#linuxcnc-devel Logs

Feb 26 2017

07:25 AM jthornton: I'll look in a bit
08:04 AM JT-Shop: https://www.maritool.com/Indexable-Tooling-90%C2%B0-Square-Insert-Shell-Mill/c50_111/index.html
08:04 AM JT-Shop: https://www.maritool.com/Indexable-Tooling-45%C2%B0-Square-Insert-Shell-Mill/c50_110/index.html
08:04 AM JT-Shop: https://www.maritool.com/Indexable-Tooling/c50/index.html
08:30 AM KGB-linuxcnc: 03Dewey Garrett 05master e29a74c 06linuxcnc 10docs/src/config/python-interface.txt docs/src/config/python-interface.txt jog syntax * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=e29a74c
08:30 AM KGB-linuxcnc: 03Dewey Garrett 05master 3910a2a 06linuxcnc 10docs/src/getting-started/updating-linuxcnc.txt updating-linuxcnc.txt section on interface updates * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=3910a2a
08:43 AM Tom_itx is now known as Tom_L
11:43 AM seb_kuzminsky: thanks JT-Shop
04:18 PM pcw_home: ipcs lists before linuxcnc started, after linuxcnc started, before halscope-rt loadrt attempt, after halscope-rt loadrt attempt:
04:18 PM pcw_home: http://pastebin.com/vXDVkqEN
05:39 PM seb_kuzminsky: pcw_home: that's helpful, thanks
05:39 PM seb_kuzminsky: note how the halscope shm (key=0x130cf406) is owned by root, not by you
05:40 PM seb_kuzminsky: that's the proximal problem
05:40 PM seb_kuzminsky: what are the perms of your $(which halscope)?
05:41 PM seb_kuzminsky: mine's owned by me and not setuid or anything weird:
05:41 PM seb_kuzminsky: -rwxr-xr-x 1 seb seb 387K Jan 23 13:47 /home/seb/linuxcnc-dev/bin/halscope*
05:50 PM pcw_home: -rwxrwxr-x 1 pcw pcw 300788 Feb 26 12:48 /home/pcw/linuxcnc-dev/bin/halscope
05:54 PM jepler: I think that the halscope shared memory area is created by scope_rt, so this may be a problem with user ID management in rtapi_app (I'm assuming uspace)
05:55 PM pcw_home: yes this is uspace
05:55 PM jepler: is this 2.7 or master? I revamped the user id handling in halscope in master branch trying to fix the mint iptables problem
05:56 PM pcw_home: this is master
05:57 PM jepler: seems like while I was travelling I saw something scroll by about the order of loading hm2_eth causing some kind of shm problem?
05:57 PM jepler: I never tried to research what was going on there
05:58 PM pcw_home: I haven't seen that
05:58 PM jepler: oh maybe that was someone else, or just a nightmware
05:58 PM jepler: does a basic test in the testsuite that uses streamer/sampler pass? e.g., runtests tests/abs.0
06:00 PM pcw_home: Runtest: 1 tests run, 1 successful, 0 failed + 0 expected
06:04 PM pcw_home: Basic LinuxCNC stuff seems to be fine with 16.04 just the halscope issue and the numpy problem with dxf2gcode
06:04 PM pcw_home: (some weird complains from the chooser also)
06:05 PM pcw_home: (too many configs for the chooser maybe)
09:45 PM jepler: > Error starting domain: Unable to preserve mac for eth1: No space left on device
09:45 PM jepler: what
09:50 PM jepler: so that was caused by /var/run being full, due to a huge file called "atop.acct"
09:50 PM jepler: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=650222
09:51 PM jepler: pcw_home: does something as simple as the following reproduce your halscope problem? halrun / loadrt threads / start / loadusr halscope
09:52 PM pcw_home: let me try
09:54 PM pcw_home: that works...
09:55 PM jepler: OK, does it reproduce when you load sim/axis and launch halscope from the menu?
09:57 PM jepler: hm I wonder what my root pw is on this vm
09:58 PM pcw_home: yes halscope works with sim
10:07 PM jepler: I assume the affected configs are loading something in the hm2 family
10:08 PM jepler: is it affecting pci or just ethernet?
10:20 PM jepler: not reproducing it with pci on debian jessie, will have to try tomorrow when I can hook up ethernet hardware again
10:40 PM jepler: pcw_home: in rtapi/uspace_rtapi_app.cc change '#ifdef RTAPI_DEBUG_PRIV' to '#if 1'. look for the additional debug information just below 'before WithRoot' and 'after WithRoot'
10:40 PM jepler: afk and goodnight
10:53 PM jepler: see http://www.mail-archive.com/emc-developers@lists.sourceforge.net/msg17154.html
10:53 PM jepler: it's cmorley who reported what I suspect is the same thing..
10:53 PM jepler: zzzz sleeping zzzz
11:24 PM nobody is now known as Guest65649
11:25 PM Guest65649: does the buildbot even work anymore? no builds for days? buildbot.linuxcnc.org/dev.html Service Temporarily Unavailable
11:36 PM pcw_home: jepler:
11:36 PM pcw_home: http://pastebin.com/pSxD5b8t