[00:01] Dont understand === _TechAway_ is now known as _Techie_ [00:06] Anyone around that is proo on proftpd [00:08] eriksson25: what's wrong with vsftpd? aside from not being able to restart [00:10] Well, I have been using proftpd for a long time. But since updating to latest the tls support dosent work. So I looked at vsftpd but its virtuel user suport is so messy to set up. And since I only need to get tls to work on proftpd for everything to be set up I wanted to give it a other chanse === _Techie_ is now known as _TechAway_ [01:00] eriksson25, I'm not proo on proftp, but the upstart message is fairly clear - its an "improvement" on the standard service starting mechanism [01:00] I have one instance of proftp working on a server , but its a simple setup === zz_sailerboy is now known as sailerboy === _TechAway_ is now known as _Techie_ === _Techie_ is now known as _TechAway_ [01:36] where do /dev/md_d*p* come from? I have /dev/md_d3 and /dev/md_d3p[1-4] for some reason, and I can't get rid of them. [01:36] mdadm --stop /dev/md_d3 will work, but I can't get rid of the p1-p4 [01:47] electrofreak: you want to remove the array? [01:52] martin-, I don't know where it came from.... so..... yes [01:56] how is the release date looking for Maverick? :) [01:58] electrofreak: mdadm --detail /dev/md_d3, mdadm --stop /dev/md_d3 and mdadm --zero-superblock the devices that are part of the array [01:59] md_d3p[1-4] are just partitions on md_d3, so they should disappear when you stop /dev/md_d3 I think [02:00] * fluvvell is annoyed that when he hit enter on "Commit partition changes and write to disk", it suddenly started installing the root system before he has his raid1 partitions allocated [02:01] martin-, they didn't disappear.. :-/ [02:03] these might be coming from some external HDD I used to play with a linear "raid" [02:03] I thought I zero'd out the whole drive when I saw done... maybe not [02:04] if you zero the superblock on the raid members and reboot, they should be gone [02:04] I'm assuming you have backups of course :) [02:05] martin-, hah, yea.... ideally. My storage volume kinda out grew the external drives... [02:05] I think I was working on playing with btrfs + compression and stuff... never got back to it though... [02:06] hehe [02:07] so, is 10.10 going to be a 2.6.35 kernel? or 2.6.34? [02:07] I haven't been keeping up with it [02:08] I have 2 drives installed: /sda and /sdb, what is dm-0??? [02:08] I just installed 10.10 desktop on my laptop, which is .35 [02:09] Dravekx: dm-0 is a mapper device used by lvm, dm-crypt etc. [02:09] martin-, ah... === harrsonk is now known as harrisonk3012 === harrisonk3012 is now known as harrisonk === harrisonk is now known as harrisonk_away === Datz_ is now known as Datz [02:41] Hi all... I'm restoring my server after having to rebuild my RAID array... I was wondering if anyone could direct me to information for ensuring that GRUB2 will recognize my new mdadm RAID array as well as what I need to do to update mdadm on the system. [02:48] aegis, the initram should have what is needed for RAID... which is built when you install a kernel... [02:48] if you install mdadm... I think by default it'll detect your array(s)... [02:49] electrofreak: the problem is I'm restoring my system from backups... the backups most definitely have different UUID's for the raid arrays since these are new. [02:49] edit /etc/mdadm/mdadm.conf [02:49] I can probably edit /etc/mdadm.conf [02:49] right [02:50] but I still think grub is going to throw a fit when I go to boot the system... actually, I think grub may not even pop up to be honest... [02:50] did you run grub-setup? [02:51] IIRC the default behaviour is to scan all partitions for md arrays, and activate any it finds. [02:51] or grub-install [02:51] Ahhhh... upstart change. :( [02:51] That is, in an initrd that was built after mdadm was installed. [02:53] electrofreak: I just rebuilt the arrays using sysrescuecd and am in the processes of restoring the filesystem from backup onto the array... That's where I am right now... When it finishes, I will have a RAID 1 and RAID 10 array with my system as it was 3 days ago... [02:53] Sorry, I tell a lie. In 8.04, at least, mdadm.conf is copied into the ramdisk and appears to only assemble pre-defined arrays. [02:54] electrofreak: So should I try to rub grub-install or grub-setup from the rescuesyscd I have up now? [02:54] twb: So do you think if I just change the UUID's in mdadm.conf that it will boot? [02:54] aegis: you will need to change the UUIDs in the *ramdisk* mdadm.conf. [02:54] I would chroot into your restored install, then run it from