[00:00] <sarnold> gartral: the first one will go a long way if your processes all need consistent accesses but don't have a common primary group in common
[00:00] <sarnold> gartral: but if one user has umask=0077, it won't help :)
[00:02] <gartral> sarnold: my issue is that I have a folder /home/minecraft/plugins and I can read/write to *THAT* folder, but when I add a plugin too the server that plugin makes a new subdirectory there in which I only have read access.. meaning I have to contact the Server Owner to have her fix the perms each time I add a new plugin >.<
[00:03] <gartral> we both agree that this is becoming a PiTA.
[00:03] <gartral> and I'm in the minecraft group on that machine
[00:04] <sarnold> gartral: how are the plugins run? does the minecraft server start them?
[00:04] <gartral> sarnold: yes, on load
[00:05] <sarnold> gartral: is the directory created at load time? or during an install mechnaism?
[00:05] <zul> adam_g:  ill have a look
[00:06] <gartral> which means that stop/starting the minecraft server process loads the plugins in /home/minecraft/plugins, most plugins as part of their initial load create a new subdir containing all of the configureation information for the plugin
[00:06] <gartral> and I'm sick today, so if my spelling is a little screwy, i apologize
[00:09] <gartral> dangit, i have a fix for GarretKajmowicz
[00:09] <sarnold> oh?
[00:10] <gartral> sarnold: yea, bootchart. it'll profile the boot proccess and let him see where the boot failed
[00:10] <sarnold> gartral: it might be hard without a rw filesystem..
[00:11] <gartral> sarnold: this is why my /log partitions are always on an NFS share :P
[00:11] <sarnold> gartral: haha :) nice, but he's stuck even without a / ...
[00:11] <gartral> sarnold: but back to my issue, any thought?
[00:12] <gartral> sarnold: ohh.. ewww.. his kernel just isn't loading right, huh?
[00:13] <sarnold> gartral: hrm, I'm having trouble finding his pastebin of dmesg.. here's his askubuntu question: http://askubuntu.com/questions/307509/upgrade-to-12-04lts-dumps-to-busybox-on-boot
[00:13] <sarnold> gartral: I'm surprised the plugins would execute as your user, rather than a minecraft user..
[00:14] <sarnold> ah, here we are, his failed boot dmesg: http://pastebin.ca/2455008
[00:15] <sarnold> gartral: is there a dedicated minecraft user account? is that user acccount starting the minecraft server?
[00:16] <gartral> sarnold: yes and yes
[00:21] <sarnold> gartral: hrm. well, you can try the setgid thing easily enough, 'chmod g+s /path/to/directory/' -- I'm not sure why it doesn't Just Work, but I'm not familiar enough with minecraft to know what to look for.
[00:22] <gartral> sarnold: it's not a minecraft specific issue, this happens with all folders in that directory no matter what made them
[00:26] <sarnold> gartral: can you get an ls -ld /home /home/minecraft /home/minecraft/plugins   output? I'm curious about user/group/permissions..
[00:31] <gartral> ls: cannot access /home/minecraft/plugins: No such file or directory
[00:31] <gartral> drwxr-xr-x 5 root      root  4096 Sep 16 19:23 /home
[00:31] <gartral> drwxr-xr-x 8 minecraft users 4096 Sep 18 15:17 /home/minecraft
[00:31] <gartral> oops
[00:32] <gartral> thought I had my browser focused
[00:33] <sarnold> no /home/minecraft/plugins ? how odd..
[00:43] <gartral> sarnold: sorry, stand by
[00:44] <gartral> sarnold: http://paste.ubuntu.com/6130558/
[00:46] <sarnold> gartral: looks sane enough... you might try: chmod g+s /home/minecraft/unl-testing/ /home/minecraft/unl-testing/plugins/
[00:47] <sarnold> that'll set both those directories to force new directories and files created within to have 'minecraft' group ownership. it's a start. :)
[00:58] <gartral> sarnold: should that be run as root or my user?
[00:58] <sarnold> gartral: it could either be run as the minecraft user or as root
[01:00] <gartral> sarnold: that still doesn't fix my access
[01:01] <sarnold> gartral: hrm, ls -ld on the newly created directories and maybe ps auxw | grep minecraft   ... or whatever the server runs as..
[01:01] <gartral> sarnold: does that chmod a+s fix already created directories?
