=== NG_ is now known as ng_ [01:14] ciao === ng_ is now known as NG_ [01:36] New bug: #782842 in bacula (main) "package bacula-director-pgsql (not installed) failed to install/upgrade: el subproceso instalado el script post-installation devolvió el código de salida de error 1" [Undecided,New] https://launchpad.net/bugs/782842 [04:43] Hi I just a did a new installation of 10.04.2 After install I did a aptitude update ; aptitude safe-upgrade [04:44] after rebooting, it still says "40 packages can be updated" "21 are security fixes" when I login [04:45] I've never seen the login dialog do that before on 10.04 server machine. [04:45] any ideas? [04:46] it's a big with motd uppdates [04:46] seems to me canonical has less than sufficient resources these days [04:47] RoyK: Hi! yes, seems like a bug to me too. [04:47] I have 20+ servers and all of them have that thing [04:47] it seems there is another bug with motd and the CPU/BIOS NXflag bit being disabled [04:48] I enabled the NX bit in the bios, and motd still tells me, "hey enable that in the bios" [04:49] it's rather alerming - a cron bug which affects a large number of users hasn't been fixe [04:49] and now when I run the suggested check that motd tells me to do, it says all is well! [04:50] RoyK: tell me about the cron bug. I already reported the root crontab bug in OI! [04:53] n2deep: it's a startup issue - cron doesn't start with a lot of setups [04:56] I see [05:03] I'm looking at /etc/cron.daily/apt on a 10.04 desktop machine that does not have this motd issue [05:03] though I'm not sure I'm looking in the right place [05:36] New bug: #782890 in backuppc (main) "no ping response on all hosts after upgrade" [Undecided,New] https://launchpad.net/bugs/782890 [08:49] Wooohooooo! [10:58] list === NG_ is now known as ng_ [11:54] Hello together...is S.o. here who can help me with mdadm? [11:56] !ask | pilleslife [11:56] pilleslife: Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) [12:00] Ok...I've got a soft-raid5. 2 out of my 6 drives degraded, but I think 5 devices are ok. Then I tried to assemble the raid and I got this messages (http://forum.ubuntuusers.de/topic/soft-raid-5-degraded/). After that I tried to create a new array over the old, but it doesn't work. If I set the disk-size mdadm says that the drive ist smaller than de given size. (I hope you understood it. If not I will try to explain it again.) [13:43] I do not beleive this is kubuntu specific [13:43] wrong channel :) [13:43] I have just installed kubuntu 11.04, when I start up, it says: error: no video mode activated. So nothing happens. I beleive it is already in grub it fails. [13:52] ok i have server a and server b in my LAN and i have 1 WAN ip adress. Server A has virtualhosts setup for webmail and a website. I wnat to have another subdomain to point from server a to server b with another web application. I that possible ? and if so how? [14:09] <_ruben> Henriquez: mod_proxy [14:30] ok thank you _ruben\ [14:30] so as far as i can see it is ProxyPassReverse === koolhead171 is now known as koolhead17 [14:46] hello guys [14:49] * RoyK wonders why the fuck the cron/upstart bug hasn't addressed yet - bug #592114 [14:49] Launchpad bug 592114 in upstart "upstart doesn't start cron automatically on boot in lucid for server on amd64" [High,Confirmed] https://launchpad.net/bugs/592114 [14:51] * IdleOne wonders why the only time I see cursing in here it is from RoyK [14:51] IdleOne: I generally don't curse too much, but this bug is rather bad, but still seems to be ignored [14:52] I had pretty much exactly that problem with squid after I upgraded to lucid [14:52] a couple reboots and manual start/stops fixed it === chuck_ is now known as zul [14:57] I'm developing a distributed system that involves multiple independent machines. I'm developing/testing with virtual machines using VMWare Fusion on a Mac. Now I'm faced by the decision how to manage the machine images. Up to now I simply duplicated a cusotm vm image I set up. However, such an Image is 9GB in size (only the bare os from the ubuntu server install image). What would you recommend in order to reduce the size of the a [14:57] (e.g. by stripping out all unnecessary packages, asf.) And is there a way to