there. [02:54] UUIDs can be found with blkid [02:55] aegis: you can do this by editing the chroot and running "update-initramfs -u -k all", or by doing it by hand with cpio [02:55] AHHHH, great idea [02:55] You can, of course, simply tell mdadm to create the array with the old UUID [02:55] I forgot about chroot... interesting... that opens up some possibilities. [02:56] twb: the arrays are already created though and LVM installed on top... the system is being restoring onto the lvm as I type... [02:56] something like mdadm assemble /dev/md0 /dev/sd[ab]1 --update=uuid --uuid=XXXX:XXXX:XXXX:XXXX [02:56] One of my failover products works by making the failover host have the same UUIDs as the "real" host. [02:57] twb: that's interesting... that sounds like that might be the easiest thing rather than to hunt throughout my system to find all the places it might be referenced... [02:58] electrofreak: so when I chroot into the system, are you recommending I run grub-install? [02:58] aegis, well... you need a bootloader.... so yes === _TechAway_ is now known as _Techie_ [03:00] electrofreak: so, grub-install versus update-grub? [03:01] either might work [03:01] cool... I'm just researching the update-initramfs command twb recommended as well... [03:01] aegis: you'll want to do it for the LV names and filesystem UUIDs, too [03:02] aegis: you can't fix the MACs of the NICs, though, so you'll want to edit or delete /etc/udev/rules/*persistent-net.rules [03:02] twb: okay, I'm a little confused now... what will I want to do for the LV names and filesystem UUID's now? [03:02] Oh -- and if you make UUIDs match, NEVER EVER EVER put both the old and new disks in the same box at the same time. [03:03] twb: these are the same disks [03:03] aegis: make them match the old box (or hunt down references to the old UUIDs and update them, e.g. /etc/fstab) [03:03] twb: the disks were just rebuilt into a new array [03:03] aegis: oh, you're restoring from tape or something? [03:03] twb: from tar... basically... [03:04] okey dokey [03:04] twb: so if I delete those rules, will they be recreated somehow? [03:05] twb: I did use mostly the same LV names... I figured that would be the easiest thing to fix by editing /etc/fstab anyway... [03:05] The persistent-net.rules file is generated by the persistent-net-generator.rules file in the same directory (or in /lib/udev/rules.d/ in newer versions) [03:06] twb: okay, the nic thing shouldn't be a problem though right? it's going in the same box... so the MAC for the NIC should be the same... [03:06] twb: do you agree? [03:06] Yeah [03:07] what does the -u -k all switch do for update-initramfs? errr, I'll just read the manpage. ;) [03:08] W: A error occurred during the signature verification. The repository is not updated and the previous index files will be used. GPG error: http://extras.ubuntu.com maverick Release: The following signatures couldn't be verified because the public key is not available. [03:08] Magic [03:08] LOL [03:08] oh, so that will affect all the kernals... [03:08] kernels rather [03:08] -k all is basically because you're probably running a live CD so "uname -r" is the wrong mojo [03:09] s/is/gives/ [03:09] twb: but even if I'm chrooted? uname-r should work for the chrooted system, right? or will it go back to the host? [03:09] Dravekx: you need to install the appropriate key [03:10] Dravekx: for official archives, that's usually "apt-get install foo-archive-keyring" or so -- for PPAs and such, you'll need to use gpg(1) and apt-key(8). Join #debian-bots and /msg dpkg apt-key XXXXXX, where XXXXXX is the key hash mentioned in the error. [03:10] uh oh... my tar just gave me an "tar: exiting with failure due to previous errors"... :( [03:10] aegis: chroot doesn't change the kernel [03:10] aegis: and now you've just learned why tar is a bad archive format [03:11] twb: any idea how to figure out what errors ? [03:11] Not offhand [03:12] no problem... thanks for your help though! I'm closer to getting this fixed... if I can figure out what files had errors I might get there. :) [03:15] Er, what tar said is "I stopped unpacking partway because I had a problem" [03:15] (I think.) [03:16] I'd be diffing the lslR you generated at archive time with the current one. [03:16] twb: roger that... I'm just trying to find out what the problem might be... if it is the fact that I created certain directories already prior to extracting, that's fine... but if it lost files, that's more serious... [03:17] twb: not exactly sure how to do that... sounds like a good idea though. They should be almost identical [03:19] Well, it assumes you did an "ls -lR / >/lslR" or so