[01:02] <sarnold> gartral: no, just new directories and files
[01:02] <sarnold> gartral: do you have any currently-existing 'broken' directories?
[01:02] <sarnold> ls -ld on them would be helpful too..
[01:02] <gartral> sarnold: yes, about 12
[01:03] <gartral> sarnold: yea, they just "fixed" it by chmod 775 /home/minecraft/unl-testing
[01:04] <gartral> sarnold: well, nvm then
[01:11] <GarrettKajmowicz> I'm back, now with a dedicated IRC client.
[01:26] <gartral> GarrettKajmowicz: so your server isn't mounting / at all?
[01:38] <sarnold> GarrettKajmowicz: gartral suggested trying bootchart, there may be a mechanism it provides to discover how far through boot you get before you're hung up, and it might provide some way to get the data off the system ..
[03:47] <mojtaba1> Hi, do you know how can I connect to my c$ windows partition from ubuntu cmd?
[03:47] <mojtaba1> They are in the same LAN.
[03:49] <mojtaba1> Hello?
[04:35] <dingo311> i cannot ssh from outside my network. canyouseeme.org sees my port, port forwarding is on. i dont know what to do. help please.
[05:49] <Quest> what doss attacks are still with no solution and i should be prepared for defending?
[05:56] <Guest65934> roundcube sucks that all i have to say ...
[05:57] <josephtacos> i started getting a red 0 in front of my prompt in byobu. some how i triggered it. any clue on what it is
[06:02] <Guest65934> I see that also today 12.04 lts server not sure it went away itself tho--- hummm
[06:07] <josephtacos> i think there exit codes
[06:17] <sarnold> Quest: there are hosting providers with unbelievable amounts of bandwidth available, e.g. cloudflare. If you're worried about the reliability of your site even in the face of modern DDoS attacks, it'd be worth having a conversation with them before you need their help.
[06:28] <Quest> hm
[07:55] <foubarre> Hi everyone. I have an ubuntu server acting as a DHCP server. I changed the gateway address. Now machines randomly receive the old or new gateway. DHCP config only contains the new address (dhcp service was restarted many times just in case). Any idea of why that can happen?
[08:00] <linocisco> foubarre, restarted on server? let clients PC restart
[08:01] <foubarre> linocisco: restarted the server, restarted client machines too. Does that since yesterday.
[08:02] <foubarre> linocisco: on the PCs, doing "ipconfig /renew" returns a mix of old and new gateway. Does the same on OSX so i excluded a local caching problem.
[08:03] <elnur> Hey. I'm trying to figure out why nginx and PostgreSQL don't start on system boot. Where do I start?
[08:03] <elnur> Other stuff like PHP-FPM *do* start.
[08:30] <Rory> elnur: Do they start correctly without errors when you start them manually?
[08:31] <Rory> elnur: If so, make a service start at startup with: sudo update-rc.d SERVICENAME defaults
[08:33] <elnur> Rory, yes.
[08:33] <elnur> Let me try that.
[08:34] <elnur> Rory, it says: System start/stop links for /etc/init.d/nginx already exist.
[08:40] <elnur> What else should I try?
[09:11] <linocisco> hi
[09:56] <elnur> I'm trying this:
[09:56] <elnur> cat < `crontab -l` <<EOF | crontab -
[09:56] <elnur> @reboot service nginx start
[09:56] <elnur> @reboot service postgresql start
[09:56] <elnur> EOF
[09:56] <elnur> But it doesn't work.
[09:56] <elnur> What I want is to append those two lines to the output of running `crontab -l` and then pipe it to `crontab -`
[09:56] <elnur> How should I do that?
[10:01] <sgran> elnur: don't do it that way
[10:02] <elnur> How should I do it then?
[10:02] <sgran> first, since you know the two lines you want to add, just do crontab -e :)
[10:02] <sgran> second, fixing why they don't come up at start up is probably the saner approach
[10:02] <elnur> I need to script this.
[10:02] <elnur> I've spend the whole day trying to fix that already.
[10:02] <sgran> ls /etc/init/*nginx* /etc/init/*postgres*
[10:03] <elnur> No such files.
