[00:07] can someone check if my apache works by clicking this ? http://96.246.33.167/ [00:07] It says it works [00:08] how do i set my ip to static [00:08] JanC, thanx [00:10] man interfaces [00:12] hm, that manual could use some better examples ☺ [00:13] vladimir_: https://help.ubuntu.com/9.04/serverguide/C/network-configuration.html [00:26] wheres the log to read what would make server make an internal speaker beep, and have networking apparently fail for 5 minutes or so, gmpc and ssh failed to connect. during this time, now its back up like normal. i see nothing in /var/log/messages or dmesg. [00:27] zroysch: You could also try syslog to check on network messages [00:28] zroysch: As for what makes the speaker beep, I don't think that's logged anywhere. [00:29] it was a single beep, like it would do on a reboot [00:29] but it has apparently not rebooted, according to uptime [00:29] One thing you can do to test your network is leave a ping or 'mtr' running. After the network drops and comes back up, check your ping or mtr and see if they show packet loss. [00:30] not seeing much in /var/log/syslog [00:30] zroysch: Maybe someone else can chime in on the uptime, but if your system clock was off and then the system rebooted, it should have been corrected by ntp...not sure if uptime is based on that... [00:34] I have a drive running on LLVM, but it's a single volume. [00:35] for some reason, people configuring Ubuntu do this a lot. [00:35] I need to pull the drive out, and put it in another box. how do I read the data? [00:35] s/LLVM/LVM/ [00:44] nevermind. lvresize. [01:34] Morning all. I've got an internal HDD that's fat32 that just will not share with users. I've chmod -Rv 0775 /media/Storage and made sure fstab is vfat defaults 0 0 .. still it is locked read only to users. What else can I look at? [01:54] NineTeen67Comet: Is that an external drive you plug in to your machine, or a internal drive that was there when Ubuntu was installed? [01:55] Internal regular IDE .. used to have MS installed dual boot on that server, now it's all Ubuntu-Server but too much on that drive to just convert it back to xfs .. [01:56] Can you copy the contents of /etc/fstab and the output of the mount command to http://dpaste.com and give me the URL? [01:58] http://dpaste.com/114838/ .. [01:58] Last line .. [01:58] The line commented out was another attempt too .. [01:59] That looks correct. Can I get a copy of the output of the 'mount' command too? [02:00] (A command I love is: apt-get install pastebinit && mount | pastebinit) [02:00] darkpixel: you mean: sudo mount -a .. or if I mounted it manually? [02:01] I didn't know about pastebinit .. that's awesome for a new guy .. I don't come around much but this one is driving me nuts and Google hasn't told me I didn't already know .. :) [02:01] If you just type 'mount' it will spit back a list of the current mounted file systems and their options. [02:01] Gotchya .. [02:02] dropping pastebinit on there .. nice stuff .. love handy tools .. [02:02] Yeah--it saves a few clicks [02:03] http://pastebin.com/f20d5083e .. friggin' awesome .. [02:04] Ok--so it looks like your drive is mounted correctly, and it's not read-only according to the kernel... [02:05] Shows rw on that too .. I just don't get it .. [02:05] Let me see if I understand this correctly: You are able to read/write, but other user accounts are unable to access it? [02:05] No users can write to that directory/hdd .. just sudo (root) .. [02:05] (Even though fstab says rw, it's possible for a drive to be mounted ro if the kernel thinks it's corrupted) [02:06] You mean /media/Storage? [02:06] aha .. could be .. [02:06] does fat32 have the same issues as ntfs if it wasn't shut down properly? [02:06] Yeah /media/Storage is /dev/hdb1 (Only partition on that hdd) [02:06] What I meant by that is that it's possible for fstab to say 'rw' but the mount command tells you the actual current state of the drive. [02:06] It could. [02:07] can I use fsck to check vfat? [02:07] Yes--but it doesn't appear necessary as the drive is mounted correctly. [02:07] trying fsck.vfat /dev/sdb1 now .. [02:08] yeah it's got me stumped .. doesn't show full, doesn't show root only and it shows rw in mount and in fstab (I've always used defaults 0 0) .. [02:10] I'll be in the states mid December, I can swing by and physically touch the box .. Maybe something is wrong with the hdd, but that's would effect all write not just users .. === Nicke__ is now known as Nicke [02:12] NineTeen67Comet: Sorry, I was afk for a minute--my five month old took a big tumble. Can you show me the output of 'ls -lha /media'? [02:12] Yup, just a moment, rebooting that box .. [02:13] Also, how are users accessing the data? Are they linux users logged into the box through SSH or the console, or are they Windows/Samba users? [02:14] Just ssh and on the box (he can login via X when he wants - I installed ubuntu-desktop for him too - ) most of the time it's just ssh and nfs but haven't gotten nfs up since we can't even write to it as it is .. you know? lol .. [02:16] nfs will be his primary way of accessing it once it's up .. it'll be his networked storage box most of the time (he just got a netbook (I put 9.10 on) that is actually his daily use computer) [02:17] Ok--do an 'ls -lhad /media/Storage | pastebinit' and that'll give me the permissions on the Storage folder. [02:17] drwxr-xr-x 16 root root 16K 1969-12-31 16:00 /media/Storage or http://pastebin.com/f38a1abf2 [02:19] That last r-x means everyone should have read access to the Storage folder. If you want everyone to be able to write to it, you need to 'chmod o+w /media/Storage' [02:19] If you want to include subdirs, make that 'chmod -R o+w /media/Storage' [02:19] that might have been it .. I've only been doing chmod -Rv 777 /media/Storage .. [02:20] We'll know momentarily .. lol [02:20] I would stay away from the numeric chmod's unless you understand the bitmasks. [02:21] yeah .. I've only got an entry level understanding of the numbers .. I have a calculator that shows me the output for what ever you want .. read/write user, group etc .. [02:21] The parameters are u g and o (for user group and other) followed by either - + or = (to remove, add, or explicitly set) permissions which can be r w and/or x (read/write/execute) [02:23] This is probably a good starting point to understand the bitmasks: http://danielmiessler.com/study/unixlinux_permissions/ [02:23] Thanks .. it's opening now .. [02:26] Awesome. From there NFS should be pretty easy. If he will be connecting in from the same IP or subnet (like both machines being on a home LAN), you just create (or edit) /etc/exports to have a line that says '/media/Storage ip.add.res.s(rw)' [02:26] drwxr-xr-x 16 root root 16K 1969-12-31 16:00 /media/Storage Yeah, no change .. this is driving me nuts .. lol [02:26] Was that after unmounting and mounting the drive? [02:27] Yeah my house is full of Ubuntu stuff (mythbuntu, desktops and notebooks .. all NFS) .. never had this issue .. [02:27] I rebooted it .. [02:27] and that's the resulting code .. [02:27] NFS has been my friend for ever, samba so/so but only when people are here with MS stuff .. [02:28] Yeah--I'll have to dig for a minute on that one. The system sets /media/* with special permissions, but I don't remember where it does that. [02:29] Actually, since it's an internal drive, you may want to consider mounting it somewhere outside of the /media folder. Maybe /mnt or create your own folder. I believe Ubuntu puts special permissions on the /media subfolders. [02:30] Yeah I can toss a mount point in /mnt (from my Gentoo days :) .. I'll try that .. [02:30] Ok. I'm willing to bet it'll hang on to those permissions. I have a 2 TB RAID 6 array mounted at /data and the permissions stick every time. [02:31] ...although I reboot it about once every few years when the next LTS comes out... ;) [02:32] darkpixel: My web server used to run pretty much from Kernel update to Kernel update .. I'm on an Air Force base now and Verizon sells us our ADSL and they don't allow anything like a web server .. so it's sitting in my server closet humming along with no users .. sad .. and I hate Verizon .. lol [02:33] NineTeen67Comet: Yeah--I know there are a lot of Comcast haters too, but they treat me pretty well. I shell out about $60/mo and I get their 20 down/5 up business service. No hassles about bandwidth and I get a block of 5 IPs from them. [02:35] When I was offbase we had Fiber .. that was supplied with PPPoE (sucks) but the speeds were GREAT and they gave me everything I neeed .. it was around 8,000 yen a month (basically $80) .. Now I pay $100 for phone and ADSL and it's slow, unreliable and did I mention restrictive as hell .. [02:38] lol--stupid telco monopolies. I used to work for a non-telco ISP years ago. I loved our service. We brought DSL into the area several months before the telco even got to it. The best part is was over their own lines because we were a CLEC. [02:38] Same same .. permissions read the same, can't write (sudo can) and it is mounted /mnt/Storage .. I've got no idea where to look next .. Might just wait until I get there in December and rebuilt the box ground up (it's an old Microtel from Wal-Marts quick attempt at a sub $200 computer that I added ram, hdd and video to) [02:38] Did you change /etc/fstab to point to the new mountpoint? [02:38] Yeah .. [02:39] UUID=4568-C05A /mnt/Storage vfat defaults 0 0 [02:39] Even tried /dev/sdb1 ........ [02:40] Skip using the mnt directory. Do a 'mkdir /Storage', set the permissions correctly on that folder, change fstab, 'mount /Storage', verify the permissions (fix if necessary), then reboot and see if it still works. [02:40] Is it normal for the UUID to be so short? On the rest of my stuff it's always been pretty long .. [02:40] I almost never touch /mnt, so for all I know it may have some strange permissions too. [02:40] k [02:41] No, I was wondering why it was so short... [02:42] Ubuntu used to have a 'vol_id' command (or something similar) that would return the correct UUID for the partition, but it doesn't appear to be on my system. [02:42] I used it a few months ago, can't remember what it is called now .. but it was pretty handy for fstab if you were going to move hdds around .. [02:44] Aah--it's been replaced. The new command is 'blkid'. You could try 'blkid /dev/sdb1' and find the correct UUID for fstab that way. [02:44] But using /dev/sdb1 in fstab works just as well at the moment. [02:45] hi got problem i instaled ubuntu 8.04 amd64 when i update i end up file being corrupted cannot download, is the us mirror have problem on connection, with i386 its fine when i test it.. [02:45] NineTeen67Comet, I think this is relevant, though unsolved: http://ubuntuforums.org/showthread.php?t=324524 [02:45] darkpixel: yeah spits out the same number anyhow .. no biggie .. [02:45] Weird. I've never seen a UUID that short. [02:45] ditto [02:46] ruben23: Have you tried running an 'apt-get update' again (or from the GUI, clicking reload)? It may have cached some invalid information. [02:46] We bought that HDD new about 3 years ago .. [02:47] coincidentally, that post is about 3 years old [02:48] NineTeen67Comet: I have no idea how bklid gets it's unique ID--that thread seems to suggest that it's something in vfat that limits the ID to 2 bytes... [02:48] darkpixel: im using CLI, im not on gui [02:48] darkpixel: tried several times, ive restarted the server, same thig [02:49] darkpixel: no problem I put it back to /dev/sdb1 .. still has the same issue .. Might have to bring another hdd with me, move all his stuff to it, reformat that one to something more nix friendly (like ext3) and press on .. [02:49] darkpixel: are there option i can change my mirror, currently its us ubuntu.. [02:49] ruben23: What is the name of the corrupted file? [02:49] darkpixel: I'm headed off to help the wife and kids (I've got a 6 month old (7yr old and 4yr old too) .. so I feel your pain .. take care and thanks for the time .. I appriciate it .. [02:50] NineTeen67Comet: The permissions getting reverted like that wouldn't be a problem with the drive. Have a good night. [02:50] darkpixel: I'll see if he's got an external I can have him hook up so I can move all his stuff off then back on after I reformat to something besides fat .. [02:50] g'night . (almost noon here) [02:51] darkpixel: ill check it 2 seconds [02:51] ruben23: If you edit the file /etc/apt/sources.list and change 'us.archive.ubuntu.com' to something else like 'archive.ubuntu.com', 'nz.archive.ubuntu.com', etc... and then to apt-get update, apt-get upgrade again it may fix it. [02:52] But I'd be curious what file is actually corrupted. It's somewhat unusual that a mirror has a corrupt file. [02:54] ruben23: can you tell me which IP you are hitting? [02:55] I run the mirror that is currently hosting us.archive, and I'm very interested in tracking any errors down [02:55] maswan, got any stats like cacti or mrtg? for curiosity sake [02:55] http://www.acc.umu.se/technical/statistics/ftp/monitordata/ [02:55] openNMS for the win [02:55] hammurabi and napoleon are the ones doing us.archive duty [02:56] darkpixel: sorry, was not able to cehck the file, but im on a remote are now..as soon as ill be in office ill message here [02:56] thx maswan i love graphs [02:56] maswan: ill also cehck on that [02:57] darkpixel: but thus squid proxy contributes to the problem, i have it on my network.. [02:57] maswan it is safe to say most recent days spike is because of 9.10 ? [02:57] ruben23: You might try clearing the squid cache. If it somehow dropped a connection mid-download, you may have an incomplete file cached. [02:58] ive isolate it tried shutting down the appication and got a reboot, but still same thing.. [02:58] vraa: yeah, well, the first spike is that. then we took over us.archive and that's the second (sustained) spike [02:58] ruben23: Alternatively, if you are able to bypass squid, that would be a good test. I've run into this issue before. [02:59] ruben23: An IP and URL would be very appriciated, so I can check if it is a fault on our side or if it was just a random occurance getting squid:ed [02:59] darkpixel: yes shutdown the squid application si what you mean right..? done it and tried again..same error, actually it got update hit, but on the last part..the error comes up [02:59] maswan: Love the graphs. Am I reading it correctly that you're peaking around 530 MB/sec? [03:00] maswan: yes ill do that...asap [03:00] ruben23: After squid it shutdown try 'apt-get clean' to clear your apt cache and then try it again. [03:01] darkpixel: yes, we only have a 4 x gigE uplink (brezhnev is on another switch though) [03:02] maswan: Nice bandwidth. The largest pipe I've ever worked with was an OC3 [03:03] darkpixel: That's the computer club, work is http://stats.nordu.net/stat-q/load-map/ndgf,,traffic,peak <- those are all 10GE links [03:05] maswan: Beautiful. Man, I gotta move out of the US. The lack of bandwidth is annoying. After all the telco talk and taxes back in the 80s and 90s, I should have 10GE to my house... ;) [03:06] darkpixel: Well, I only have 100mbit to my house. [03:06] maswan: Ooh--never noticed that before. You're IPv6 enabled. [03:06] maswan: It must be tough... ;) [03:07] In .au you get a maximum download cap, and a theoretical maximum speed of 22mbps, but effective only about 4mbps. [03:07] twb: I read about that a few months back. That seriously sucks. At least where I am, Comcast caps home connections, but if you pay $10-$20 more per month you get an uncapped business connection. [03:08] And that's in the middle of the city. Country areas have satellite or 56kbps lines. [03:08] twb: Can they get ISDN at least? [03:08] Dunno. I haven't heard of ISDN being widely used in .au. [03:08] But then, how WOULD I hear from people in the country? [03:10] twb: lol. I had a friend that (until two years ago) had a box out at the road that had a WebRamp so he could shotgun two 56k modems, and there was a WAP in there that shot wifi back to his house. He regularly used it to play Halo2 on Xbox Live. (Every time he hosted a game, we lost because he was so laggy...) [03:18] where can I talk about uec, ubuntu enterprise cloud ? [03:29] axisys: Probably this channel, but I won't be of any help since I haven't touched it yet. [03:30] anyone installed UEC? i cannot login to the webgui with admin/admin === debio264_ is now known as debio264 [03:46] hmm.. it is working now [03:47] i guess some virtualbox glitch === marcelcohrs is now known as cupCakeBoy [04:12] hello, i just installed ubuntu server in a virtual machine (virtualbox) as a guest. i installed it in a super-minimal way and now i want to add an ssh server to interface with it. i tried sudo apt-get install openssh-server but it installs a whole lot of X11 things, i guess just for x11 forwarding. [04:13] can i install an openssh server that is built without X11 so i can avoid picking up all of these X11 packages? i'd like to squeeze as much as i can out of the hard drive space and also avoid having to go through all these X11 things during updates [04:14] something equivalent of the nox versions of emacs [04:17] Ganymede: what X11 packages did you install with openssh-server? [04:18] snth: openssh-server recommends xauth, if that's any help [04:19] And xauth IS just for X forwarding. [04:19] You'll note that while x11-common and friends are quite a lot of packages, they all up only total a few MB of space. [04:19] hmm...you're right, maybe i should looked closer [04:19] You can also opt out of recommendations using aptitude -R or apt-get --something [04:19] i just saw it install a whole bunch of things so i assumed it was pulling in much of Xorg with it [04:20] but it looks like very few X11 packages actually [04:20] X *clients* never pull in X servers [04:20] If they do, that's a bug and you should report it [04:21] libx11-data, xauth, libxmuu1, libxcb1, libxau6, x11-common, libx11-6, libxext6 [04:22] i don't remember if libxkb-data and libxdcmp6 were there before [04:22] Shrug [04:29] twb: I didn't know about the -R for aptitude. This is cool. I am looking up the whole apt.conf stuff now. Thanks :) [04:30] oh no, apt-get install recommends by default, it seems [04:30] but i am too rebellious to accept its recommendations [04:30] Ganymede: -R tells aptitude to NOT treat recommendations as dependencies [04:31] Do you guys know of a url listing statistics about the current linux/unix server deployments? [04:31] snth: yes, but i've been using apt-get thus far, and i never noticed it was installed recommendations for me [04:32] Something like what's the precentage of the servers using Redhat vs freebsd vs ubuntu-server vs whatever. [04:32] snth: i will be using aptitude -R in my deployment scripts from now on [04:32] Ganymede: Sometimes recommendations save you some headache :). But sure. [04:33] Ganymede: the Debian policy manual basically defines Recommends as being appropriate for an opt-out policy. [04:33] snth: netcraft used to do metrics like that for webservers and browsers [04:33] The problem is that until very recently, apt-get treated it as opt-in, so lots of stuff that belongs in Suggests was dumped in Recommends because DDs didn't know any better. [04:38] "DDs"? [04:39] Debian developers [04:39] fallous: I found the numbers for linux vs microsoft vs BSD on netcraft [04:39] I wish they would have provided a bit more detail about the distributions. [04:39] twb: thanks for this information [04:40] snth: does it really matter who's using what? [04:41] twb: I don't care about WHO is using what. I wanted to know what is being more than what :) [04:41] s/being/being used/ [04:44] Shrug [04:54] surprised nginx outnumbers lighttpd by so much, although I guess china telcom is distorting that quite a bit [05:04] It's only people running webservers anyway [05:27] jo letz [05:27] is how do icheck in ubuntu if curl is added to my php? [05:28] without using the phpini [05:28] sorry [05:28] phpinfo [05:51] cupCakeBoy: dpkg -l php5-curl [05:51] aaaah [05:51] thank bro [05:51] i love linux [05:51] SOOOO easy to istall packages extensions etc [05:51] its a dream that came true [05:51] and i never wanna wake up [05:51] ;) [06:41] Do you guys know if there are any plans to support Dracut in Ubuntu? [06:48] Never heard of it [06:48] Nor has Wikipedia. [06:49] http://fedoraproject.org/wiki/Dracut [06:50] They have a link to the man page. [07:28] "Unlike previous initramfs systems, Dracut aims to have as little as possible hard-coded into the initramfs." [07:29] Sounds like Fedora's playing catch-up [07:29] Ubuntu's initramfs already uses udev and is customizable by packages and by sysadmins [07:29] "It's likely that we'll grow some hooks for running arbitrary commands in the flow of the script" [07:30] ...Ubuntu already has that. [07:30] twb: I honestly don't know much about that. Glad to learn though. Thanks for the info. [07:31] It's possible I've misunderstood, but it sure looks like they're trying to reinvent initramfs-tools [07:32] Fuck. I just realized the bookkeeping I spent the last four hours doing has a mistake right at the start. [07:38] h how do i copy content files from my /var/www server1 to another local /var/www server2 with rsync through ssh, what command should i used.. [07:50] ssh server1 rsync -a /var/www/ server2:/var/www [07:50] Assuming you can ssh directly from server1 to server2. [08:15] is there a way to install phpmyadmin without requireing apache ? *i allredy have lighttpd setup and going* apache didint useto be a requirement of that package :( [08:24] Zodling: phpmyadmin doesn't require apache. It only recommends it. You can use aptitude -R to ignore the recommendations. [08:25] in 9.10 if you use aptitude to install phpmyadmin it will force install apache [08:25] oh err sorry didint read it all [08:26] thanks ;) sorry agan [08:26] :) No problems. [08:28] didint do the trick :( [08:28] aptitude -R install phpmyadmin [08:30] Use pastepin :) [08:30] kk [08:30] It'll only install apache if you let it pick the default choice for libapache2-mod-php5 | php5-cgi | php5 [08:30] s/pastepin/pastebin/ [08:31] http://paste.ubuntu.com/306499/ [08:48] Zodling: try something like aptitude install -R php5-cli phpmyadmin [08:48] Zodling: or using aptitude's GUI [08:49] Even aptitude install -R --schedule-only phpmyadmin; aptitude --schedule-only purge ~ainstall~napache; aptitude install -R might work [08:56] no dice :( [08:58] Zodling: just use the GUI [08:59] yah giveing that a try now [09:05] i think i got it in the GUI thanks :) pity it cant be done by script so ez anymore :( [09:13] Zodling: try this. sudo aptitude install -R phpmyadmin apache2: php5-cgi [09:14] I meant to have the colon. Try it as is. [09:18] Of course, even if it works you'll lose, because you're doing php and mysql [09:18] o.0 why is that ? ;( [09:19] Zodling: This simply says don't install apache2 and choose php5-cgi as alternative to libapache2-mod-php5 [09:19] Zodling: I think that would work for your script or whatever. :) have fun. [09:19] Ah, I see I mistyped php5-cgi above [09:20] Night guys [09:20] yah thanks ii i was asking why about the php mysql part ;) [09:20] Because they're software written by idiots for idiots. [09:21] twb: pretty big claim right there buddy. I'll keep this conversation for another night :) [09:21] haha . good night guys. [09:22] eah.. if you say so [09:22] later snth [09:30] i have a server powered on, but i forgot its ip address, is there a way to send a ping to every computer of my network so that it would appear with "arp -a" ? [09:30] arp -l ? [09:31] twb, it returns nothing [09:31] Hm, maybe just "arp" [09:31] lemme find a real machine that I can test it on [09:31] twb, i meaan this syntax is invalid [09:31] Yeah, if you just run "arp" without arguments, it should emit the entire arp table [09:32] But I guess you realize that. [09:32] You could do a broadcast ping, but I dunno if that would populate the arp table. [09:33] twb, how to do a broadcast ping ? [09:33] ping -b -c 1 192.168.155.255 [09:33] It actually seemed to work here [09:34] Where 192.168.155/24 is your network [09:46] difference betweem lightweight resolver and a regular one [09:46] ? [09:47] twb, thanks [09:57] Fenix1: resolving what? [10:00] I mean are the terms used for each other or are they different matters all together [10:00] synonyms for each other* [10:06] New bug: #468197 in samba (main) "ufw blocks samba in karmic default installation" [Undecided,New] https://launchpad.net/bugs/468197 [10:06] :s [10:07] hm? how's that a samba bug? [10:08] Fenix1: I have no idea what you're talking about. [10:08] it isn't, it's probably a ufw bug [10:08] yeah, but now I'll have another email about it [10:08] kblin: if you can't handle useless email, get off the internet [10:08] :-) [10:09] twb: it's not as if I didn't get enough real bug reports :) [10:09] erm, it sounds like it isn't a bug in either of them. it sounds like a configuration issue to me. [10:09] well yeah [10:09] * kblin nods [10:10] but the question is, which package sets that configuration [10:10] IMO it's a *good* thing if services only bind to 127.0.0.1 by default [10:10] that's a bit pointless to run samba then [10:10] It prevents admins from getting distracted and forgetting to tighten it [10:10] Because you can forget to tighten interface binding, but not forget to loosen it (because it obviously won't work) [10:11] twb: I don't even want to know how many bug reports we'd get then [10:11] kblin: I have no problem with alienating users who can't read a debconf prompt [10:12] I'm trying to figure out how I should report this bug in VLC [10:12] kblin: yes, a little ott perhaps :p [10:12] not server related, it's not preventing the screen saver [10:12] twb: and I don't have an extra person triaging bugzilla for all these NOTABUG bug reports [10:12] i think they way things are now, that most packages come with a reasonably secure, but real-world config out the box is nice [10:12] DrNick_: the actual problem case I have in mind is OpenSSH [10:13] Which allows passwords from anyone from anywhere OOTB [10:13] go too far the other way we risk becomming like OpenBSD - so paranoid *nothing* ever works without tweaking [10:13] twb: having that bind to 127.0.0.1 is not a good fix, though [10:13] kblin: hand waving [10:14] kblin: it doesn't specifically have to be done by interface binding [10:14] tbh the better fix would be to set it to do key-based auth only per default [10:14] The important thing is that it services don't talk to any remote hosts until the sysadmin explicitly says "yes, these hosts should be allowed to use this service" [10:14] kblin: yeah, that'd be a damn good start [10:14] of course there's a bootstrapping problem there [10:15] kblin: or having debconf prompt for an AllowUsers [10:15] because usually I first use my password once to copy over a ssh key, and once that works, I turn off password-based auth [10:22] are the terms lightweight resolver and resolver in conjuction [10:22] used in conjuction that is [10:22] Fenix1, what do you mean by resolver? [10:22] DNS [10:23] presumably a lightweight one is smaller, faster, or uses less resources [10:23] okok [10:23] like dnsmasq or the like [10:23] i.e. not BIND [10:25] Is there any ETA on PCC 9.10 Server release? [10:26] what's PCC? [10:27] powerpc... [10:27] oh, ppc [10:27] :D [10:27] damn [10:27] oh well little typo [10:27] I thought PPC wasn't officially supported by Ubuntu [10:28] sorry, no idea if/when it's planned [10:28] there [10:28] is desktop and alt releases :) [10:28] from 9.10 [10:28] s0liddi: that doesn't mean it's supported [10:28] i know [10:28] It could just be some cowboy within canonical wanted his old macbook to work [10:30] i think it'll pop out someday :) [10:30] well, you could always install desktop and purge ubuntu-desktop, networkmanager, etc [11:23] if i upgrade my kvm server to karmic will app armour stop my vm's from running as i keep them under srv? [11:28] hello [11:28] I've just upgraded to 9.10, and my mail server isn't allowing connections since then [11:31] I got it, that was because my /etc/resolv.conf was pointing at the router [11:32] New bug: #468343 in autofs (main) "Automounted NTFS disks with Unicode in label results in invalid path after upgrading to 9.10" [Undecided,New] https://launchpad.net/bugs/468343 [11:53] hi all [11:53] I'm having an issue with getting my i.p. address static [11:54] I have edited the /etc/network/interfaces [11:54] fil [11:54] file* [11:54] and the /etc/hosts [11:54] file [11:54] but no joy :( [11:55] is there anything else I need to do? [11:59] Sorell, what does your interfaces file look like? [12:00] # This file describes the network interfaces available on your system [12:00] # and how to activate them. For more information, see interfaces(5). [12:00] # The loopback network interface [12:00] auto lo [12:00] iface lo inet loopback [12:00] # The primary network interface [12:00] auto eth0 [12:00] iface eth0 inet static [12:00] address 192.168.1.8 [12:00] netmask 255.255.255.0 [12:00] network 192.168.1.0 [12:01] broadcast 192.168.1.255 [12:01] gateway 192.168.1.1 [12:01] dude, pastebin [12:02] i bet you could well still have dhclient running [12:03] lololol [12:03] what is paste bin? [12:03] comment out the auto eth0 [12:03] www.pastebin.com [12:03] ty [12:03] and [12:03] http://pastebin.ubuntu.com/ [12:03] sigh [12:04] do i dare upgrade my kvm server to 9.10 [12:04] takes 9-10 hours right now [12:04] i just upgraded 2 servers, but they are just compile nodes [12:04] I'm trying to make my server secure by only allowing administration services from localhost. I'm using a ssh tunnel to access things like phpmyadmin from other computers. If I use lynx to go to localhost directly on my server it works. If I try to use a browser through the SSH tunnel it doesn't work. I've flushed all my iptables rules. I got this working before on my old ubuntu server. Any suggestions? [12:05] seems a little over kill to me [12:06] Baversjo did you check the apache error log? [12:06] nope let me try that :P [12:06] cuz you have to use stunnel to go to port 443 [12:07] ala udo stunnel -c -d 80 -r https-for-this-eg-web-page.co.uk:443 [12:07] sudo* [12:07] setup the ubuntu apache ssl [12:08] bam, secure like a bank [12:08] i thought he was just getting his ssh tunnel settings wrong [12:08] [Sun Nov 01 12:48:51 2009] [notice] Apache/2.2.12 (Ubuntu) PHP/5.2.10-2ubuntu6 [12:08] [Sun Nov 01 12:57:16 2009] [notice] caught SIGTERM, shutting down [12:10] Let me try something and get back :P [12:10] k [12:24] do I need to uninstall dhcpd3 as well? [12:25] to get the static ip working? [12:25] Sorell did you comment out the auto line and then do a "sudo /etc/init.d/networking" restart/ [12:25] yes [12:26] all that did was make it so that when I do an ifconfig I see nothing. :( [12:26] just go here [12:26] http://www.cyberciti.biz/tips/howto-ubuntu-linux-convert-dhcp-network-configuration-to-static-ip-configuration.html [12:27] or [12:27] http://www.ubuntugeek.com/how-to-set-a-static-ip-address-in-ubuntu-810-intrepid-ibex.html [12:27] plus you never pastebined your config, so i dont know how you expected anyone to fix it [12:29] I will type it into paste bin one sec. [12:35] http://pastebin.ubuntu.com/306647/ [12:35] k [12:36] do you need dns servers? [12:37] I am using my router. [12:37] as one. [12:37] no [12:39] http://pastebin.ubuntu.com/306652/ [12:44] :) TY [12:45] yw [12:51] New bug: #468461 in mysql-dfsg-5.0 (main) "cannot install mysql-server 5.0 on Ubuntu 9.10" [Undecided,New] https://launchpad.net/bugs/468461 [13:02] can anyone here help me with a small postfix problem? [13:02] !ask | dinger1986 [13:02] dinger1986: Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) [13:04] ah ok sorry [13:04] Nov 1 11:59:49 floserver postfix/smtpd[16575]: NOQUEUE: im just setting up a postfix server and i know this is very basic but i cant send from a remote machine what do i need t add the config to allow me to do that? i get reject: RCPT from host212-183-132-77.uk.access.vodafone.net[212.183.132.77]: 554 5.7.1 : Relay access denied; from= to= proto=ESMTP helo= [13:04] in the log file [13:12] anyone know? [13:20] dinger are you using it as your smtp server or sending from one smtp server to another/ [13:22] the gist of it being you need to authenticate with the smtp server [13:23] cuz its not an open relay by default [13:24] http://www.postfix.org/SMTPD_ACCESS_README.html [13:24] I'm trying to make my server secure by only allowing administration services from localhost. I'm using a ssh tunnel to access things like phpmyadmin from other computers. If I use lynx to go to localhost directly on my server it works. If I try to use a browser through the SSH tunnel it doesn't work. I've flushed all my iptables rules (Using policy accept for input, forward and output). Getting the following TCP error when I try to connect using [13:24] mysql query browser: " No connection could be made because the target machine actively refused it" over the ssh tunnel. Getting the same TCP error when I try to connect to localhost in Firefox over the ssh tunnel. I got this working on my old ubuntu (using 9.10 now). Feels like there's something about permissions or iptables... [13:29] sshd is on port 443 === AndIrc is now known as Guest5651 [13:35] No changes in apache access.log or error.log when I try to connect. [13:37] Baversjo did you get http working? [13:37] Ok w8 I got it working now (almost). [13:37] Baversjo http://www.stunnel.org/examples/mysql.html [13:38] stunnel has to listen on one port and redirect to another [13:38] The tunnel is working I'm tunneling mysql query browser. [13:38] The only thing now, is the apache configuration. I'm not so good at that [13:38] bah i give up you are all over the place [13:39] Just saying that I got the tunnel working ;) [13:40] bogeyd6: i am looking to use it as the main smtp server [13:41] i would be quite happy to have it as an open relay, obviously with usernames and passwords tosecure it [13:47] bogeyd6: Got it working now :D Problem was that the tunnel was resolving DNS locally. Now it's resolving on the server :) When I connect to localhost I get to phpmyadmin and when I connect to my hostname outside lo I get to the regular homepage YEY :D [14:16] dinger1986 open relay is bad, mmmmkay [14:16] users should be set to authenticate to use SMTP [14:17] can it be authenticated by the email doman or email address which is sending? [15:13] hello to all. i have just installed gallery on my ubuntu and would like to know how i can run it as a vhost ? [15:15] xperia2: Do you mean, how to configure a virtual host in apache? [15:16] snth: no. i have allready installed mutiple vhost on the server and they run just fine. now i have installed gallery with apt-get install gallery and cant acess gallery becouse of the vhost [15:17] i need to setup a vhost for gallery but i dont have the folder [15:18] maybe i need however a vhost for the /var/www folder only. this way it would work [15:19] but i fear that it will have conflicts with the other vhost site as they are also in the /var/www folder [15:22] it looks like that the ubuntu gallery package can not be runned as a vhost. could it be that true ? [15:32] xperia2: I have never used it before, but it seems that you have vhost pointing to /var/www/albums [15:32] xperia2: /usr/share/gallery/setup is where you set it all up I guess. [15:33] yes the path for the install folder is right [15:33] but i dont think /var/www/albums is the right path for the vhost [15:34] as i dont have any scripts in this folder [15:34] the folder /var/www/albums is empty [15:34] /var/www/albums is created when you install this package. [15:34] hmm [15:34] yes that is right but for me this folder looks like that is only a storage folder [15:35] xperia2: are you sure that it doesn't get populated after you run your setup? [15:36] well for setup i need it to call it from the web browser if i am not wrong [15:36] and exactly here maybe is the problem [15:37] what i am asking me is first if i needmaybe to create a vhost with the path /usr/share/gallery/ [15:38] till yet however i have allways used only the folder /var/www as vhost path folder [15:38] xperia2: what's in /etc/gallery/apache.conf? [15:40] snth: hmmmm this loooks very interessting. it has even a example for vhost [15:40] let me put it on pastebin [15:41] http://pastebin.com/d130c1052 [15:42] xperia2: Cool :) so, that's it. [15:44] snth: thank you a lot. i will change now the file and restart apache ! [15:44] now it must work ! [15:46] New bug: #468809 in vm-builder (universe) "Support adding custom debs" [Undecided,New] https://launchpad.net/bugs/468809 [16:01] New bug: #468837 in dhcp3 (main) "dhcp3-server could not be installed/updated during the distribution upgrade" [Undecided,New] https://launchpad.net/bugs/468837 [16:26] New bug: #468881 in postfix (main) "package postfix 2.6.5-3 failed to install/upgrade: " [Undecided,New] https://launchpad.net/bugs/468881 [16:34] hi all. I've just rebooted my dev computer with ubuntu-server on it and it's trying to check my 750GB filesystem ("primary superblock features are different from backup, check forced") [16:34] it's been about 15 minutes and it's only gotten 20% through. I need this system up now, but can't find a way to cancel the check [16:46] New bug: #468928 in openldap (main) "no prompt to enter the admin passwort upon package installation" [Undecided,New] https://launchpad.net/bugs/468928 [16:47] cameron: Unfortunately, you're stuck. Your filesystem is corrupted. Continuing to run without checking/fixing the errors could corrupt data even further. [16:48] darkpixel: i'm pretty sure it's from doing a fs resize. Honestly i don't care about corrupting my filesystem further.. i'm mostly trying to check if another hard drive is broken or not [16:49] cameron: I feel your pain. I have a 2 TB RAID 6 array at home. Checking takes the better part of an hour. And you can't skip it. I know it's been discussed in the past to let people skip the automatic ~32 mount check, but I'm not sure where that left off. [16:50] i've just had one of my machines start refusing to boot :( [16:50] the hard drive isn't being detected on the BIOS, on either drive controller [16:50] swapped cables and all that, and i think it's broken.. annoyingly it's 1 year 2 days old [16:52] hard drive = the