#emc | Logs for 2008-05-31

[01:04:43] <klickrr> You know how you set limits on the Axis with MIN_LIMIT MAX_LIMIT, well yesterday I set those up properly with my machine, they worked great, when jogging around the machine wouldn't let me exceed those coordinates. Now today for some reason, it allows them to fly right past them, during jogging, was I in some special jogging mode yesterday? I can't figure this out
[01:06:05] <klickrr> If i run a gcode it still tells me before I run if it exceeds them or not, but I kinda liked jogging and not worring about hitting the sides..
[01:06:18] <jmkasunich> are you homed?
[01:06:27] <klickrr> or tripping a limit switch and forcing a fast stop
[01:06:59] <klickrr> well, that's an interesting question, do I have to hit "Home Axis" before it will make those work? I thought if I just setup the coordinates it woudl adhere to them
[01:07:11] <klickrr> i setup the position.txt file so that I wouldn't have to re-home each time
[01:07:21] <klickrr> although i realize it's not 100% accurate
[01:07:35] <jmkasunich> do you have a home switch?
[01:07:38] <klickrr> yes
[01:07:43] <klickrr> homing sequence works
[01:07:57] <jmkasunich> if you have a home switch, why not home the machine when you power up?
[01:07:59] <klickrr> and yer right, when I said it worked, was after I did the homing sequence
[01:08:18] <klickrr> well, it might take a few minutes if it's way on the other side
[01:08:29] <klickrr> i home slowely
[01:08:34] <jmkasunich> anyway, I'm not sure if that is the cause or not - it might ignore soft limits when you aren't homed
[01:08:48] <jmkasunich> because it doesn't know where you really are, so the limits could be all screwed up
[01:08:48] <klickrr> yea, that makes sense though
[01:08:56] <klickrr> right, that makes sense
[01:09:29] <klickrr> i think yer right, because I did all the homing stuff, then setup position.txt figuring it would just save the data... knowing it wouldn't be 100% accurate but for me it would be close enough between homing
[01:10:07] <klickrr> but after just starting without homing, it would exceed the soft limits
[01:10:10] <klickrr> ok, cool, thanks
[01:11:49] <dmess> try homing a 500 tool t/c when the last operator left it a 492 and it single-pot-steps to pot # 1
[01:12:40] <toastydeath> looooooool
[01:13:03] <toastydeath> come back at lunch
[01:13:05] <toastydeath> is what that means
[01:35:31] <dmess> or the end of the first shift... if something screws up...
[01:36:33] <Guest404> vvfr66..
[05:22:26] <K`zan_emc> Got the new machine (old machine) up with the 8.04 installed. It only gives this old agp card 800x600 though.
[05:22:47] <K`zan_emc> It is like standing 2 feet from a billboard :-).
[05:23:22] <K`zan_emc> Wonder if I can just copy the old ~/emc directory in as well as the .emcrc? Anyone know?
[05:31:17] <fenn> yep
[05:31:24] <K`zan_emc> delays on the old AMD Thunderbird at half the speed of the celeron is like 1/10th of what the celeron was doing. Wow!
[05:31:30] <K`zan_emc> fenn: Thanks!
[05:31:33] <fenn> try running 'sudo dpkg-reconfigure xserver-xorg' to fix the monitor issue
[05:31:44] <K`zan_emc> renesis_: Latency-test
[05:32:13] <K`zan_emc> err re: got turned into renesis, sorry about that.
[05:32:39] <K`zan_emc> fenn: THAT is what I was trying to remember, thanks!
[05:34:54] <K`zan_emc> Weird, the reconfigure blows out on: FATAL: Module battery not found.
[05:35:00] <renesis_> eh
[05:35:01] <renesis_> o
[05:35:04] <K`zan_emc> lemme try that again.
[05:36:32] <K`zan_emc> Same thing, the only battery on this box is cmos battery, wtf?!?
[05:39:09] <K`zan_emc> the word battery doesn't even exist in the xorg.conf file...
[05:39:18] <K`zan_emc> Oh well.
[05:44:01] <K`zan_emc> Nothing on google other than other folks asking the same question with no resolution. Guess you take whatever it decides to give you :-/.
[05:44:13] <K`zan_emc> I guess I can stand across the room to read it LOL!
[05:44:57] <K`zan_emc> Well, enough for the day. Night all.
[07:42:00] <pjm_> good morning
[07:53:38] <Poincare> morning
[12:54:53] <BigJohnT> gezr: http://imagebin.ca/view/rEQYTqIr.html
[13:25:44] <rayh> BigJohnT, stack the yz on the x slider and you've got Mazak's vertical traveling column.
[13:29:24] <SkullWorks_PGAB> Bridge mills are a pain.
[13:30:09] <BigJohnT> hi rayh that is for gezr
[13:31:19] <BigJohnT> why is that SkullWorks_PGAB
[13:33:18] <SkullWorks_PGAB> ever use one?
[13:34:23] <SkullWorks_PGAB> on some the whole toolchanger/tool magazine moves with the gantry
[13:34:59] <BigJohnT> no
[13:35:43] <BigJohnT> he almost has it done http://imagebin.ca/view/sDR2uUxo.html
[13:36:08] <SkullWorks_PGAB> you are limited in material and fixtures that must fit on the table - nearly within the work envelope - at least on one axis
[13:37:29] <BigJohnT> ok
[13:37:59] <SkullWorks_PGAB> loading large work is an extra pain
[13:38:53] <SkullWorks_PGAB> I still think a Knee mill is the most versatile
[13:39:49] <SkullWorks_PGAB> you can have material hanging out both ends and in the front.
[13:40:36] <SWPadnos> if you might have a 30-ton workpiece, moving the machine around is a more predictable load than moving the work
[13:40:46] <anonimasu> yep
[13:40:51] <anonimasu> that makes sense
[13:43:19] <SkullWorks_PGAB> true - but if your doing parts of that size you would likely also have a gantry crane built into the building and/or dedicated forklifts and winches (to drag the part into position).
