IRC log for #fredlug on 20070626

01:10.03plarsenhowdy howdy
01:11.25plarsenohh, just me and the stickster again ...
01:11.46plarsenstickster: do you follow novalug?
01:12.03sticksterI pick and choose threads; some I read, some I don't
01:13.02plarsendid you check out the dailycutoftech link ?
01:13.49plarsenhttp://www.dailycupoftech.com/2007/06/19/how-photos-should-be-viewed/
01:13.54plarsenThat is WAY too impressive!!!!
01:22.53sticksterYeah, I remember hearing about that project last year when I was in Seattle
01:23.09sticksterMS Research does, and buys, a lot of interesting things
01:29.04plarsenyup - they invest - but doesn't innovate ;)
01:30.14plarsendo you know if iptables has somekind of "debug" option - or do I need to ethereal everything up and look over 1-2 hours of traffic to see why things aren't working?
01:30.21plarsenI've got some wierd rejections
01:30.42plarsenThey're not considered part of an eixsting connection, but they should have been (it's the return/answer message being rejected)
01:31.57plarsenrejected IN=eth1 OUT= MAC=00:18:8b:3c:0c:c1:00:15:f2:39:8d:b2:08:00 SRC=209.132.177.100 DST=10.1.1.15 LEN=52 TOS=0x00 PREC=0x00 TTL=46 ID=36007 DF PROTO=TCP SPT=443 DPT=2533 WINDOW=3129 RES=0x00 ACK URGP=0
01:32.44plarsenthat is a return message from rhn ... not sure what "df", "res" and "urgp" is about; but it's a simply response to an outgoing message
01:36.55*** join/#fredlug jsmith (n=jsmith@72-255-77-7.client.stsn.net)
01:39.51plarsenahhh, the 3rd musketeer joins the gang!
01:40.32plarsenhey, who cares if you end up with the ladies :)
01:40.48jsmithI guess I do :-)
01:43.21plarsenjsmith: got any ideas of iptables "debug" tools?
01:43.42plarsenI've got an intermiddent problem here ... that comes every 30-60 minutes and rejects ONE package that shouldn't be rejected.
01:43.51jsmithYou mean, besides tcpdump?
01:44.25jsmithTo be honest, I've been spoiled in my short little career, and had hardware firewalls that cost more than my house.
01:44.38plarsenhehe
01:44.39jsmithAs such, I'm not a big fan of iptables
01:44.51jsmithI really ought to do more with it... but I'm lazy
01:44.52plarsenI have that, but I still lock down boxes on the internal network
01:45.06jsmithSure...
01:45.14plarsenhmmm - I'm a pretty big fan; what do you think runs inside most of those fancy firewalls of yours ;)
01:45.26plarsenBut I hate complex network routing problems :(
01:45.29jsmithtcpdump is your friend, and wireshark is worth every ounce of energy you put into learning it
01:45.48jsmithI can guarantee that the firewalls I was using didn't have iptables on the inside
01:46.07jsmithMost had ASICs (application-specific integrated circuits) doing the routing and firewalling
01:46.25plarsenhmmm - nope ;) If I wanted to do that I would use ethereal ... tcpdump doesn't give you an option to post-filter packages to investigate. But I'll rather not track 30-60 minutes (at least) worth of traffic
01:46.46jsmithUnfortunately, iptables has a hard time keeping up with several hundred megabits per second of traffic on the line
01:46.59jsmithplarsen: Capture with tcpdump, analyze later in wireshark
01:47.24jsmithplarsen: And if you know what types of packets to look for, you can filter down the capture so that it doesn't get too large
01:48.03jsmithplarsen: Or, you could setup a setup circular ring-buffers to keep the last 45 minutes worth of data... and then wait for the problem to happen, and stop logging then.
01:49.15plarsenjsmith: ohhh - that last bit sounds interesting. Got any references?
01:50.40jsmithCheck out the -W flag to tcpdump
01:50.58jsmith(or, if you can get wireshark on the box that's having the problems, look at the capture options window)
01:51.34jsmithFor example, you might do something like:
01:51.55jsmithtcpdump -C 50 -W 10 -w filename -i eth1 port 4569
01:53.18jsmithThat's saying "capture all the traffic to or from port 4569 on eth1, and keep 10 files worth of data, with 50 megs in each file"
01:54.05jsmithAfter file number 10, it will go back to file number 1 and begin writing there again.
01:54.45jsmithplarsen: So, here's my proposal... you teach me all about iptables, and I'll teach you about tcpdump
01:56.44plarsenlol - I've used tcpdump a lot ... but I've stuck to ethereal to do package analysis
01:56.45plarsen:)
01:56.53plarsenbut sure - that's a deal
01:56.56plarsenI'm willing to learn more
02:00.47jsmithngrep can be useful too... but again, you have to know what you're looking for first
02:02.27plarsenexactly!
10:43.05*** join/#fredlug jsmith (n=jsmith@72-255-77-7.client.stsn.net)
13:00.37*** join/#fredlug jsmith (n=jsmith@208.253.92.75)
16:42.25*** join/#fredlug bit4man (n=plarsen@c-24-125-211-129.hsd1.va.comcast.net)
16:48.25plarsen_workhowdy doodie
16:50.06plarsen_workAnyway to re-read xorg.conf without restarting it?
