=== kentb is now known as kentb_out === Narwhal is now known as IdleOne [00:32] I'm trying to connect my windows xp machine directly to an ubuntu server with a crossover cable. I thought this was going to be easy. It doesn't appear to be. When I configure the server with a static IP with a netmask, I get "eth0: ERROR while getting interface flags: No such device ... SIOCSIFADDR: No such device" [00:33] I'm hoping the fix for that is something that's obvious to someone with more experience than I. [00:35] It also says "SIOCSIFNETMASK: No such device" and "Failed to bring up eth0" [00:41] New bug: #619540 in samba (main) "package smbclient 2:3.3.2-1ubuntu3.5 failed to install/upgrade:" [Undecided,New] https://launchpad.net/bugs/619540 === erichammond1 is now known as erichammond [01:04] well I figured it out... for some reason my two network cards are eth3 and eth4 [01:05] did you change cards or motherboards? [01:05] ubuntu remembers the id's of the cards, and assigns new numbers for new cards [01:06] the hard drive is from another machine... that could be it. [01:06] now however I think my problem is that my crossover cable isn't really a crossover cable [01:08] mm... I dunno... the definitely have different wiring between the two plugs [01:17] whew... I rebooted and eth0 is now eth3... [01:17] er... I mean eth3 is now eth0 [01:19] hai [01:19] hey, can someone help me out with PPTPD server? [01:19] i'm having a problem with it. :( [01:19] ND-movie: no one knows if they can help you because you haven't told us the problem [01:19] a fair point [01:20] basically i'm having trouble getting the darn'ed thing to work right, i have it all set up, but i'm getting errors when it tries to handle dhcp [01:22] here's pastebin [01:22] http://pastebin.com/31qdtXsp [01:23] like i can communicate with the VPN server, its not refusing the connection [01:23] but *shrug* [01:23] i think i have the internal and external IP's set up incorrectly [01:23] xxx.xx.31.59 i have set as my internal (which is the IP of my server) [01:23] and i have 192.168.1.1-254 set as my external [01:24] thats probably the wrong idea, i'm guessing? [01:31] any suggestions === rmk_ is now known as rmk [02:32] Hi all [02:33] I'm trying to setup Virtual Machine using JeOS & vmbuilder [02:34] The VM seems to have been installed, but I don't know how to start it :( === dendrobates is now known as dendro-afk [02:35] running a virsh -c qemu:///system list doesn't display anything so I'm not that sure the VM has been installed [02:36] any VM helper there ? [02:53] guest9876543210: (i'm not sticking around, but one suggestion) do virsh list --all [02:53] else you don't see not-started vms [02:53] español? [02:54] fundacion : nope, I'm French [02:55] I give --all a try ;-) thanks [02:55] no body speak in spanish [02:55] or where i can find a irc about server in spanish? [02:56] you seem to type a good english ! can't you ask for help this way ? [02:58] but, i don't understand the englis at perfection [02:59] I'm not aware of Spanish server room, I'd suggest to ask for help there ... [03:02] I'm having some issues with Raid. I have a raid6 array and all of my drives all of a sudden went to spares. mdstat reads this: http://pastebin.com/c8Li0hHj I've tried rebootin and reassembling the array, but don't want to mess anything up the best I can get is 5 devices to try to assemble, but I need at least 6. [03:10] Placebo : do you have critical datas on your RAID array ? [03:14] guest9876543210, I do, loosing it would be bad [03:15] n00b Q: I want to put my wife's teaching docs online for her to access from home/school/wherever... Most files are on ubuntu server box right now. Is WebDAV the way to go here? [03:15] I assembled the array with: mdadm --assemble --force /dev/md0 /dev/sd{b,c,d,e,f,g,h,i} [03:15] and it started with 6 devices [03:16] Placebo : and ?? what's wrong then ? [03:16] guest9876543210, there are 8 devices, and I can't add the other two [03:17] guest9876543210, also, now one of the 6 is marked as Failed [03:17] new mdstat http://pastebin.com/CcrgJQs4 [03:19] Plecebo : it might worth to test your HDs, using smartmontools [03:20] guest9876543210, I can do that, do they need to be tested one at a time? [03:21] also, should I take my raid array offline? [03:21] Plecebo : yes one disk at time (2mn per disk), you can leave your RAID on [03:22] sudo smartctl -t short /dev/sda [03:22] then after 2mn : [03:22] sudo smartctl -a /dev/sda [03:22] for each HD ... [03:23] guest9876543210, ok, i'll do that and report back, thanks for your help *fingers crossed* [03:25] guest9876543210, what am I looking for? how will I know there is a problem? [03:26] Plecebo : you'll see errors in the last lines [03:29] Placebo : you can run to know to number of hours your HDs have been up : [03:29] for sda : [03:29] sudo /usr/sbin/smartctl -d ata -a /dev/sda | egrep '^ 9' | tr -s ' ' | cut -d' ' -f11 [03:31] /dev/sdb looks clean http://pastebin.com/sCFPcESs [03:32] 8851, is that minutes? [03:32] in hours [03:32] oh hours? nice [03:32] lol [03:32] nearly 1 year [03:33] seems about right [03:33] thats about how long i've had them, nearly always on [03:34] due to RAID (1) crash, I now change HDs before 30 000 hours (3 years) [03:35] between 20 000 and 25 000 [03:36] Plecebo : it might worth to check your HDD temp too : [03:36] sudo hddtemp /dev/sd[a-l] [03:37] mine have often failed because of high temp [03:37] guest9876543210, I know I recently moved them to a hotter part of my house and I was copying data to them when the array crashed [03:37] /dev/sdc looks ok http://pastebin.com/vxpFKmTd [03:38] yeap [03:39] guest9876543210, between 34C and 41C [03:39] at the moment [03:39] but now I know how to monitor them [03:40] FYI : 47° has always killed my HDs [03:41] guest9876543210, well seeing as how its a lot cooler today then it was the day the crashed i wouldn't be surprised if they hit something that high [03:41] guest9876543210, thanks a bunch for your help [03:42] Plecebo : you're welcome .. help each other ;-) [03:42] guest9876543210, yea, i'm writing down these commands so i'll know next time and someday can pass on the info :) [03:44] Plecebo : I've installed a MediaWiKi on my server, so I can record all theses usefull commands and access them from anywhere with www connection, I use it everyday as I can't recall everything ;) [03:44] sdd looks good http://pastebin.com/90HPtcdj [03:45] guest9876543210, that is a great idea [03:45] guest9876543210, as long as your server is up eh? [03:45] Plecebo, do you mean uptime ? [03:46] guest9876543210, if your MediaWiki is on your server and your server is what you are trying to fix... could be an issues [03:46] opps, yeap, that's what I first type .. as long as your server is alive ;) [03:47] it was on this server I have experienced a RAID crash :( [03:48] Maybe I could keep tomboy notes and have them sync with ubuntuOne [03:48] lost 1 week / 15H/day typing invoices and so on ... [03:48] guest9876543210, ewww no fun there [03:48] Now backups are daily [03:48] guest9876543210, at least it wasn't two weeks LOL [03:48] (they were wekkly before) [03:50] sde looks ok http://pastebin.com/UkspLJaj [03:50] Plecebo : I haven't yet found how to use UbuntuOne (neither understood what it can do) [03:50] sdi sdh sdg might be the culpits [03:51] (as they're shown as _ instead of U in mdstat) [03:52] guest9876543210, I have only begun to use it, I think eventually they want it to be like a roaming profile in the cloud... but I think there are some kinks before they get there [03:52] guest9876543210, yea my suspects as well, testing f now [03:53] FYI : I don't use anything that RAID1 for my sensitive datas now : 2 HDs in use, 1 for spare [03:54] yea, I should consider moving mine around a bit [03:54] re-thinking [03:55] I'm using ecryptfs over Ubuntu One, bc I keep switching laptops etc. No problems so far, after 2-3 weeks of heavy daily use [03:55] course all it takes is one snafu on the backend :) [03:55] as sidekick users know [03:55] guest9876543210, most of my 6TB is used for ripped copies of movies I own that I could easily replace (just take me a few months to re-rip them) [03:55] hallyn, you are happy so far? [03:55] very [03:55] I install a new laptop, hook into ubuntu one, wait a few mins, and everything is there [03:56] (bouncing between lucid and maverick) [03:56] hallyn, i've used it for the music stuff, but there is so little space i usually pull it off right away [03:56] Plecebo : for the ripping thing datas, I use RAID5 (never tried the 6 level) as I don't care datas [03:56] again, i suspect i"ll get bitten at some point. but