produce system "diffs" ... so that once the base vm image changed, I can simply move that "diff" to another base vm image? [15:00] 9GB sounds rather a lot for a base install - it should be 10% of that or so [15:00] yeah [15:00] perhaps a bit more [15:00] I've installed ubuntu server inside 2GB before [15:00] with not much extra room, but still [15:01] * RoyK runs Debian on a guruplug with 256MB flash and still has 100MB or so free [15:02] ubuntu base includes a bunch of things you don't really need that much - in the install page, you can choose to install the minimised version [15:02] if you're trying to shave megabytes off a configured image, you can remove logs and /var/cache/apt/archives/*.deb [15:02] or the JEOS version if you like [15:03] but if you're at 9GB, something is wrong [15:03] apt-get clean [15:03] 9GB is large for even a basic workstation setup [15:04] a JEOS install should fit into 100-200MB [15:04] and a fresh off the disc, regular server install should be under 900MB === koolhead17 is now known as koolhead17|afk [15:04] peta: check the logs - du -sh /var/log [15:06] Thought that too ... 9GB is pretty large. I just created a minimal vm in Fusion and did a default install from ubuntu server cd image. [15:07] peta: is the image 9GB? whatdoes df say? === ng_ is now known as NG_ [15:09] RoyK: http://pastebin.com/NSA8CVwz [15:10] so 3 gigs in use... [15:10] RoyK: strange ... [15:10] that's still a lot [15:11] did you update or install any extra packages? [15:11] peta: try apt-get clean [15:11] anyway - last I checked, vmware images grow when needed, but never shrink [15:12] aah ... maybe its a Fusion-specific behaviour that the virtual hdds has a given size at minimum [15:13] I'll try apt-get clean first ... [15:14] great ... now I have a usage of ~900MB [15:14] peta: most hypervisors allow for either pre-allocated images or growable ones [15:14] that's more like it [15:15] RoyK: just checked the Fusion settings ... the virtual hdds are growable and chunked at 2GB parts [15:15] peta: if you install JEOS (F6 and something in the initial grub menu IIRC), it'll be even lower [15:17] RoyK: "Just Enough Operating System" .. sounds useful. WIkipedia says that its included in the vanilla server version since 8.10 ... must I select the JEOS option in the installer? [15:18] make that F4 - "Install a minimal system" [15:18] * RoyK tests to see how large that one gets :) [15:18] whats the device name for /dev/sda in natty for amazon ec2/ebs? [15:18] * RoyK has no friggin' idea [15:19] I would guess /dev/vda, but I may be wrong [15:23] peta: just tested 'minimal install' on a vbox vm here - 630MB [15:23] vbox image size is 832MB [15:24] that's with lucid [15:24] okay. stupid question: how did you chose "minimal install" ... I don't have this option in my installer menu (10.04, 64bit image) [15:25] press F4 [15:26] i did ... only the choice "Normal" [15:27] I'll work that out by myself ... thanks for your help [15:28] it really should show up.... [15:29] http://karlsbakk.net/ubuntu-install1.png [15:33] wtf!? when language is set to german, only the option "Normal" appears :D [15:33] now I chose english and it works [15:33] lol - file a bug :) [15:34] and btw, choose minimal _virtual_ system [15:34] * RoyK loves his mac - after the upgrade to an SSD instead of spinning rust, it's blazingly fast :D [15:35] RoyK: I did [15:36] ubuntu base install in 3 minutes :D [15:36] minimum virtual (+ssh) is now down to 515MB [15:37] Which SSD do you use? [15:37] it's a Crucial C300 [15:37] make that Micron - they bought them recently [15:38] not top-of-the-line, but still very much better than spinning rust ;) [15:38] nice. performance values look good [15:39] I'd like to buy one too, however, i need at least 500gb (using macbook pro) ... and its definitely too expensive [15:40] indeed - I had to clean up some old stuff before going down to 256GB [15:40] I saw a classic hdd that uses 4GB of integrated SSD cache ... [15:40] it's well worth it, though [15:40] just get a home server or something for the large stuff :) [15:41] we have some QNAP boxes at work for that - small linux-based things [15:41] stuff some 2TB drives in them and you have room enough for some time :) [15:42] my