Re: [Machinekit] Machinekit not starting: seems to be attempting to start incorrect rtapi app

[schoone...@btinternet.com]

On 20/03/17 09:06, schoone...@btinternet.com wrote: Just an update on this It is just possible there is a corruption in the machinekit repo, which is causing problems with recent package builds too We have just tried a build from a fresh clone of the repo and get all sorts of errors Brian - If you know how, are you able to revert packages back to something produced at around 8th March and try those? http://deb.machinekit.io/debian/pool/main/m/machinekit/machinekit_0.1.1488973272-1mk.travis.master.git187524c9~1jessie_i386.deb and the matching flavor package Simon - likewise, but don't screw up your system if you are not sure We are looking at it , but it may take a while to resolve The problem has been found and a correcting PR made. There was an unexpected knock on from another recent change, which did not show up in test builds and even allowed packages to build without errors, despite them not having setuid on required files and flavor segfaulting and not getting the proper kernel flavour, so it defaulted to posix. The build process will take a while, I will advise once new packages are available, until then do not update Reverting to the packages I suggested will work if you are desperate to use your system and have the non working packages installed On 20/03/17 08:33, schoone...@btinternet.com wrote: Can you try running the flavor binary in /usr/libexec Running flavor on its own should give the kernel flavour Also run flavor -m , flavor -b and flavor -d This seems similar to Simon Mali's problem when he upgraded, the kernel flavour was not being detected and a default of posix was being tried. On 20/03/17 03:17, Brian Schousek wrote: I'm trying to get a basic Machinekit up and running on an older PC. I've apparently successfully installed and started an RT-PREEMPT kernel (see ref 1), properly installed the appropriate machinekit packages (see ref2), and tried to start things up with various configurations. Ref 3 shows a representative error (through iterations I determined that just trying to run 'realtime' yields the same errors as starting the whole package, and errors out much quicker.) My interpretation of ref 3 is that it is trying to run rtapi_posix (not installed) instead of rt-preempt. As can be seen in ref 4 there is an executable for rtapi_app_rt-preempt but no rtapi_app_posix so it's no great surprise there is an error in ref 3. What simple thing am I missing here? I've been unsuccessful in finding a solution in this forum and on the machinekit website.  Ref 1 { brian@surely:/usr/libexec/linuxcnc⇒  uname -a Linux surely 4.1.19-rt22mah #1 SMP PREEMPT RT Tue Mar 15 09:12:20 CET 2016 i686 GNU/Linux } Ref 2 { brian@surely:/usr/libexec/linuxcnc⇒  apt list --installed | grep machine machinekit/stable,now 0.1.1489960089-1mk.travis.master.git7df8d183~1jessie i386 [installed,automatic] machinekit-dev/stable,now 0.1.1489960089-1mk.travis.master.git7df8d183~1jessie i386 [installed] machinekit-rt-preempt/stable,now 0.1.1489960089-1mk.travis.master.git7df8d183~1jessie i386 [installed] } Ref 3 {brian@surely:~⇒  sudo realtime start                                                                     Warning - /usr/libexec/linuxcnc/rtapi_app_posix not setuid 'sudo make setuid' missing? /usr/bin/realtime: line 111: test: =: unary operator expected /usr/bin/realtime: line 174:  4065 Segmentation fault      ${rtapi_msgd} --instance=$MK_INSTANCE $NAME_CMD --rtmsglevel=$DEBUG --usrmsglevel=$DEBUG --halsize=$HAL_SIZE $MSGD_OPTS rtapi_msgd startup failed - aborting /usr/bin/realtime: line 192: /usr/libexec/linuxcnc/rtapi_app_posix: No such file or directory rtapi_app startup failed - aborting halcmd: cant connect to rtapi_app: -1 (uri= uuid=a42c8c6b-4025-4f83-ba28-dad21114744a): rtapi_rpc(): reply timeout halcmd: the rtapi:0 RT demon is not running - please investigate /var/log/linuxcnc.log halcmd: the msgd:0 logger demon is not running - please investigate /var/log/linuxcnc.log } Ref 4 { brian@surely:~⇒  ls /usr/libexec/linuxcnc  -l                                                            total 200 -rwxr-xr-x 1 root root  14888 Mar 19 17:52 flavor -rwxr-xr-x 1 root root   5960 Mar 19 17:52 inivar -rwxr-xr-x 1 root root  14268 Mar 19 17:52 pci_read -rwxr-xr-x 1 root root  14288 Mar 19 17:52 pci_write -rwsr-xr-x 1 root root 101368 Mar 19 17:52 rtapi_app_rt-preempt -rwxr-xr-x 1 root root  43960 Mar 19 17:52 rtapi_msgd } -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to machinekit+unsubscr...@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to machinekit+unsubscr...@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to machinekit+unsubscr...@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to machinekit+unsubscr...@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout.