[02:28] <pentest> Host OS Xubuntu 13.04 is the LTSP server. 1 wifi card connected to Internet. I have got wlan0 (addr:192.168.1.2) and virbr0 (addr:192.168.122.1). Thin Client Guest OS installed in Virtual Box. Is it possible to have Thin Client boot-up with this LTSP server plz using virbr0? Note I have 1 physical NIC and 1 virtual NIC.
[03:12] <lesterc> I knew there would be a server channel! :)
[03:31] <lesterc> so anyone here runs ipsec+xl2tpd inside a lxc container? :)
[03:32] <stgraber> I think I actually saw a bug report about that recently, I've never tried it myself but it sounds potentially tricky due to the way ipsec works with the kernel...
[03:33] <stgraber> if it was a simple userspace daemon using a tap/tun device, that'd be easy (and openvpn tends to work great thanks to that) but ipsec is much trickier with some bits in the kernel and some bits in userspace and my fear is that not all the kernel bits are namespace aware
[03:39] <lesterc> stgraber: yeah I am running openvpn inside lxc as we speak. :)
[03:50] <shotoflove> Hi, anyone have experience with kerberos,ldap and ad?
[03:53] <lesterc> What's the question shotoflove?
[04:01] <shotoflove> lesterc: I'm trying to follow https://help.ubuntu.com/community/ADWin2k8KerberosLDAP but after getting my kerberos ticket I'm unable to getent passwd with data from Windows AD.  I'd be willing to pay for support getting this setup.
[04:04] <lesterc> shotoflove: sorry - no experience with recent windows stuff.
[04:05] <lesterc> I'm sure you can get people to do a one-off job for you if you are willing to pay.
[04:05] <shotoflove> Where at?
[04:05] <lesterc> check your local linux user group. :)
[04:06] <lesterc> to answer your question - if you managed to get a AD ticket issued to your linux host all you need is to configure nsswitch.conf (i think)
[04:07] <shotoflove> Yeah, I modified nsswitch to use ldap for passwd and group
[04:07] <shotoflove> My issue is syslog says nslcd is unable to login with ldap to my windows ad
[04:07] <shotoflove> Says invalid credentials
[04:07] <lesterc> check your libldap.conf (or equalvilent?)
[04:08] <lesterc> brb
[04:11] <shotoflove> lesterc: I used /etc/nslcd.conf binddn CN=User,OU-Users,DC=domainname,DC=edu
[04:12] <shotoflove> I set my Windows AD up as name.domain.edu
[04:12] <shotoflove> I tried binddn CN=User,OU-Users,DC=name,DC=domainname,DC=edu
[04:13] <shotoflove> Nothing seems to want to authenticate
[04:13] <lesterc> IIRC AD requires login for lookup no?
[04:14] <shotoflove> I think so
[04:14] <lesterc> I'm **really** the wrong guy for these sort of questions...
[04:14] <shotoflove> When i used likewise it also joins my computer to AD.  With a kerberos ticket I'm not seeing my computer joined to the domain.  I'm wondering if that might be the issue as well.
[04:14] <shotoflove> hah
[04:14] <shotoflove> All good
[04:15] <shotoflove> Sorry that bother you
[04:15] <lesterc> just trying to help. u might get a better answer from others on the channel however.
[04:15] <lesterc> but if I were you I'd check binddn and password in ldap.conf
[04:16] <shotoflove> Hm that guide had me remove that file
[04:16]  * lesterc is happily Windows free for the last decade.
[07:36] <pentest> Guys I am stuck. This is LTSP related. Please help..here is the details
[07:36] <pentest> Host OS Xubuntu 13.04 is the LTSP server. 1 wifi card connected to Internet. I have got wlan0 (addr:192.168.1.2) and virbr0 (addr:192.168.122.1). Thin Client Guest OS installed in Virtual Box. dhcpd.conf edited with these values:
[07:36] <pentest> authoritative;
[07:36] <pentest> subnet 192.168.122.0 netmask 255.255.255.0 {
[07:36] <pentest>     range 192.168.122.20 192.168.122.250;
[07:36] <pentest>     option domain-name "example.com";
[07:36] <pentest>     option domain-name-servers 192.168.122.1;
[07:36] <pentest>     option broadcast-address 192.168.122.255;
[07:36] <pentest>     option routers 192.168.122.1;
[07:36] <pentest> Thin client virtual network is this: Attached to: Bridged Adapter, Name: virbr0.
