[00:06] hackeron: Thanks for answering. But can't you get the same with other 'server' versions of other distros? [00:08] hi guys, may a ask a newbie question ? 1) how should i configure my apache2 to be able to access the directory list on http://someipaddress/repo 2) how can i access my website index.php on http://someipaddress/web ? === chuck_ is now known as zul [00:10] ruben23: http://lmgtfy.com/?q=ekiga+asterisk [00:11] PoorNewb23: sure, the rest is personal preference really - which community you like best, how bleeding edge you want to be, how security updates are handled, etc [00:12] I keep getting "2003 can't connect to mysql server on 'IP' (10061)" when trying to connect to my mysql server remotely even though I allowed the port through ufw [00:13] PoorNewb23: many like centos because it's similar to redhat enterprise, I find it much easier to work with ubuntu personally and everything is a lot more up to date, yet still rock stable === dendro-afk is now known as dendrobates [00:23] Greetings. I was wondering if CD/DVD burning was going to be included with LTSP. I found this page: https://blueprints.launchpad.net/ubuntu/+spec/ltsp-cd-burning but it is quite old.. === dendrobates is now known as dendro-afk [01:06] Can anyone help me install the perl module ExtUtils::Embed, It says I need a version of perl newer than in apt? [01:11] Hiya. A while back I had to install sendmail on one of our servers. I tried using apt-get install sendmail to install it, but got a message saying couldn't find sendmail. So I downloaded the tar file from sendmail's website. Anyway, my boss asked me to remove it completely, and so I tried apt-get remove sendmail (I forgot the apt-get install didn't work), and that didn't work. Is there [01:11] another way to uninstall applications from Ubuntu? Or do I have to "locate sendmail" and delete the files and folders it finds? Thanks in advance for all your help. === dendro-afk is now known as dendrobates [01:30] TommyC, you can only automate removal if you used a package manager to install [01:30] Yeah. The manual install needs to be done. Any particular way I can go about doing that? [01:30] for future reference, you should use checkinstall or similar to build packages from source, rather than just installing directly with 'make install' [01:31] All right. [01:31] that way they can be easily removed [01:31] but if you just ran a make install, you will have to manually locate and remove the files [01:31] Gotcha. [01:31] it may have made it easy on you and installed to /usr/local/ [01:32] but it depends on the compile options [01:33] All right, thanks. [01:42] anyone ever tried aborting a fsck prematurely? [01:42] Yes [01:42] how did that go for you? [01:42] I didn't say it was me [01:42] how did it go for whoever? [01:42] But it'll depend on the filesystem [01:43] forgot to run the damned thing on screen, and I need to close the SSH terminal [01:43] For ext3 you are probably OK [01:43] ext4, late kernel [01:43] I assume ext4 isn't much different [01:43] ext4 should be better [01:43] we're on hour 8 of this stupid thing [01:43] I mean, I wouldn't interrupt it unless you have a choice [01:43] sounds like mine :) [01:43] gtaylor: is this after rebooting, because it hasn't been fscked for 30 days? [01:44] twb, what is nice is when I do a bios reset, or have a bad bios battery, fs timestamp in future, fsck disk NOW [01:44] Oh yeah, I've seen that [01:44] twb: It's a secondary mount that corrupted independently of the root partition, so it's un-mounted and fsck'ing [01:44] and 10tb and millions of files later :) [01:45] twb: So I'm not worried about the root partition. It's just got a lot of valuable media files in it. [01:45] and no, not porn [01:45] what a waste of a server [01:55] RoAkSoAx: if the fix is small, sure [01:56] SpamapS: sommer appeared from within the server community, not the docs community. I'd recommend someone go hunt him down and ask pretty please with a cherry on top what would it take to get you to come back .... [01:57] (just read the backscroll from today's meeting) [01:57] Heya kirkland. [01:57] ScottK: hi [01:57] I appreciate all the work you're doing getting things sorted. [01:58] ScottK: sure, you're welcome [02:40] ScottK sommer said he was busy with his new job [02:41] Oh. OK. [02:41] yeah he said he was gonna be around by beta [02:41] hopefully he will [02:41] Just about time then. [02:42] kirkland could u please upload new powernap? the fix os in the trunk [02:43] ScottK indeed :) === WinstonSmith_ is now known as WinstonSmith [03:01] New bug: #740666 in beautifulsoup (main) "package python-beautifulsoup 3.1.0.1-2build1 failed to install/upgrade: error writing to '': Success" [Undecided,New] https://launchpad.net/bugs/740666 === robbiew1 is now known as robbiew === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [04:45] tftp [04:49] ideaman: what abou tit [04:52] just trying to get familiar with it when it comes to using it to back up IP phone configs [04:53] just out googling and looking for good articles relating specifically to that [04:53] curl tftp://phone1/config.txt ? [04:55] so will that set it up so anytime the phone reboots, it'll pull that config file and overrite anything someone's tried to do? [04:57] Oh, the phone acts as a TFTP client? [04:57] yes. [04:58] Then you just need a read-only TFTP server for the phones to be pointed at [04:58] The setup I have now was done before I came along and tftp was already installed on my server, so I only know how to edit it after the fact [04:58] just learning how to do it from nothing [04:58] If you also use DHCP, then dnsmasq is the easiest way to do it [04:59] Yea, that's what I've been told, so everytime they reboot, TFTP sees the DHCP request, then loads it's config if it isn't already loaded [04:59] I don't know about dnsmasq though [04:59] That's not how it works [05:00] Yea, I know I'm missing pieces of it [05:00] or even the whole process [05:00] The phone is a DHCP client. It asks for a DHCP response. The DHCP response will contain "next-server" and "filename" options, which are e.g. 1.2.3.4 and /phone.conf respectively [05:01] Then the phone makes a TFTP GET request to 1.2.3.4 for file /phone.conf and then, presumably, applies it [05:01] ...at least, that's how it would normally work. Your vendor may be silly in some way [05:01] I am the vendor/IT guy so we're good [05:02] That makes alot more sense [05:03] so once I've apt-get tftp, how do I set it up initially so those linksys/polycom phones are looking for that .conf file [05:03] No, tftp is the client [05:03] option in the phone intially where I need to point the phone to the IP of the TFTP server? [05:03] You need a tftp *server* [05:04] Either dnsmasq or tftpd-hpa are good choices [05:05] I guess I just need to put that tftpd-hpa on my box and get familiar with it [05:06] It is not very different from an FTP server [05:06] In terms of setup, I mean. The protocol is quite differrent [05:06] It doesn't seem like there's to many parts [05:07] *too [05:08] Alright, thanks for your help, I'm outta here! [05:45] I was using itunes for my ipod classic. I there any software for syn. ipod with my ubuntu PC [05:45] ? [05:47] zee313 https://help.ubuntu.com/community/PortableDevices/iPod is probably a good start; otherwise I'd suggest that -server's not the most obvious channel for that, and you may have more luck in #ubuntu proper [05:50] For an second- or fourth-gen iPod, they were just HFS+ filesystems with a single metadata file. There is a simple tool to generate the metadata file. [05:51] Those gens were around in, like, 2004, so new ones are probably completely different. [05:54] I had an 'ipod photo', which was 4th or 5th. it was happy with vfat, but same idea for metadata. rhythmbox & co never had a problem with it tho [05:56] 4th was monochrome LCD [05:59] Oh, apparently there were 4G color units [05:59] Maybe I had a third-gen [06:01] According to http://en.wikipedia.org/wiki/IPod_Classic I had a 3G and a 4G. I know my first one had a non-mechanical scroll wheel (which was awesome). [06:04] I guess the good news is that they're still amongst the best supported on linux. the 'classic' line has remained largely stable since they went heard first into 'iOS' [06:06] Cool [06:44] goodmorning [06:52] * aliverius is going to rebuild a raid-0 array for the 1st time [06:54] <_ruben> raid0 doesn't really exist [06:57] Hi, i have hired a vps with ubuntu -server installed on it and suddenly i can't connect to it using sftp and can't import Sql database. [06:58] <_ruben> complain to the vps provider? [06:59] is it something caused by me or them? [07:00] You expect us to know? [07:01] _ruben: last time I looked, RAID0 existed [07:01] _ruben: OK, it wasn't *R*edundant, so I guess it's better called "AID0" [07:16] Can I upload files from my computer to the vps server using terminal? i am connected to it via ssh. [07:19] If you have shell access over SSH, then you almost certainly can use SFTP or scp. [07:19] twb: When I try connecting to it using sftp it just jeeps loading without showing me any files or folders. [07:21] *keeps [07:22] <_ruben> twb: exactly (wrt (R)AID0) :) [07:22] <_ruben> Syria: do the logs show anything odd ? [07:23] Syria: that