before the problem occurred [03:19] That is, because you made the tarball [03:19] That is, BEFORE you made the tarball [03:19] well, it's a backuppc archive [03:19] I'm not familiar with backuppc [03:19] I created a tar from that [03:22] I guess you could try something like "diff -u <(find / -xdev -ls | sort) <(tar tvf /dev/rmt | sort)" [03:22] Where / is the chroot dir and /dev/rmt is your tarball [03:22] The trick will be to get the line format to match so diff doesn't have a whole lot of false positives [03:30] twb: I'm trying to just chroot into it now... not working out so well... lol Backuppc did give me this report when creating the tar: bin$ /usr/share/backuppc/bin/BackupPC_tarCreate -t -n -1 -h localhost -s '/' / > /mnt/data1/try/restore.tar [03:30] Done: 119909 files, 50597613663 bytes, 18531 dirs, 9994 specials, 0 errors [03:44] anyone have a tutorial on auto mounting a second drive on startup? [03:46] Dravekx: put it in /etc/fstab [03:48] twb, do I need to add the UUID for that? or can I simply specify /dev/sdb ? [03:49] Either will do [03:49] great. [03:49] But note that drive names are increasingly allocated dynamically [03:50] so the UUID would be a better recommendation? [03:51] If you can assume udev (which you can for Ubuntu), then UUID or LABEL is more future-proof than drive name === sailerboy is now known as zz_sailerboy [03:53] ok thx [03:54] i'm trying to find where my databases are stored in sql. the only way i've ever been able to transfer them is through export feature in phpmyadmin. is there a way i can manually recover them? [03:58] knolls: mysql databases typically live in /var/lib/mysql [03:59] Obviously you cannot simply copy the files while the database is active, because they're incoherent. [04:00] Either stop the database before copying the files, or use the mysql-specific dump command. #mysql or the Ubuntu Server Guide are probably good places to learn about that. [04:04] oh woes of 10.04 server. Can anyone tell me what the virtual machine support is that is referenced in the install ? [04:05] fluvvell: probably KVM [04:06] Are you looking at the tasksel screen in d-i? === zz_sailerboy is now known as sailerboy [04:07] way past that twb, have a running system. So much is different to hardy. [04:07] I've botched the raid install, was trying to do it by hand [04:07] Might be better to start again. [04:08] Well, I think I may have figured out part of my problem... I was trying to chroot from a 32 bit live cd into a 64 bit system... I think that probably doesn't work so well. [04:15] Wait, my raid1 array has changed its name, its not /dev/md0 anymore, its... /dev/md_d0 ? and only has one element Whaaat ? Is this normal behaviour? [04:15] FSVO well = at all [04:16] fluvvell: oh, that. You're screwed. I don't know how to stop that [04:16] fluvvell: basically what's happening is that post-install, the two RAID1 partitions are recognized as a single RAID1 *disk* with two partitions *inside* the array [04:16] I "solved" that problem by installing 8.04 instead === sailerboy is now known as zz_sailerboy [04:20] twb, :-S [04:21] twb, I've managed to --auto-detect and start the original array, but the second hard drive is the one thats showing up as the md_d0 and md_d1 [04:21] doesnt mdadm have a way of knocking off an array ? [04:26] I don't know, man [04:27] I guess I need to find the changelog of software raid. [04:27] Why do they screw around with stuff like this without telling us????? [04:28] Because where Debian is run by a committee of conservative pedants, Ubuntu is run by a handful of Canonical employees [04:30] So they Just Do Stuff, and sometimes it works great, and sometimes they go "oh, oops, that actually totally breaks one in eight machines. Oh well, they aren't desktops" [04:30] Not that I'm bitter... [04:30] no, not a bit lol. [04:31] Having seen how well per's conservative startpar work WORKS on Squeeze, I'd put all of upstart into the latter bucket [04:32] err, ok. I'm not really following debian squeeze, but there is times when ubuntu's pressing forward really bugs me too. btw, I just tried the grml disk that someone recommended the other day, and .. well it left me less than satisfied. [04:33] particularly that I couldn't chroot into my system, it grizzled on about zsh stuff [04:33] Well, grml is a fork of Debian [04:33] yeah, got that part. [04:33] Try Debian Live, [04:33] it's *not* a fork [04:34] yeah, well I'm either past that... 10.04 is booting (very quickly I might add) but just the raid stuff is a pig. [04:34] so I either reinstall, or [04:34] Oh I might as well. [04:43] great, the raid configuration utility in the