#emc | Logs for 2004-10-22

[00:17:32] <paul_c> * paul_c spots a security hole on the compile farm
[00:19:55] <paul_c> * paul_c realises trp has been lurking for a couple of hours...
[00:20:03] <trp> 08paul_c, good afternoon. I think I have things set up for you today
[00:20:04] <paul_c> Hi Tony.
[00:20:21] <trp> thought you may be busy and waiting patiently
[00:20:59] <trp> If you use the same port you can get on the original box and port 34 for the test box
[00:21:55] <trp> hope lunch was good. I usually wait till noon though
[00:22:35] <paul_c> OK... In on p32
[00:24:00] <paul_c> Going to check the GUI into CVS before I do anything else
[00:32:51] <trp> jim had to configure the connection on the test box manually. is there something wrong with the networking or internet?
[00:34:34] <paul_c> Which CD did you install from ?
[00:43:51] <trp> I installed from the cd that was burned yesterday
[00:45:45] <paul_c> Oh...
[00:46:28] <paul_c> networking is borked after installing from that CD, and the ssh server isn't installed
[00:48:16] <paul_c> Have you got an rc46 disk to hand ?
[00:48:31] <trp> Paul, it was our understanding from Ray that this was a proprietary interface for us here at Smithy. This is not supposed to be put into the CVS
[01:23:48] <alex_joni> hello... me back
[01:47:50] <trp> crashed again, gotta love this network
[06:39:44] <ksb> Hi Paul and Tony and Ray - now that we're all here, let's figure out where we are.
[06:40:06] <rayh> Is there a better place for this than emc?
[06:40:09] <paul_c> Sorry guys - Didn't see you all pile in.
[06:40:28] <ksb> The more the merrier
[06:40:45] <paul_c> #emc-smithy
[06:41:45] <rayh> You'll need a /join in front of #emc-smithy
[16:22:46] <CIA-9> 03jmelson * 10documents/lyx/Int_ppac.lyx: correct many errors, flesh out text a bit
[21:50:49] <alex_joni> hey paul... didn't notice you joining the channel ;)
[22:26:30] <paul_c> Afternoon Alex
[22:26:52] <alex_joni> how are you?
[22:27:06] <paul_c> Switching between desktops & X-servers at the mo...
[22:27:12] <alex_joni> oh... ;)
[22:27:44] <paul_c> Been thinking...
[22:28:12] <paul_c> Some of this autoconf stuff could be applied to the main EMC code base without too much effort
[22:28:34] <alex_joni> I found that thinking usually helps ;)
[22:28:44] <alex_joni> and discard the .def?
[22:28:49] <alex_joni> or generate the .def?
[22:28:49] <paul_c> * paul_c reaches for the wet kippers
[22:29:22] <paul_c> auto-gen realtime.def and nonrealtime.def
[22:30:10] <paul_c> modify the run script so that the symlinks are not deleted and created
[22:30:45] <paul_c> and have a single script (bash or shell) in the top level emc directory to fire everything off
[22:31:12] <alex_joni> sounds great...
[22:31:20] <alex_joni> but I think we should wait a bit...
[22:31:26] <alex_joni> t have autoconf for emc2 stable
[22:31:31] <alex_joni> hi ray
[22:31:36] <paul_c> It was something I had suggested at the emcFest
[22:31:41] <alex_joni> to have autoconf for emc2 stable
[22:31:43] <rayh> Hi Alex.
[22:31:58] <alex_joni> although autoconf for emc is a little bit different than emc2
[22:32:11] <alex_joni> (a lot of test shoul dbe the same thou..)
[22:32:18] <paul_c> true - But the same rules can be used
[22:32:26] <alex_joni> * alex_joni is having problems typing
[22:32:34] <paul_c> ditto
[22:32:49] <alex_joni> yup...
[22:33:22] <alex_joni> * alex_joni is wondering why autoconf still fails on the compile_farm... and why rc46 is gone
[22:35:06] <paul_c> probably down to no X-libs flags in the Makefile.inc
[22:35:43] <paul_c> also looks like ulapi.o failed to compile
[22:36:58] <alex_joni> looks like ./configure fails on TNG
[22:37:42] <alex_joni> I'm wondering if AC_CHECK_HEADER([$KERNELDIR/include/linux/version.h],...
[22:37:59] <asdfqwega> heheh...
[22:38:04] <alex_joni> shouldn't be simply AC_CHECK_HEADER(linux/version.h,...