[10:04] <sgran> ls /etc/init.d/*nginx* /etc/init.d/*postgres* /etc/rc?.d/*nginx* /etc/rc?.d/*postgres*
[10:04] <elnur> # ls -l /etc/rc?.d/*nginx
[10:04] <elnur> lrwxrwxrwx 1 root root 15 Sep 19 21:23 /etc/rc0.d/K20nginx -> ../init.d/nginx
[10:04] <elnur> lrwxrwxrwx 1 root root 15 Sep 19 21:23 /etc/rc1.d/K20nginx -> ../init.d/nginx
[10:04] <elnur> lrwxrwxrwx 1 root root 15 Sep 19 21:23 /etc/rc2.d/S20nginx -> ../init.d/nginx
[10:04] <elnur> lrwxrwxrwx 1 root root 15 Sep 19 21:23 /etc/rc3.d/S20nginx -> ../init.d/nginx
[10:04] <elnur> lrwxrwxrwx 1 root root 15 Sep 19 21:23 /etc/rc4.d/S20nginx -> ../init.d/nginx
[10:04] <elnur> lrwxrwxrwx 1 root root 15 Sep 19 21:23 /etc/rc5.d/S20nginx -> ../init.d/nginx
[10:04] <elnur> lrwxrwxrwx 1 root root 15 Sep 19 21:23 /etc/rc6.d/K20nginx -> ../init.d/nginx
[10:04] <elnur> # ls -l /etc/init.d/*nginx
[10:04] <elnur> -rwxr-xr-x 1 root root 2091 Dec 17  2012 /etc/init.d/nginx
[10:05] <sgran> that looks fine.  Can you paste the contents of /etc/init.d/nginx at paste.ubuntu.com
[10:06] <elnur> sgran, http://paste.ubuntu.com/6131972/
[10:07] <elnur> The script works when I run it manually.
[10:07] <elnur> Just doesn't work on boot.
[10:07] <sgran> is there anything in the logs?
[10:08] <elnur> Nothing in syslog about nginx and nothing in nginx's error.log.
[10:08] <sgran> anything in upstart's log?
[10:08] <elnur> Let me check.
[10:09] <elnur> Which file?
[10:12] <sgran> hmm.  I see upstart only logs jobs, rather than overall status
[10:12] <sgran> disappoint
[10:14] <elnur> Meanwhile:
[10:14] <elnur> (crontab -l; echo "@reboot service nginx start") | crontab -
[10:14] <elnur> (crontab -l; echo "@reboot service postgresql start") | crontab -
[10:14] <elnur> \o/
[10:15] <elnur> Of course, I'd like to fix the root issue.
[10:15] <elnur> I'm just not that good with advanced linux stuff.
[10:15] <sgran> do you see anything in boot.log ?
[10:16] <elnur> I'm recreating the VM. Will tell you in a couple of minutes.
[10:16] <elnur> Vagrant.
[10:17] <elnur> BTW, can Vagrant/VirtualBox be the reason?
[10:18] <sgran> I'm guessing at this point that they're failing to start because of some network or other related configuration that isn't there when they start at boot time, but are there later when you run the script by hand
[10:18] <sgran> so, maybe?
[10:18] <elnur> Would I be able to find this information in any log file?
[10:18] <elnur> I'll check boot.log soon.
[10:19] <elnur> BTW, PHP-FPM starts okay.
[10:19] <elnur> It's nginx and PostgreSQL failing for some reason.
[10:25] <elnur> The VM is up.
[10:26] <elnur> Where's the boot.log file?
[10:26] <sgran> in /var/log
[10:26] <elnur> Nope.
[10:26] <elnur> `locate boot.log` gives nothing.
[10:29] <elnur> Any other ideas? :)
[10:31] <sgran> hmm
[10:31] <sgran> I see a /var/log/boot and a /var/log/boot.log on precise, here
[10:31] <sgran> does vagrant give you a console log from bootup?
[10:33] <elnur> Let me see.
[10:35] <elnur> Hmm.
[10:35] <elnur> I'm looking at `vagrant up` logs and it looks like it's adding network interfaces after the machine boot up.
[10:36] <elnur> [default] Machine booted and ready!
[10:36] <elnur> [default] Configuring and enabling network interfaces...
[10:36] <sgran> so that'll be the problem, I'd guess
[10:37] <elnur> Hm. I'll try another network configuration.
[10:54] <elnur> Well, I tried different network configuration, but that didn't help.