cd is clearlyl screwed. [04:49] so, my file server locked up again last night, and even though I have auditd installed, there was nothing in /var/crash [04:50] am I right in assuming it pretty much guarantees hardware fault? [04:54] I'm not familiar with auditd [04:54] But obviously if you have e.g. a kernel panic, it will be hard for it to then write stuff via the filesystem [04:55] I don't think it's a kernel panic, but I don't know for sure, it just drops off the network with a blank screen and no keyboard response [04:55] any time I've had a kernel panic on screen the keyboard flashed too [04:55] but this one just halts, basically [04:56] one time it responded to sysrq commands to sync and reboot [04:56] this last time it didn't [05:02] hey everyone [05:02] anyone know anything about how to setup a PXE network boot server? If so do you know a 'how to' so I can learn how to do it myself? [05:03] how do I keep iftop from figuring out that my local IP address is myserver.local without the -n option because I still want to see the reverse lookups on the remote hosts, I just need to distinguish between each IP address while looking at it, because now they are all lablled "servername.local" [05:03] reggi: apt-get install di-netboot-assistant [05:04] it used to do what I wanted, just showed my the local IP address that was making the connection but it randomly decided it wanted to start just telling me that useless "servername.local" crap [05:04] demonspork: edit /etc/hosts or wherever your nsswitch.conf is configured to reverse-resolve those IPs [05:05] hey twb thanks for replying. i'll check it out - by the way, do you have experience with network booting? [05:05] so I could put a name on them? could I name them after themselves? like " 1.2.3.4 1.2.3.4" [05:06] reggi: http://prisonpc.com <-- is me [05:07] IOW I run some diskless netboot farms [05:07] demonspork: er, no. You'd remove the entry entirely [05:07] demonspork: of course, that will almost certainly break OTHER things, that DO need to reverse-resolve your local IPs [05:08] twb - nice work man [05:08] you could just pipe the results through a sed for easier reading [05:08] qman__: eh? [05:09] Oh, ntop. I think ntop is a curses application, not a stream. [05:09] oh, ok [05:09] And I *think* his problem is that he has mapped all his addresses back to a single name, so he can't tell which is which [05:09] so why would it have suddenly changed from just seeing the IP address to resolving to the servername.local [05:09] twb, yeah, that is what is happening [05:10] e.g. his ntop has both 127.0.1.1 and 10.0.0.1 mapping to wankfest.example.net [05:10] I can't tell which is which [05:10] ah [05:10] iftop actually [05:10] well, the solution would be to unmap them or map them to different names [05:10] but why would it have changed? [05:10] demonspork: you could also change the mapping such that the canonical name is something like 1.1.0.127.wankfest.example.net [05:10] but you need to make sure that won't break anything [05:11] either your name resolution has changed, or that function was broken for some reason and an update fixed it [05:11] even BandwidthD used to give " 1.2.3.4 - Configure Reverse DNS for this IP" [05:11] but now it says "servername.local [05:11] " [05:12] well, the solution is to modify your name resolution so that the different addresses have different names [05:12] if I used the /etc/hosts file does the name being applied have to be a fully qualified domain name, or does it not matter [05:12] depends on what your applications want [05:12] apache needs an FQDN [05:13] hey twb are you free for a couple of minutes so I can pick your brain on something related to network booting? [05:13] does a FQDN need to have a true reverse lookup set up on the IP, or can I just set it in the hosts file and leave the actually reverse DNS setup for later [05:14] By convention, one would put in /etc/hosts "IP FQDN [ CNAME ... ]" [05:14] demonspork: that depends on the application [05:14] reggi: ask away [05:14] thanks twb [05:15] twb, i'm a computer technician, work from home. i do lots of virus repairs / windows reinstalls. [05:16] i would like to be able to network boot a computer and have the option (in a menu) to choose to install xp, vista etc over the network, or run diagnostics (UBCD). [05:16] my current setup is as follows... [05:16] 1 gateway which acts as the DHCP [05:17] 1 linux laptop (ubuntu desktop 10.04) [05:17] I don't know about installing Windows or UBCD over the network, but otherwise, that's possible. [05:17] a couple of pc boxes. [05:17] ok twb, my question is this [05:18] is it possible to setup