16:58.29jsmith-teachingNope
16:58.49jsmith-teachingEdit, restart X, edit again, restart X, edit again, restart X... rinse and repeat
16:59.16jsmith-teachingNow, if you're just changing resolution/orientation, and you have the xrandr module loaded, you can use xrandr to change the resolution and rotation without a restart
16:59.30jsmith-teachingAnything else requires a restart, as I understand it
17:10.21plarsen_workjsmith-teaching, Well, that's basically what I'm trying to do. This monitor (HP w22) isn't listed, and I'm trying to setup the 1680x1050 resolution required by it.
17:10.27plarsen_workKeeps getting overwritten though ....
17:10.46jsmith-teachingWhat happens if you run xrandr?
17:11.13jsmith-teaching(without any options)
17:12.31plarsen_work[plarsen@oem ~]$ xrandr
17:12.32plarsen_work<PROTECTED>
17:12.32plarsen_work*0   1600 x 1200   ( 542mm x 406mm )  *60
17:12.32plarsen_work<PROTECTED>
17:12.32plarsen_work<PROTECTED>
17:12.32plarsen_work<PROTECTED>
17:12.34plarsen_work<PROTECTED>
17:12.36plarsen_work<PROTECTED>
17:12.38plarsen_work<PROTECTED>
17:12.42plarsen_work<PROTECTED>
17:12.44plarsen_workC
17:12.46plarsen_workCurrent rotation - normal
17:12.48plarsen_workCurrent reflection - none
17:12.50plarsen_workRotations possible - normal
17:12.53plarsen_workReflections possible - none
17:15.24jsmith-teachingOk... sounds like you need to add the resolution to your xorg.conf, and possibly add a modeline for that monitor
17:16.45plarsen_work<PROTECTED>
17:16.45plarsen_work<PROTECTED>
17:16.53plarsen_workI've added that - no difference.
17:17.03plarsen_workDoing a full update now, when it's done I'll restart X and see if it helps.
17:17.53jsmith-teachingDo you have any other "Modes" lines in your xorg.conf?
17:18.00jsmith-teachingIf not, you may need a "modelines" line
17:18.46plarsen_worknope, no other modes
17:19.02plarsen_workEhh, sorry yes I do
17:19.10plarsen_workSection "DRI"
17:19.10plarsen_work<PROTECTED>
17:19.10plarsen_work<PROTECTED>
17:19.10plarsen_workEndSection
17:21.12plarsen_workSince the monitor wasn't listed, this is what I ended up with:
17:21.15plarsen_workSection "Monitor"
17:21.15plarsen_work<PROTECTED>
17:21.15plarsen_work<PROTECTED>
17:21.15plarsen_work<PROTECTED>
17:21.15plarsen_work<PROTECTED>
17:21.16plarsen_work<PROTECTED>
17:21.18plarsen_work<PROTECTED>
17:21.20plarsen_workEndSection
17:26.12jsmith-teachingHave you checked the manual to see what your actual Horizontal Sync rates are?
17:28.09plarsen_workvert is to be 60 - that's what it says
17:28.17plarsen_workhori isn't set.
17:28.22plarsen_workRestarting X now :)
17:28.25plarsen_work<PROTECTED>
17:32.38*** join/#fredlug plarsen_work (n=plarsen@c-24-125-211-129.hsd1.va.comcast.net)
17:32.54plarsen_workinteresting ... it changed resolution, but it's not listed in the "display" mangement.
17:46.55jsmith-teachingFigures...
17:47.07jsmith-teachingMy screen does the same thing... 1280x854 is not exactly standard
17:47.43*** join/#fredlug plarsen_work (n=plarsen@c-24-125-211-129.hsd1.va.comcast.net)
17:47.49jsmith-teachingMy screen does the same thing... 1280x854 is not exactly standard
17:47.57jsmith-teaching(repeated for plarsen's sake)
17:48.42plarsen_workhmmm - where does it get the list from? I thought that was what the screen/monitor driver was all about?
17:49.28jsmith-teachingWho knows?!?
21:31.48jsmith-teachingNot nearly long enough...
21:32.01jsmith-teachingYou can do tricks like using media converters to push it over Cat5 or fiber.
21:32.23jsmith-teachingOtherwise, you get strange artifacts in the video with anything longer than about 18 feet
21:41.23plarsen_workhmmm - 18 feet?? I think it's 10 feet and it looks like shadows on the screen :(
21:42.32jsmith-teachingYeah... poor cables
21:42.52jsmith-teachingYou'll especially notice it with vertical lines where there's a high contrast between light and dark
21:42.53plarsen_workhmmm ... true true. It wasn't exactly expensive
21:43.20plarsen_workthere's a slight vertical line(s) moving upwards ... I've never seen that on LCDs before ...
21:43.26plarsen_workSo I'm begining to think it's the cable.
21:43.36plarsen_workI bought an extension so I can access my KVM from my desk
21:44.35jsmith-teachingYeah...
21:52.47plarsen_workGot ideas on where to get quality cables???
21:56.38jsmith-teachingMy best ones came from a company that specifically did audio-video type services
21:56.53jsmith-teachingThey used coax cables in the middle to eliminate the noise on the lines.
22:02.51plarsen_workyeah, I saw several with different filters on them, but I didn't know which if any I needed
22:03.01plarsen_workI use(d) cable'n'more
22:03.05plarsen_workNever been disappointed there before
22:04.45plarsen_workups, correction - cables'n'more
22:04.53plarsen_workwww.cablesnmore.com

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.