| |
|
|
|
lccdt checks a command definition table.
The tools indicate either successful completion or syntax errors in the
parsed file.
On successful completion the diagnostic message is:
'Successful compilation: x commands'
This message gives the total number of commands defined in the file.
On error, the tools stop on the first error with the following
diagnostic message:
'file:line number: error description'
where error description is one of the following:
'syntax error': syntax error detected by the parser,
'command name too long': command name has more than 7 characters,
'bad value type': parameter value does not match given type,
'command name or synonym already exist': same name given twice,
'invalid display format': display format does not match reply
parameters.
lccit checks a command interpreter table.
The tools indicate either successful completion or errors in the
parsed file.
On successful completion the diagnostic message is:
'Successful compilation: x commands'
This message gives the total number of commands defined in the file.
On error, the tools stop on the first error with the following
diagnostic message:
'file:line number: error description'
where error description can be one of the following:
'syntax error': syntax error detected by the parser,
'command name too long': command name has more than 7 characters,
'invalid priority': task priority must be between 0 and 255,
'invalid run-time option': run-time options are defined in the
VxWorks manuals,
'invalid stack size': task stack size must be a positive even
integer.
lccdb checks the syntax of a database point configuration file or a
complete database branch.
Options
-r Recursive. With this option, the argument is the name
of the root point [directory] of a database branch.
All Point Configuration files found below the given
directory are checked. The Point Configuration file of
the root point itself is also checked if it exists.
For each checked, the tool indicates either successful completion or
syntax errors in the parsed file.
On successful completion the diagnostic message is:
'fileName: successful parsing: x lines parsed'
On error, the tool stops on the first error with the following
diagnostic message:
'file:line number: error description'
where description is the error message associated to:
lccERR_DB_SYNTAX : syntax error detected by the yacc parser,
lccERR_LENGTH : point, attribute or field name has more than 19
characters,
lccERR_DE_TYPE : value type does not match element type,
lccERR_MAP_MAXIMUM : too much attributes in point, or too much
fields for a table attribute,
lccERR_INV_RECEL : initialization outside vector or table limits.
lccBoot() enters the local environment, sends the <LCCBOOT> command
to the lccServer process on the <environment> and waits <time-out>
seconds for the final reply reporting a successful reboot.
Received replies or/any abnormal conditions are printed to stdout.
The utility terminates after the last normal reply has been received
or the timeout has elapsed.
<environment> lcu environment to reboot
<time-out> time-out value in milliseconds to wait for reboot
completion; 0 means msgNO_TIMEOUT (WAIT FOREVER)
OPTIONS
lccBoot() accepts following options:
-v causes lccBoot() to run in verbose mode
-n forces lccBoot() to skip command checking.
By default lccBoot() checks the validity of the command,
hence it requires a loaded CDT of the process the message
is sent to. To avoid this check the option -n can be used.
lcustatCheck monitors the status of a list of LCU environments
and updates a table in the WS database with its information.
The table is located under the lcustat application branch
(:Appl_data:lcustat), the attribute name is 'lcuEnvs'. To
configure the table use lcustatSetup(n)
When starting, lcustatCheck checks and set its lock attribute
(:Appl_data:lcustat.lock), if the lock is already set, then another
lcustatCheck process is already running and the current one
exits. If the lock was left set by a died server (it might happen if the
process gets a KILL command) then the lock can be released by a newer
lcustatCheck process by specifying the -unlock argument in the command
line.
The server checks and updates the status of the LCU environments
as follows: The UNIX ping command is used to detect if the
node is reachable, then if the LCU node is alive, a PING command is
sent to the lccServer on the LCU environment running in that
node.
The status attribute in the table used to store the LCU's status is
updated depending of the result of the previous checks with the
following values:
0 - UNIX ping command failed, host is down
1 - UNIX ping succeed (host ok), lccServer not responding
2 - UNIX ping succeed and there is communication with lccServer
The server caught the signal SIGTERM to clean temporary files
used to store the output of the UNIX's ping command. If the server
gets a KILL command or a SIGSTOP signal a number of small files <=40
named as LCU environments could be left.
This files might be get deleted by the next running server
(if its configuration does not change).
lcuwdTrig will read a list of LCU environment names from the point
:Appl_data:lcustat.lcuEnvs and will send the command WDSTRT with
<alarmTimeout>, <rebootTimeout> parameters to all of them which have set
the wdStart attribute.
Every <alarmTimeout> seconds it will trigger the LCU's
watchdog by sending the command WDTRIG to the LCU environments it
watches.
lcuwdTrig relies in the CCS time facilities to send a TRIGGER
command to itself every <N> seconds. If it cannot schedule the
TRIGGER command a WDSTOP command is sent to all its LCUs and the
program exits.
A minimum value of 30 seconds for the parameters
<alarmTimeout> and <rebootTimeout> is (silently) enforced.
 Quadralay Corporation http://www.webworks.com Voice: (512) 719-3399 Fax: (512) 719-3606 sales@webworks.com |
| |
|
|
|