the network boot server on a virtualbox machine (on my linux)? [05:18] IME virtualbox's networking leaves something to be desired [05:19] netbooting requires broadcast, so basically you need to be on the same network [05:19] or will it not work because the dhcp is already provided by the gateway [05:19] That *probably* means that it will work with a vbox virtual server iff you're bridging, but NOT if you're routing or masquerading [05:19] for that reason among others I'd suggest breaking this section of your network off from the rest by a router [05:19] ya i always set the virtualbox network adapter to bridge mode [05:20] where is the setting to allow php after chrooting sftp? [05:20] i see [05:20] yeah that was suggested actually, to create a separate network for the purpose of network boot server [05:20] you could accomplish the same thing by putting two NICs in your netboot server [05:21] Dravekx, there is no such thing as a "setting to allow php" [05:21] oh yeah so putting two nics = two seperate networks [05:21] you need to configure file permissions correctly [05:22] oh well looks like i'll have to set up a seperate network [05:23] although it would have been handy to stick on the existing network and use my Nas (freenas) for storing images [05:23] you can still route between the two [05:23] though I'd be careful about that, since you mentioned you're dealing with viruses [05:23] qman__, php is disabled in home directories after setting userdir. [05:24] i can? would that mean my existing network would be 192.168.0.0 and the PXE network could be 192.168.1.0? [05:24] reggi, yes [05:25] ahah [05:25] qman__, I found it. :) [05:25] you don't need NAT, just regular routing [05:25] and I suggest a firewall [05:25] hmm [05:25] you don't want your infected windows machines doing anything nasty [05:26] just basic internet access, plus access to your file store [05:26] maybe what I'll do is learn to setup this PXE in a separate network before joining networks. [05:27] have you already squared away the part about booting windows installs? [05:27] I've not seen that done and am somewhat interested [05:28] yeah I actually saw a youtube vid of someone who had setup a neat little network boot with menu [05:28] where you can choose which version of windows to install [05:28] so rather than putting the install CD in the drive (which takes a long time to complete the installation process) [05:29] you basically install over the network, so basically the relevant version of windows ISO is transferred to the computer as it is required. [05:29] lemme look for that youtube vid. [05:30] http://www.youtube.com/watch?v=-Xi0xRVlxbk [05:30] thanks [05:30] http://www.youtube.com/watch?v=zUajpWii_c0 [05:34] nah those vids arent it [05:34] it's missing the menu...lemme look for it [05:42] lol, reading back, a few months ago I slapped together a virtual box and bound it to one of the ethernet ports on my laptop and installed Ubuntu via a PXE boot on a laptop with a bad CDROM that did not support USB booting [05:44] FOUND IT! [05:44] http://www.youtube.com/watch?v=mJZsDQtmQvg&feature=related [05:44] I know it's pretty easy to do with linux [05:44] but installing windows is trickier [05:45] unfortunately the creator of this utube vid does not provide a 'how to' [05:46] yeah, he shows it off, but doesn't actually say what software it is or how to set it up [05:46] wait a sec [05:46] oh, UDA [05:46] what are those two links in the description? [05:46] is that it? [05:47] that utilities menu [05:47] has links to installing different versions of windows, accessing acronis, and other utilities === zz_sailerboy is now known as sailerboy === sailerboy is now known as zz_sailerboy [06:24] i need a bit of help [06:25] im looking to consolidate all my network gear into one low power computer running xen or kvm [06:26] i'd like to know if it is possible to dedicate a NIC to a xen or KVM instance, without the "host" being accessible on that interface [06:26] i know in vbox/vmware you can do things like NAT or bridged, but the host OS is visible on that NIC [06:26] echosystm: AFAIK it is [06:26] i want the host to be invisible to the outside world [06:26] Don't assign an IP to the host OS on that iface [06:26] it should look like a collection ofphysical devices, not one [06:27] rightyo [06:27] Note that's just SPECULATION [06:27] ok [06:27] I plan to do that, but I haven't actually tried it yet [06:27] anyone with xen/kvm experience able to confirm? [06:28] It will also matter if you're running libvirt, since that gets its greedy fingers into the network configuration [06:28] howso? [06:28] Because obviously