[07:36] <pentest> When booting up says searching for server ip (DHCP)....No IP.No IP.No IP.
[07:45] <Andre_Gomes> Hello
[07:46] <Andre_Gomes> i'm new to ubuntu server and i think i maked a big mistake, i typed pure-pw mkgroup -l > /etc/group and pure-pw mkpasswd -l > /etc/passwd now my system wont boot
[07:49] <Andre_Gomes> i can only boot to recovery, i have searched but no lucky
[09:05] <lesterc> does anyone knows about the status of bind10? :)
[09:06] <sgran> hi, is this a reasonable place to ask about the cloud-archive?
[09:07] <sgran> I'd like to poke a few bug reports, but I'm not sure where to prod
[09:09] <xnox> sgran: typically it's a straight backport from current ubuntu release to the LTS, so just open bugs against the affected package with $ ubuntu-bug package, it should file the bug correctly I believe.
[09:09] <xnox> there is little chance that it doesn't affect the current ubuntu release as well.
[09:10] <sgran> for instance: https://bugs.launchpad.net/cloud-archive/+bug/1233178
[09:10] <sgran> or https://bugs.launchpad.net/cloud-archive/+bug/1220692
[09:10] <xnox> sgran: that seems reasonable place to file those bugs.
[09:11] <sgran> if those look correctly filed, then I guess I'll just wait.  I was hoping to push it forward a bit, but I'm not sure how to help
[09:12] <xnox> smoser: jamespage: ^^^ can you please take a quick look at the above bugs?
[09:41] <koolhead17> hi all
[09:41] <koolhead17> jamespage: around
[09:42]  * koolhead17 knows zul must be sleeping
[10:02] <serapath> hello
[10:02] <serapath> i'm using koding.com. It offers a VM that i can use from a terminal within the browser
[10:02] <serapath> thats my question: http://askubuntu.com/questions/352224/howto-start-scripts-before-after-startup-reboot-suspend-hibernate-power
[10:03] <serapath> if i go offline and come back later after more then 20 minutes, my VM has been restarted or returns from suspend or something, because all my daemons or services (e.g. dropbox, nodejs server) are not running anymore and i have to restart them manually
[10:03] <serapath> i would like to have a script which is automatically executed when i return, because its a bit cumbersome to always start that stuff manually
[10:04] <serapath> how would i do it?
[10:04] <serapath> i'm not exactly sure if my VM is suspended, but if thats the case, how can i do that manually, so that i do not have to wait 20 minutes? (pmi action suspend) doesnt work
[10:45] <jamespage> koolhead17, yes
[10:46] <koolhead17> jamespage: how have you been sir?/
[10:46] <jamespage> koolhead17, very well thanks - and you?
[10:46] <koolhead17> jamespage: am rocking thanks
[10:47] <jamespage> xnox, ack - will do
[10:48] <xnox> jamespage: thanks a lot.
[10:58] <rbasak> zul: any comment on bug 1231970? "nova-novncproxy requires websockify version 0.5.1" - does this need an FFe and sync?
[10:59] <jamespage> rbasak: yes I think so
[10:59] <rbasak> jamespage: shall I target to Saucy then?
[10:59] <jamespage> rbasak: please do
[11:00] <jamespage> sgran, re those two bugs; one will be fixed at havana rc1 - the other should not be a problem in grizzly - but see convestation above re havana
[11:00] <rbasak> done, thanks
[11:37] <serverchicken> Installing ubuntu server 12.04. The CD does not include a precompiled module of the computer's network card
[11:37] <jamespage> zul, can you take a look at bug 1231970?
[11:37] <zul> jamespage:  yep will do so when i get in
[11:38] <serverchicken> I have compiled one though. is it possible for me to load the module while installing?
[11:38] <serverchicken> Jump to a shell and do a modprobe [module]
[12:51] <zul> jamespage:  https://code.launchpad.net/~zulcss/nova/lxc/+merge/187796
[12:53] <smoser> random neat little thing i just tested as functional
[12:53] <smoser> cat ~/.ssh/authorized_keys ~/.ssh/id_rsa.pub > mykey.pub
[12:54] <smoser> nova keypair-add --pub-key mykey.pub default
[12:54] <smoser> nova boot --key-name=default --flavor=m1.small --image=$IMG my-server-name
[12:55] <smoser> then i can ssh in with any of the keys there.
[12:55] <smoser> ie, your pubkey that you upload can have more than one pubkey in it.