[13:44:00] <fenn> heh having applied sharpie markers isn't "almost done"
[13:44:11] <SWPadnos> positioning isn't what I was talking about. tuning servos to move a table with 0-30 tons (for example) is a harder problem than tuning servos to move a 5.0-5.1 ton gantry
[13:44:23] <SkullWorks_PGAB> almost conceived
[13:44:26] <SWPadnos> (depending on tool weights and such)
[13:44:40] <fenn> i dont see how you're going to cut the Y ways out of the block using a shaper
[13:44:48] <SkullWorks_PGAB> I understood
[13:45:05] <fenn> maybe i dont know enough about shapers.. wouldnt it get all jammed up because of the blind groove?
[13:45:10] <SWPadnos> fenn, it's separate layers
[13:45:12] <SWPadnos> (I think)
[13:45:26] <SWPadnos> top way surface, way sides, lower way surface, base
[13:45:29] <fenn> hmm all i see are some blocks
[13:45:40] <fenn> doing it as layers would be the sensible thing of course
[13:45:57] <fenn> and cutting out dovetails would be smarter if all you have are some blocks..
[13:46:06] <fenn> to each his own mistakes
[13:58:04] <JymmmEMC> SWPadnos: yo
[14:48:50] <SWPadnos> hiya
[14:50:15] <JymmmEMC> SWPadnos: Hey
[14:50:24] <SWPadnos> whassup?
[14:50:48] <JymmmEMC> SWPadnos: I'm thinking about getting a quad box and tossing it the DC and running VM's.
[14:51:05] <SWPadnos> DC?
[14:51:10] <JymmmEMC> DataCenter
[14:51:24] <SWPadnos> ah - tossing it in the DC? :)
[14:52:02] <JymmmEMC> SWPadnos: Replacing the existing (older) 1U box with a quad dual core box instead.
[14:52:05] <SWPadnos> quad core or quad CPU?
[14:52:17] <SWPadnos> oh, interesting
[14:52:23] <SWPadnos> AMD or Intel?
[14:52:36] <SWPadnos> (I'd suggest AMD for multi-chip motherboards)
[14:52:43] <JymmmEMC> SWPadnos: Then just run VM's on it. Probably Intel, whatever Dell surplus has on sale
[14:52:58] <SWPadnos> what's your budget?
[14:53:12] <JymmmEMC> They have a couple 1U versions of their 2940's
[14:54:17] <SWPadnos> 1U brings in several problems that you don't have with 2U or 4U though
[14:54:29] <JymmmEMC> SWPadnos: It'll really depend on what they offer, I'd like something with removable drives
[14:54:36] <SWPadnos> power supplies, short memory sticks, short cooling devices, ...
[14:54:38] <JymmmEMC> SWPadnos: No real choice in it.
[14:54:46] <SWPadnos> oh, bummer
[14:55:14] <JymmmEMC> SWPadnos: The idea is to start making revenue to eventually get to a point of getting a full rack
[14:56:13] <SWPadnos> well, that's a good plan :)
[14:57:29] <SWPadnos> I don't see any 4-socket 1U systems
[14:58:00] <SWPadnos> it's unlikely you'll need more than 1 or 2 cores anyway, unless you're getting gigE connection rates (average)
[14:58:33] <JymmmEMC> SWPadnos: They're for the VM's primarily
[14:58:57] <SWPadnos> I guess it depends on how much work you expect each VM to do
[14:59:19] <SWPadnos> memory is more important than CPU though, if there are going to be a bunch of web servers or lightweight DBs in the VMs
[14:59:44] <SWPadnos> if you wanted people to be able to run compile servers or compute servers, that would be a different thing
[14:59:58] <JymmmEMC> ppl write fucked up code, so have to limit cpu usage (easy in a VM), still working on bandwidth limitation
[15:00:42] <JymmmEMC> $1400 http://www.dell.com/content/products/productdetails.aspx/pedge_1950?c=us&cs=28&l=en&s=dfb
[15:00:54] <JymmmEMC> dual quad core
[15:01:11] <JymmmEMC> just wish it had four hdd's instead of two
[15:02:03] <SWPadnos> ok, so dual quad-core
[15:02:22] <JymmmEMC> SWPadnos: only becasue I cna't find 4 cou on their website =)
[15:02:32] <JymmmEMC> s/cou/cpu/
[15:02:47] <SWPadnos> I didn't see any listed - I was on the overview page that shows all the rack servers
[15:03:10] <SWPadnos> the only 4-CPU units are 4U
[15:03:20] <JymmmEMC> SWPadnos: Yeah, so scratch that.
[15:04:04] <SWPadnos> what's there now?
[15:04:09] <SWPadnos> (that you would be replacing)
[15:04:10] <JymmmEMC> POS
[15:04:17] <SWPadnos> how slow a POS? :)
[15:04:27] <JymmmEMC> P3
[15:04:37] <JymmmEMC> I got it for $1, what can I say
[15:05:14] <SWPadnos> if everything the customer sees is a VM, that gives you a perfect way of upgrading hardware without anyone noticing
[15:05:17] <JymmmEMC> Not sure on this FBD dimms though.
[15:06:14] <SWPadnos> http://www.siliconmechanics.com/i8188/quad-opteron.php :)
[15:07:19] <JymmmEMC> SWPadnos: Yeah, but I haven't tried to get any clients due to it the existing box. With VM's at least there will be their own environment
[15:07:43] <SWPadnos> and all you ahve to do is copy the VM files over to the new machine, and all is well (in theory anyway)
[15:07:45] <SWPadnos> have
[15:08:12] <JymmmEMC> No VM now, this is the plan though
[15:08:27] <JymmmEMC> THEN, I'll be able to cp over the VM's easily enough.
[15:10:12] <JymmmEMC> Hmmm, I think supermicro has a 1U box with 4 hotswap drives.