[10:54] <elnur> I also checked out PostgreSQL logs and found this:
[10:54] <elnur> 2013-09-20 10:42:31 UTC LOG:  database system was shut down at 2013-09-20 10:42:30 UTC
[10:54] <elnur> 2013-09-20 10:42:31 UTC LOG:  database system is ready to accept connections
[10:54] <elnur> 2013-09-20 10:42:31 UTC LOG:  autovacuum launcher started
[10:54] <elnur> 2013-09-20 10:42:32 UTC LOG:  incomplete startup packet
[10:54] <elnur> 2013-09-20 10:49:34 UTC LOG:  received smart shutdown request
[10:54] <elnur> 2013-09-20 10:49:34 UTC LOG:  autovacuum launcher shutting down
[10:54] <elnur> 2013-09-20 10:49:34 UTC LOG:  shutting down
[10:54] <elnur> 2013-09-20 10:49:34 UTC LOG:  database system is shut down
[10:54] <elnur> Nothing interesting, I guess.
[10:55] <elnur> I'll go with @reboot cron job for now.
[10:55] <elnur> sgran, thanks for the attempt. :)
[11:17] <sgran> no sweat :)
[12:33] <sarthor> HI, I did apt-get update and then apt-get upgrade, Now when I rebooted, machine is not booting in normal way, I went through recovery mode, and then Network.. So I can use my machine, .. when machine is booting in normal way, it says..  udevd[468]: timeout: killing '/sbin/modprobe -bv pci:v00008086d000027D8sv00001028sd000001DEbc04sc03i00' [473] .. and never boot...HELP please.
[12:46] <sarthor> HI, I did apt-get update and then apt-get upgrade, Now when I rebooted, machine is not booting in normal way, I went through recovery mode, and then Network.. So I can use my machine, .. when machine is booting in normal way, it says..  udevd[468]: timeout: killing '/sbin/modprobe -bv pci:v00008086d000027D8sv00001028sd000001DEbc04sc03i00' [473] .. and never boot...HELP please.
[12:47] <newhite> What is a good socket monitor for Ubuntu?
[12:48] <greppy> sarthor: !repeat
[12:48] <sarthor> greppy: OK. can i quit?
[12:49] <greppy> sorry, was trying to trigger the bot, out of practice.
[12:49] <greppy> !repeat
[12:50] <sarthor> I love this channel. I got a lot of help here, You guys helped alot. Thanks a lot
[12:50] <greppy> sarthor: if no one is awake that can help you, you won't get any help, sorry.
[12:56] <newhite> greppy, that is very logical.  Haha :)
[13:05] <Rory> sarnold: Any time
[13:15] <zul> jamespage:  https://code.launchpad.net/~zulcss/cinder/housecleaning/+merge/186612
[13:17] <jamespage> zul, wrap-and-sort -s required
[13:18] <zul> jamespage:  fixed
[13:33] <jamespage> zul, looks good, built tests OK
[13:33] <zul> jamespage:  cool ill merge it
[13:33] <jamespage> zul, wait
[13:34] <jamespage> zul, one thing - oslo-config needs an epoch
[13:34] <zul> doh
[13:35] <zul> fixed
[13:51] <xerxas_> utlemming: I somehow cannot find how you built the ubuntu cloud vagrant images
[13:51] <xerxas_> do you provide any script, or any doc ?
[14:14] <Diegonat> can i create a link file in two different devices?
[14:26] <bjf> on a maas server what does "node added but never seen" mean?
[14:26] <bjf> i'm trying to enroll a single system
[14:27] <bjf> i'm on a saucy installed system
[14:27] <bjf> if i look under "Nodes" i see the node, it's status is "Declared" (whatever that means)
[15:13] <zul> jamespage:  im re-enabling the tests for glance as well
[15:22] <jamespage> zul, good luck
[15:22] <jamespage> please don't melt my laptop again :-)
[15:24] <zul> jamespage:  ill try not to ;)
[16:20] <zul> adam_g/jamespage: https://code.launchpad.net/~zulcss/glance/housecleaning/+merge/186836
[17:15] <zul> adam_g: i dunno if you saw this https://code.launchpad.net/~zulcss/glance/housecleaning/+merge/186836
[17:18] <adam_g> zul, i did, gonna look at it in a bit
[17:19] <zul> adam_g:  cool thanks
[17:19] <adam_g> zul, please merge you changes into the ubuntu-server-dev branches instead of pushing them directly
[17:19] <zul> adam_g: yep
[17:42] <zul> adam_g:  fixed
[17:46] <adam_g> zul, its still not wrapped to 80 and there are extra newlines
[18:14] <zul> adam_g:  https://code.launchpad.net/~zulcss/keystone/housecleaning/+merge/186849
[18:26] <hamnstar> hey folks - getting that pesky "waiting for network configuration" message on startup (ubuntu server 12.04).... it's odd, because my configs both work (2 NICs each with a static IP, only one default gateway)... is there anything other than /etc/network/interfaces that i should look into?