[13:00] <herol3oy> Hi. I have got a pen driver and it has got the "New Folder.exe" virus and now I want to search for it in terminal and then delete them immediately? how can i do it? thnkx
[13:04] <ikonia> herol3oy: what do you want to search for ?
[13:04] <herol3oy> ikonia:  I want to find all the "New Folder.exe" files and then delete them!
[13:05] <ikonia> herol3oy: find them on what ?
[13:06] <smoser> herol3oy, find / -name "New Folder.exe" -delete
[13:06] <smoser> maybe with 'sudo' first.
[13:06] <herol3oy> ikonia: u know i don't wanna format my pen drive so i need to find all these files and then delete them so my friends computer will be safe when I put the pen drive in their usb drive.
[13:07] <smoser> herol3oy, find /media -name "New Folder.exe" -delete
[13:07] <ikonia> herol3oy: ok, so mount the pendrive, then use the command smoser suggested replacing / with the mount url
[13:07] <ikonia> mount point (not url)
[13:07] <smoser> yeah, and just go from /media unless you'd *want* to leave 'New Folder.exe' files around.
[13:08] <smoser> you could also do:
[13:08] <smoser> well, if you drop the '-delete' it will print them.
[13:08] <smoser> so do that first, be satisfied that its not doing anything you dont want
[13:08] <smoser> then run with '-delete'
[13:20] <sgran> jamespage: so that means it will get uploaded to saucy and then pulled to the cloud-repo?  If so, cool, and thanks
[13:20] <sgran> I'm doing a test deploy now, so I'm trying to report and track any bugs before we need it for anything important
[13:21] <herol3oy> ikonia: hey dude. I run this command line:     find /media/herol3oy/usbdrive/ -name "New Folder .exe" -delete      but it only delete the file which was located in root. how can i delete them all also in other folders and sub-folders?
[13:22] <ikonia> herol3oy: it should only search /media/herol3oy/usbdrive
[13:24] <jamespage> sgran, that is the case yes
[13:24] <jamespage> sgran, are you using the havana-staging PPA?
[13:26] <sgran> deb http://ubuntu-cloud.archive.canonical.com/ubuntu precise-{updates,proposed}/havana main
[13:26] <jamespage> sgran, urgh - Ok - thats not quite as up-to-date as ppa:ubuntu-cloud-archive/havana-staging
[13:27] <sgran> can it be :)
[13:30] <soren> jamespage: I'm around now if you want to discuss https://code.launchpad.net/~soren/python-jenkins/add-crumb/+merge/188155
[13:33] <jamespage> soren, sorry - looking now
[13:33]  * jamespage got distracted by some openstack issues yesterday
[13:51] <soren> jamespage: I know that feeling.
[13:51] <herol3oy> ikonia: thanks dude. i did it ;)
[14:03] <rbasak> smoser: we'll want uvtool in main eventually I presume? Does this block juju at all?
[14:08] <smoser> juju is universe
[14:08] <smoser> right?
[14:08] <smoser> (thats what apt-cache policy tells me)
[14:09] <smoser> i do not have any reason for why uvtools should be main.
[14:09] <smoser> not that i can think of.
[14:09] <rbasak> Oh. Of course juju isn't in main yet.
[14:09] <rbasak> Thanks
[14:11] <excalibr> Is discard directive in rsyslog broken in Raring? I put & ~ at the end of my iptables log config and i still see its log messages being logged to /var/log/syslog
[14:12] <excalibr> :msg, contains, "iptables: " /var/log/iptables.log
[14:12] <excalibr> & ~
[14:19] <jamespage> soren, that should fix the 'you need todo POST' error message from jenkins right?
[14:19] <jamespage> I'm still seeing that
[14:22] <soren> jamespage: "you need todo POST"?
[14:22] <jamespage> soren, "You must use POST method to trigger builds. (From scripts you may instead pass a per-project authentication token, or authenticate with your API token.) If you see this page, it may be because a plugin offered a GET link; file a bug report for that plugin."
[14:23] <soren> jamespage: Uh. No.
[14:23] <soren> jamespage: This fixes authentication when CSRF is enabled.
[14:23] <soren> jamespage: Your problem seems related to your not using the API token, but perhaps your password?
[14:25] <soren> So, if CSRF protection is enabled, you need to include a crumb in your requests (to validate that you haven't just been tricked into posting something from a malicious web site).
[14:25] <soren> This patch fetches that crumb and adds it to your request.
[14:25] <stemid> I have this weird issue on an old 10.04 system where /tmp suddenly uses 4.4G with no large files inside it. usually this means some process has an open file handle on /tmp but lsof |grep inode number of /tmp shows nothing.