[15:11:12] <JymmmEMC> woohoo 8 hotswap drives in a 1U http://supermicro.com/products/system/1U/1025/SYS-1025C-UR.cfm
[15:11:13] <SWPadnos> http://www.siliconmechanics.com/i9226/quad-core-opteron.php
[15:11:27] <JymmmEMC> SWPadnos: yeah, your links can bite me! =)
[15:11:30] <SWPadnos> heh
[15:11:37] <SWPadnos> $2k - are the Dells much less?
[15:12:38] <JymmmEMC> SWPadnos: I can call dell and they will be in the DC within 4 hours =)
[15:13:07] <SWPadnos> well la-ti-da
[15:13:42] <JymmmEMC> Heh, well if your in BFE, you know your stuff is being taken care of, and tha'ts a good thing.
[15:13:55] <SWPadnos> BFE?
[15:14:07] <JymmmEMC> Bum Fuck Egypt (in the middle of nowhere)
[15:14:12] <SWPadnos> ah
[15:14:34] <JymmmEMC> You don't even have to be in the country and you know things are being fixed.
[15:15:02] <SWPadnos> is that a function of Dell or the data center?
[15:15:09] <JymmmEMC> Dell
[15:15:20] <SWPadnos> even on refurb/outet stuff?
[15:15:34] <SWPadnos> (or is that an extra cost warrantee?)
[15:15:42] <JymmmEMC> if you buy warranty dell will do anything
[15:16:19] <SWPadnos> that's a $400 option at that site I linked, for 3 years
[15:16:27] <SWPadnos> about the same as Dell, I think
[15:16:41] <JymmmEMC> On site within 4 hours for $400?
[15:16:45] <SWPadnos> yep
[15:16:51] <JymmmEMC> 24/7/365?
[15:16:54] <SWPadnos> 24x7 phone and 4 hours onsite
[15:17:00] <JymmmEMC> ah cool.
[15:18:18] <SWPadnos> I just configured a mega-system: 2x 2GHz quad core, 16GB RAM, 4x1TB Barracuda HD, with the big 3yr warrantee - $4239
[15:19:07] <JymmmEMC> 288Watts though.
[15:19:37] <SWPadnos> 378, as configured
[15:19:42] <SWPadnos> (that's a cool feature!)
[15:21:22] <SWPadnos> changing to HE CPUs and WD hard drives takes out 75W, down to 303
[15:21:54] <JymmmEMC> Yeah, I think I'm screwed there. Need around 100W (I think)
[15:22:09] <SWPadnos> oh, then you can't do dual-chip
[15:22:37] <SWPadnos> and you probably need to use an Opteron EE, which is a 65W quad-core
[15:22:41] <SWPadnos> dunno if the Inels are any lower
[15:22:52] <JymmmEMC> AMd's are power hungry
[15:23:06] <SWPadnos> not the EE series, they're quite low power
[15:23:23] <SWPadnos> also, be careful of Intel power specs - I think those are averages, not maximums (like AMD)
[15:23:31] <JymmmEMC> ah
[15:23:37] <SWPadnos> (and of course the AMD number includes the memory controller)
[15:24:12] <SWPadnos> I'm not positive about the Intel specs, but I do know that you are running the chip out of spec if you disable the thermal management SMI interrupts
[15:26:42] <JymmmEMC> Maybe I'll have to find something else. I'm not sure how much power is available, but I sure know it aint 800 Watts.
[15:28:38] <JymmmEMC> SWPadnos: Oh, check this out....
[15:29:14] <JymmmEMC> SWPadnos: http://denyhosts.sourceforge.net/
[15:39:40] <jmkasunich> JymmmEMC: interesting
[15:39:48] <jmkasunich> that prompted me to look at auth.log
[15:40:15] <jmkasunich> it has 4300 lines in it - all from today
[15:40:42] <JymmmEMC> jmkasunich: Yeah one of the guys at work turned me on to it.
[15:40:58] <JymmmEMC> and i trust when he suggests something.
[15:41:25] <JymmmEMC> His KungFu is good =)
[15:43:54] <jmkasunich> 785 attempts to log in as root
[15:47:45] <JymmmEMC> which port?
[15:47:49] <JymmmEMC> 22?
[15:48:40] <jmkasunich> yes
[15:48:57] <JymmmEMC> Yeah, thus why I don't have sshd on port 22 =)
[15:49:04] <jmkasunich> I think anyway - I'm not looking at firewall logs, I'm looking at sshd logs
[15:49:40] <JymmmEMC> no ftp either...scp ftw!
[15:50:14] <alex_joni> ftp is the top problematic thingie
[15:50:24] <alex_joni> just after bad php codes on a webserver
[15:50:32] <JymmmEMC> alex_joni: not to mention pop3/imap in the clear too
[15:50:54] <alex_joni> well.. yeah, that's just misconfigured :D
[15:51:00] <alex_joni> especially if no-TLS
[15:51:18] <JymmmEMC> screw tls, ssl baby =)
[15:53:19] <JymmmEMC> Hmmm there's TLS-PSK too, that might be interesting.
[15:53:38] <alex_joni> I thought PSK wasn't that secure
[15:54:14] <JymmmEMC> alex_joni: No, that' be the RNG in debian/ununtu =)
[15:54:17] <JymmmEMC> lol
[15:55:08] <JymmmEMC> OpenPGP certs in the works. That be cool
[15:55:10] <Vq^> JymmmEMC: there was nothing wrong with the RNG!
[15:55:25] <JymmmEMC> Vq^: Yeah, uh huh
[15:55:56] <JymmmEMC> We found over 32,000 duplicate keys within the first hour.
[15:56:54] <Vq^> probably because the RNG wasn't used when generating the keys
[15:57:19] <JymmmEMC> You keep believing that
[15:57:57] <Vq^> not necessarily
[15:58:59] <Vq^> but my understanding of the problem was that the RNG got disabled
[15:59:03] <Vq^> am i wrong?
[16:01:03] <JymmmEMC> Does it really matter how it occured? be it a bug, or human error. It happened and in a very bad way. Had it been the GPG keys that got duplicated Debian/Ubunutu would have had to start the "web of trust" all over again from scratch as ALL distributed pkgs would have been null and void.