slowest and least reliable part of a computer. [16:52] thus... raid. [16:55] hi, I installed UEC and all goes well, but when I want to install an image from the Store tab in the webinterface, I get the error message "error 28: connect() timed out!" [16:56] is this a known issue or temporary situation? [16:57] I can't see which server it actually tries to connect to [17:19] hi [17:19] amule is using many and any port, how can i set a rule in iptables to make it send its packets from a given NIC ? [17:40] I have sshd for sftp installed and I have a user/password that matches my user/password on the ubuntu server... i am guess that they are not the same because when I set priviledges as the ubuntu user I cannot see them via sftp... how do i set up sshd to match my ubuntu priviledges? [17:51] bdelin88, sshd uses system privileges === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [18:01] qman__: hmm [18:01] wonder why my user wasn't able to view the folder then? [18:36] can postfix be authenticated by the email domain or email address which is sending? [18:36] [18:46] I'm trying to reinstall mysql-server and mysql-client. I don't get the directory "mysql" in /etc. Why? [18:50] I am using the runtime cd and need to figure out if ubuntu will recgonize netgear or belkin adapters without any additional modications (out of the box) [18:50] I uninstaled mysql-server. When I reinstall it I don't get the /etc/mysql directory. Why? [18:52] can I use netgear or belkin wifi adpaters with the runtime ubuntu cd? [19:24] is there anyway to allow emails sent from a specific domain to be relayed without putting in all the ip addresses? [19:25] Domains are forged easily [19:25] Are you using Exim as your MTA? [19:25] no using postfix [19:26] do i dare upgrade my kvm server to 9.10, if i upgrade will i have to modify my apparmour config to support my setup? [19:26] i have put in smtpd_recipient_restrictions = permit_mynetworks, check_relay_domains, check_sender_access hash:/etc/postfix/sender-access [19:26] into my main.cf file but its still not working [19:40] dinger1986: If you do this based on a domain (for example: example.com), I can easily set my mail program to pretend to be 'example.com' and start spamming through your server. There are really only two good ways to relay mail. Either put the IP address of the mail server or machine that needs to relay, or require users to authenticate. [19:40] well they will be autheicated by a password as well [19:41] Then don't use 'check_sender_access', use 'permit_sasl_authenticated' instead. That pretty much says "Anyone who uses a valid username/password can relay mail" [19:41] darkpixel: do you know how to tell postfix to send this way? [19:42] i tried that, it stops working [19:42] dinger1986: Do you mean you have one postfix server that needs to relay mail to another postfix server? [19:43] no i want to be able to send from my mobile phone via postfix === Adri2000 is now known as Guest44663 === Authority is now known as Guest70752 === tonyyarusso is now known as Guest1320 === Guest1320 is now known as tonyyarusso === Guest44663 is now known as Adri2000` === Adri2000` is now known as Adri2000 [20:00] dinger1986: Sorry, not sure what happened to freenode, but it kicked a lot of people. Anyways, the easiest way to get authenticated SMTP with postfix is to also setup POP3/IMAP. I install dovecot on my servers. Once dovecot is installed and setup, you can point Postfix to authenticate off it's username/password list. [20:00] ok [20:01] i managed to finally get it to send from my laptop but still struggling to send from the phone but ill get it [20:01] i am getting Helo command rejected: Host not found; [20:01] It's somewhat annoying, but Postfix doesn't support authentication out of the box, it requires you to setup another system that uses authentication and connect the two. [20:03] That's usually something in smtpd_helo_restrictions like reject_invalid_helo_hostname [20:04] If you have an 'smtpd_helo_rejections = ' line in your main.cf, copy and paste it in here and I can see what's blocking you. [20:11] darkpixel: i have smtpd_helo_required = yes [20:11] [20:12] done have smtpd_helo_rejections [20:12] sorry nt done meant dont [20:20] dinger1986: That's a bit odd. If you don't have any helo restrictions, postfix shouldn't be complaining about a host not being found... [20:21] hmmm well this is from my logs Nov 1 20:16:58 floserver postfix/smtpd[26970]: NOQUEUE: reject: RCPT from host212-183-132-19.uk.access.vodafone.net[212.183.132.19]: 453 4.7.1 : Helo command rejected: Host not found; from= to= proto=ESMTP helo= [20:22] You might try 'sudo apt-get install pastebinit && cat /etc/postfix/main.cf | pastebinit'. That will post your main.cf file to the pastebin website and return a URL. If you post that URL here I can take a look at the config and help you. [20:22] That's strange that it says helo= unless your machine is actually named 'Inbox' [20:25] http://pastebin.com/f705028d8 [20:26] Two changes: smtpd_recipient_restrictions = permit_sasl_authenticated, permit_mynetworks, check_sender_access hash:/etc/postfix/sender-access, check_relay_domains [20:26] heya! can somebody help me setting up ip aliasing? [20:26] And delete: smtpd_sender_restrictions = reject_unknown_hostname [20:26] its not working, even despite it should ... [20:26] I need ideas to troubleshoot [20:26] Hello everyone.. I've been all day trying to work this out, but without success.. I installed Ubuntu Server 9.10 with egroupware. Now i'm trying to install a pear component with pear install Auth_SASL, but after when i do pear list it says not installed. Can anybody help me please? [20:27] vertigo: have you restarted apache? [20:27] Vertigo: Have you done an /etc/init.d/apache2 force-reload? [20:28] dinger1986 and darkpixel: i only did apache2 restart.. will try force now. [20:28] dinger1986: Beat me to it. If you make those two changes and restart postfix, you should stop getting that error. Also the 'relay_domains' parameter should be deleted. [20:29] what relay domains parameter? [20:29] dinger1986: http://pastebin.com/f705028d8 (line 48 in the dump you sent) [20:30] dinger1986: Sorry I can't stay and assist. I just got word my uncle is in the hospital. Gotta drive 2 hours to go see him. I'll be on later tonight if you still need assistance. [20:30] aye sorry i had deleted that anyway as i thought it was wrong [20:30] ok