[14:26] <stemid> so far the only solution has been reboot
[14:26] <soren> stemid: Why would it show the inode number of /tmp?
[14:26] <soren> stemid: Doesn't it say "/tmp/blah (deleted)" or something to that effect?
[14:27]  * soren forgets the exact format
[14:28] <soren> stemid: Yup, it certainly does for me.
[14:28] <soren> stemid: sudo lsof | grep tmp.*deleted
[14:30] <stemid> soren: lsof shows inodes too, I've obviously already tried greping for /tmp or I wouldn't be searching for inodes now
[14:30] <stemid> and I just tried a grep -E '(every|inode|in|/tmp)'
[14:30] <stemid> just to be sure
[14:32] <jglanz> Hi all, never chat with the sparkfun guys online!
[14:33] <Pici> jglanz: okay? this is #ubuntu-server though.
[14:33] <jglanz> I had a random question, I'm building my first project enclosure which includes several components, I purchased an enclosure, and board from you guys...question is should I be soldiering or wire wrapping, etc?
[14:33] <jglanz> sorry wrong room
[14:36] <soren> stemid: If that really yields nothing, I'd start killing processes one at a time until the space was released.
[14:46] <brendand> hello, i'm trying to use mount on a cloud image but it's failing - the same command i use for desktop/server images works
[14:49] <aandy> hi guys, couldn't find a BIND/named chan, so hope it's ok i ask here. any users of bind9 who can review a config of mine and help me figure out why views won't work properly?
[14:50] <Sling> try #bind
[14:50] <Sling> (and /msg alis list *bind* to search for all channels matching the name 'bind')
[14:50] <aandy> ah, i joined on the wrong network, that explains it. d'oh. thanks Sling ;)
[14:51] <Sling> np
[14:57] <jamespage> zul, re lxc mp - one test failed for me when test building
[15:01] <zul> jamespage:  yeah fixing it now
[15:02] <BrixSat> how do i resize a partition on a live server?
[15:02] <BrixSat> i mean i know its with resizefs but im kind of afraid.
[15:05] <ivoks> hallyn: have you seen issues with running qemu inside kvm, with both acpi and apic enabled? :)
[15:05] <ivoks> hallyn: acpi and apic enabled in qemu
[15:07] <hallyn> ivoks: no.  smb is investigating some kvm-in-kvm nesting issues...
[15:07] <hallyn> you're not trying qemu64 in kvm32 I assume?
[15:07] <ivoks> it's 64 in 64
[15:07] <ivoks> same behavior is observed with qemu inside of esxi
[15:08] <ivoks> once i disable acpi and apic, system boots
[15:08] <ivoks> with it, softlockups or blkid timeouts
[15:09] <hallyn> is there an open bug for this?  which releases/kernels?
[15:09] <smb> There is something subtly broken with events/traps in nested vmx, but I have not really gotten down to it
[15:09] <hallyn> smb: but ivoks is talking about unaccelerated qemu inside kvm
[15:09] <ivoks> hallyn: there isn't; i'm still trying to come up with a stable reproduces
[15:09] <ivoks> sometimes it works, sometimes it doesn't
[15:09] <hallyn> ivoks: love those
[15:10] <smb> hallyn, Ah ok this would be different then
[15:11] <PaulePanter> Hi. The server in the data center does not restart running Ubuntu 12.04.
[15:12] <PaulePanter> The problem is I only have network connection and do not know what is going on.
[15:12] <BrixSat> This is my gparted http://paste.ubuntu.com/6179921/
[15:12] <PaulePanter> I can ping it but I cannot log in using SSH.
[15:12] <PaulePanter> Sending Ctrl + Alt + Del using the hoster’s robot interface works too.
[15:12] <PaulePanter> How do I know that there is no fsck going on?
[15:13] <ivoks> PaulePanter: how long was it up?
[15:13] <PaulePanter> Maybe half an hour.
[15:13] <ivoks> PaulePanter: before reboot
[15:13] <PaulePanter> Four days.
[15:13] <ivoks> PaulePanter: do you have raid on it?
[15:14] <PaulePanter> In the rescue system, running fsck manually it says everything is fine.
[15:14] <PaulePanter> ivoks: Yes, it is a RAID.
[15:14] <ivoks> PaulePanter: and both disks are working?
[15:14] <PaulePanter> They should. The rescue system could mount it.