[16:02:18] <cradek> people have been knocking on ssh's door for years
[16:03:10] <Vq^> bug or human error? :)
[16:03:15] <jmkasunich> understood - but the more they knock, the greater the chance they will get lucky
[16:03:27] <JymmmEMC> cradek: doens't mena you need to let them in on the property =)
[16:03:59] <JymmmEMC> (aka No Trespassing)
[16:05:13] <jmkasunich> if nothing else, switching the ssh port will save a few megs of disk space
[16:05:30] <jmkasunich> (the space occupied by logging the thousands and thousands of attempted breakins
[16:05:46] <Vq^> it was a bit weird that the patch could live outside openssl for so long
[16:08:46] <gezr> mornin, and whats all this about?
[16:08:53] <SWPadnos> note that the bug was found on both Debian and Ubuntu systems
[16:09:24] <SWPadnos> I seem to remember someone saying that they were switching to Debian over this problem ;)
[16:09:47] <gezr> what was the problem?
[16:10:57] <SWPadnos> SSH key generation problem
[16:11:13] <gezr> exploit?
[16:12:02] <cradek> there were some weak ssh keys out there that were generated on newish debian-based systems
[16:12:21] <SWPadnos> so potentially eaasier to guess at keys
[16:12:48] <cradek> SWPadnos: very easy to brute force. there's a working exploit in the wild.
[16:12:49] <gezr> why just debian-baised?
[16:13:01] <cradek> gezr: a debian person made the programming error
[16:13:13] <gezr> damn it batman
[16:13:39] <cradek> the good news is it's fixed
[16:13:58] <SWPadnos> the latest Ubuntu (and presumably debian) packages have a utility called ssh-vulnkey that lets you check your keys against the list of vulnerable ones
[16:14:16] <SWPadnos> http://www.ubuntu.com/usn/usn-612-2
[16:24:59] <JymmmEMC> cradek: Not weak keys, but duplicate keys, as there were only 32,000 potential available.
[16:25:19] <cradek> yes that's why the brute force was so easy
[16:25:42] <JymmmEMC> I wouldn't even call it brute force, more like a slight nudge and you're in.
[16:26:30] <cradek> brute force is commonly used to describe an attack where you can try 'every possibility' in order to find the one that works. it doesn't really mean that it's hard
[16:28:05] <JymmmEMC> Yes, I'm fully aware of the meaning. I do it all the type on my boxes. But when you have ALL the keys on your keyring and you KNWO for sure that one of them will work, there's not much brute force envolved, it's more like you forget which key it was.
[16:30:02] <anonimasu> random testing stuff is still a bruteforce attack
[16:30:17] <JymmmEMC> who says it was random?
[16:30:33] <anonimasu> well, going through a list until you find something that works..
[16:30:35] <JymmmEMC> (eeesh havne't you ppl read all the comics strips yet?)
[16:32:11] <DanielFalck> systematic
[16:32:18] <JymmmEMC> http://www.random.org/analysis/dilbert.jpg
[16:38:55] <JymmmEMC> jmkasunich: you do python (a bit) don't ya?
[16:39:19] <jmkasunich> a very very small bit
[16:40:56] <JymmmEMC> jmkasunich: It's all good. But that DenyHosts script is pretty cool. Once I get that going, gonna stop using passwords too, and just keys. I found out that if you shell into one box, then from there into another, the key is automatically passed forward for you.
[16:42:03] <JymmmEMC> And, well... you jsut can't beat a 4K "password" key, compared to a 12 to 64 char password.
[16:42:21] <jmkasunich> just keys is independent of whether or not you have denyhosts running (and far more effective)
[16:42:45] <jmkasunich> but it means you can't log in from a machine that doesn't have your key on it
[16:43:04] <JymmmEMC> jmkasunich: Sure, but thumbsticks are far too cheap these days.
[16:47:53] <lerman_> lerman_ is now known as lerman
[16:48:07] <gezr> hmm, those tools wernt installed with the new update
[16:53:03] <awallin> some machining pics from last weekend and today: http://www.anderswallin.net/2008/05/finderscope-rings/
[16:57:30] <awallin> could someone go here http://www.anderswallin.net/2008/05/steel-bulb-nr3/ and tell me if they see 150pixel thumbnails or full-size pics all overlapped and messed up ?
[16:58:11] <gezr> sure
[16:58:19] <DanielFalck> awallin: overlapping here
[16:58:52] <awallin> there's something that broke with the thumbnails in wordpress then...
[16:58:53] <gezr> overlapping for me as well, firefox on a winbox
[16:58:54] <JymmmEMC> overlapping hugemungo images
[16:59:10] <jepler> same here
[16:59:12] <lerman> And here. Also firefox on windoze.
[16:59:37] <jepler> ff3 beta
[16:59:44] <gezr> the grasshopper looks cool
[16:59:51] <awallin> the friendly advice from #wordpress is to try to rebuild the gallery. kind of like rebooting the pc when something doesn't work
[16:59:57] <JymmmEMC> awallin: It's the stylesheet that's the problem. disbale it and no overlapping, but just large images.
[17:00:52] <alex_joni> awallin: same here.. Opera on doze
[17:01:01] <JymmmEMC> awallin: So, is this still true? proudly powered by WordPress
[17:01:08] <JymmmEMC> * JymmmEMC snickers
[17:01:41] <alex_joni> proudly bugged by WordPress
[17:02:09] <JymmmEMC> Sounds liek the standard response whne they dont know wtf the problem is or how to fix it.
[17:02:15] <awallin> JymmmEMC: hrm... it worked until today
[17:02:25] <alex_joni> wtf is wrong with freenode?
[17:02:31] <alex_joni> didn't see a wallop
[17:02:51] <jepler> all I saw was this: 11:49:41 [freenode] -!- WALLOP tomaw: Hi all, some scheduled maintenance at one of our sponsors caused some users to split there, please keep with us while they reconnect.