is probably by design. You need to talk to your VPS vendor. [07:23] Thank you very much guys. [07:26] if they've given you a ssh account (on the host) that attaches to a tty (on the guest), it's probably locked down to do just that. it's usually intended to give you a last-resort access in case you goof up, not a usable sftp session [07:28] shauno: hum, fair enough [07:28] <_ruben> then either the vps provider changed something, or the initial problem description was misleading [07:28] * _ruben points at "suddenly" [07:30] the 'terminal' bit threw me off. the ssh my vps provider offer, goes straight to a stty. it's like having a serial console on a physical machine. [07:31] the sshd on the guest itself is my job, and the provider doesn't touch it [07:31] yeah right [07:34] was a big "if" I guess, but the most sensible reason I can think of for a lack of sftp [07:34] Sigh. [07:34] Which monkey wrote lucid's /etc/init/squid.conf [07:35] grepconf2()? I mean really [08:14] twb: It's from the init script, actually. [08:14] twb: It's been there since at least 2004. [08:14] Yeah, I guessed [08:14] FWIW I replaced it all with a normal upstart job and I haven't had any problems [08:15] http://paste.debian.net/111679/ [08:15] What prompted me to do this was that I rebooted my LXC server, and the squid container was hanging around for minutes -- preventing the entire system from rebooting promptly. [08:16] http://paste.debian.net/111675/ [08:22] New bug: #740777 in samba (main) "smbd.conf needs to wait for network up event" [Undecided,New] https://launchpad.net/bugs/740777 [08:37] Hello there, [08:38] how can i disable the system information "motd" after an ssh login for specific users? [08:42] ok.. i finally found it... just by adding .hushlogin [08:46] BlackDex: that information is in the fortunes-ubuntu-server package, which provides helpful hints in the MOTD :-) === ogra is now known as Guest42038 [09:42] Talking of which fortunes-ubuntu-server needs some love... === Guest42038 is now known as ogra_ [09:42] anyone want to have a go with it? :) [09:46] it's about adding new smart quotes to it? ;-) [09:50] when i installed ubuntu i chose a raid0 setup. i had one disk atm but it still accepted to do the raid0. now that i connected the 2nd hard disk, how easy will it be? is there an automated way? [09:58] aliverius, I'm not sure if mdadm allows to add drives to a RAID0 [09:58] (Google seems to say it doesn't) [09:58] what is your need, and have you considered using LVM striping instead? [09:58] <_ruben> because raid0 isn't realy raid, it'll work with just one disk just fine .. as for adding extra disks, probably: sudo mdadm /dev/m0 --add /dev/sdX [09:59] <_ruben> then again, raid0 is useless especially when running lvm on top of it [09:59] yep [09:59] raid0 is pretty useless in a lot of situations [10:00] and software raid0 is a waste of ressources in most cases (especially with one drive) [10:00] Hello, I can't edit anything in phpmyadmin, i have hired a vps with ubuntu 10.4 installed on it. the support team at the company told that i have to fix this. [10:00] <_ruben> i wonder what the person whas smoking who "created" raid0 in the first place [10:00] _ruben, well, LVM didn't exist at the time I believe [10:01] maybe the guy who created raid0 was running Windows and couldn't attach partitions/drives to mount points... [10:02] _ruben, there is a point about raid0 though, which is performance [10:02] <_ruben> raphink: that's not the point .. raid0 contradicts itself, it *can't* exist [10:02] <_ruben> you can stripe disks for performance, sure, but it's not raid [10:02] you have more writing heads on the same logical volume [10:03] I think calling it RAID Level 0 is exactly appropriate as a description of what it is -- the 0 matters :-) and it has to be part of the RAID family, for, e.g. RAID 10. [10:03] that said, you put the data of two (or more) disks at risk with raid0 [10:03] since if one disk crashes, you destroy and array and lose the data on both disks afaik [10:04] jkg, right [10:04] it makes sense to use it in this context [10:07] What should I do if I want to delete and re-install phpmyadmin and all the data bases please? [10:08] apt-get remove --purge phpmyadmin [10:08] But this will not remove the data bases right? i want to remove and re-install everything. [10:09] not sure if that removes the DB. If the DB is myisam, you can delete the files associated with it in /var/lib/mysql [10:09] raphink: I can't edit any value from phpmyadmin after clicking "Go" the page keeps loading so do you think that this might solve the problem? [10:10] certainly, bleaching the