if you're doing it via libvirt you're stuck with whatever libvirt can do [06:29] Whereas if you set it up by hand, you aren't [06:29] ah yeah [06:29] what happens at the moment is this... [06:30] (wireless adsl modem)--wireless--(my access point)--(switch)--(my computers) [06:30] because i share a unit with other people [06:31] i need to create a DMZ, because im setting up a few servers here... so it would look like [06:32] (modem)--wireless--(AP)--(unsecure stuff / router)--(safe stuff hanging off router) [06:32] theres going to be cables going everywhere [06:32] so... what i want to know is if it is possible to just go like this... [06:33] (modem)--wireless--(xen/kvm server)--(safe stuff) [06:33] does that make any sense at all? [06:34] my AP/router/servers/whatevers will all go in that xen or kvm box [06:43] I have done that before [06:43] with vmware server 1.x [06:43] simply don't assign the host an IP [06:43] effectively completely inaccessible [06:44] you would, however, still have to worry about kernel bugs and exploits for the host OS [06:44] And ARP [06:44] ARP? [06:44] <_Techie_> any good tutorials for setting up cups for an HP laserjet 2100m or similar ( lpt port ) ? [06:44] MAC <--> IP resolution [06:44] In IPv4, ARP is used for that and it happens below the IP layer [06:45] doesnt it somehow magically handle that too? === Barre_ is now known as Barre [06:45] oh wait [06:45] nevermind [06:45] sorry [06:45] the host kernel would still be vulnerable to ARP poisoning attacks [06:45] yep [06:45] echosystm: it only matters if you have a rogue ARP poisoner on the local network [06:45] what is the worst that can happen from an ARP poison though? [06:46] echosystm: MITM attacks on your online banking [06:46] oh. [06:46] do not want [06:46] anything from complete network outage to MITM snooping whatever they wish [06:47] is that any more or less likely with xen/kvm though? [06:47] Of course, if your bank actually a fucking clue how to do security, and your users had a clue about not just clicking through SSL errors, you probably wouldn't be vulnerable [06:47] it's got nothing to do with them [06:47] You *can* turn off ARP and hard-code MAC resolution, of course. [06:47] static ARP is the workaround, if you have to worry about that sort of thing [06:48] i see [06:48] an attacker needs to be on the same switched network as you to do it [06:48] Or switch to IPv6 only, which replaces ARP with IPsec-secured ICMPv6 [06:48] i think ipv6 would be a source of brain hurt, so i might give that a miss :P === _Techie_ is now known as _TechAway_ [06:49] Sorry, I forgot the ";-)" [06:50] so just to clarify, i would definitely be no more or less secure from ARP attacks than if i wasnt virtualizing everything? [06:50] im a bit confused === _TechAway_ is now known as _Techie_ [06:51] yes [06:52] excellent [06:53] thanks :) [07:46] can someone nmap my ip? i'm trying a firewall [07:47] echosystm: ARP poisoning is pretty bottom-of-the-barrel as far as attacks go [07:47] Especially since you'll probably be running PHPMyAdmin and wordpress and other random, totally insecure, services [07:48] phpmyadmin and wordpress are totally insecure? [07:48] Well, they sure get lot of CVEs [07:49] And obviously anything that allows arbitrary users on the internet access to your database using single-factor authentication is bound to be Bad News [07:49] well, for plugins..not itself.. [07:49] registering is disabled in wp [07:51] http://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=wordpress [07:53] seen..none about version 3.0 [07:53] is syslog replacing klogd [07:54] ? [07:56] mmhh..when starting /etc/init.d/klogd it remains in "* Starting kernel log daemon.."...any clue? [07:56] i need it to log firewall rules [07:59] Erm, historically syslogd and klogd were separate daemons provided by the same sysklogd package. [07:59] As at 10.04, rsyslogd (I believe) performs both actions [08:00] i've 10.04 server..csf firewall tell me "syslogd appears to be running, but not klogd which logs kernel firewall messages to syslog. You should ensure that klogd is running" [08:00] cause i have problem tracking port scan..that's why i've asked someone to portscan me [08:00] If rsyslogd isn't installed then somebody fucked with the defaults [08:00] It could be that your "csf firewall" is reporting a false positive [08:01] syslogd is running.. [08:01] rsyslogd, not syslogd [08:02] dunno what to do.. [08:02] Complain to your VPS vendor that they've messed with the Ubuntu defaults? [08:03] and what should i say? [08:03] That the 10.04 default syslogd is rsyslogd, but it appears you've got sysklogd, and its