[15:17] <ivoks> PaulePanter: did you check logs while in rescue system?
[15:18] <ivoks> if you can ping it, it at least mounted filesystem and set up networking
[15:19] <PaulePanter> ivoks: I checked the logs. It looks like it was up as `/var/log/wtmp` had a recent time stamp.
[15:19] <PaulePanter> ivoks: /var/log/upstart/syslog strangely did not have something from the last reboot though.
[15:20] <ivoks> PaulePanter: i wouldn't expect anything in that file
[15:20] <ivoks> PaulePanter: since that file will contain only errors on starting the syslog daemon
[15:21] <PaulePanter> ivoks: Ah.
[15:21] <PaulePanter> How do I disable services from a rescue system?
[15:21] <PaulePanter> Or do I go into a chroot and do it from there?
[15:22] <ivoks> ubuntu rescue mode or hetzner's rescue image?
[15:22] <PaulePanter> Hetzner’s rescue image.
[15:23] <ivoks> you have to enter chroot
[15:25] <PaulePanter> Thanks. Did that now.
[15:35] <jamespage> soren, ok - I'm confused
[15:35] <jamespage> soren, I turned on CSRF protection
[15:35] <PaulePanter> So in the rescue system looking at the time stamps of the files in `/var/log/*`, only `wtmp` has an updated time stamps and in there, the reboots (Ctrl + Alt + Del) are noted.
[15:36] <jamespage> and python-jenkins worked without your patch...
[15:36] <PaulePanter> All other files have the time stamp of time it worked the last time.
[15:37] <zul> jamespage:  lxc patch fixed
[15:47] <jamespage> zul, time to chat about horizon and writing to /etc/openstack-dashboard?
[15:48] <zul> jamespage:  always
[15:49] <jamespage> zul, well
[15:49] <zul> jamespage:  dont do that? :)
[15:49] <jamespage> the code out of the box does this generate_or_get call
[15:49] <jamespage> zul, I think that should be pointing at /var/lib/openstack-dashboard
[15:49] <jamespage> with www-data permissions?
[15:49] <zul> sounds like a plan
[15:49] <jamespage> OK
[15:51] <zul> jamespage:  are you going to patch it up?
[15:51] <jamespage> zul, just trying to figure out how the compression is working now as well
[15:52] <zul> jamespage:  ack...im just in the middle of fixing keystone and going to look at this oauth business again
[15:59] <jamespage> zul, lesscpy is borked
[15:59] <jamespage> bin/lesscpy uses python3
[15:59] <jamespage> but it uses argparse in the code
[15:59] <zul> lovely
[16:00] <zul> ill take a look
[16:04] <zul> roaksoax/jamespage/adam_g: https://code.launchpad.net/~zulcss/keystone/oauth2-refresh/+merge/188639
[16:24] <adam_g> zul, is oauth2 now completely optional?
[16:24] <zul> adam_g:  yeah
[16:24] <adam_g> cool
[16:24] <zul> its still in the requriments.txt file though but if you dont have it installed it will be ignored
[16:25] <jamespage> zul, raising a bug for python-lesscpy
[16:25] <zul> jamespage: ack
[16:43] <jamespage> zul, bug 1233749
[16:43] <zul> jamespage:  ack
[16:46] <jamespage> zul, is there a precendence for sneaking a source package from Debian into Ubuntu?
[16:48] <zul> jamespage:  not that i remember
[16:48] <zul> check on #ubuntu-release
[17:01] <jamespage> zul, adam_g: https://code.launchpad.net/~james-page/horizon/fixup-secret-storage/+merge/188659
[17:03] <zul> jamespage:  looks good to me, i think the generate_or_read_from_file should be upstream as well
[17:34] <mgriffin> can i determine if some package has a command (or perhaps i can only know it is a file) in it called "foo"?
[17:34] <mgriffin> i don't have the foo command and want to know if some package can provide it
[17:35] <mgriffin> oh, i forgot that was built in to ubuntu by default :D
[19:02] <soren> jamespage: Well, then your CSRF protection is broken.
[19:03] <soren> jamespage: It's supposed to reject the requests. My Jenkins certainly does.
[19:03] <jamespage> soren, hmm - I'll poke it again tomorrow
[19:03] <jamespage> somethings wonky
[19:18] <soren> jamespage: np
[20:37] <Hexch> hi
[21:11] <Overand> Hey - question.  How can I figure out what stuff I've got on this system that i installed via apt. vs installed via .deb files?