machine will not resolve the problem as far as clicking "Go" [10:10] heyyyyyyyyyyyyyyyyyyyyy did i say raid0???????!!! i meant raid 1 so let me repeat what i said [10:10] you won't have a "Go" button anymore if that helps ;-) [10:10] aliverius, I think you did ;-) [10:10] when i installed ubuntu i chose a raid1 setup. i had one disk atm but it still accepted to do the raid1. now that i connected the 2nd hard disk, how easy will it be? is there an automated way? [10:11] raphink: and how am i supposed to save to changes? [10:11] there you go, i corected it [10:11] aliverius, _ruben answered you then [10:12] Syria, I'm no phpmyadmin expert (I am more of a cli user) [10:12] raphink: Thnx anyway. [10:13] removing the database is probably not the best solution to your issue though [10:14] Syria: what error message, if any, do you get when it fails? [10:15] jkg: The page keeps loading and i don't get any messages. [10:15] Syria, is there anything in the apache logs? [10:15] ok guys thanks [10:15] i will come back with more questions i am sure [10:16] raphink: http://paste.ubuntu.com/584185/ this is the error.log [10:16] raphink: And this is the access.log http://paste.ubuntu.com/584186/ [10:19] Syria, there's not much there ... [10:20] raphink: Is there anything that I can do? [10:20] anything else? [10:21] if one drive fails and is removed will the other drive boot normally? [10:21] or is brub installed on only one drive [10:22] grub [10:22] does ubuntu know how to setup both grubs at the same time? [10:28] raphink: I have removed and reinstalled phpmyadmin using apt-get --purge-- remove phpmyadmin $ apt-get install phpmyadmin [10:28] but now i can't log in :( [10:41] Syria, you mean you can't log in on phpmyadmin, right? [10:43] raphink: Yes. [10:43] Syria, have you configured it? [10:44] yes, sudo dpkg-reconfigure phpmyadmin is this correct? [10:44] that, or looking at the config.php or whatever else it uses [10:52] jhunt: around for a quick upstart question? [10:52] jamespage: hi [10:53] jhunt - lemme just pastebin the config [10:54] jhunt: http://pastebin.com/1W0HDbLS [10:55] so this is the upstart config for jenkins [10:55] couple of things I wanted some guidance on [10:56] I noticed that sometimes a restart fails - it looks like the java process running jenkins was not fully shutdown before the new process was started. [10:56] do I need to use any extra config as the java process gets run through su? [10:57] second - also noticed odd stuff with overrides [10:58] so I use AJP_PORT in JENKINS_ARGS - however if I use an override file to set the AJP_PORT the value in JENKINS_ARGS is still -1 [10:58] i.e. from the jenkins.conf file [10:58] help much appreciated jhunt [11:03] hmm - not sure why the override isn't working. Are you sure you created it as jenkins.override (two "r"'s)? [11:03] I've just tried a quick test and it works for me. [11:05] to see the override file being applied, run, "sudo initctl log-priority debug". Then do, "touch /etc/init/jenkins.override" and look at the messages in /var/log/syslog [11:05] also, I'm a bit confused by the restart - how do you know the process wasn't fully shutdown out of interest? [11:10] Hello. I was told ny my manager to look into cloud computing and I have questions. If I want to deploy a Apache server that is going to take some heavy loads, would it be wiser to load balance it in a more "traditional" way (several servers running apache) or put up a Ubuntu cloud and deploy it there? [11:11] sorry for all the typos! [11:12] using cloud computing doesn't prefvent you from scaling "the traditional way" [11:12] you can choose one big instance or several small ones [11:12] it's still up to you to choose your architecture [11:12] But what would be easiest to scale when I need to expand? [11:12] but cloud computing will be more profitable with several small instances, since you will have more fine grain to scale up and down [11:13] cloud computing will make it easier to scale up and down, using small instances [11:13] because you can launch more instances when you need them, and shut them down when you don't [11:13] if you choose one big instance, you can't scale down and then it's probably cheaper to get a dedicated server [11:14] Can I run one big instance and scale it by just adding more nodes? [11:14] sure [11:14] heh? [11:14] you can't run an instance that is larger than one node [11:15] you can add more instances and load balance [11:15] I think that's what haaga means, no? [11:15] he clearly said, one big instance and scale by node :) [11:15] that's what he said ;-) [11:15] what do you mean haaga ? [11:15] ohh, so