thanks alot [20:33] any ideas on the ip aliasing issue? I have centos5 vmware server 1.x host with a range of ips. two ubuntu vms that work fine. I tried to add ip alias to one ubuntuntu guest, but it doesn't work [20:33] when I try to reach the ip, the host answers, instead of ubuntu guest [20:34] peugi: How did you add the ip alias? [20:35] snth: sudo ifconfig eth0:0 192.168.1.11 up [20:35] also in the interfaces file [20:35] and promptly restarted the service [20:36] peugi: did you add a routing rule? [20:36] snth: I guess not? I didn't know I had to [20:37] cat /proc/net/aliases [20:37] What's the output? [20:37] snth: you mean something that the ISP adds to the routers? [20:38] snth: or something I have to do on the centos5 host? [20:38] peugi: http://www.faqs.org/docs/Linux-mini/IP-Alias.html [20:38] oooo let me try ... [20:42] snth: I added the routes, but still nothing [20:42] snth: localhost sees its own ip address, but can't access it from the outside ... [20:43] hmm [20:43] pastebin your config files. [20:43] /etc/network/interfaces [20:46] snth: http://pastebin.com/m2cd13af1 [20:46] the two ips are on different ranges [20:47] but my centos5 host sees them all [20:49] When you ping 124.28 does it come from centos? [20:50] yes [20:50] I mean I tried to ssh [20:50] peugi: is it up? what's output of ifconfig? [20:50] peugi: IM me. [20:54] snth: sent you IM [21:12] Ah, this is really fun. I have a server that has a simple built in SSD drive and a 1TB drive connected via USB. Installing /boot to SSD and the rest of thesystem to the TB drive - I have a feeling GRUB is going to not going to be very happy as previous attempts lead to grub error 2. [21:30] Ash-Fox: that is a very simple configuration, grub will have no problem with that === dendroba` is now known as dendro-afk [21:33] I just want to let everyone know that snth is DA MAN!!! [21:34] ? [21:34] * peugi bows to snth's l33t networking skillz :) [21:34] peugi: haha .. thanks. [21:36] kids remember, *never* alias an ip twice! [21:36] :) === dendro-afk is now known as dendrobates [21:42] ikonia, considering I'm just getting error 22 and error 2, depending on how I switch things around in the BIOS boot priority, I don't think it's so simple. [21:44] Ash-Fox: What's your grub.conf look like? [21:45] I have no idea. The netboot installer doesn't seem to let me mount anything in the terminal when I Try to do things manually. [21:45] Ash-Fox: good tip is to not change the bios order, it messes grub up [21:45] ikonia, it was messed up before changing it. [21:45] Ash-Fox: yes, but changing it is adding more complexity to it [21:46] I have on occasion had success with getting things to boot that way. But in this case, I think I'm just going to have to disable the built in ssd. [21:51] Ash-Fox: when you try to boot off of the SSD, what error do you get? [21:56] error 2 [21:57] What's the file system of your /boot partition? [21:58] I set it to ext2 [21:59] I /think/ I know what happened, I think the system is trying to boot off the built in cardreader (which has nothing in it) on the motherboard. [21:59] Ash-Fox: did you try grub2? I had this old dell and switching to grub2 fixed my boot problem [22:00] Nope, but I'm confident I'll figure this out soon enough. I just hate dealing with weird configurations like this. [22:00] Ash-Fox: Yeah, make sure you are actually booting off of the ssd that has the boot partition. This can definitely be it. === dendrobates is now known as dendro-afk [22:03] I believe what is happening is that it is booting off the SSD, but the grub configuration is trying to read the card reader, although it's quite difficult to find out exactly what grub is doing since I can't get into it's config yet. I need to stick a live linux on a usbkey. [22:10] testing... quiet here! [22:12] Quite amusing, it seems to be getting the better of me. Time to try the LVM trick. [22:31] anyone running karmic yet? is it worth doing a complete reload and setup ext4? [22:33] peugi, only if you need the additional performance [22:34] personally, I don't trust my important data to ext4 yet, maybe in a year or two [22:34] but it's great for non-critical filesystems [22:39] Hello everyone. I'm having some difficulty installing pear auth_SASL in my ubuntu server.. after a pear install auth_SASL, it keep saying that it's not installed. Can anybody help me please? [22:39] Is there a cache of DNS information that would not be cleared by restarting BIND? Yesterday I was having a heck of a time trying to access something, and 'dig' returned the result I wanted, but 'w3m' still went to the wrong place. [22:40] tonyyarusso, dig doesn't use the standard way of resolving things, but nslookup does [22:40] not sure exactly how it works, but I think there's a local cache [22:41] Oh. Didn't know dig was different. Lovely. [22:41] I didn't try nslookup, so I don't have that data poitn. [22:41] *point [22:43] of course I need the ext4 performance gains ... who doesnt :) [22:44] and the data corruption issues ext4 had in testing phase is making me cringe :) [22:47] peugi, ext4 is still getting bug fixes, though they're much more minor than the data corruption problem [22:48] perhaps I should put my home directory on ext3 and the rest on ext4, to speed stuff up? [22:48] I want my firefox to open 1 second faster :) [22:49] peugi, then don't use firefox ;) [22:49] haha [22:49] but yeah, ext4 is fine for most things but I wouldn't keep any important data on it [22:49] or at least keep good backups [22:50] whats a good webkit browser [22:50] epiphany? [22:51] hmm, I think in 9.04 epiphany still uses gecko ... [22:51] as a backend ... [22:53] correct [22:53] 9.10 changed it [23:11] Hello everyone. Can you help me understand why my ubuntu server is not installing pear auth_SASL? [23:11] root@neptuno:/# pear install auth_SASL [23:11] downloading Auth_SASL-1.0.3.tgz ... [23:11] Starting to download Auth_SASL-1.0.3.tgz (5,724 bytes) [23:11] .....done: 5,724 bytes [23:11] root@neptuno:/# pear list auth_SASL [23:11] `auth_SASL' not installed [23:11] root@neptuno:/# [23:11] Any help is very appreciated.. [23:12] that's a neat trick [23:13] sorry, I can't really help [23:13] qman__: thanks anyway. [23:46] !info php5-sasl [23:46] php5-sasl (source: php-sasl): Cyrus SASL extension for PHP 5. In component universe, is optional. Version 0.1.0-1 (karmic), package size 9 kB, installed size 88 kB