then i have backups on several laptops, implicitly :) [03:56] Plecebo: i want quick syncs, so i don't put music on there [03:56] not yet anyway [03:57] if I could do priorities, "sync this first", then I might start [03:57] maybe i should hop over to #ubuntuone and suggest that :) [03:57] SDG seems fine http://pastebin.com/QbUXawbV [03:57] hallyn, yea the distributed idea of it is nice [03:57] Plecebo : yeap no error shown yet [03:58] guest9876543210, its hard to move around 3TB of data though while you restructure everything [03:59] Plecebo : I know ... :( [03:59] hallyn, priority would be nice I havent heard but excited to see what the devs have in store for 10.10 [04:00] Is one of you good with Virtual Machines ? (JeOS & vmbuilder) [04:00] SDH looks fine http://pastebin.com/mZGZPQHr [04:01] guest9876543210, havent played around with it yet, but i'm interested. I'm considering attempting a small office install with ebox towards the end of sept and want to divide up the server a bit [04:01] guest9876543210, but like I said, not tried playing with it yet [04:02] Plecebo : I'm trying right now .. but it seems I have to read a lot ... not able to start a VM yet [04:03] SDI looks ok... http://pastebin.com/PGeKNUR8 [04:05] Plecebo : once you've tested all drives, try to remove then re-add sde1 to md0 [04:07] mdadm /dev/md0 --fail /dev/sde1 --remove /dev/sde1 [04:07] mdadm /dev/md0 --add /dev/sde1 [04:07] and watch the raid re-construction [04:08] watch cat /proc/mdstat [04:08] try to do this for the other drives after (but as i told : never tried the RAID6 level) [04:08] guest9876543210, it added e as a spare [04:09] Plecebo : maybe because of the --force you typed [04:09] guest9876543210, and it doesn't seem to be rebuilding either [04:10] guest9876543210, could be [04:10] from the last mdstat you sent, sde1 if marked as FAILED [04:10] so, try to remove & re-add [04:10] (without force) [04:11] guest9876543210, remove and readd the array? [04:11] guest9876543210, mdstat now doesn't have b and c either, so let me try to readd them [04:11] Plecebo : remove sde1 from array then re-add it (see previous commds I've sent) [04:12] guest9876543210, when I tried to remove/re-add sde1 using the commands you gave they were added back as spares [04:12] arg ... [04:13] guest9876543210, adding sdb and sdc with the same command added them as spares as well [04:13] hmm .. [04:15] it seems you need 4 drives for RAID6 [04:15] as the array already contains 4 valid drives, I think that's why they are added as spare [04:16] from what i understood raid6 is like raid5 with the ability to lose 2 drives [04:16] I'd try a mdadm --assemble --scan [04:16] so my 8 drive array should be able to assemble with 6 drives [04:17] it says my array is active and it can't be restarted [04:18] I suppose this array is only for datas (no boot nor OS on it) ? [04:19] guest9876543210, just data, correct [04:19] so you can stop the array then re-assemble it [04:20] sudo mdadm --stop /dev/mdo [04:20] then [04:20] sudo mdadm --assemble --scan [04:21] mdadm: /dev/md0 assembled from 5 drives and 3 spares - not enough to start the array. [04:21] LMAO [04:21] closer though [04:21] maybe if i alter the mdadm.conf and restart [04:22] what is your mdstat right now ? [04:27] guest9876543210, it says again that all of my drives are spares (S) [04:28] all the 8 ? [04:28] yea [04:29] md0 : inactive sdf1[0](S) sde1[10](S) sdb1[9](S) sdc1[8](S) sdd1[6](S) sdi1[3](S) sdh1[2](S) sdg1[1](S) [04:30] and if you try to assemble the array you're told not enough drives ? [04:31] mdadm --assemble --scan [04:31] mdadm: /dev/md0 assembled from 5 drives and 3 spares - not enough to start the array. [04:43] Plecebo : Sorry, I'm a bit confused, I really don't know what to do ... [04:44] guest9876543210, ok well thanks for the effort, ill keep plugging away here trying to see what I can come up with === oubiwann-away is now known as oubiwann [05:46] guest9876543210, oh no, i think i might have goofed [05:46] ? [05:48] I removed and recreated the array [05:48] and it started rebuilding [05:48] but it failed at the end ... that's it ? [05:48] but then I got freaked out that I wasn't able to mount the array, which i've been able to do in the past [05:49] and then two of the drives failed out of the array [05:49] and the sync stopped [05:49] but im worried that when I created the array I overwrote the superblock [05:50] arg, sincerely I'm not a that good mdadm user, but I think that overwriting the superblock was probably a bad idea [05:51] yea I'm thinking the same [05:52] I think it's time to think : go for a new mdadm aventure and forget your past array datas ... [05:56] guest9876543210, maybe, but i'm not at that point yet [05:57] courage is a good thing ! :) [06:00] I'm trying the 4th installation of a Virtual Machine on this server ... [06:01] well the info i'm seeing leads me to believe i'll be fine, if I can get the array to rebuild [06:01] I'm shutting it down for a bit and i'm going to try to cool things off in here then have it rebuild again [06:01] good luck with your vm project, i'll be back in a bit [06:47] hi all, anyone used to JeOS & vmbuilder ? [06:48] where does trasmission save torrent files? [06:52] alex88, ~/Downloads [06:54] lol..i already created that dir..so i didn't search in there :) thank you :) [07:55] guest9876543210, gah! it won't rebuild because two of my drives keep failing [08:07] Plecebo, your array should be ok with 6 drives out of 8 [08:07] just remove the bad ones and keep only the good ones [08:08] guest9876543210, but the array keeps wanting to sync [08:09] so it won't stay online [08:09] do you mean on its mountpoint ? [08:09] hey guys is it bad practice if you have a raid 1 array to have swap also mirrored? [08:10] eagles0513875|2 : it works but it is a better idea to have a separate swap on each drive [08:11] i have one main drive and one drive which is mirrored onto [08:12] eagles0513875|2: what guest9876543210 is trying to say is: you don't want to mirror your swap [08:13] O_o [08:13] doh [08:13] eagles0513875|2 : do you mean you have only a / and a swap on the drive and the whole drive is mirrored (using mdadm) to another one ? [08:13] guest9876543210: exactly i have the primary drive and using mdadm its mirroring / and swap :( [08:13] eagles0513875|2: you want 2 swap one on each disk [08:13] ok so it being mirrored is ok gentooxer [08:14] eagles0513875|2: it doesn't harm anything [08:14] gentooxer: it seems you haven't created your RAID array at system install ... [08:15] isnt it better to have swap mirrored just in case one needs to rebuild the array they will have the swap partition already there [08:15] sorry, this was for eagles0513875|2 .. [08:15] and guest9876543210 i did create the array at system install [08:15] eagles0513875|2: the swap is only temporary data like ram [08:15] eagles0513875|2: its garbish after a reboot [08:16] arent you up the creek though if you have to rebuild the array [08:16] eagles0513875|2: why bother an mirror it? [08:16] eagles0513875|2 : datas in swap file are erased after a reboot [08:16] ok then i guess ill re install again :( [08:16] so what would you guys mirror [08:16] the bios boot partition and / [08:17] eagles0513875|2: Ah I see, you have the option to mirror partitions not the whole disk [08:17] just partitions [08:17] i have 2 mirrors using mdadm [08:17] they are /dev/md0 and md1 [08:17] eagles0513875|2: so your RAID doesn't even see the swap, so you are save at rebuild ... [08:17] md1 is the swap [08:17] and md0 is the / partition [08:17] eagles0513875|2 : you should create your partition at install, mark them as RAID, then create the RAID array(s) [08:17] guest9876543210: i did [08:18] eagles0513875|2: just erase md1 and create 2 seperate swap [08:18] eagles0513875|2 : you just have to delete your md1 and create swap on each drive instead [08:18] humm ok [08:18] eagles0513875|2: but you don't need to reinstall [08:19] im new to creating raid arrays in linux [08:19] this was my very first one tbh [08:19] eagles0513875|2 : it doesn't hurts to have swap mirrored (my first server was like this and it has always run fine) [08:19] eagles0513875|2 : just not the best way to go ... [08:19] guest9876543210: humm k :( [08:20] eagles0513875|2: you can deactivate the swap with sudo swapoff [08:20] eagles0513875|2: unmount the swap [08:20] ok [08:20] eagles0513875|2: an create 2 new partitions [08:20] why 2 [08:20] with sudo swapon the system searches for your swap and activates it [08:21] eagles0513875|2 : one on each drive ;) [08:21] eagles0513875|2: because of the 2 disks [08:21] ahhh [08:21] primary disk has swap on it already [08:21] eagles0513875|2: or do you have RAID on a single disk? [08:21] 2x2tb hdds [08:21] raid 1 [08:22] thats kinda pointless though [08:22] to have it on one disk there is no redundancy [08:22] eagles0513875|2: so you create 1 swapspace on each disk [08:22] ok [08:22] eagles0513875|2: you can use the space of md1 [08:22] what does : cat /proc/mdstat answers yuou ? [08:23] hold on let me pastebin [08:24] guest9876543210: http://pastebin.com/HnYujTjc [08:25] gentooxer : are you used to JeOS and vmbuilder ? [08:26] eagles0513875|2 : sorry I have troubles to access pastebin.com :( [08:26] O_o [08:26] guest9876543210: what kinda trouble :( [08:26] paste there ? [08:27] strange [08:27] --- pastebin.com ping statistics --- [08:27] 9 packets transmitted, 0 received, 100% packet loss, time 7999ms [08:27] O_o [08:27] works fine for me [08:27] was working ok 30mn ago .. [08:27] guest9876543210: where are you located? [08:28] never seen the .nc before [08:28] South Pacific [08:28] New-Caledonia [08:28] kool kool === rmk_ is now known as rmk [08:28] guest9876543210: ill pm u and paste it to you that way [08:28] just paste here .. easier [08:29] ill spam out of the network [08:29] or at least the channel [08:29] !paste | eagles0513875|2 [08:29] eagles0513875|2, please see my private message [08:29] guest9876543210: does http://paste.ubuntu.com/479797/ work for you [08:30] eagles0513875|2 : yeap ok [08:31] Good morning! [08:31] eagles0513875|2 : just unmount the swap as previously explained, delete md1 and recreate separate partition on each drive [08:31] ok guest9876543210 :) will do [08:31] moin \sh ;) [08:31] moin binBASH [08:32] moin eagles0513875|2 [08:32] hi binBASH [08:32] hi guest9876543210 [08:32] <\sh> hey binBASH [08:32] luckily irc client has tab ;) [08:33] is one of you used to Virtual Machine : JeOS & vmbuilder ? [08:33] binBASH: konversation? [08:33] eagles0513875|2: I didn't read it yet, I just attached to screen. [08:33] nice [08:34] and the very first thing to do this morning is, having a look why MySQL replication lag on the server is at 16.000 [08:34] ouchie [08:34] got another question for u guys [08:35] im trying to install ubuntu server on another machine but for some reason when it loads language selection on the menu after booting off the cd it hangs [08:35] ok it's at 0 again. That was fast ;) [08:35] lol [08:35] man speaking of mysql i love mysql work bench [08:36] eagles0513875|2 : check RAM, CD integrity, CD player [08:36] guest9876543210: cd works fine [08:36] cd integrity [08:36] ya ill check the cd player and reseat the ram [08:36] and try again [08:36] eagles0513875|2: Well, I hate it to manually shard with Mysql [08:37] ? binBASH [08:37] eagles0513875|2: I have to change our company's application so it can use horizontal scaling (db sharding) [08:38] it practically means, changing of around 1500 source files........ [08:38] binBASH: have you tried to see if you could make ur life easier by using the mysql workbench gui [08:38] eagles0513875|2: Yup [08:38] binBASH: what i love about it is that you can reverse engineer and forward engineer the db [08:39] binBASH: humm let me ask in the workbench channel really quick [08:39] eagles0513875|2: We have very big tables here with a lot of entries ;) [08:39] MyISAM....... [08:39] wow [08:39] well im still fairly new to db's [08:39] studied them as part of my course this past yr [08:40] Tables with 19800000000 entries ;) [08:40] O_o [08:40] and that's only for one customer :D [08:40] nice [08:42] im slowly starting to setup my IT solutions business [08:42] eagles0513875|2: You know our company hosts services for image agencies like Getty Images for example. [08:42] nice bi Nafallo [08:42] whoops [08:42] wrong person meant binBASH [08:42] and we also write software to manage their images ;) [08:42] nice nice :) [08:42] im hoping to branch my biz in that direction as well software development [08:43] i have a few ideas i want to work on [08:46] eagles0513875|2: It's a good business I think. [08:46] ya [08:46] im a comp science major hehe [08:47] I never studied. [08:47] ;) [08:48] eagles0513875|2: I started coding 15 years ago [08:49] nice i started 3 yrs ago [08:49] ;) [08:49] in the middle of a java based project atm [08:49] I started coding in the age of 14 [08:49] VGA Graphics and Sound Stuff... [08:50] wow [08:50] i really wanna beef up my programming knowledge [08:50] i dont know much :( [08:50] and i really wanna start developing my own games [08:50] on iphone? [08:51] I didn't write an app for iphone yet. Though some other smart phones ;) [08:52] When I worked in Karlsruhe / Germany I had to code for mobile devices exclusively. But that is now 3 years ago. === rmk_ is now known as rmk [08:54] binBASH: on all platforms [08:54] im not that far from you binBASH [08:54] geographically lol [08:56] eagles0513875|2: Well I'm in Zurich / Switzerland ;) [08:56] binBASH: join me in hi_jack_This its a tech channel [08:57] I have to code now and fix software bugs :) [08:57] PHP.......... [09:00] nice binBASH i need to learn php and java script [09:03] learn python imho :D [09:04] but well, a good coder can use any language ;D [09:08] damn, since upgrading from 8.04 to 10.04 the serial console redirection seems not to work anymore [09:22] binBASH: back [09:40] damn, i fiddled with menu.lst and grub hangs now haha... will have to revert from recovery... sucks [09:47] g0rd0n: hope u took a backup of it [09:48] g0rd0n: no, but i remember what i changed... so as soon as i mount the boot partition i can modify the file [09:48] but still i wonder why console redirection doesnt work anymore [09:48] cause the parameters were not changed in the kernel line [09:48] same as the old kerne [09:48] g0rd0n: 8.04 used grub1 [09:48] if im not mistake [09:48] yes, i still have grub1 [09:48] it doesnt get upgraded [09:49] ya 10.04 uses on a clean install grub2 [09:49] no it doesnt [09:49] !grub2 | g0rd0n [09:49] g0rd0n: GRUB2 is the default Ubuntu boot manager since Ubuntu 9.10. For more information and troubleshooting for GRUB2 please refer to https://help.ubuntu.com/community/Grub2 [09:49] that will tell ya how to upgrade [09:49] i know, but i failed clean installing, cause my video redirection doesnt support framebuffer [09:49] i tried upgrading grub2, but it doesnt work... he doesnt like the boot partition [09:49] i followed the instructions on that page [09:53] i dont remember the exact error message, but i googled and it looks like grub2 wants the boot partition to start on block 63 which mine apparently does not [09:53] but thats ok, i can live with grub1 [09:58] nub question how does one change the motd [09:58] hm... http://www.google.is/search?hl=en&client=firefox-a&hs=2VC&pwst=1&rls=Palemoon:en-US:unofficial&nfpr=1&sa=X&ei=c_JpTP6_GoK88gaZgOmyBA&ved=0CBYQBSgA&q=0+degrees+celsius+%2B+0+degrees+celsius&spell=1 [09:58] i changed the file in /etc/motd but after logging in to the system again it did have the message i gave it [10:00] eagles0513875, the motd is now automatically generated with a series of scripts [10:01] qman__: how are you supposed to for instance put a legal disclaimer on the motd [10:01] ahhh wait [10:01] motd isnt what i want [10:01] located in /etc/update-motd.d/ [10:01] how can i put a legal disclaimer or have one pop up when someone tries to login to the machine and or ssh into the server [10:02] SSH has its own banner setting [10:02] I don't recall off the top of my head where the local terminal banner setting is [10:02] ok [10:02] not necessarily terminal [10:02] gets a login screen in kubuntu would it be posisble to get the disclaimer from a remote server prior to logging into that server? [10:04] if you want a message at a graphical login screen, you'll need to investigate kdm for kubuntu and gdm for ubuntu/xubuntu [10:05] as for remote logins, if you're using SSH, you want the 'banner' directive in the sshd_config [10:07] bear in mind, however, that many SSH clients ignore the banner anyway [10:07] so you will want your legal disclaimer in the motd as well [10:08] ya im noticing that [10:10] :) ssh banner set :) [10:13] qman__: would i need to create another script and put the motd message there [10:14] yes, or modify an existing one [10:14] the scripts are rc style, smaller numbers execute first [10:14] ok [10:15] :) this is gonna be interesting never worked with bash scripting much but im eager to learn [10:15] well, yours should be simple [10:15] start with a shebang, #!