[19:20] <zul> adam_g:  https://code.launchpad.net/~zulcss/neutron/housecleaning/+merge/186870
[19:25] <adam_g> zul, looks okay but i'm not approving it unless you promise its going to get pushed to ubuntu-server-dev as a merge and not a direct commit
[19:30] <zul> adam_g:  uh?
[19:30] <adam_g> .:adam_g:. zul, please merge you changes into the ubuntu-server-dev branches instead of pushing them directly
 adam_g: yep
[19:30] <zul> adam_g:  i thought i was dong that
[19:31] <adam_g> zul, doesn't look like it
[19:31] <zul> frig sorry
[19:32] <zul> adam_g:  so you want me do bzr branch the neutron branch and then merge my branch
[19:32] <adam_g> zul, ideally, yes
[19:32] <adam_g> it can be automated
[20:26] <th299792458> Hi guys! I'm new to linux. I would like to run ubuntu server on a machine that I would also like to use as a media center. What do you guys think is the best way to do so?
[20:27] <th299792458> Ubuntu server with a gui would be somewhat weird, right? But a media center without a gui is also somewhat weird
[20:27] <ikonia> th299792458: why do you need ubuntu server for this ?
[20:28] <ikonia> th299792458: why not use the desktop CD for this ? it comes with a gui
[20:28] <th299792458> Yes but I would also like to run ubuntu server to host a site for example
[20:28] <Pici> Also, you need to define 'Media Center' many media centers have a gui.
[20:29] <ikonia> th299792458: you can do that on the desktop install
[20:29] <ikonia> th299792458: it has the same access to packages, such as apache web server
[20:38] <qman__> the only real difference between the server and desktop editions is the set of which packages are installed; you can install any package you like on either version, so choose what suits your application best
[20:57] <nikolaj_basher> is there a way I can buck an error of af php site through my apache log
[21:10] <ikonia> buck ?
[21:11] <arthur789> im trying to edit my hosts file to allow acess from my ddns inside my network. im not sure exactly what to put into the hosts file.
[21:12] <ikonia> why are you changing your hosts file ?
[21:13] <arthur789> ikonia: from reading that my router has no NAT or loopback service. i thought the next step was the hosts file.
[21:13] <ikonia> why ?
[21:13] <ikonia> nat has nothing to do with your host file
[21:14] <gholms> utlemming: If I reconfigure cloud-init's upstart bits with "console output" and systemd bits with "journal+console" can we just let those send stdout to the console instead of trying to open it explicitly?
[21:14] <gholms> utlemming: I'm trying to deal with containers that don't have /dev/console and whatnot.
[21:15] <arthur789> ikonia: then i probably have my terms confused. connected to my local lan i can not use my ddns to access my machine. i can only connect from outside the network
[21:15] <ikonia> arthur789: that's not controlled by host file
[21:17] <arthur789> ikonia: then i am severly confused. i enabled port forwarding and have a dynamic dns set up. sitting at home, connected to my local LAN, i want to type buffet.dlinkddns.com:4040 and get what i want, not 192.168.0.107:4040. how do i do this?
[21:17] <ikonia> arthur789: you need a dns service and/or you need to edit the host file on the CLIENT
[21:20] <arthur789> ikonia: i have dns service. So i need to edit a file for every computer on the LAN? i have done this before and that doesnt seem right. right now im editing the hosts file on the server machine.
[21:21] <ikonia> arthur789: you don't have a dns service for your internal lan
[21:21] <ikonia> arthur789: editing the hostfile on the server is going to do nothin
[21:21] <ikonia> nothing
[21:21] <arthur789> ikonia: i will stop messing with the hosts file then.