[17:04:00] <jmkasunich> logger_emc: bookmark
[17:04:00] <jmkasunich> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2008-05-31.txt
[17:04:10] <alex_joni> jmkasunich: network is acting up
[17:06:26] <jmkasunich> yeah
[17:07:22] <jmkasunich> on my end all I got was a timeout (well after I actually dropped off the channel)
[17:13:50] <gezr> sweet, space shuttle launch in a few hours
[17:14:24] <alex_joni> if you're quick you might still catch it
[17:14:32] <pjm_> gezr over on another irc net we're tracking the comms that support the shuttle launch, and we'll have direct comms from the orbiter when it flys over head
[17:14:52] <gezr> thought it was a 5pm est launch?
[17:14:59] <gezr> pjm_: really?
[17:15:40] <jmkasunich> pjm_: what channel?
[17:15:49] <pjm_> its #hearsat on starchat IRC
[17:15:57] <gezr> i just fired up nasa tv is all
[17:16:18] <gezr> hey, any of you guys know a lot about enamel paints?
[17:17:07] <pjm_> http://www.uhf-satcom.co.uk/sounds/ has some of the sound samples we've recorded of shuttle comms
[17:22:05] <[2]a-l-p-h-a> [2]a-l-p-h-a is now known as a-l-p-h-a
[17:22:52] <gezr> yeah, nasa tv shows them in the van on the way to the shuttle, very cool
[17:23:17] <pjm_> gezr so anyway if u are under the ground track u can hear direct comm from the orbiter during ascent
[17:23:39] <gezr> very cool
[17:26:53] <awallin> how long till the launch?
[17:27:21] <pjm_> it goes at 2102UTC
[17:27:29] <gezr> sorry, just heard it, and ah 502pm eastern
[17:27:47] <gezr> brain wasnt in listen mode, I have been reading about paint
[17:27:51] <jmkasunich> bummer - I don't get a pic from nasatv, only audio
[17:28:13] <awallin> so it's still around 3.5 hours
[17:29:21] <pjm_> jmkasunich u can try the other stream types, there is also realvideo and windows media etc
[17:29:26] <gezr> hahaha, talking about the damn van, "they used it during apolo"
[17:29:34] <pjm_> i'm using VLC on the windows media stream and its fine
[17:29:58] <gezr> jmkasunich: your just missing a video codec i bet
[17:30:10] <jmkasunich> probably
[17:30:13] <jmkasunich> I hate codexs
[17:30:18] <jmkasunich> codecs
[17:30:28] <gezr> they take the free right out of fun
[17:30:40] <gezr> try a different type of broadcast
[17:32:30] <gezr> its my understanding that part of this mission is to replace broken parts on the russian toilet?
[17:35:03] <gezr> I wish space flight was treated with a few more risks I guess, 10 years ago, we could have loaded a module with a few prisoners, and sent them straight to mars
[17:36:12] <pjm_> hehh that would be bad as mars could end up like austrailia
[17:36:15] <gezr> its just not treated from the lewis and clark mind set, were going to go do this, and we dont know what we face. Both of them could have been killed, eaten, starved whatever, just cause they didnt doesnt mean that the chance wasnt there, and they didn't care
[17:36:24] <pjm_> better to send them to a mission to observe the sun
[17:36:47] <gezr> oh man
[17:37:55] <gezr> so this canned spray paint says apply all coats within 2 hours or so, and I have another enamel paint to put in my gun to do the fine details, however I dont know, ah hell, I should just do it to it eh?
[17:38:50] <gefink> alex_joni: Hello, remember my duron800-latency problem. I solved a part of it
[17:39:25] <alex_joni> gefink: cool.. what was it?
[17:39:54] <gefink> like someone in this chat sayed the cdrom
[17:40:19] <gefink> anything is permanent accessing block mayor11
[17:40:39] <dmwaters> {global notice} Hi all, we've had some maintenence going on with one of our sponsors. We also found a small problem with services that we're in the process of working on. Thank you for your patience, and thank you for using freenode!
[17:41:13] <gefink> if i change sr_mod in the aliasfile to off and then unload sr_mod the problem is gone
[17:41:49] <gefink> after reboot sr_mod is then only loaded at start
[17:42:11] <SWPadnos> gezr, the paint that says to apply within 2 hours may form a hard coat that doesn't allow later layers to stick - you may essentially end up with several separate layers on top of each other
[17:42:31] <SWPadnos> (like the Hoover Dam - it was done in a continuous pour so it would all stick together :) )
[17:42:37] <gezr> yeah, thats what I was affraid of, within 2 or after 48...
[17:42:55] <SWPadnos> oh - after 48, interesting
[17:43:18] <gezr> it may have to do with a shrinking or something, if you put on after 2, part is dry, and the other isnt, so one does something that affects the other
[17:43:28] <gezr> maybe ill just be paitent
[17:43:38] <gefink> i will try to find who is permanently using mayor11. i think its gnome or any automounter but lsof dont help
[17:43:52] <SWPadnos> it may flow better before 2, be har dafter 2, but get "globby" in between
[17:44:44] <SWPadnos> ii is SCSI CD-ROM
[17:44:46] <SWPadnos> 11 is
[17:44:55] <SWPadnos> http://www.kernel.org/pub/linux/docs/device-list/devices.txt
[17:45:01] <gefink> no its an ide-cdrom
[17:45:11] <SWPadnos> device 11 is SCSI CD-ROM :)
[17:45:26] <JymmmEMC> Yeah... SCUM-C CD-ROM
[17:45:29] <SWPadnos> most CD-ROMs are run with the scsi layer anyway, regardless of the connection
[17:45:42] <gefink> but under ubuntu 8 all ide devices are scsi bounded
[17:46:19] <alex_joni> gefink: if you boot with "all-generic-ide" I think they are not
[17:46:57] <gefink> i dont tried that because the boot want a scsi-harddisk
[17:47:26] <gefink> but i can try root=....
[17:48:08] <gefink> hmm fstab also
[17:48:34] <SWPadnos> if it's installed already, the hardware detection may have set things up for SCSI devices
[17:49:06] <SWPadnos> you might try reinstalling after booting the CD with the all-generic-ide boot parameter
[17:49:51] <gefink> its less problem to change some files....
[17:50:23] <gefink> or try to remove the automounter
[17:50:52] <alex_joni> depends on who's doing the work :)
[17:51:44] <gefink> may be
[17:53:39] <gefink> Block11 is called verry often because after rmmod sr_mod it will reloaded faster then a second
[17:54:42] <SWPadnos> are you sure the module actually gets unloaded?
[17:55:04] <gefink> yes. i checked the order
[17:55:12] <SWPadnos> ok
[17:55:17] <gefink> after rmmod its top
[17:55:40] <gefink> when the alias is changed to off its not reloaded
[19:01:06] <gezr> using an air brush to fill in the small details was a great idea, however the end result isnt as good as I expected
[19:01:32] <gezr> I should buy a small touchup gun and just spray the whole thing with a good coat in a few days
[19:01:55] <gezr> or just use the last can of the good stuff, and call it a done deal monday night
[19:02:50] <gezr> its not bad right now, heck its just a frame :
[19:03:01] <gezr> and im not making a masterpiece, just want a motorcycle
[19:23:50] <jmkasunich> * jmkasunich just made 30 holes
[19:24:08] <jmkasunich> CNC is handy sometimes
[19:24:28] <alex_joni> heh :)
[19:28:42] <gezr> wo wo wo wo wo, openssh == ssh?
[19:29:31] <gezr> no they dont, they are not the same
[19:31:47] <gefink> gezr: it shold work together
[19:32:00] <alex_joni> same protocol
[19:32:13] <gezr> yeah, but the packages are not the same...
[19:32:34] <gefink> hmm. is one the sshd?
[19:33:01] <gezr> yeah, I guess they are intermixxed
[19:33:11] <gezr> im trying to check the current state of debian
[19:35:04] <gezr> openssh-server_1%3a4.3p2-9_i386.deb
[19:36:19] <gezr> yeah, i should be good to go
[19:49:36] <awallin_mibbit> anyone done gear-hobbing with emc?
[19:51:54] <anonimasu> awallin_mibbit: got the video online yet?
[19:52:14] <anonimasu> of thoose rings?
[19:55:12] <rayh> awallin_mibbit, I've talked with Roltek a bit about gear hobbing. He does a lot of manual hobbing for transmissions.
[19:57:55] <rayh> He believes it is difficult to get the arc second accuracy needed between the work spindle and hob.
[19:58:30] <rayh> There are a bunch of patents on various approaches.
[20:04:18] <gezr> 13min till launch
[20:04:22] <dmess> if you can rigid tap you should be able to hob
[20:05:05] <gezr> tapping is 1 axis slaved, hobbing requires 2 to be slaved to each other right?
[20:05:22] <dmess> your just sync'ing 2 rotary axes instead of 1 linear 1 rotary
[20:05:22] <rayh> two rotary yes
[20:05:39] <rayh> The real issue is the accuracy between them.
[20:06:13] <anonimasu> hm, then it's one of thoose problems where if you throw enough cash on them it dissapears
[20:06:13] <anonimasu> :p
[20:06:15] <dmess> we do it on our WFL-Millturns.. with siemens 840D
[20:06:35] <gezr> but the siemens 840D is an uber control
[20:06:46] <dmess> so is EMC2
[20:07:00] <rayh> Most of the big control makers have a blackbox add on for hobbing and gear grinding.
[20:07:40] <dmess> siemens has it built int the 840d... just learn to use it
[20:07:52] <rayh> I'd like to see someone try it.
[20:07:55] <gezr> yeah, they are not controled via pcc, its all offloaded to the drives
[20:08:27] <awallin_mibbit> anyway I need a rotary table for that. has anyone done a cnc-conversion of a manual dividing head?
[20:09:07] <rayh> Yea. Matt Shaver for a clocker maker in New Jersey USA.
[20:10:00] <rayh> http://www.arcadianclock.com/our_workshop/index.html
[20:10:20] <gezr> hmm, couldnt emc easly do it, if both hobb and gear were rotated via a gear drive?
[20:10:31] <gezr> or would the accuracy still be lost?
[20:10:51] <dmess> thats OLD school...
[20:11:04] <rayh> Sure if they were geared together
[20:11:09] <dmess> i could do it without CNC if i did that
[20:11:14] <rayh> But then you loose flexibility.
[20:11:17] <gezr> yeah, I guess thats true
[20:11:35] <rayh> Emc could crown and set angles.
[20:11:42] <dmess> what about electronically gearing them together??
[20:12:05] <rayh> Yes.
[20:12:24] <gezr> thats where the issues develop I think, maybe in a black box, but I dont think the pc could update fastenough, is that where the problems lie pretty much?
[20:12:28] <rayh> But to get arc/second accuracy you would need some high count encoders.
[20:12:38] <dmess> using the feedback from the first drive to run the following drive??
[20:12:59] <rayh> Exactly.
[20:13:05] <rayh> gotta run
[20:13:13] <lerman> Also any gears between the encoder and the table would have to have arc second precision.
[20:13:33] <dmess> so add a rotary scale to both might work..
[20:14:14] <lerman> Yes. One piece of good news is that backlash isn't important in this application.
[20:14:43] <anonimasu> howcome?
[20:14:56] <dmess> understand arc-second precision is needed to get to mars.. here at home arc-minutes are fine..
[20:14:56] <gezr> dmess: you know, i bet it would be good enough to make gears comprable to cheap imports
[20:15:01] <lerman> The axis doesn't reverse direction.
[20:15:21] <anonimasu> ah ok
[20:15:39] <gezr> in fact I think it would be able to make decent gears
[20:15:47] <dmess> gezr.... look at what you get off shore.... ITS CRAP
[20:16:07] <gezr> yeah, thats why I revised to say decent
[20:16:26] <dmess> i see no reason why NOT... correct tooling.. speeds & feeds...
[20:16:46] <gezr> if you have minimal torque problems, I dont see why simple encoders, and proper cuts wouldnt make a good gear
[20:17:19] <dmess> its all in the programming.. :)
[20:17:28] <gezr> yeah
[20:17:51] <gezr> and you still need some fine control of the motor timmings
[20:18:25] <gezr> but you could make good enough gears to make a hobbing machine to make perfect gears, to make all the gears youll ever need
[20:18:40] <dmess> we are currently getting ready to cut the gear on the v22-osprey's new nose landing gear... it should be fun... partial sector..
[20:19:10] <gezr> dmess: I really like the 840 control by the way
[20:19:34] <dmess> so do i.... she's MY sweetheart...
[20:19:49] <gezr> she has a ppro chip in her if im thinking correctly
[20:19:53] <gezr> 200mhz
[20:20:12] <dmess> as well as any # of other control/machine combinations i've worked on
[20:20:18] <gezr> they have probably upgraded that since I was list playing with one
[20:20:34] <gezr> fagor uses linux in the background
[20:20:35] <dmess> dual 200 ppro's
[20:20:47] <gezr> yeah thats right
[20:21:01] <dmess> fagor... uses linux???
[20:21:10] <dmess> cool
[20:21:19] <gezr> yep
[20:21:32] <gezr> by far the most user friendly control ive ever used
[20:22:09] <gezr> comercial contol that is
[20:22:10] <dmess> i know a friend who did a few retrofits with fagor's and said it was a dream... but never got into the specifics
[20:22:20] <gezr> they are just awsome
[20:22:53] <dmess> can you re-use existing drives and the likes??
[20:22:59] <gezr> I believe so
[20:23:27] <dmess> i may have to chat with Vits again.. to get the source..
[20:23:40] <gezr> oh oh oh, uber
[20:24:14] <gezr> I dont know if its emc, or if they just use linux to talk to the dives, they are all fiber connected,
[20:24:19] <dmess> i have a knack of absorbing software..
[20:26:16] <gezr> i bet gear cutting would be rather simple on a control like the grex
[20:26:40] <gezr> well, thats probably wrong
[20:26:43] <gezr> I dont kow
[20:27:04] <gezr> I need to check emc out at some point in time today, or get to work cutting the ways for my little machine
[20:27:18] <gezr> I cant paint, I cant assemble the motor, cant do nothing on the bike till the frame is complete really
[20:27:25] <dmess> its all about the SAME hardware on the 2 timed axes..
[20:28:08] <dmess> http://www.snkamerica.com/niigata/niiuaxis/u-axis.shtml
[20:28:13] <alex_joni> good night all
[20:28:19] <dmess> check that out
[20:28:23] <BigJohnT> good night alex
[20:28:27] <dmess> gnite Alex
[20:29:43] <gezr> yeah the nigatas are nice
[20:30:48] <dmess> ive run a few... and set up at a show beside them... HARD act to play with... But i still winned
[20:31:20] <dmess> i was with Toshiba machine then..
[20:31:32] <gezr> BigJohnT: you think that your method of the topped U hold is better then a straight inverted L method?
[20:31:55] <gezr> or interlocking L's is what I was going to go with
[20:32:30] <BigJohnT> should be the same
[20:33:03] <gezr> the [-----] may be easier to machine though
[20:33:06] <BigJohnT> If it was a machine that would get a lot of use I would prefer seperate pieces so only the worn ones can be changed
[20:33:19] <BigJohnT> flats are always easier to make
[20:33:30] <gezr> I may have a chip issue with that though
[20:33:53] <gezr> if its L7 then the ways are semi covered
[20:35:40] <gezr> hmm, crap, I just discovered a flaw
[20:37:17] <gezr> oh well, the ways are just goign to be mega ultra small
[20:39:09] <BigJohnT> that's why I like Solidworks I just change things till they are correct... then make the parts. Of course it cost $5k so it's not for eveyone...
[20:39:23] <SWPadnos> don't forget that CAMWorks is extra ;)
[20:39:30] <anonimasu> http://www.snkamerica.com/niigata/niihor/hnc.shtml
[20:39:35] <anonimasu> I love that design
[20:39:38] <SWPadnos> or SolidCAM or whatever they're calling it now
[20:39:46] <BigJohnT> SWPadnos: everything is extra with solidworkds
[20:39:53] <anonimasu> yep
[20:39:59] <SWPadnos> yes, I know
[20:40:26] <toastydeath> we have a bunch of machines like that at work, but smaller
[20:40:38] <dmess> im planning on a u-drive and quill type dual pallet machine
[20:40:52] <dmess> nigata's?
[20:40:57] <toastydeath> nah
[20:41:00] <toastydeath> some no-name brand
[20:41:10] <dmess> tiawanese??
[20:41:16] <toastydeath> no idea, but probably
[20:41:26] <dmess> any good??
[20:41:32] <dmess> cheap??
[20:41:32] <toastydeath> yeo
[20:41:34] <toastydeath> *yep
[20:41:38] <toastydeath> but i have no idea on $$$
[20:41:40] <dmess> diaposably?
[20:41:42] <toastydeath> Saelio, i believe
[20:41:56] <dmess> spanish..
[20:42:05] <dmess> or brazil..
[20:42:06] <toastydeath> 15x15x15 envelope
[20:42:15] <toastydeath> dual pallet, two 6k rpm machines and one 10k
[20:42:34] <toastydeath> they work good, no real complaints
[20:42:38] <dmess> i need 1.5Mx 1.5m x 1.5m or better..
[20:42:45] <toastydeath> mori sekiki!
[20:42:54] <toastydeath> mori makes a quad pallet machine that big.