the instances can never be bigger than one node? [11:16] the advantage of cloud computing, is the ability to divide and reconfigure things, without reinstalling hardware [11:16] i meant that i'll put one apache instance in a cloud and then when the load increases, add nodes and let them be a rouserce to exisiting apache [11:16] not start new instances, just scale the exisisting one [11:17] no, you would have to have more instances running apache [11:17] and loadblaance them [11:17] you should think of an instance like a real computer [11:17] and that instance can never grow beyond the capabilities of the node hardware it's running on [11:18] now, what it will make nice though [11:18] humm, it would be sweet just to being to scale the instance over several nodes [11:18] if you need to make an instance more powerful, is installing a better node [11:18] and moving the instance to that more powerful node [11:18] without reinstalls [11:19] so, it's basiclly the same as vmware? [11:19] like using an EBS, detaching it from the old instance and attaching it to the new one [11:19] moving from one host to a better host [11:19] yep [11:20] although adding instances is more of the spirit afaik [11:20] raphink, depeds on the workload [11:20] right [11:20] duplicating cache memory is extreemly wasteful [11:20] don't the load balancers on aws support sticky sessions? [11:21] oh right, cache [11:21] ;-) [11:21] sorry [11:21] :-) [11:21] i would be sweet just to scale them over the nodes and when demand increases, just add more nodes and let the instance scale over them too? [11:21] it would* [11:21] haaga, not really [11:21] networks can't keep up with that speed [11:22] humm, well.. it would be sweet in a perfect network world [11:22] :) [11:22] and in that perfect world, you have infinite amounts of cash :) [11:23] ofcourse! [11:23] heh [11:23] this is why, if your workload permits, you just have puppet auto-configure a new backend instance, and the loadbalancer adds it into the pool [11:23] I know many places that do that [11:24] so, in my case, the only thing i would save on running in a private cloud would be power consumption? [11:24] going as far as to turn nodes on and off dynamically throughout the day, based on load [11:24] instead of buying several servers and just load balance them [11:25] and in my case, when running vmware, i wouldn't save anything? [11:25] heh? [11:26] if you run a private cloud, you still need to pay for the physical machines... [11:26] so I don't see how you would save power [11:26] unless you have several services to scale on it [11:27] Well, i was thinking that you proberly utilizes the machines more when using cloud/virtualization [11:29] jhunt: well the override is working - as long as I use it outside of a env declaration i.e. the use of a env declaration in an override within a env declaration in the main conf file does not seem to work. [11:29] well, it all depends on the workload [11:29] if you can balance cpu, ram, and disk usage [11:29] true [11:29] you can build one huge powerful node, and run it all [11:29] and it would be more efficient [11:29] jhunt: so with regards to the restart - I can still see messages being generated in the jenkins log from the old process while the new process is in the process table. [11:32] jamespage: well, as you haven't specified the "instance" stanza, Upstart won't allow >1 of these jobs to run, so I'm not sure. Can java apps fork? If so, Upstart won't be following the fork as there is no "expect daemon" or "expect daemon". [11:32] would it be possible to move the instances to a minimal amount of nodes when the load decreases and power off the extra nodes, then power them on a spread the instances again when load increases? [11:32] jamespage: you could convince yourself by maybe putting a "ps" in the pre-start and logging that to a file [11:33] jhunt: ok lemme try that [11:33] haaga, yes, even vmware does that, and ubuntu [11:34] $ sudo mdadm /dev/md0 --add /dev/sdb [11:34] and this is what i get http://pastebin.com/bMjMR1PQ . does this mean the array is being rebuilt? [11:34] raphink _ruben ^^ [11:35] and isnt it a bit slow? [11:37] jhunt: def still around - see http://paste.ubuntu.com/584235/ [11:37] <_ruben> aliverius: it's probably spitting all current blocks over the 2 disks, as for the speed, 38MB/s aint that bad for a single disk [11:38] jhunt: and when running - http://paste.ubuntu.com/584236/ [11:39] _ruben: and how i sync and use the array? [11:40] jamespage: so it's forked by the looks of it. Upstart doesn't know about that pid since no "expect" stanza [11:40] <_ruben> aliverius: the