:30] wizardslovak: sudo tasksel install lamp-server [02:30] (tasksel is a front-end for apt-get/aptitude, actually) [02:30] wizardslovak: or tasksel --list-tasks for the options [02:31] whats the difference btw tasksel and apt-get and aptitude [02:31] cjwatson: which is a dependency resolving front-end for dpkg [02:31] Damm: there's no wikipedia article for Chef, so it's obviously no good! [02:31] p_quarles: I'm aware of that, but I think it's relevant that it isn't a front-end to dpkg directly, since that makes it clear that anything you can do with tasksel you can also do with apt-get [02:32] cjwatson: true; tasksel is only useful if you want to deploy a common default; for wizardslovak's purposes, though, the default lamp stack should be ideal [02:32] wizardslovak: tasksel is a simplified interface that just offers you the ability to install or remove sets of packages corresponding to common tasks (or profiles, whatever you want to call them). apt-get and aptitude are full package managers that give you manual control over which packages are installed. [02:33] I basically agree with p_quarles though, just nitpicking :-) [02:33] :) [02:33] twb, www.opscode.com [02:35] it's amazing that tasksel is pretty much the same as it was in Debian Potato [02:38] That's because nobody uses tasksel except newbies [02:38] Damm: pretty different internally though :) [02:38] twb: Hmm, newbies eh? [02:39] newbies?? [02:39] i always used apt-get [02:40] cjwatson, I would hope so. [02:40] if the code stagnated from old versions of libc5 and still works on current glibc [02:40] owh: yeah; tasksel's main job is to provide newbies an easy way to opt-in to getting Xorg and gnome installed :-P [02:41] well you'd have to give that person a medal [02:41] twb: ROTFL [02:41] Damm: it's in perl, so libc is not all that relevant [02:41] rofl [02:41] that figures shows you how much I use tasksel [02:41] Damm: and potato didn't use libc5 [02:41] potato had elf [02:41] that i do recall [02:42] Damm: well, I noticed that ubuntu-server has some fancy different tasksel tasks [02:42] twb, i hasn't noticed [02:42] fai kinda gets in the way of ubuntu being useful [02:42] There's now tasks like "I want lots of security holes", i.e. LAMP [02:43] the libc6 transition was in bo, two releases before potato [02:43] err, hamm, not bo [02:43] right [02:43] but ole fashioned libc5 [02:43] or old libc was potato [02:43] no it wasn't! [02:43] not this new fangled stuff that Ulrich sold us on [02:43] potato was two releases after Debian switched to libc6; it did not use libc5 [02:44] not for anything of any importance anyway [02:44] used libc4? [02:44] it used libc6. 5+1. [02:44] I'm not sure how to be any clearer :) [02:44] * Damm runs in terror [02:44] you can't [02:44] and I've hit the moment that I can't takes no more [02:44] * Damm remembers why he ran to freebsd from Debian now [02:45] Hey so guys, a stock ubuntu-server 8.04 install, and I deliberately break filesystem mounting (by removing /srv's disk). [02:45] I get dropped into a recovery shell... in which /bin is not in $PATH [02:45] Sorry, /usr/bin is not in $PATH [02:45] /usr/bin/ls? [02:45] Yes, obviously [02:45] Damm: (http://www.debian.org/doc/manuals/project-history/ may help to keep track of the codenames) [02:45] But the question is: how did it get to that point? [02:46] PATH is /sbin:/bin [02:46] there are some code paths where you only get /bin and /sbin I think [02:46] I usually just stick /usr/sbin and /usr/bin in there and don't worry about it [02:47] I vaguely suspect that Ubuntu has been "clever" when it juggled stuff like /etc/environment and /etc/profile around, or just that this hasn't started a login shell for some silly reason [02:47] the unified PATH in /etc/environment is only effective for PAM sessions [02:47] anything else gets to fend for itself so you may notice some desynchronisation [02:47] * Damm laughs [02:48] well that's a horrible way of making /etc/profile cleaner then [02:48] cjwatson: this is clearly a bug, because if I manually run "bash -l", it also lacks PATH [02:48] *lacks a proper PATH [02:48] twb: it'll just inherit it from the parent [02:48] oh, -l? not sure. [02:48] cjwatson: I expect the system-wide PATH default to be set in /etc/profile or some other similar place [02:48] sulogin should probably sort it out. [02:48] the problem with /etc/profile was that we got a different kind of desynchronisation [02:49] /etc/profile only applies to Bourne-style shells; there are a number of processes that start up