Table of Contents
PCT -- Process Control Thread
pct [-nid num -pid num ] [-D] [-S [1|0]] [-L [1|0]] [-daemon]
The PCT (Process Control Thread) is a process running on each compute
node of the Cplant. The PCT accepts the parallel application executable
image from yod and starts the user process. The PCT can respond to bebopd
requests for status information about it's state and about the user process
it is hosting.
- -nid node-number -pid portal-id
- The PCT normally registers with the
bebopd managing the virtual machine. If an alternative bebopd is running
for test purposes in addition to the regular bebopd, and you wish this
PCT to register with the alternative bebopd, then provide the node number
and portal ID of that bebopd with these arguments.
- -D
- This option causes
the PCT to output information about what it is doing. Repeating the -D option
on the command line increases the amount of information. Normally the information
goes to the log file (/var/log/cplant) on the compute node. This can be
changed with the -S and -L options.
- -daemon
- By default the PCT runs as
a foreground process. Use -d to run it in the background.
- -S [0|1]
-
The PCT outputs warnings and errors, and, if the -D option is used, status
information. The 0 switch turns off all output from the PCT to std
err. The
1 switch turns it on. By default, the PCT doe
s not write to stderr.
- -L
[0|1]
- The PCT outputs warnings and errors, and, if the -D option
is used, status information. The 0 switch turns off all output from the
PCT to the log file. The 1 switch turns it on. By default, the PCT does
write to log file.
The PCT reports error conditions in the log file (/var/log/
cplant).
It will not exit unless the error makes it impossible for the
PCT to do
it's job, for example if it has no memory to allocate a por
tal. Errors in
loading applications are reported back to yod.
On receiving a SIGUSR1 or SIGUSR2, the PCT will log the rou
tine it
is in now, what routines it was in most recently, and the state o
f the
application process it is hosting. It will also display a loop c
ounter,
which should increment on subsequent displays, indicating whether
the
PCT is stuck somewhere or has been progressing through it's main
service
loop. On receiving a SIGHUP, the PCT will close and reopen it's l
og file,
re-read the site file, and re-read the cplant-host file and send
a status
update to the bebopd listed there.
- /cplant/cplant-host
- This file identifies the location of a bebopd
daemon.
- /etc/local/site
- This file defines site specific information
that may be required by the bebopd.
- /var/log/cplant
- This is the log
file where Cplant daemons and utilities log status.
pingd
bebopd
yod
site
Let us know if you locate any (cplant-help@cs.sandia.gov).
Table of Contents