plan's to move the hdd into the optical bay, and ditch the dvdrom. have a fast drive for things that benefit from speed, and a slow drive for things that require bulk [15:42] I know people that have done that - should work rather well [15:42] * RoyK still wants the optical for watching^Wripping movies etc [15:43] I used to think that, until I broke the dvd on my last laptop. never bothered replacing it for 3 years [15:44] my old MBP, 1st gen, eventually lost its optical - checked with the apple vendor and they told me a new one would cost NOK 5k from apple! [15:44] sure, small fileservers at home/office are great for loading off data. some nas devices even offer media streaming capabilites ootb. [15:44] got one from ebay for 6% that price [15:45] peta: this qnap was rather nice - no limits - ssh directly into it and I was root :) [15:46] yeah ... there are many ecosystems with 3rd party plugins for qnap atl. [15:46] seems it's using raid1+0 for the system partitions (on the data drives) and whatever you choose for the data [15:46] however, mature solutions (with redundancy and failure checks) are still a pricy affair [15:47] for power redundancy etc, you'll need a bit more [15:47] * RoyK uses ZFS for big storage [15:47] openindiana with a truckload of drives [15:47] we have a couple of 80-drive boxes at work - 100TB net storage each [15:49] wanted to buy me a thecus n4200 for some time now ... but as soon as i had the money anything else happened ^^ [15:49] RoyK: http://www.thecus.com/product.php?PROD_ID=41 this thecus even has an mini ups included [15:49] and usb ports where one can plugin a usb stick and get them backed up automatically .. [15:50] not badly priced either :) [15:50] and no hdds included yet :D [15:51] but with 2TB drives below the $100 mark, that shouldn't be much of a problem ;) [15:54] hmm ... let's see. my next invest is definitely a larger hdd for my mbp (~15GB free space right now). maybe with internal ssd cache as said above. [15:57] peta: better use external storage for the big stuff [15:57] btw, anyone seen the pandaboard? http://www.pandaboard.org/ [15:57] looks rather amazing :) [15:59] indeed. but check out "Raspberry Pi" a minimal platform for 18 bucks http://www.raspberrypi.org/ [16:01] wtf can I get that? [16:04] dunno. the platform was released to the public this week. they should be broadly available soon. [16:11] RoyK: haaa, my new minimal 10.04-64bit installations has 576MB :D [16:17] :) [16:17] why 64bit if it's minimal? [16:17] not that it matters much... [16:24] well, the later production machines will use 64bit os too and in order to replicate the production env in my vm for testing/developing it chose this [16:24] k [16:26] Hi Guys, I installed ubuntu server 10.04 LTS 64bit and it completed successfully however when i try to boot i can't get passed the "Starting Apparmor profiles [OK]" the system just seems to hang (i do have encryption on my volumes enabled). [16:35] Proz01d: The next service that should be initialized is cpufreq (cpu throttling, energy saving, asf.). So maybe its a problem related to that. [16:37] At least in all 10.04 32/64bit versions i have installed. I know that because inside my vm supervisor cpufreq isn't supported and thus this service always fails on sys init [16:38] ok i am struggeling with mod_proxy from the apache package in ubuntu 10.04 server and have looked all over to find an answer. Where do i put the ProxyPass and ProxypassReverse? in the virtualhost file? in the apache2.conf file? [16:38] do i need to change something else like hostnames to point to the server that the request is forwarded too? [16:39] and do i need to create another virtualhost file on the server acceptin the request? [16:39] peta: is there a way to bypass it? How do i get around this? [16:40] Henriquez: at the beginning of every directive sections, it is told where this directive may appear (vhost conf, server config): http://httpd.apache.org/docs/2.2/mod/mod_proxy.html [16:41] Proz01d: do you run your installation as vm or on a real box? [16:41] vm [16:41] which one? [16:41] thank you i will examin the document more closely then (already found that one) [16:42] Henriquez: np [16:42] peta: may i pm? [16:42] sure === Henriquez is now known as henriquez === henriquez is now known as Henriquez === medberry is now known as med_out === NG_ is now known as ng_ [18:26] ubuntu server 64bit with encrypted partitions... should i still be able to run the CLI only? [18:26] (with only 128mb of ram [18:26] ) [18:27] 128ram? i have more on my phone ;/ [18:27] /msg frostschutz hi, jemand da? [18:28] yes... i'm running a vps. trying to keep cost down [18:33] Proz01d: I don't see why not. [18:40] Proz01d: still the same problem? [18:45] Proz01d: I just tested a minimal VM here - boots ok with 32MB RAM [18:45] no encryption, though [18:47] RoyK: it's weird, because he uses a vanilla default-vm provided by his hosting company. the system starts up, AppArmore is started (status message shown) but after that it freezes. [18:47] I said that normally directly after AppArmor cpufreq is initialized ... [19:11] peta: what hypervisor? [19:11] xen [19:12] no idea - sorry [19:12] * RoyK doesn't use xen [19:12] Where is the best place to get support for serial passthrough for the libvirt-bin/qemu-kvm versions on server 10.04 LTS? I have been unable to get my guest's fax software working with USB or serial passthrough. === ng_ is now known as NG_ [19:26] Superchicken: from virt-manager, it's rather trivial [19:26] Superchicken: just add physical host device, choose the serial port and you should be done [19:27] * RoyK likes easy configs [19:28] Superchicken: from there you can choose a PCI ID - you'll find that in lshw [19:43] Just an updated.... i had my hosting provider assign another 128mb and it started up correctly. [19:43] (total of 256mb) [19:46] prolly the encryption part needing some ram [19:46] not all can be swapped out... [19:47] I can connect it, but I've tried: USB passthrough with another USB fax modem, USB passthrough with a Keyspan, direct serial passthrough for the serial port on the host, and serial passthrough via the tcp option with ser2net running on the host for my USB modem, my serial modem, and with a Keyspan. I have differing results with differing methods and the TCP is most promising with what appears to be a perfect fax document (versus di [19:47] rect serial passthrough which causes artifacts and and fax job errors), unfortunately even with ser2net, something goes wrong and my fax program on the host reports an error from the modem. I believe it's a timeout issue, which seems to indicate that nothing is making it back to my host from the modem after the data part of the transmission. [19:47] * RoyK is currently programming a 16MHz processor with 4kB RAM and 16kB flash and is having good fun with it [19:47] i heard all vps providers have access to the HDs and can look at the files without your knowledge so to protect the data on there i encrypted it.. [19:48] theoretically, they can, of course [19:48] but not with encryption right? [19:48] but then, as long as the VPS is running, they can always barge in and read the data from the mounted filesystem [19:48] so then this is pointless to have encryption [19:49] depends - if you're paranoid, add encryption for certain parts and mount them when needed [19:50] if a volume stays mounted, you can always get in, given console access [19:50] that is - not really... [19:50] you'll need passwords etc [19:51] Proz01d: If you don't trust them, move to something else. ;-) [19:51] but if you're automounting the volume, you need to hardcode the key, and that's like hanging our doorkey at your door [19:52] jpds: indeed - or get your own box :P [19:52] * RoyK has a private server [19:52] hello [19:52] or two... [19:52] i looking for tool to make statistic what country my serwer have most connections [19:53] web server? [19:54] no [19:54] shell [19:54] perhaps you can hack awstats to do it [19:54] it works well with other services [19:55] no, i havent logs to network [19:55] maybe vnstat? [19:57] wmp: That only does data traffic, not countries. [19:57] true, in this database isn't ip [19:58] maybe nestat? [20:01] maybe iptables? :) [20:05] iptables??? [20:05] the lastlog should be sufficient for shell access [20:06] as in /var/log/lastlog [20:06] RoyK: how to read lastlog? [20:07] RoyK: hmmm, ubuntu bug: May 15 06:38:20 cardinal rsyslogd: [origin