/bin/bash [10:15] and then echo "disclaimer" or cat a file containing it [10:16] then chmod +x [10:16] New bug: #619712 in keepalived (main) "keepalived vrrp race condition and fix (versions 1.1.17 and 1.2.0 but perhaps all?)" [Undecided,New] https://launchpad.net/bugs/619712 [10:17] qman__: O_o create a variable $disclaimer [10:17] im confuse dlol [10:17] i think lunch would fix that soon hehe [10:19] heh, it's 5 am where I'm at [10:20] bit of an insomniac [10:20] but you really shouldn't need any variables, just a simple echo or cat [10:20] qman__: your saying i make another script called disclaimer [10:20] then call it from the 00-header script [10:20] no [10:21] create a script called /etc/update-motd.d/92-legal [10:21] and it should contain [10:21] #!/bin/bash [10:21] echo "here is my disclaimer, by using this system blah blah blah give me your immortal soul blah blah blah" [10:22] if you want it to appear closer to the top, name it 01-legal [10:28] qman__: im trying to add it to the script 00-header [10:28] and its not working i did as u said [10:28] echo "disclaimer" then it outputs the kernel and ubuntu version [10:29] well, that should work [10:29] I forget what triggers it to update the actual motd though [10:30] I think it's a cron job [10:30] no, guess not [10:31] cuz im running cat ../motd and its not showing up [10:31] a relative path could be the problem [10:31] use absolute [10:32] its working [10:32] i tried to login directly to the server [10:32] and it shows up [10:32] instead of using ssh [10:32] so both ways are working [10:35] the other option here is to have it cat a file [10:35] ok [10:35] in that way, you could have both SSH and the motd point to the same file [10:35] so when you need to change it, it changes both [10:40] qman__: im using the motd message for ssh in the file /etc/issues.net [10:40] then when i login to the physical machine i added the echo u mentioned to the 00-header file [10:48] look in /etc/motd [10:48] look at your ssh config (not sshd_config) [10:48] check if it's using a login shell also [10:49] that's the common stuff [10:56] HOw DO I find my PoE switch IP address that ive just plugged into my server [11:05] <_ruben> garymc: read its manual? its bound to list the default ip address [11:06] <_ruben> or hope it uses dhcp by default so can just look at your dhcpd logs [11:06] yeah im a newbie. It says if you plug into a DHCP server (thats what I plug it into) then type "SHOW NETWORK" [11:06] show network doesnt work [11:07] where are the dhcpd logs stored? [11:11] <\sh> garymc: /var/log/syslog [11:11] <_ruben> /var/log/daemon.log [11:14] while we are on the topic of logs here where does iptables log information to if i have it set to log ? [11:28] _ruben : Thats not showing the Netgear switch [11:28] garymc: switches dont provide dhcp [11:28] you need a router [11:28] to do dhcp and NAT [11:29] probably atm you have only one ip which means without a router you cant get any other pcs on ur network on the internet [11:38] no but the Server does yes? [11:44] garymc: is dhcp setup on the server? [11:44] yes [11:44] its an LTSP server [11:44] Linux Terminal Server Project [11:44] ok [11:45] I want to swap my current switch to my new one with PoE [11:45] just swamp it [11:45] swap [11:45] you shouldnt need to config anything [11:45] well it says I do? [11:45] ? [11:45] maybe not then [11:45] here is the model number [11:45] switches arent that smart [11:45] so to speak [11:45] they arent like routers where u have a web interface they are just plug and play [11:45] Apparently it has a WEB GUI I need to configure [11:46] humm [11:46] what make [11:46] hold on [11:46] it could be you have a managed switch in that case it will have a web gui [11:46] Netgear FSM7326p [11:47] So I have it connected to the server via a Cat cable [11:47] give me sec [11:48] garymc: ya thats a managed switch i havent really worked with them or switches that much [11:48] ok [11:48] not sure if you have taken a look at the knowledge base [11:48] well anyway I could find out its IP? [11:48] http://kb.netgear.com/app/products/model/a_id/2408 [11:48] it should have a default ip that comes setup on it from the factory [11:48] that link has all the documentation [11:49] administration to commands etc [11:51] yeah i cant get into that either [11:52] well it says to check my DHCP list by typing SHOW NETWORK [11:52] but that command doesnt work [11:54] Anyone knows when a lucid-kernel with the ext4/snapshot bug will be released? Ref. https://bugs.launchpad.net/ubuntu/+source/linux/+bug/605551 and https://bugs.launchpad.net/ubuntu/+source/linux/+bug/595489 [11:54] Launchpad bug 605551 in linux "lvm2 hangs when creating snapshot, requires reboot to resolve (dup-of: 595489)" [Undecided,New] [11:54] Launchpad bug 595489 in linux "lvm snapshot causes deadlock in 2.6.35" [High,Fix committed] [11:56] humm garymc a call to their support might help more then i [11:56] Ah, that's what's killing my servers! I knew there was something going wrong with the snapshots, but I didn't know exactly what. (reboot needed) [11:56] I just need to find out its allocated IP address.... does anyone know How I could do that? [11:57] garymc: the documentation doesnt have it [11:57] the administration documentation [11:58] I thought UBUNT would tell me [11:58] *Ubuntu [11:58] See I have lots of Terminals attached to the server, there ips are something like 192.168.0.25 [11:59] I have 10 comps on there [12:00] garymc: [12:00] then in that case [12:00] do an ifconfig and see what the gateway ip is [12:00] wait [12:00] that wont work :( [12:00] no i know [12:01] it just shows 192.168.0.254 [12:02] why not tail syslog and unplug/replug your device [12:05] yeah ill give it a bash [12:05] what is SPIP config? [12:05] I found something on 192.168.0.40 [12:06] it asks for username and pass [12:06] admin and blank password not working [12:07] admin netgear [12:15] no its not that [12:15] SPIP is the Voip phone config screen [12:16] I plug and unplug and nothing changes in deamon.log or syslog [12:17] alvin: Hehe, yeah, it's pretty horrible. [12:17] alvin: Our backup software vendor issued a big fat warning not long ago. [12:19] I have had no such warning. Couldn't log into the servers on monday fo two weeks. I wonder how these issues make it into official releases. [12:22] alvin: ? what happened [12:23] eagles0513875: It's about a kernel issue. Create lvm snapshot is now equal to: crash your server immediately. [12:23] O_o [12:23] oh my [12:23] what bothers me though is the short 6 month release cycle [12:23] tbh i would rather have fewer releases [12:23] and a stabler system like debian [12:24] then a new release every 6 months [12:24] Same for me. I often think Ubuntu Server would be better off when based on Debian stable instead of unstable. [12:25] i think the dev cycle seriously needs to be revised [12:25] how is the dev cycle split i would like to know [12:25] eagles0513875: ever heard of the LTS releases? ;) [12:25] joschi: yes [12:25] and im on them here at work [12:25] LTS doesn't mean 'stable'. It means: longer support. [12:25] exactly joschi [12:25] alvin [12:26] alvin: it means: "converges to stable in a few months" basically [12:26] That's on Windows (service packs) [12:26] alvin: check pm m8 [12:36] New bug: #619769 in samba (main) "package samba-common 2:3.4.7~dfsg-1ubuntu3.1 failed to install/upgrade: el subproceso script post-installation instalado devolvió el código de salida de error 10" [Undecided,New] https://launchpad.net/bugs/619769 [12:53] lol..if you need a longer release cycle switch on debian! === garymc_ is now known as garymc [12:56] hello, can anyone point me to information how much is ubuntu server install and maintainance cost ? [12:57] anyone know how I connect to a Netgear hub via console? [12:57] what is a console? [13:00] hey, i just did a regular update to my ubuntu 10.04 system and now i can't logon. it says 'permission denied' i have two user accounts and both are denied. it actually takes a long time to process and sometimes even times out trying to figure out if i typed the right password or not. [13:00] i feel a slight panic beginning to happen. [13:00] seeing how it's 8am :) [13:01] so i'm running samba and ltsp [13:01] m1r: well, it's free [13:01] this is the first time i've rebooted in maybe a couple of weeks. i'm gonna go see what's going on at the console. [13:02] garymc: you can access control panel (if is there) via lyn [13:02] oh wait, it's logging me in now. [13:02] *lynx [13:02] alex88: i understand OS is free, but setup and configuration gotta have some price for time invested ? [13:02] AndyGraybeal: lol :) [13:02] alex88: shush, i just had a heartattach [13:02] m1r: it depends by how much the person who configures for you asks.. [13:03] m1r: you can do everything by yourself for free [13:04] alex88: is there any "recomended" price for this work ? [13:04] m1r: it depends where are you and how much ppl asks for it [13:04] if you do for profession or time for time [13:05] alex88: for friends home server [13:05] m1r: friends? 