[21:12] <sarnold> Overand: I guess you'll need to compare /var/log/dpkg.log* files against /var/log/apt/history* files. There may be a better option, but that's the first that comes to mind..
[21:12] <Overand> woof.
[21:12] <Overand> Yeah.
[21:13] <Overand> Maybe there's a way I can look at what's installed vs. what's avaialble via apt - while not 'technically correct' i think in this case it might work.
[21:13] <sarnold> Overand: aha, try 'apt-cache pkgnames' for a starting point there.
[21:18] <Overand> sarnold: it's not even an ubuntu-server machine, it's a legacy debian box.  most of my curerent machines are ubuntu-server
[21:40] <pentest> can some one help plz http://ubuntuforums.org/showthread.php?t=2178061
[21:56] <sarnold> pentest: do you have a dhcp server listening on virbr0 to hand out IP addresses? do LTSP guests need IP addresses? Or a dhcp server? do you need to configure NAT on your machine to forward packets for those guests? Or is that the role of LTSP? (Can you tell I know nothing about LTSP? :)
[22:03] <pentest> sarnold: let me try to answer...
[22:03] <pentest> if you see the dhcpd.conf setting.. I have changed the ip settings there..
[22:04] <pentest> so technically there should a dhcp server.. however when I booted up lubuntu guest OS ..I cannot get an ip
[22:04] <pentest> so that is the prroblem..
[22:05] <pentest> yeah..virtual box do not seem to communicate well virbr0 virtual adapter it seems
[22:05] <sarnold> pentest: check netstat -lnp output, make sure that the dhcp server is listening on the correct interfaces / addresses
[22:05] <sarnold> pentest: check iptables -L output to make sure that the guests are allowed to communicate with the host?
[22:06] <pentest> tcp        0      0 192.168.122.1:53        0.0.0.0:*               LISTEN      2261/dnsmasq
[22:07] <pentest> http://pastebin.com/QehtnebV
[22:07] <pentest> seems good for me here
[22:08] <pentest> I think key will be able to use virtual box created adapter rather than libvirt
[22:20] <gartral> ok all, i'm gearing up to do a drive swapout from a dying drive to a new one, would it be easily do-able to just dd the drives from current to new?
[22:21] <sarnold> gartral: if the drive is truly going, you might need to use something like dd_rescue or myrescue, in case IO errors prevent plain dd from working
[22:22] <sarnold> gartral: (maybe conv=noerror is good enough these days? there has to be a reason for those tools, perhaps they were writte nbefore conv=noerror..)
[22:22] <sarnold> gartral: anyway I've had success with dd to move filesystems around before, that part ought to work fine.
[22:31] <pentest> just ignore me.. testing something
[22:33] <pentest> just ignore me.. testing something
[22:34] <pentest> just ignore me.. testing something
[22:35] <gartral> sarnold: the drives' internal error recovery is working fine, but the drive is slowly going due to old age..
[22:35] <sarnold> pentest: you may wish to /join #test if you're going to be doing too much more..
[22:35] <sarnold> gartral: oh good :)
[22:39] <pentest> just ignore me.. testing something
[22:41] <pentest> just ignore me.. testing something
[22:44] <pr3d4t0r> Greetings.
[23:43] <metasansana> Is there a way to listen for syslog events from an application?
[23:53] <qman__> metasansana, you can configure syslog to log a given application's events to a specific log file, and then watch that log file with inotify
[23:55] <metasansana> qman__, I'm try to write a small monitor that will react to certain syslog events.
[23:55] <qman__> there's also the quick, dirty, and performance-impacting way of: tail -f /var/log/syslog | grep myapp
[23:55] <qman__> metasansana, if it's for one specific application, doing it the way I first mentioned is probably best
[23:56] <qman__> if it's for multiple, use the main syslog file
[23:56] <qman__> with inotify
[23:56] <qman__> tail -f adds unnecessary disk activity and should really only be used interactively
[23:57] <metasansana> I supposed I will have to parse the file after being notified.
[23:57] <qman__> yes
[23:57] <metasansana> ick
[23:57] <qman__> alternatively you could write your application to accept the syslog data directly, and have syslog log to that file handle
[23:57] <metasansana> Can I mock a file with an application? Like an input stream or something?
[23:57] <qman__> yeah
[23:58] <metasansana> ah yes
[23:58] <metasansana> qman__, thanks! I'll google how to do that
[23:58] <qman__> not sure whether unix sockets or named pipes or something else would be better (not a programmer) but it's definitely possible to do it that way