in other ways and need a PATH [02:49] cjwatson: yeah, you're right, this problem is caused by PATH setting being moved from /etc/profile to /etc/environment in Ubuntu [02:49] cjwatson: that's a fair point [02:50] and this actually was a practical problem - see http://wiki.ubuntu.com/OneTruePath [02:50] Thanks, looking [02:50] cjwatson: btw, has this issue propagated into the Debian BTS? [02:51] twb: I don't think so [02:51] sulogin does NOT fix it, BTW [02:52] when I said "should", I meant "ought to be made to" rather than "I believe it does so now" [02:52] stupid English language [02:52] Sorry, I misunderstood. [02:53] actually, it's curious that this happens at all, because upstart sets PATH [02:54] and apparently has done since August 2006 [02:54] I wish upstart hadn't been deployed until metainit was actually ready. [02:55] I just don't see the benefit of upstart when it has to then go through a sequential sysvinit-compat layer [02:55] metainit couldn't have been made ready without experience with the initial deployment of upstart [02:55] and the deployment was a lot easier this way [02:55] Scott does actually know what he's doing [02:55] (I'm not sure whether we'll actually use metainit, but.) [02:56] I really meant "deployed as the default" [02:56] It had to be the default to get tested. [02:56] Double-edged sword, really. [02:56] On an unrelated note: upstart's shutdown(8) has no way to force a fsck on reboot... [02:56] infinity: yeah, I guess... [02:56] We needed the intermediate sysvinit-compat rollout to be able to use it at all. [02:57] twb: These days, isn't "forcing fsck on boot" more a function of marking the filesystem dirty, since "running fsck" does pretty much nothing on a journalled FS that claims to be clean. [02:58] Even fsck -f? [02:58] It takes a long time to do nothing... [02:58] all that shutdown -F ever did was to touch /forcefsck before shutting down. [02:58] I figured it was something like that [02:59] And, FWIW, checkroot.sh still honors that file. [03:00] infinity: ah, cool [03:00] I just didn't remember what the exact name for the file was, or I'd have just touched it [03:00] As it is I'm plugging the root filesystem into my laptop to fsck -f -- it's a USB key :P [03:00] -f touched /fastboot, and -F touched /forcefsck [03:00] Both are referenced in /etc/init.d/checkfs.sh [03:00] would be worth a bug to add those options, since they're trivial [03:01] Though it might be nice for you to file an upstart bug to get those options back in. *shrug* [03:02] Is there actually a way to defragment ext2? [03:03] Some people have written some tools. Nothing that I've ever trusted, mind you. [03:04] ext* are fairly fragmentation-resistent anyway, though if you regularly redline between 90% and 100% usage, you'll be as screwed as the next FAT32 user... [03:04] You could do one that just called sleep(size-of-disk) :-) [03:07] amazing ubuntu did something right with libtool [03:07] yay [03:09] cjwatson: regarding bash; I suggest making the default /etc/profile say something like "if path is , source /etc/environment" [03:10] Since AFAICT that would be exactly a no-op for login shells that *do* go through pam_environment.so [03:10] twb: /etc/environment is not guaranteed to be sourceable by a shell [03:10] twb: it may happen to be in many cases, but the format is not the same as shell [03:11] so it gets a bit tricky [03:11] I'd rather fix the (few) places that are still showing up with broken paths [03:11] cjwatson: it's not just lines of X=Y? [03:11] Yes, obviously actually fixing things is preferable :-) [03:11] the quoting rules are not the same as shell. [03:11] cjwatson: ah, fair enough [03:12] httpd: pthread_mutex_lock.c:87: __pthread_mutex_lock: Assertion `mutex->__data.__owner == 0' failed. [03:12] well damn it's still there [03:12] i bet this is something in apache === UdontKnow is now known as \0 === \0 is now known as UdontKnow [03:58] hi everyone [04:01] i have some questions on how i should setup a new server, for a apache mysql server [04:15] !ask [04:15] Please don't ask to ask a question, simply ask the question (all on ONE line, so others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) [05:21] I have a question about setting up LAMP if anyone can help. [05:45] Good morning [05:45] Anyone available for a networking question on Ubuntu Desktop 8.04.2 ? [05:45] I am looking for some assistance on obtaining a DHCP IP address on Ubuntu 8.04.2 over a Linksys wireless bridge 802.11g, model number WET54G ver 3.1 [05:46] I am able