[21:23] <arthur789> ikonia: in the past, when things where set up correctly i could get to my server through the buffet.dlinkddns.com while connected to the wifi router
[21:23] <arthur789> ikonia: things went bad and now im trying to re-set-up and this is being a headache
[21:27] <arthur789> ikonia: this is what i have been googling, mainly 'ddns not working inside network'
[21:32] <ikonia> arthur789: it's totally possible, but it's awkward to use an external dns service to resolve internal addresses
[21:32] <ikonia> arthur789: as you're advertising private addresses that are probably going to be dynamically allocated from a private source (your internal router) to a public service
[21:34] <arthur789> ikonia: well, i dunno what i did in the past, but i could do what i am speaking of. editing the hostfile seems familiar... even though you say it has no effect....
[21:45] <arthur789> ikonia: thanks for your help, i need a reboot
[23:02] <arthur789> hi all. in the past i have been able to access a single machine behind a wifi router no matter if i was away from the LAN at a friends or at home, on the LAN by using my dynamic dns. the computer went south and i am on a fresh install of server 12.04. for some reason i cannot get this going again. ikonia tried to help, but im sure i edited the /etc/hosts file in the past. either way im pretty sure i need to edit a file because my rout
[23:21] <hamnstar> arthur789: you likely need to poke a hole in your routers built-in firewall.  Any previous rules might have been associated to a past (DHCP?) IP of the previous install
[23:37] <arthur789> hamnstar: thanks for the response, this is a new router too. there is no firewall setup on the routers config page.
[23:38] <hamnstar> arthur789, any port forwarding settings?
[23:38] <hamnstar> on the router
[23:40] <arthur789> hamnstar: yes, three. all can be seen on canyouseeme. all can be accessed via say cell phone on 4g. but once im home and on wifi i must use the local ip
[23:41] <hamnstar> ohhh yeah routers dont like routing "out" and back "in"
[23:41] <hamnstar> iirc
[23:41] <hamnstar> you could use a hosts file on your LAN devices to get around that?  is that what you had going on before?
[23:46] <arthur789> im pretty sure. when i came in here before asking about esiting it someone else didnt think that was correct. im not sure how to config the /etc/hosts file.
[23:49]  * gholms does exactly that "out and back in" routing all the time, is now confused
[23:49] <arthur789> i think the hosts file needs work, from the googling i did
[23:51] <hamnstar> gholms: looks like i dont recall correctly then... i think some routers might prevent routing a LAN address to wAN and back into LAN to prevent cyclic routing??
[23:51] <arthur789> gholms: i think its cause my router isnt advanced enough? missing some technology.
[23:52] <hamnstar> arthur789, /etc/hosts would statically map the dynamic dns name to a LAN ip... so as soon as you're off your lan, you would lose ability to contact again
[23:52] <arthur789> when i google dyndns not working inside network i get most results refering to hosts file, im just too noob to know what to put
[23:52] <hamnstar> but would work in the home :P
[23:54] <hamnstar> yeah from some googling it looks like some routers do not handle "out and back in" aka "loopback NAT"
[23:54] <arthur789> i have connectbot on my phone. previously, i could ude the same preset no matter where i was or what network i was on. thats my end goal
[23:54] <gholms> hamnstar: Could be.  What I end up with is traffic going to the public IP on the router regardless of which interface it's coming from and getting forwarded to the appropriate box.
[23:55] <gholms> But maybe the router needs some sort of magic to make that work or something.
[23:55] <arthur789> hamnstar: correct, and from my understanding that means the hosts file needs edit
[23:55] <arthur789> gholms: got any extra magic laying around?
[23:55] <gholms> arthur789: Not if you need something that'll work on linux.  :(
[23:56] <hamnstar> arthur789: a more elegant solution would be to set up a dns on your local lan.... quite a bit more involved
[23:56] <arthur789> hamnstar: well, i can barley handle the easy stuff. but im up for it.
[23:57] <hamnstar> arthur789, using the hosts file will always resolve the name of the server to the same ip.... which is not what you want, as you need a local lan address when on your lan, and your routers publicly routable ip when off your home network
[23:58] <jrwren> what? no mdns love?
[23:58] <gholms> Heh
[23:58] <hamnstar> I can't help you too much in setting up a DNS server... but, what you can basically do is set your primary DNS to a local dns server (managed by you) and secondary DNS to your ISP/preferred DNS server
[23:58] <arthur789> wow. i cant belive i ever had it working
[23:59] <hamnstar> I'm gonna guess your old router handled the "loopback NAT" stuff that some routers do not
[23:59] <arthur789> well, i thought i upgraded. guess i need to get my old one back