[20:42:59] <anonimasu> hehe
[20:43:03] <anonimasu> $$$
[20:43:04] <toastydeath> er, seiki
[20:43:20] <dmess> how much... does it have a quill
[20:43:22] <toastydeath> i don't think you can find 1.5m work envelope for less than $$$$
[20:43:34] <anonimasu> hehe
[20:43:43] <toastydeath> i've never asked mori, because there's no way in hell we'd ever need a 1.5m horizontal pallet machine
[20:44:09] <dmess> $$$$$ isnt an issue... WE NEED to stay in buisiness
[20:44:26] <toastydeath> and i don't know about quills on them, i just know they have a massive pallet changing hmc
[20:45:03] <toastydeath> but it sounds like you're looking for a pallet changing boring mill more than an HMC?
[20:45:22] <dmess> im lookin' at 3M x-axis stuff... floor type table on the left and a rotary on the right...
[20:45:28] <toastydeath> and i don't know about those other than they exist somewhere
[20:45:41] <toastydeath> ah
[20:45:57] <toastydeath> can't really help =(
[20:46:09] <toastydeath> used or new
[20:46:22] <dmess> Toshiba makes them... but the are older technology search Awia machine tools
[20:47:01] <toastydeath> there are a bunch of smaller hbm manufacturers now
[20:47:10] <toastydeath> not small mills, but small companies
[20:47:45] <toastydeath> i'll bbl
[20:48:10] <dmess> yeah.. but im leary... this stuff need years of heritage..
[20:51:32] <gezr> final count down has begun
[20:51:52] <gezr> oops will resume in 2min
[20:51:56] <jmkasunich> heh
[20:52:20] <gezr> they say clear to launch, then they say "countdown will resume in 2min 30s
[20:53:01] <jmkasunich> clear to launch means the go/nogo poll is ok - they want that completed before the end of the scheduled hold
[20:54:17] <jmkasunich> 9 min and counting
[21:00:07] <jmkasunich> 3:00
[21:00:18] <SWPadnos> are you getting video?
[21:00:38] <jmkasunich> no, just audio
[21:01:14] <SWPadnos> ok. it looks like I might get video (at least my right monitor just blanked out)
[21:01:29] <jmkasunich> blanking is easy - I can do that
[21:01:51] <SWPadnos> but does your video driver do that automatically when it's about to show video fullscreen on that monitor? :)
[21:02:03] <jmkasunich> yes
[21:02:10] <jmkasunich> then it proceeds to not show the video
[21:02:12] <SWPadnos> oh, in that case. hmmm
[21:02:19] <SWPadnos> heh - it looks like I get the same response
[21:02:44] <jmkasunich> 0:25
[21:02:56] <jmkasunich> 0:10
[21:03:10] <jmkasunich> liftoff
[21:05:07] <jmkasunich> 1:45, 23 miles away aready
[21:05:51] <jmkasunich> 37 miles alt, 52 miles downrange, 5200 mph
[21:06:17] <dmess> what site?
[21:06:24] <jmkasunich> http://www.nasa.gov/multimedia/nasatv/
[21:06:37] <jmkasunich> 52 alt, 100 downrange
[21:06:53] <dmess> shes movin'
[21:07:28] <jmkasunich> 4:15, 62 alt, 170 downrange
[21:07:37] <dmess> shit no plug in
[21:08:55] <dmess> i HATE when that happens
[21:09:02] <jmkasunich> tell me about it
[21:09:28] <jmkasunich> 6:15, 397 downrange, 10000mph
[21:10:14] <dmess> mach 14 or so
[21:10:40] <dmess> alt??
[21:10:46] <gezr> high
[21:10:57] <gezr> 1 minute of engines on left
[21:11:00] <jmkasunich> 700 downrange, 14,000 mph
[21:11:27] <jmkasunich> they're not quoting alt anymore - going orbital is mostly about horizontal speed
[21:11:35] <dmess> you call it Jmk
[21:11:46] <jmkasunich> meco
[21:12:06] <jmkasunich> tank separated
[21:12:24] <gezr> nice launch
[21:12:33] <anonimasu> I missed the liftoff -_-
[21:12:35] <dmess> and there's the pitch
[21:12:41] <anonimasu> crap.
[21:13:13] <gezr> ?
[21:13:50] <dmess> he' rounding first.... going for 2'nd...
[21:15:36] <dmess> what are we launching... anyway??
[21:15:41] <jmkasunich> shuttle
[21:16:15] <anonimasu> the japanese module for ISS
[21:16:27] <dmess> oh.. i've been pre-ocupuied... woman stuff..
[21:17:52] <dmess> BTW.. women engineers are way under rated... and good lookin too
[21:20:29] <anonimasu> haha
[21:23:44] <dmess> new office layout has me facing a PERFECT 26 yr old MODEL..... and I have to walk and talk with her... train/introduce her and all... whats a guy to do...
[21:38:25] <fenn> jmkasunich: http://www.nasa.gov/images/content/234212main_launch-m_1600-1200.jpg
[21:39:31] <pjm_> that was a cool launch, i saw it pass over here in the south of the UK
[21:39:37] <pjm_> i could see the shuttle and the ET
[21:41:39] <fenn> nice, did you have any radio gear? or just on irc
[21:42:13] <pjm_> yeah i was tracking the sts on 2217.5MHz too
[21:42:22] <pjm_> and also copied a couple of overs on UHF from the orbiter
[21:44:10] <pjm_> http://pjm.dyndns.org/sband/ant/trk1.jpg shows the antenna that can track in az & el
[21:44:55] <anonimasu> looks like a death ray
[21:45:18] <pjm_> hehh
[21:45:29] <pjm_> it could be converted to one fairly easily
[21:47:31] <fenn> so uh, what does a space shuttle look like from a couple hundred miles?
[21:50:24] <pjm_> of course u cant see any detail but i could see it as a massivly bright white dot, and also about 1" behind it i could see the red ET
[21:51:40] <pjm_> but if u look at http://mysite.wanadoo-members.co.uk/satcom_transits/March2005.html u can see some images of the ISS/ Shuttle made with a small telescope
[23:06:15] <gezr> Im pretty sure a chery lime aid has made me sick