Tiny Core Linux
Tiny Core Extensions => TCE Bugs => Topic started by: SamK on July 26, 2013, 02:33:25 AM
-
TC: v4.7.5
gpm: OnBoot (default configuration used)
Following a reboot, commands run as root
Obtain status of gpmps | grep gpm
... root grep gpm
gpm is not running.
Start gpm via init.d/usr/local/etc/init.d/gpm start
gpm is already running
ps | grep gpm
... root grep gpm
Expected message, "Starting gpm."
gpm is not running.
Start gpm at the command line using standard configurationgpm -m /dev/input/mice -t ps2
ps | grep gpm
... root gpm -m /dev/input/mice -t ps2
... root grep gpm
gpm is running
Kill the X-server and load Midnight Commander.
Mouse works as expected.
Stop gpm via init.d/usr/local/etc/init.d/gpm stop
Killed
ps | grep gpm
... root grep gpm
gpm successfully stopped.
Again stop gpm via init.d/usr/local/etc/init.d/gpm stop
Killed
Expected message, "gpm was not running, no process killed."
Edit: added expected message
-
Resurrecting this one since I have experienced this issue and have a fix:
The init.d/gpm script is broken on lines 9 and 27.
Currently it reads:
if ps | awk '{print $3}' | grep -e "gpm" >/dev/null; then
This is incorrect, since the script is also called "gpm"! So it will find its own entry and abort.
Instead, the code should read:
if ps | awk '{print $3}' | grep -e "/usr/local/sbin/gpm" >/dev/null; then
This will ensure searching for current gpm processes identifies the actual gpm executable, not the init.d script. This is in keeping with the many explicit references to /usr/local/sbin/gpm in the rest of the script.
Otherwise, gpm will never actually be started. Spent a minute scratching my head over this one.
I am happy to submit this fix more formally - is posting in the forums the official method, or is there some place I can submit a pull request?
Thanks!
-
Aha! Thanks.
For submitting extensions see here:
http://wiki.tinycorelinux.net/wiki:creating_extensions#submission
-
Ah thanks, submitted.
-
Resurrecting this one since I have experienced this issue and have a fix:
The init.d/gpm script is broken on lines 9 and 27.
Currently it reads:
if ps | awk '{print $3}' | grep -e "gpm" >/dev/null; then
This is incorrect, since the script is also called "gpm"! So it will find its own entry and abort.
Instead, the code should read:
if ps | awk '{print $3}' | grep -e "/usr/local/sbin/gpm" >/dev/null; then
This will ensure searching for current gpm processes identifies the actual gpm executable, not the init.d script. This is in keeping with the many explicit references to /usr/local/sbin/gpm in the rest of the script.
Otherwise, gpm will never actually be started. Spent a minute scratching my head over this one.
I am happy to submit this fix more formally - is posting in the forums the official method, or is there some place I can submit a pull request?
Thanks!
hi famousbirds,
I think I have solved a similar situation by adding "| grep -v grep |".
regards
Greg