30euro/h? [13:05] alex how? [13:05] alex88: ^ [13:06] alex88: manY thanks! [13:06] garymc: sudo apt-get install lynx && lynx ip_of_the_router [13:06] usually you use browser to access it? [13:06] m1r: np :) [13:06] yes but it just times out [13:07] garymc: so there is a router problem [13:07] tried to reset [13:07] ? [13:07] cant reset this without going into its CLI via the IO port [13:08] IO port is serial port? [13:10] also..why here in ubuntu-SERVER? [13:14] hmm now i'm having issue with running 'sudo' - i type my password and it is taking forever to do anything -- i did 'sudo ls' just to do it and i'm still waiting for it to recognize my password [13:15] mmhhh..that's a bigger problem.. [13:15] tried sudo su? [13:15] just did after your suggestion, and still waiting on auth [13:15] oh it works. [13:15] it took a while [13:16] that's strange..it takes time when you mistype the pass [13:16] not when it's correct === mathemonkey is now known as darkk^ [13:21] alex88: my 'sudo ls' eventually came up also [13:21] i just rebooted so maybe things are just settling in though [13:22] i'm on my first half cup of coffee oto [13:22] :) i hate coffee [13:23] yea, i go back and forth. today i told myself no coffee... alas.... [13:23] it's late for that :) [13:36] alex88: aah italy :) nice [13:37] i'm on the eastern side of the US; just getting started [13:42] AndyGraybeal: seen in whois :) [13:42] ? [13:43] btw, which city? i'll be in NY in the first week of next year [13:50] is there any specific place on ubuntu where sockets are regularly stored? [13:53] sockets? [13:53] uwsgi.sock.. like mysql.sock I figure /var/run [13:53] They're not really stored. They're a figment of the kernel's imagination. [13:55] :) nice explaination [14:00] anyone setup Netgear PoE switches? [14:04] jdstrand: hm, there's no bzr tree for libvirt-bin, right? [14:05] you always work with the packaging itself? [14:05] alex88: yes [14:06] alex88: i live in southeastern ohio; athens ohio [14:08] Ok getting back.... I connect Cable to console from Laptop. What software do I use to connect to it? === e-jat is now known as ejat [14:19] smoser: this is what I get back from get-console-output: http://pastebin.ubuntu.com/479922/ [14:19] garymc: you want to log in to a Netgear hub via the command line e.g. telnet? They usually have web interfaces. [14:19] right. hggdh is there a bug ? [14:19] if not, we need to open one, and mark it high [14:19] heh? netgear hubs don't have web or telnet [14:20] the managed switchs do, at different levels, depending on if it's fully managed, or just a smart switch [14:20] smoser: it sounds like we have... do you remember something similar? I dimly do [14:20] I'm thinking switch. But anyway ... trying to get more infomation. [14:20] no. i've never seen this bug on eucalyptus. [14:20] smoser: OK. Opening a new one [14:20] there are other "euca-get-console-output doesn't work in some way" [14:21] but this one is new. [14:22] hggdh: Are the tests you run against Eucalyptus described anywhere that I can see? [14:23] * soren is curious how far OpenStack is from passing those tests at this stage [14:25] soren: partially on the blueprint, partially in the code. You can get the code from /~uec-testing-scripts-dev/uec-testing-scripts/trunk/ [14:26] soren: basically: hammer down as many instances as possible, on all types and images available; (2) create/attach/allocate/use/reuse volumes [14:27] soren: er. If you happen to test OpenStack... mind telling me the results? ;-) [14:27] hggdh: Sure :) [14:27] hggdh: How many physical boxes are you using? [14:28] soren: six on this rig [14:28] each with 16 cores [14:28] Wow. [14:28] * soren glances at his feeble set of test hardware [14:28] and I would still like to have more ;-) [14:28] ...and sobs. [14:29] well, if you looked at my personal set of hardware... you would really cry [14:29] Well, there's a reason I've taught OpenStack to use user-mode-linux as its backend instead of KVM. [14:29] hallyn: not a separate bzr tree for Ubuntu, no. Debian uses git iirc and there is the distributed development one for Ubuntu, but that hasn't worked well for me in the past [14:29] so I can test stuff on rackspace cloud servers or whatnot. [14:30] soren: cool! [14:30] hggdh: It's not a replacement for full integration testing, of course, but it does a long way in testing OpenStack itself. [14:30] *goes* a long way.. [14:31] I agree [14:31] hggdh: Your servers have 16 cores? [14:31] binBASH: these, yes [14:31] I hope not 8 real and 8 ht [14:31] :D [14:31] New bug: #619843 in eucalyptus (main) "euca-get-console-output returns one single line" [Undecided,New] https://launchpad.net/bugs/619843 [14:32] hggdh, thanks. [14:32] binBASH: no needs to worry ;-) [14:32] hggdh: For instance, the uml backend doesn't support get-console-output. [14:32] hggdh: Yet, anyway. [14:32] hggdh: I'm searching here server solution as well. We run out of cores :) [14:32] hggdh: I just submitted a patch to libvirt this morning that'll make it happen. [14:33] need those for imagemagick thumbnail creation and ffmpeg video processing [14:33] hggdh: ..but hotplugging volumes and such... I doubt that'll work at all. [14:33] soren: nice. I get you are working heavy on OpenStack? [14:33] hggdh: That's all I do. [14:33] oooohhhhh [14:33] hggdh: http://www.unicorner.de/index.php?option=com_content&view=article&id=188%3Afujitsu-primergy-cx1000&catid=53%3Aunicorner&Itemid=139 [14:34] Maybe will buy this. [14:34] hggdh: Well, that and fix dependencies around it, like libvirt in this case. [14:35] binBASH: you have more money than I do ;-) [14:35] soren: *very* cool. [14:35] hggdh: it costs 80.000 USD [14:35] binBASH: Do you really need that sort of processing power continuously? === e-jat is now known as ejat [14:36] binBASH: I very much doubt I can justify $80k for tests [14:36] soren: Yeah, Getty Images is processing much stuff....... :) === garymc_ is now known as garymc [14:37] hi guys.. how often are updates for versions of programs in packages in universe? [14:37] binBASH: Ok. [14:38] soren: We're running for them atm 3 servers for processing, with 8x2,53 Ghz each. It's not enough [14:38] * patdk-wk spanks binbash [14:39] and we host only one of their sites right now. They plan to come with 5-8 more sites :) [14:39] binBASH: I'm not questioning whether you need the processing power. I'm questioning whether you need it *continuously*. [14:39] in other words, elasticity... [14:39] soren, sure, how will he do seti? :) [14:40] $80k gets you a lot of CPU cycles with $CLOUD_PROVIDER. [14:40] soren: We measured what cloud processing would cost per customer. [14:40] and it was more expensive. [14:41] I mean, if you buy such a 80K system and you use it for 36 Month it will cost 4000-5000 USD / month. [14:41] including connection....... [14:42] and if we put one customer to the cloud eg. amazon ec2 it will cost like 1500 USD / month. [14:44] 5000/month can get you between 1600 and 2000 cores if you play your cards right. [14:44] hmm.. i'm trying to get "gnome-session-save --forced-logout" in a script called 'logout' in /usr/lib/xscreensaver - it's referenced by /usr/share/applications/screensavers/logout.desktop.. it doesn't appear to work though when it's activated; so i'm a little confused. [14:44] soren: Maybe spot instances. [14:44] AndyGraybeal: Wrong channel :) === garymc_ is now known as garymc [14:44] binBASH: Nope. [14:44] Spot Instance is not really an option ;) [14:45] soren: I just looked at Amazon, I didn't check other cloud solution providers. [14:45] binBASH: It depends on storage and memory requirements, too, of course. I all you were doing was adding numbers, so very little need for storage and memory, the smallest instance on Rackspace's cloud platform still gives you access to 4 cores @~2.6 GHz. [14:46] binBASH: And those cost ~$11/month. [14:46] storage, hmm. 10TB+ [14:47] Well, that's the thing... Do you need to it all in the same place as you do the processing? [14:47] It's not a magic silver bullet thing. You need to design for it to get the real benefits. [14:48] soren: The storage can be outside cloud. [14:48] however also need to think transfer costs will be higher [14:49] binBASH: Certainly. [14:49] binBASH: As I said: If you just want to add numbers... :) [14:50] binBASH: If you need more than that, it's a more complex equation. [14:50] binBASH: I [14:50] Need more than only add numbers ;) [14:50] soren: k thanks [14:50] images need to be send to ftp servers from partners etc... [14:51] New bug: #619855 in php5 (main) "session.gc_probablity=1 in /etc/php5/apache2/php.ini conflicts with permissions on /var/lib/php5" [Undecided,New] https://launchpad.net/bugs/619855 [14:53] binBASH: There's also the fact that you pay as you go rather than buying expensive hardware up front, you can scale up and down as needed (I'm sure you have "dead" periods of the day, week, and year), and don't pay for power for the equiment nor cooling. [14:54] binBASH: I'm not trying to sell you anything, honest :) I'm just trying to help you save money. [14:54] soren: Yeah I know :) [14:58] kirkland, http://www.dreamhostapps.com/ [14:58] starting the worker machines via the API only when needed could be a better solution really [15:01] soren, that is easy to do if he buys a system also, he can shutdown systems and turn on based on load [15:01] soren: We will have conference in company about solutions tomorrow, so we can discuss and calculate what is best ;) [15:01] binbash, maybe you need to see how much each task you do does :) [15:01] like webservers, vs ftp, vs thumbnail creation [15:02] might be best for your own webservers, but to use like ec2 for thumbnail creation [15:02] patdk-wk: He has still paid for it. If it's off, he's not getting full value of his investment. [15:02] soren, if he needs it during the day but not at night? [15:02] patdk-wk: It may still be more cost-efficient. I'm just saying your analogy is faulty. [15:03] didn't know I made an analogy at all [15:03] patdk-wk: Your hardware vendor is not going to give you a discount because you tell him you're going to turn it off at night. [15:03] :D [15:03] But datacenter is ;) [15:03] no, but they normally give a discount if you guy 50, vs 20 :) [15:04] and if he needs 40 [15:04] patdk-wk: So you'll still have paid through the nose for the hardware, but since it's off the money you paid for the hardware isn't being put to good use. [15:04] though turning servers off at night is not an option when you have customers all over the world :D [15:04] binbash, I'm sure you don't have an even load on your systems in all hours of the day [15:05] binBASH: No, but perhaps during the weekends. Or during summer or particular holidays. I've yet to meet anyone whose load is static 24/7/365. [15:10] soren: That's not an option for us :) [15:11] patdk-wk: Atm we have load there all day long :) Customers push images, queues are overloaded :) [15:23] hggdh, for the record "empty console output" , you may have seen this, but we see it on ec2 [15:24] on ec2, though, there is no console output, we believe as a result of kernel bug (or xen hypervisor borkage). here, in eucalyptus, thta is not hte case, i've verified that the console log is present in the /var/lib/eucalpytus/... [15:26] smoser: so for Eucalyptus we have the full console in /var/lib/euca*, but only one line is shown -- correct [15:26] ? [15:26] i just commented that in the bug, yes. [15:27] thanks === garymc_ is now known as garymc === dendro-afk is now known as dendrobates [17:11] New bug: #619947 in autofs5 (main) "initialization of autofs is erratic (no nis maps available)" [Undecided,New] https://launchpad.net/bugs/619947 [17:28] re preseed, is there some way to enforce that old-style partition nomenclature be used on target system (sda? as opposed to uuid)? === robertpayne_ is now known as robertpayne [17:39] pmatulis: I'm curious what your reasoning is behind using SCSI device ordering instead of UUID. [17:40] SpamapS: me too [17:46] New bug: #619970 in eucalyptus (main) "suddenly no instance starts anymore - walrus issue?" [High,New] https://launchpad.net/bugs/619970 [17:56] pmatulis: I could see one reason which is creating raid volumes.. never done that in preseed [18:14] How can I get a system back to having only the base packages installed? [18:15] I'd like to just go into aptitude - tell it to mark all for removal, then go through and pick what I want to keep - like ubuntu-minimal and openssh-server [18:21] i just installed aide and noticed that some files in /etc/aide/aide.conf.d/ are given +x permissions while others are not. Why is that? [18:33] smoser: ping [18:34] RoAkSoAx, hey [18:35] smoser: howdy!! just saw your bug report. What's that support for UEC images for testdrive? you already have a patch, what does it require? [18:35] i was going to start writing a patch. [18:36] just putting together the changes to the images that it will need. [18:36] smoser: where are they available? [18:36] uec-images.ubuntu.com [18:37] there is no builds available that have the stuff needed at the moment, but i'm about to kick one off. [18:38] http://bazaar.launchpad.net/~ubuntu-on-ec2/vmbuilder/automated-ec2-builds/annotate/head%3A/README.files describes the members that are in a .tar.gz file [18:38] of interest here is the -floppy file [18:39] smoser: so it would be jsut matter of downloading the tar.gz, uncompress it, ad then run the .img with kvm? [18:41] hello guys [18:41] im having some issues [18:41] RoAkSoAx, yeah. [18:42] installing ubuntu from usb and without internet [18:42] could you give me a trick ? [18:42] RoAkSoAx, the important piece being boot wiht '-fda' floppy. [18:43] thats where the tricks are (providing kernel command line and such) [18:43] smoser: ok, let me think how we can do that. I have to run for lunch. when I get back we can discuss that. [18:44] k [18:44] somker [18:44] smoser [18:45] could u help me ? [18:45] im having problems with installing ubuntu without internet [18:45] what is rong? [18:45] i cant step over choose-mirror screen [18:45] wrong [18:45] im having problems with installing ubuntu without internet [18:45] i personally have never tried such a thing. it does'nt give you any option like choosing a cdrom as the mirror ? [18:46] not at all [18:46] and im installing it from a usb stick [18:48] itterbium, i'm sorry, i don't really know. [18:48] anyone have a solution for pci_add_option_rom: failed to find romfile "pxe-virtio.bin" in kvm, I have found some bug reports [18:48] ok, no problem, thanks anyway [18:49] i had no internet cuz the kernels modules are buged [18:49] i had an RTL 8111 [18:49] im really fuck up [18:50] !language [18:50] Please watch your language and topic to help keep this channel family friendly. [18:51] sorry [18:51] my bad [18:57] smoser: ok i'm back. So, anyways, we'd only need to run 'kvm' with custom options right? [18:58] basically, yes. [18:59] smoser: so something like: testdrive -u uec-image.img --uec-image [18:59] in 30 minutes or so, i'll be able to put something at http://uec-images.ubuntu.com/server/maverick/20100818.1 that you can poke at. [19:00] RoAkSoAx, well, kirkland was suggesting to just trigger off the .tar.gz in the invocation [19:00] would indicate "oh, that is a uec image" [19:00] but you could possibly do better. === nxvl_ is now known as nxvl [19:01] smoser: yeah, could possibly use tar -t to list contents, grep for something uec-y [19:01] tar -t is going to be badly slow [19:02] and generally a waste. [19:02] to -t a tar archive, you'd have to uncompress and read the whole thing (slow/io and cpu intensive) [19:04] hi! my ubuntu install is missing the /etc/inittab file, but i need it to specify a serial console redirection... can i just add that file or is there some package i need to install? [19:05] kirkland: smoser we can do something like manually download all the desire uec images and place them in $TESTDRIVE_CACHE/uec, and then run 'testdrive --uec', and that will display a list of the available UEC disk images to run [19:06] itterbium: you still there? [19:07] and further along the road we can automate the process of syncing the .tar.gz, uncompressing, and making the disk images for testdrive [19:14] RoAkSoAx, well, i'd ideally like it to be easy just like testdrive, and "just work". [19:14] at very least i dont want to require the user to download something, extract the image into a cache . and then run. [19:15] in that case, there is little to no value in testdrive. [19:15] i've tried a update-rc.d -f mysql remove and also mysqld but it has no links found..how can i prevent mysql to start on startup? [19:16] alex88: edit /etc/init/mysql.conf and change the line 'start on xxxxxx' to 'start on never' [19:17] alex88: assuming you're on 10.04 [19:17] smoser: of course. but I'd first provide the support for actually running the disk images, then I'd provide the method to automatically sync the tar.gz, untar it, and make it available to run. This will change depending on what front-end we would like it for (-cli or -gtk, or even both) === EvilPhoenix is now known as EvilTrek [19:19] SpamapS: why it don't use rc like others? [19:20] alex88, upstart is the new way. things are being transitioned off sysvinit [19:21] i think i'll need to read something about that...thank you :) [19:25] i've installed apache 2 with mod_php5, on restart it says "Apache/2.2.14 (Ubuntu) PHP/5.3.2-1ubuntu4.2 with Suhosin-Patch configured -- resuming normal operations" but it still download the index.php file.. [19:26] oh sorry..now it works.. [19:29] alex88, you ran into https://bugs.launchpad.net/ubuntu/+source/apache2/+bug/603192 [19:29] Launchpad bug 603192 in apache2 "install of libapache2-mod-php5 may not result in enabled php" [Low,Triaged] [19:31] i ran into, noob that don't clear browser cache....-.-' [19:32] alex88: -. --- --- .--- [19:32] yeah... [19:34] if i've got a file owned by a user and i want to give other users permission to execute it. should i just make it 774 then add other users to the owner's group? [19:34] sorry.. [19:37] if you want to be selective for users that need to execute..yes.. === teddy_ is now known as teddymills [19:46] I'm trying to follow this guide (http://nginx.localdomain.pl/wiki/FcgiWrap) but when I get to make install it decides it hates me and says no make file exists [19:47] then you should probably ask the writer of that. [19:52] which torrent is the "basic" 10.04 server edition? [19:52] i386? [19:57] * Ose downloads that one [20:03] smoser: those uec images are also daily images correct? [20:03] there are dailies and released. [20:03] smoser: which ones are the ones desired to testdrive, both? [20:04] well, no releases are available yet that would be suitable. [20:04] but i would suspect that if you test drive stable releases, you'd want to test drive stable uec images. [20:04] (once they're there) [20:04] basically, i'd think it would be the same as the other things that you can "test drive" [20:05] smoser: yes but for TestDrive itself, what is the main test purpose, daily images? [20:05] s/test purpose/test target/ [20:05] i would think that it is no different for uec images compared to "other" [20:06] that make sense ? [20:06] they're just like releases of something else that ubuntu makes. [20:07] what is "the main test target" for daily ubuntu desktop isos ? [20:07] smoser: yes, but the way TestDrive handles it depends on the .manifest and .manifest-daily files [20:07] which one is on releases.ubuntu.com and the other on cdimage.ubuntu.com [20:08] so you just want to know "is this thing a released image, or is it a daily" ? [20:08] i was unaware of .manifest-daily versus .manifest. [20:09] smoser: yes I just wanted to know that since at the end the isos are handled the same way, the onyl thing that would change would be obtaining the manifests [20:11] well, you can easily do it based on the a [20:11] a.) url [20:12] smoser: anyways, so it would be something like: 1. Obtain UEC .tar.gz list automatically and make it avaialble through TestDrive. 2. Sync .tar.gz, and prepare image (untar files). 3. Provide kvm with related parameters. 4. launch image [20:12] b.) presense of published-ec2-daily.txt versus published-ec2-release.txt [20:12] c.) you could actually just read http://uec-images.ubuntu.com/query/ for clean informatoin [20:12] RoAkSoAx, yeah. thats about right. [20:13] i'd say you might want to do: 2.5. create qcow2 image using backing image of the .img file (so it stays 'clean') . see man qemu-img and '-b' [20:14] smoser: but first of all i'd need a manifest similar to: http://cdimage.ubuntu.com/.manifest-daily, which instead of listing ISO's, it will list .tar.gz files [20:15] oh. [20:15] query does provide you taht. [20:16] smoser: should prolly be in http://uec-images.ubuntu.com/.manifest-daily [20:16] kirkland: any thoughts? ^^ [20:16] i'm not terribly opposed to providing that, but /query is the data that we're using elsewhere. [20:18] smoser: i think i'd rather have a manifest like those used in cdimage or releases.ubuntu given that will allow for easy integration to TestDrive [20:18] and allow me to re-use most of the code [20:20] and just say something like "if it is from uec category, sync .tar.gz instead of iso, and when finished, untar file and prepare image", so that launching would be done easier [20:20] RoAkSoAx: smoser: yeah, i think it best if smoser generated a very similar manifest [20:21] and whne launching i'd do "if from UEC category, use this custom KVM_ARGS" [20:22] well, i can write the conversion from /query to whatever data format you're getting out. [20:22] i just don't want to write more data at this point. [20:27] There are 3 suspiciously similar new samba bugs... bug #619769, bug #619773, and bug #619540 [20:27] Launchpad bug 619769 in samba "package samba-common 2:3.4.7~dfsg-1ubuntu3.1 failed to install/upgrade: el subproceso script post-installation instalado devolvió el código de salida de error 10" [Undecided,New] https://launchpad.net/bugs/619769 [20:27] hallyn: Error: Could not parse data returned by Launchpad: list index out of range [20:27] Launchpad bug 619540 in samba "package smbclient 2:3.3.2-1ubuntu3.5 failed to install/upgrade:" [Medium,Incomplete] https://launchpad.net/bugs/619540 [20:27] but i can't reproduce... [20:27] smoser: having a manifest in http://uec-images.ubuntu.com/.manifest-daily with this http://pastebin.ubuntu.com/480079/ would simplify the integration with TestDrive a lot. [20:28] anyone see anything obvious why samba-common --configure would suddenly fail for 3 people in different locales? [20:29] RoAkSoAx, alright. i'll work on getting that created. [20:30] smoser: awesome then. Thanks :)! [20:36] hallyn: which version? [20:38] smoser: btw... DOes http://uec-images.ubuntu.com support rsync? [20:41] zul: chasing a gremlin hanging out in my wifi card, need a minute... [20:41] hallyn: okies [20:41] zul: one is Configurando samba-common (2:3.4.7~dfsg-1ubuntu3.1) ... [20:42] RoAkSoAx, yes. [20:42] also bug number :) [20:42] RoAkSoAx, one thing, though, is that the rsync of a .img.tar.gz is not as nice as it is with a .iso. [20:43] zul: another was /var/cache/apt/archives/smbclient_2%3a3.4.0-3ubuntu5.6_i386.deb [20:43] zul: bug #s above: bug #619769, bug #619773, and bug #619540 [20:43] as it is a partition image of a ext4 filesystem rather than a iso filesystem. each day, the layout of the filesystem varies, and randomly so, while the creation of a .iso is more ordered, resulting in higher rsync friendlyness. [20:43] Launchpad bug 619769 in samba "package samba-common 2:3.4.7~dfsg-1ubuntu3.1 failed to install/upgrade: el subproceso script post-installation instalado devolvió el código de salida de error 10" [Undecided,New] https://launchpad.net/bugs/619769 [20:43] hallyn: Error: Could not parse data returned by Launchpad: list index out of range [20:43] Launchpad bug 619540 in samba "package smbclient 2:3.3.2-1ubuntu3.5 failed to install/upgrade:" [Medium,Incomplete] https://launchpad.net/bugs/619540 [20:43] smoser: smoser well by default TestDrive uses rsync to sync the .iso, so I thought that I'd also use to sync the .tar.gz [20:43] zul: i dunno, it does look like it's all just 3 ppl having very similar bad luck at the same time [20:44] yeah, and that will work. [20:44] i just want to make sure i'm not ignoring a much deeper error that'll keep cropping up [20:44] the difference is just that tomorrow you wont get as large a benefit from already having a .tar.gz as you would from having a .iso [20:44] heh, there's a php5 bug for SpamapS too [20:45] hallyn: heh interesting [20:45] smoser: TestDrive does, and will do something like: rsync -azP rsync://uec-images.ubuntu.com/uec-images/server/maverick/current/maverick-server-uec-amd64.tar.gz . [20:46] yeah. it just wont get as much benefit if maverick-server-uec-amd64.tar.gz was here from yesterday. [20:46] and, in that case, you should probably remove -z if you could that'd be better. [20:47] as the data is already compressed. [20:47] just wasted cpu [20:47] anwya [20:47] hggdh: fwiw, r1232 uploaded to maverick [20:47] let me get to produ cing that dat afor you [20:48] kirkland: ack, thank you === jeiworth__ is now known as jeiworth [20:48] smoser: k, that's just sync tweaking that can be handled later [20:54] RoAkSoAx, so are you thinking you can do some to fthis ? or are you expecting me to ? i was expecting to... but i'm definitely open to you doing so :) === unreal_ is now known as unreal [20:59] does anybody know of a good DNS/BIND tutorial