klogd component isn't running [08:03] At least, that's the evidence you've reported to me -- you may be misinterpreting it. [08:04] Hi, When ever i transfer a file to or from my media server i seem to lose my network connection after a few seconds. Any ideas as to what might cause this? [08:17] I have an "issue" I always have more than 1 apache processes running in my server [08:17] http://pastebin.com/R3Mss13y [08:17] it this an issue? [08:22] tydeas: no it's not [08:26] alex88: why this happen? [08:27] cause apache uses more server executables to process more requests.. [08:27] you can change that value in /etc/apache2/apache2.conf === SuperL4g_ is now known as SuperL4g === foobarbaz is now known as SuperLag === SuperL4g is now known as SuperLag [09:27] Daviey: around ? [09:46] hi, i'm running ubuntu 9.04 and grub 0.97, and attempting to configure grub to boot from my raid partition. i have grub installed on the disk, but in menu.lst, i need to specify the uuid for the /boot and / partitions. i need these to be md partitions, but they're not loaded/mounted at boot time so grub cannot find them. is it enough to put raid1 in /etc/modules? or does this mean it's loaded after the kerne [09:47] hi i bought domains in name.com and i am a new bie. am using an ubuntu server. what do i do so that my domain refers to /var/www/app/trunk folder? [09:51] anyone here? [10:08] Zeu5: you need to add virtualhosts in apache [10:08] Zeu5: i'm a newb too...but what do you mean "so that my domain refers to /var/www/app/trunk"? do you want to serve web pages? [10:26] where iptables log goes? [10:27] i've some iptables rules like "17 216 12888 LOG tcp -- * * 0.0.0.0/0 0.0.0.0/0 limit: avg 30/min burst 5 LOG flags 0 level 4 prefix `Firewall: *TCP_IN Blocked* ' [10:27] i can see those in dmesg, but neither in any file in /var/log [10:29] <_Techie_> what can i use to proxy non SSL IMAP connections, to a server that only allows connections with SSL? [10:30] proxy to? [10:36] <_Techie_> yes [10:37] <_Techie_> proxy from a non ssl capable mail client, to a mail server that only supports connections with ssl [10:38] if it only ssl you can't use non ssl clients [10:39] <_Techie_> alex88, theres a reason why im after a way to proxy the connection [10:40] well, so you're asking for a proxy that encrypts to ssl and then connects to server? [10:40] <_Techie_> yes [10:40] ttx: Yes.. sorry - irc didn't hilight me - oddly :/ [10:41] Hi everyone ... I have a really strange samba problem ... I have a smb domain @ lucid with printers via cups. Accessing cups without samba works, adding printer drivers too, connecting and adding to windows too, but printing itsself does not work [10:41] rpc_server/srv_spoolss_nt.c:8386(_spoolss_AddForm) [10:41] _spoolss_Addform: denied by insufficient permissions. [10:42] i get this error which cant be found via google which is the strange part of it [10:42] (using log level >= 2) [10:43] anyone has a suggestion where to search? samba does not give any hint which permission might be the problem, just this error. which makes me a little helpless. [10:47] with cat /proc/kmsg i can see the iptables log, how can i transfer to syslog? [10:51] is any of you using collectd for server statistics ? I am interested to know which tool you're using to graph the rrd files that collectd gathers. [11:05] I have a small 10.4.1 server running here and I installed the latest updates on Saturday which included a kernel update (I guess it must have been 2.6.32-24.43). [11:07] Now the system will not boot because /dev/mapper/severname-root does not exist. [11:08] I assume this must be an error in the initrd but I don't really know how to fix that. [11:08] tried to bood from old kernel? [11:08] Also, I don't see a grub prompt before that. How do I get into the grub menu to anyting? [11:08] alex88: ^ ;) [11:09] I don't get a chance to pick a kernel. [11:12] henninge: press Esc to get into grub menu at boot time...by default, you get 3 seconds... [11:12] henninge: you can get to the prompt from there [11:12] kinygos: let me try that. I admit, I am a bit rusty on grub usage ... [11:12] * RoyK is impressed - Ubuntu on Hyper-V is _fast_ [11:13] btw i'm trying to catch iptables logs, they're shown fine in /proc/kmsg but i need they go into syslog, in /etc/syslog.conf i've added line kern.warn -/var/log/syslog, but there are no logs in there, /var/log/kern.log shows "No module symbols loaded - kernel modules not enabled" [11:22] kinygos: no luck :( I can press (hold, repeat) ESC for as much as I like, I don't get any output (just a blinking cursor) until it says "Gave up waiting for root device". [11:23]