syncing is what the progress"bar" indicates and how to use it, well, that's up to you :) [11:40] jhunt: ok I'll try that [11:42] _ruben: excuse me but doesnt it r/w on both disks automatically? [11:43] <_ruben> it does [11:45] so what is up to me? [11:46] do i have to do any extra configuration? [11:47] also will both disks be bootable? [11:47] grub info is inside the raid partition [11:49] jhunt: hmmm - tried that out - not so greart [11:51] upstart seemed to lose track of things - start and stop just hung - tried expect daemon as well [11:58] <_ruben> aliverius: how you gonna use the striped volume is what is up to you [11:58] <_ruben> and they'll be bootable if you made 'em so [11:59] <_ruben> also, why would they both need to be bootable? [11:59] <_ruben> you'd boot from one, and if that one dies, you're screwed anyway, the 2nd disk would be pretty much useless [12:00] jhunt: ignore me on that first issue - must have made a type or something as its working now... [12:00] jhunt: still having trouble getting upstart to track the right process though. [12:04] _ruben: raid 1 [12:09] ruben, no, normal is to have both disks bootable [12:09] and if one disk fails, the system will boot from the second disk [12:10] but putting grub INSIDE the raid won't work, cause your bios doesn't understand the raid [12:11] when grub detects you are installing into onto a raid, it normally installed itself into the mbr of each raid disk [12:19] <_ruben> hadn't noticed the raid1 part in the pastebin, as all the discussions thusfar had been about raid0 [12:22] I didn't see anyone talking about raid0, except you :) [12:22] ^^ [12:22] binBASH: Error: "^" is not a valid command. [12:23] <_ruben> 10:50 < aliverius> when i installed ubuntu i chose a raid0 setup. i had one disk atm but it still accepted to do the raid0. now that i connected the 2nd hard disk, how easy will it be? is there an automated way? [12:23] hmm, I must of missed that line :) [12:23] _ruben: i corrected it a bit later [12:23] I wonder if it's possible to let uvirtbot execute a forkbomb :D [12:24] Hi patdk-wk btw. [12:24] heyyyyyyyyyyyyyyyyyyyyy did i say raid0???????!!! i meant raid 1 so let me repeat what i said [12:24] [12:10:19] when i installed ubuntu i chose a raid1 setup. i had one disk atm but it still accepted to do the raid1. now that i connected the 2nd hard disk, how easy will it be? is there an automated way? [12:24] ^^ _ruben sorry for the misunderstanding [12:24] aliverius: Error: "^" is not a valid command. [12:25] _ruben sorry for the misunderstanding ^^ [12:25] uvirtbot, shush [12:25] patdk-wk: Error: "shush" is not a valid command. [12:25] <_ruben> ah, that line got lost between the lines of a discussion i wasnt following [12:25] i am sorry :) [12:25] <_ruben> in that case, just configure grub to install itself in both mbrs and be done with it [12:28] _ruben: now, lets say i remove one drive. will it boot? and when i reinstall it, will i have to sync by hand? and whatever has changed in the first disk will be applied to the second? [12:28] <_ruben> aliverius: it will, *if* you installed grub properly on both drives, and if you add another disk again, you'd do the same as you just did: add it to the raid and it'll sync again [12:42] hggdh: are you ok to move over to the new jenkins server this afternoon? [12:43] hggdh: I will move the IP address but your slaves will need to be flushed of the offending slave.jar file [12:44] jamespage, no prob, all we need is a moment of peace ;-) [12:44] jamespage, actually, like now, if you want [12:44] hggdh: OK - I'm just going to grab a sandwich - OK if I start in about 45 minutes? [12:45] jamespage, certainly [12:45] great === jjohansen is now known as jj-afk [12:51] thanks a lot _ruben. i will test my setup to see how these happen :) [12:53] any managed to get cobbler working on ubuntu 10.04 ? === jdstrand_ is now known as jdstrand [13:25] <_ruben> aliverius: also, if you configured the raid in the installer, the installer has asked you what to do in case of a degraded array, unless that question is suprressed when only 1 disk is avail [13:26] no the question was there and i said it should boot from the degraded array [13:26] i supposed it would automatically use the good disk [13:28] morning all [13:28] hey hey hey! [13:28] RoAkSoAx: Morning :) [13:29] kim0: ready for clouds days huh? :) [13:29] RoAkSoAx: hehe indeed :) [13:29] I need some help fixing my server to re-allow ssh clients to connect. I think the issue is elated to "ipkungfu" (a firewall). I recently modified the settings to allow ssh (allowing incoming on port 22) and that didn't seem to help [13:30] anadon: iptables -I INPUT -p tcp --dport 22 -j ACCEPT [13:30] would that help :) [13:32] if I knew what they were...this is my first time getting a server off the ground [13:33] anadon: use sudo that command .. and see if it helps you ssh [13:33] kk [13:34] ipkungfu might mess it up again later though if it's badly configured [13:34] invalid command--doesn't like the "-I" [13:35] fuuuuuuuu [13:35] sec [13:36] ok, now it's just outright refusing my connections [13:37] anadon: sudo restart ssh [13:38] hggdh - I'm going to shutdown both instances now and re-sync the data. [13:39] "restart: unknown instances: " [13:40] anadon: can u try it now .. canit ssh ? [13:41] jamespage, OK, and I will reboot the systems [13:42] jamespage, ping me when the new instance comes up so that I can restart the slaves [13:48] kim0: still refused [13:52] hggdh: all done - you can access on http://jenkins.qa.ubuntu-uk.org or http://hudson.qa.ubuntu-uk.org (thanks Daviey) [13:53] hello? [13:55] hggdh: hold on that - something odd happened to the elastic ip - just re-doing [13:55] jamespage, ok, bringing up the slaves [13:55] anadon: http://paste.ubuntu.com/584289/ try to flush all the iptables rules with these commands. [13:55] jamespage, no problem, unbringing up the slaves [13:58] jamespage: i can't access jenkins.q.u.o, but can access hudson.q.u.o [13:58] thesheff17: did it, connections still refused [13:59] hmmm - something odd going on - looking at it now. [13:59] anadon: is ssh running ps aux | grep ssh [14:03] thesheff17: wha??? [14:04] thesheff17: newbie here, don't know about that or how to check, besides that in involves running "sevice" [14:05] hggdh, marjo: I think its OK now; both slaves have re-connected [14:06] jamespage, yay! Thank you dear sir, forever in your debt [14:06] jamespage, which means you get a free beer in Budapest :-) [14:06] hggdh: I will keep an eye on it this afternoon - the elastic IP kept dropping its association with the instance. [14:06] that was weird [14:07] I will shutdown the old instance but leave around on disk for a few weeks. [14:09] anadon: was ssh working before? if it was you have a config problem. If it was never working make sure it is installed. apt-get install ssh. Then make sure the service is running by using ps aux | grep ssh === dendrobates is now known as dendro-afk [14:14] that option that has turned up in the installer in 10.04, install security updates automatically, is there a way to dpkg-reconfigure that to be enabled? I didn't select it on a new install and It's on a faily unimportant machine [14:14] i changed the settings in /etc/export. how do i restart the daemon? [14:14] would prefer it to apply security updates automatically.. can't find anything about it anywhere (that being said my search skills are miserable ;) [14:14] for nfs4 [14:15] thesheff17: it was working before and through my firewall install, but every connection after that hasn't worked. the thing is, I stopped the firewall service and it's still not working. [14:17] thesheff17: it outputted, but I don't know how to read it. [14:18] found the solution in the wiki [14:18] anadon: you should see /usr/sbin/sshd -D [14:18] hggdh: now I would really like to get some ec2 tests into that jenkins instance..... [14:19] if anyone has worked on cobbler can you look as this paste bin for me? It's just be installed from the fedorahosted git repo - the one the ubuntu team were commiting to in Jan. http://www.pastie.org/1704023 [14:25] thesheff17: ok, but what does it mean or what do I do with it? [14:25] jamespage, good idea... === dendro-afk is now known as dendrobates [14:27] When compiling a project that uses the perl headers, I get EXTERN.h : No such file or directory, even with libperl-dev installed. the makefile also calls "perl -MExtUtils::Embed -e ccopts" in compilation. Can anyone help? [14:27] anadon: well if you see /usr/sbin/sshd -D that means the service is running correctly. Why it is not accepting your connections I don't know. I would check your networking stuff. Especially your default gateway. [14:29] how do i let with '/etc/export' ,any host to connect to my nfsv4 server? [14:41] hi everyone, i have two problems, first one when i start apache2, i get this libcrypto.so.0.9.8 no version information available === twister004_ is now known as twister004 [14:44] the second i can't connect to the server with ssh http://pastebin.com/7ysDq3M3 [14:44] argh I taek it back I found! https://help.ubuntu.com/10.04/serverguide/C/automatic-updates.html [14:44] felimwhiteley: is that for me? [14:45] upp: no sorry I was solving my own problem from earlier [14:45]