to assign a static IP address and DNS server information and connect just fine [05:46] WEP is enabled on both the Verizon FIOS wireless router and Linksys wireless bridge [05:47] I've tried #ubuntu and they suggested I try here [05:50] anthony1: all I can think of to try is packet sniffing [05:50] I don't understand how that would help [05:51] anthony1: it will tell you if the DHCPREQUEST is hitting the router [05:51] I'm assuming you actually have a shell on your DHCP server / router [05:52] If so, I am unable to access via telnet or SSH [05:52] There's a web interface but that's all I have access to [05:52] I'd also obviously try turning off networkmanager, which always breaks everything. [05:52] lol [05:52] And I'd try running dhclient manually and watching its output [05:59] twb, if you're still there, I got randomly disconnected [05:59] You can see that I'm here from /names [05:59] Thanks [05:59] could anyone tell me how to add a trusted SSL certificate? I'm using dreamhost for email and their ssl cert doesn't seem to work (I added it to /etc/ssl/certs but I still get an error that the certificate is not trusted) [06:00] twb: did you see the flood of output from dhclient? [06:00] anthony1: no. [06:00] anthony1: you should not be flooding the channel anyway [06:01] anthony1: use a pastebin [06:01] Long story short ....... No working leases in persistent database - sleeping. [06:01] That means it gave up [06:01] My other suggestions remain [06:01] Say it again? [06:02] anthony1: make sure network manager isn't running or installed. ssh into your DHCP server / router and packet sniff there. Packet sniff locally. [06:03] Oh, and look for suspicious output near the bottom of dmesg [06:03] Thanks [06:03] Ok === Zaraphrax is now known as Zaraphrax|Laptop === Zaraphrax is now known as Zaraphrax[DESKTO === Zaraphrax[DESKTO is now known as Zaraphrax|Deskto === Zaraphrax|Deskto is now known as Zaraphrax [07:41] New bug: #356818 in bacula (universe) "package bacula-server 2.2.8-5ubuntu7.2 failed to install/upgrade: dependency problems - leaving unconfigured" [Undecided,New] https://launchpad.net/bugs/356818 [07:42] New bug: #352391 in dhcp3 (main) "AppArmor prevents DHCP from getting address" [Undecided,Incomplete] https://launchpad.net/bugs/352391 [08:02] hello, i'm trying to do something very easy: to configure a machine with ubuntu server as firewall that does masquerading. I configured shorewall and i can do pings to the net from a workstation but when i try to access some web there seems to be no data transfer... i'm i missing something evident? [08:29] Can anyone tell me how to manually edit /templates/sources.list.tmpl to point add a seperate repo into sources.list during vmbuildr runtime? I'm trying to --addpkg a package which isn't in --mirror [08:29] or any other workaround I can try to get the same end result? [08:41] New bug: #355091 in mysql-dfsg-5.0 (main) "package mysql-server-5.0 5.0.67-0ubuntu6 failed to install/upgrade: /etc/mysql/conf.d/old_passwords.cnf: No such file or directory" [Undecided,Incomplete] https://launchpad.net/bugs/355091 [08:46] New bug: #356851 in samba (main) "Sharing cups disables entire system with segmentation faults" [Undecided,New] https://launchpad.net/bugs/356851 [08:58] Anyone know how to install jre with vmbuilder? [08:59] oh_noes: icedtea or sun? [09:00] sun-java6-jre [09:01] Basically, if I try to install it vmbuilder fails because of the deb package prompting for user to accept the license agreement [09:04] oh_noes: provide the answer to that prompt in a preseed file [09:05] debconf-set-selections <<<'sun-java6-jre shared/accepted-sun-dlj-v1-1 boolean true' [09:05] debconf-set-selections <<<'sun-java6-jre shared/present-sun-dlj-v1-1 note' [09:11] Thanks. The problem is I dont know how to provide that answer with vmbuilder [09:24] oh_noes: me either; sorry. === Zaraphrax is now known as Zaraphrax_dsktp === Zaraphrax_ is now known as Zaraphrax === andol is now known as andol_ === andol_ is now known as andol [10:17] Hi all, I have a bunch of java processes that use all the physical ram in my server and then hang it without swapping or terminating. By hang I mean most processes half work but can't do much because there is not enough memory. I can't even run ps because it tells me the process can't fork. I would assume from what I've read about kernel memory management that the processes should just get killed but this does not seem to be the case. My swappiness [10:18] is set to 60. free -m shows the free swap to be 1913 and used to be 0. Any one seen something like this before? Any ideas? === mdz_ is now known as mdz [10:46] CrummyGummy: killall java takes care java programs [10:53] wow, thats amazing. [10:55] "fix the bugs" would've been just as useful. All I'm worried about is my servers not going down when the memory management in the kernel should keep it up. [11:05] CrummyGummy: well. no idea why it dosent swap, bu killing stuff helps to solve problem [11:06] CrummyGummy: like see dmesg and seei f there has been anyproblems [11:07] CrummyGummy: what kernel you use, i had problem with 2.6.27-7-server that it would oops when using dmraid0 swap drives [11:10] 2.6.24-23-server but I've had this problem for years, I do use raid on my swaps. Id your problem go away with an upgrade? [11:10] erm s/Id/did/ [12:01] how do i install glib-2.0 ? [12:20] quizme: sudo apt-get install libglib2.0-dev (assuming you're trying to compile something and it's listed glib-2.0 as a missing requirement) [12:25] I have ata_aux 100% CPU usage after /etc/init.d/apache2 stop; pkill -9 apache2 what should I do now? I even can not restart server because reboot command not doing its job... [12:25] cjwatson yes exactly. thanks! [12:26] cjwatson: Couldn't find package libglib2.0-0-dev [12:27] Will reboot -f help? I have only remote terminal available... [12:37] quizme: read what I said very carefully. You appear to have mistyped. [12:37] cjwatson ok thanks [12:53] quizme: do you mean /etc/init.d/apache2 start dosent work ? [12:55] quizme: not 100% sure about ubuntu, but you could try zap or force-restart argument with init.d script or remove apache.running stamp from /var/cache/ ??.. not sure where ubuntu keeps track if service is runnign [13:51] http://pastebin.com/m37cc52feNo I have soft lockup problem ata_aux process usage is 100%, I can login into ssh but I can not even restart server. Thank You for answers! [14:13] !seen ivoks [14:13] I have no seen command === adam7_ is now known as adam7 [14:35] hello [14:36] my ethernets configured incorrectly it missed out eth1 and skupped numbering the interfaces directly eth2. How can I rename it to the right order? [14:36] *skipped [14:40] hi here, anyone has a very cheap drive bay (16TB) to suggest in order to do drive to drive to tape backups ? [14:46] fevel: Have a look at /etc/udev/rules.d/70-persistent-net.rules [14:48] youre the man maxb [14:49] maxb: how can I reset this file gracefully? [14:50] I think you can just edit it [14:50] ok [14:51] You can delete entries from it if you want it to regenerate them at next boot === paul_ is now known as orudie === Zaraphrax is now known as Zaraphrax[Away] [15:04] I got a e-mail from the Ubuntu server mailing list saying there is a meeting in #ubuntu-meeting at 15:00UTC. I am not sure what time that would be in US eastern Time. I spent 30min trying to figure out when it is [15:06] jamesrfla: 11:00am [15:07] thanks sommer it looks like there is some other meeting going on in there right now. Be back at 11AM [15:22] hi - im new to linux servers and I was wondering if there is something akin to the windows task manager in ubuntu, where I can get some idea of what the server is doing, and what kind of load it is under? [15:22] Doble: "top" [15:22] (in a terminal window) [15:23] ball: ahh cool, cheers! [15:24] Doble: you're welcome. That's common on most unix/Linux/BSD systems [15:26] ball: great, reading the man page now [15:27] man pages are a good thing. [15:30] Doble: you might also be interested in installing "htop" [15:30] it's more featureful than top [15:30] * ball wasn't aware of htop [15:30] I would die without it [15:30] http://htop.sourceforge.net/ [15:31] giovani: thanks i will check it out, I eventually plan to try out SNMP monitoring with nagios, but im still learning and just wanted to see what my system was up to - htop looks cool too! [15:31] Doble: nagios monitoring is hardly a replacement for a process viewer [15:32] but, it's definitely something you should incorporate also [15:33] giovani: yes indeed, cheers [15:34] well after a couple of days of resisting taking the 'easy' route with webmin and such, I now have a little home server doing DNS, file serving, and a public FTP! [15:35] what's the public ftp for? [15:35] Why public FTP? [15:36] when I say public, I mean its accessible from the internet, you still need a user/pass to access it :) [15:36] why ftp at all? [15:36] I wanted to transfer some files from work [15:36] ah, no, not again... :) [15:36] use scp/sftp [15:36] or webdav [15:36]