[22:38:24] <paul_c> * paul_c sees where BDI-2.18 auto_tools is failing
[22:38:31] <asdfqwega> You ever been in the middle of writing code, and have to redo it 'cuz you thought of something better?
[22:38:40] <paul_c> yup
[22:38:44] <alex_joni> hmmm... let me think...
[22:38:59] <alex_joni> very rarely not in that case...
[22:39:04] <paul_c> alex_joni: No - Need to use the kernel header, not the standard library header
[22:39:17] <asdfqwega> I'm wrting a python program to take an image and output g-code for laser etching
[22:39:20] <alex_joni> yup... but KERNELDIR probably failed
[22:39:57] <paul_c> * paul_c looks at the log
[22:40:30] <alex_joni> I think $($RTS --dump | grep -i linuxdir | awk -F'[[ ="]]+' '{print $2}') failed...
[22:40:42] <paul_c> certainly looks like the KERNELDIR check failed
[22:42:28] <alex_joni> hmmm... if I commit something on CVS, should the farm recompile?
[22:42:45] <paul_c> it checks for the linuxdir string and does a pwd based on the value
[22:43:12] <paul_c> If the kerneldir is NULL, the pwd picks up on the usr home dir.
[22:43:50] <alex_joni> yup :(
[22:43:58] <paul_c> If the farm is powered up, it should run again in an hour or so
[22:44:29] <alex_joni> I see...
[22:44:58] <alex_joni> I think we should add some debug output if ./configure fails
[22:45:52] <paul_c> agreed - a verbose output to start with, then cut it down once it is working
[22:48:05] <paul_c> It could be that John has taken Slot 4 down untill the defs work for rc46
[22:48:56] <alex_joni3> I got disconnected...
[22:49:36] <alex_joni3> problem is... in configure.in I don't think there is a way to tell if ./configure fails or not
[22:49:51] <alex_joni3> maybe modify the script jmk is using
[22:50:38] <paul_c> Something to raise with him on Sunday ?
[22:50:54] <alex_joni3> surely...
[22:51:09] <alex_joni3> problem is... don't know when I'll be available on Sunday
[22:51:24] <alex_joni3> I gotta drive to a customer, gotta be there first thing on Monday...
[22:51:30] <alex_joni3> so I'll be on the road
[22:51:32] <alex_joni3> :(
[22:51:48] <alex_joni3> * alex_joni3 is mad, cause he can't find his mobile-data-cable
[22:52:02] <paul_c> I'm sure the farm and auto_configure will be discussed
[22:52:35] <alex_joni3> output variables at the end of ./configure are pretty helpful
[22:53:35] <paul_c> As are comments within configure.in describing the expected results of each test
[22:53:37] <alex_joni3> KERNELDIR='/home/John'
[22:53:48] <alex_joni3> RTAI='1'
[22:53:48] <alex_joni3> RTDIR='/home/John'
[22:53:48] <alex_joni3> RTFLAGS=' -DRTAI=1'
[22:54:07] <paul_c> $(cd " " ; pwd)
[22:54:24] <alex_joni3> yup
[22:54:40] <alex_joni3> guess $RTS is misbehaving
[22:54:54] <alex_joni3> found /usr/src/rtai-24.1.10/scripts/realtime-config
[22:55:27] <alex_joni3> probably rtai hasn't been installed :-?
[22:57:18] <paul_c> It has been installed - From a BDI-TNG disk
[22:57:45] <paul_c> so I need to look at the RPM to see why it is failing.
[22:58:00] <alex_joni3> ok...
[22:58:10] <alex_joni3> I think messages to the user need to be more clear
[23:00:26] <paul_c> I'll look at it over the weekend (once I have current work out of the way)
[23:15:38] <alex_joni3> I'll try to be there too...
[23:15:52] <alex_joni3> tomorrow no problem...but sunday will be tricky ;)
[23:19:31] <alex_joni3> alex_joni3 is now known as alex_joni
[23:34:03] <asdfqwega> A quick question about comments in .ngc files?
[23:34:30] <asdfqwega> What works? I've looked in the cvs and () are used
[23:35:08] <asdfqwega> Do I have to use ()'s?
[23:37:02] <paul_c> As I recall, yes
[23:37:23] <asdfqwega> thanx
[23:42:11] <alex_joni> I'm gone paul... feeling really shitty today (very tired...)
[23:42:15] <alex_joni> we'll talk tomorrow
[23:42:20] <alex_joni> and on sunday
[23:42:22] <alex_joni> bye...
[23:42:41] <paul_c> OK - Later.