software="rsyslogd" swVersion="4.2.0" x-pid="2735" x-info="http://www.rsyslog.com"] rsyslogd was HUPed, type 'lightweight'. [20:07] RoyK: no, lastlog isn't good idea [20:08] or /var/log/auth.log [20:08] that should show the IPs etc [20:08] i have torrents on server, and i want to check this transfer ;) [20:08] then you'll need some support in the app [20:08] transmission or whatever you're using [20:09] but i have http, ftp... I need this on network level ;) [20:10] http://software.inl.fr/trac/wiki/EdenWall/NuLog [20:10] well, I don't know [20:10] I'm sure there is a way to log all IPs with netfilter [20:11] JanC: netfilter generate big load? [20:11] i have 200mbps IO [20:11] iptables -i INPUT --ctstate NEW -j LOG [20:11] perhaps [20:12] I missed the initial question, but argus generates a summary of all network connections, ra is the client to view the info it creates... is that what you are interested in? [20:12] netfilter is the part of the kernel that you use to set up a firewall with iptables ;) [20:13] it's basically a framework for inspecting IP packets & acting on their properties [20:13] ok, i must read about it [20:14] I'm pretty sure iptables can create logs that you can parse to generate the info you need [20:15] @royk and @jpds i'm just a little paranoid..... I have no reason not to trust them [20:15] but there might also be something that causes less load... [20:15] except i don't know them [20:16] what sort of data is it there? planned terrorist atttack? :) [20:18] Anyone got experience running nagios on windows? [20:18] the nsclient++ [20:18] * RoyK uses nsclient++ quite a lot [20:18] I keep getting an access is denied error when I try to edit the NSC.ini file [20:18] I'm an administrator, I've taken ownership of the entire directory and sub objects [20:19] I shut down the systray icon, and stopped the service in admin tool [20:19] s [20:19] Is there something I'm missing? Do I have to reboot into safe more to edit the file? I'm over RDP incase that matters [20:19] try manually with setfacl or something [20:20] Its windows 7 too btw [20:20] I heard of cacls, not setfacl though [20:20] there are some tools for showing which files are open too - don't remember their names, though, but available from http://www.sysinternals.com/ [20:20] erm - sorry - cacls, yes [20:21] * RoyK doesn't use windoze too much these days [20:21] setfacl is linux/posix [20:22] ah, I see [20:24] yeah, I remember SysInternals had a good free tool to see open files in the past [20:24] that seems to have done it. I did it from cmd directly on the folder (I tried just program files it was giving an access denied error as well, but not the nsclient folder) [20:24] One thing I'm a bit uncertain about... [20:25] JanC: Process Explorer. There is also an open source equivalent called ProcessHacker on SourceForge now. [20:25] But this is #ubuntu-server ... :) === NG_ is now known as ng_ === ng_ is now known as NG_ [21:20] I just got it to properly monitor my services on windows. Thanks for your help. === NG_ is now known as ng_ === ng_ is now known as NG_ [22:06] New bug: #783193 in openssh (main) "package openssh-server 1:5.8p1-1ubuntu3 failed to install/upgrade: subprocess new pre-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/783193 [23:27] hi [23:27] i heard this was the kick ass chan for server help [23:29] im getting an error running virtual servers - after getting the servers to start and run using virsh i cannont connet to them via ssh or virsh console any help would be apreciated [23:30] virsh # console ubuntu [23:30] Connected to domain ubuntu [23:30] Escape character is ^] [23:30] error: internal error cannot find character device (null) [23:31] addict@TeraFora:~/virtual$ virsh 'list --all' [23:31] Id Name State [23:31] ---------------------------------- [23:31] 1 addicts running [23:31] 2 ubuntu running [23:31] any help would be awsome [23:31] ty [23:50] anyone have experience with online backup services à la spideroak or crashplan ? [23:54] Mean_Admin: I'm using spideroak currently. [23:55] greppy: thumbs up? [23:57] greppy: or rather, have you any reason to not recommend it ? I'm happy if it just backs the data up, not looking for fancy interface (as it'll run on a server, obvioiusly)