for Ubuntu server edition? [21:00] preferrably known to work well with 10.04 [21:03] smoser: I can do some of the work (if not all). However, right now I'm on vacation in Peru, so prolly will work on it on the weekends. [21:03] alright. [21:04] enjoy your vacation RoAkSoAx [21:06] smoser: thank you :)! just let me know when the manifest is ready and I'll start the integration [21:08] * hallyn waits for SpamapS to yell at him [21:10] is adding ldap.log to the logrotation as simple as adding /var/log/ldap.log to /etc/logrotate.d/rsyslog ? [21:11] superbrad, probably the one in the ubuntu server guide [21:12] https://help.ubuntu.com/10.04/serverguide/C/dns.html [21:13] wow, adding the phrase server guide to the google search makes a heap of difference (feel free to point fingers and jeer ;) [21:14] qman__: thanks! [21:15] I have a UPS now and the data cable is connected to the server. How can I read information from the UPS? [21:16] MTecknology, apcupsd is the only software I know of for that [21:16] if your UPS isn't supported by it, and it didn't come with linux software, you're pretty much SOL [21:17] qman__: thanks :) [21:17] unfortunately there isn't a UPS communications standard in wide use [21:20] qman__: anything really simple to follow to start using it? [21:20] not really, pretty much have to just read through the configuration file and hope you can tell which model yours is [21:22] took me a while to get my backUPS pro working, and after all the effort it doesn't even tell me battery life [21:23] it only knows when power is lost and resumed [21:23] ouch [21:24] should have spent the extra $20 on the smartUPS model [21:24] What did you spend on it? [21:24] Ours cost $65 [21:24] about $150 after shipping [21:24] it's a refurb, APC BackUPS Pro 1400 [21:24] it's otherwise a great unit, long live, easy battery replacement, 24v system [21:25] it'll run about 40 minutes with a dozen machines on it [21:25] nice [21:26] replacement batteries are about $70 too, not too bad [21:27] New bug: #620085 in libvirt (main) "libvirtd does not clean up pid file" [Undecided,New] https://launchpad.net/bugs/620085 [21:29] I think it's working :D [21:30] It even tells me the model [21:30] APCMODEL : Back-UPS ES 550 [21:30] qman__: Do you know if I can tell it to run a script instead of shutting down? [21:30] namely - telling each server hosted on it to shut down before it goes down [21:39] anyone use landscape? [21:40] yeah, i print it all the time ;) [21:40] sounds like something i would be interested in though, what is it? [21:43] I got a landscape service for trial and since my last reboot I don't see any update at landscape page [21:43] Probably need some Canonical channel for that. [21:44] It's not supported in Ubuntu. [21:44] zul: I'm pushing openssl and apache2 to -proposed for all releases to fix CVE-2009-3555, fyi [21:44] mdeslaur: The TLS protocol, and the SSL protocol 3.0 and possibly earlier, as used in Microsoft Internet Information Services (IIS) 7.0, mod_ssl in the Apache HTTP Server 2.2.14 and earlier, OpenSSL before 0.9.8l, GnuTLS 2.8.5 and earlier, Mozilla Network Security Services (NSS) 3.12.4 and earlier, multiple Cisco products, and other products, does not properly associate renegotiation handshakes with an existing connection, which allows man-in-the [21:50] mdeslaur: cool [21:51] zul: please let me know if anything explodes horribly with them [21:52] mdeslaur: dont worry i will use rfc 6660 - electric shock protocol :) [21:52] hah! === skoef_ is now known as skoef [22:12] New bug: #620102 in samba (main) "package samba-common 2:3.4.7~dfsg-1ubuntu3.1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/620102 === niekie_ is now known as niekie [22:21] I just tried to boot a -virtual install into single user mode. It prompted me for the (nonexistent) root password. Is that somethign specific to the -virtual initrd/kernel? Normal installs give me a nice little menu and offers to drop me into a shell... === dendrobates is now known as dendro-afk [23:12] i have a question about setting up BIND on my server, the serverguide is a little confusing [23:12] i'm following the instructions at https://help.ubuntu.com/10.04/serverguide/C/dns-configuration.html [23:13] about 4/5 down the page, it says: [23:13] " [23:13] Next, on the Secondary Master, install the bind9 package the same way as on the Primary. Then edit the /etc/bind/named.conf.local and add the following declarations for the Forward and Reverse zones: " [23:14] this doesn't make sense. i had to create files - obviously overwriting won't work to add the secondary master [23:15] so do i make new files for the secondary master or insert extra records at the end of the files i already created (e.g., db.mydomain.com) [23:16] superbrad, a secondary master is a separate server [23:17] so the files you already created would not exist on it [23:17] if you only have one server, you only need the primary master configuration [23:17] i see ... my vps came with 2 IP addresses, I assumed I would use one for master and one for secondary master [23:18] you can, but you don't need to configure it again [23:18] the first configuration will listen on both [23:18] well, provided it's configured to [23:19] do i need to mention the "slave zone" in the named.conf.local file? [23:19] I didn't read through the whole guide so I don't know if it's binding to a specific address [23:19] no, that's only for secondaries [23:19] if you only have the one server, that part is skipped entirely [23:20] k [23:20] if you've got a minute, I'm a little stumped on the next part also [23:20] sure [23:21] i recently got this vps and then got a domain name. when i bought the domain name, i didn't have a spot to enter the IP address of my server. how do i hook the domain to my IP? [23:22] you will have to do that from the company where you registered it [23:22] they should have a web portal where you can configure it [23:22] if you want to resolve it yourself, you can have them delegate authority to your server [23:22] most places offer the full DNS resolution though, so you wouldn't need to run BIND at all [23:25] all i can find in the domain area when I log in at my service provider is textboxes for four nameservers [23:25] the only situation I can think of where resolving it yourself would be more advantageous is if you wanted to point subdomains to different IPs [23:25] yeah, probably won't be [23:27] ok, looking at one of my domains [23:27] when I go to the configuration I have two options [23:27] i'm using burst.net, in case that helps. [23:27] "Update Name Servers" and "Register Name Server" [23:28] to run your own DNS, you would click "Register Name Server", give it a name like ns1.yourdomain.com, and type in the IP [23:29] if you want the registrar to host it for you, you would have to look elsewhere in the panel [23:29] for mine, it'd be "register services for domains" [23:30] i'd be fine with whatever way is easiest ... i contacted them on their instant chat, but the guy told me with VPS servers I had to set up my own nameserver [23:30] well, since you already went through setting up BIND, might as well just go with that [23:30] I don't have any of those options that you mention in my control panel [23:31] is your domain registered through the same company your VPS is hosted by? [23:32] yeah [23:32] well, somewhere there has to be an alternative option to just updating the name servers, which is what you are seeing === gnoob is now known as xgorg [23:33] I'd contact them and ask where you register a name server [23:33] if they don't, they're missing a huge step in the DNS process [23:33] yeah, hopefully i get someone more than a sales guy this time ... not very impressed with burst.net so far, other than the price [23:33] I've got domains through mydomain.com and byet [23:33] both have been good so far [23:35] mydomain has a better website, byet's are spread all over a bunch of them and sometimes kind of hoakey [23:36] but they make up for it with good tech support [23:36] yeah, i'll keep that in mind next time. [23:36] i've used godaddy in the past, but they were a little too spendy for vps compared to other options [23:37] I don't have a VPS through byet, just shared web hosting [23:37] i know they have that section very clear and easy to use in their control setup, i've used it before. [23:37] and domains [23:37] thanks a billion for your help [23:37] much appreciated [23:37] no problem [23:56] New bug: #620155 in samba (main) "package winbind 2:3.4.7~dfsg-1ubuntu3.1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 2" [Undecided,New] https://launchpad.net/bugs/620155