Usage: LJSCAN is designed for periodical scanning, comparing and
e-mail reporting about lumberjack logged data.
It reports the integrated, average, minimum, maximum values,
total number of points logged, numbers of limit violations
and border times within which those happened.
The specified time interval is taken within the last 24 hours.
The information about user e-mail addresses, input data files,
time schedule and intervals and options is located in a job
file which is updated by D44.
Restart: This process needs no restart. It was started by a batch job
SUBMIT_LJSCAN in OP$USR1:[VAXAPPL.LJSCAN].
I. The input file format.
the format is a list of tags and values in the form {tag}={value}.
if value string contains blanks it should be enclosed in quotes.
The keywords for tags are as follows:
I.1. "name" - the name of the first device in the expression, 12 characters long;
I.2. "lj" - the name of the lumberjack node to log data, 6 characters long;
I.3. "min" - the value of the lower limit, float point number;
I.4 "max" - the value of the upper limit, float point number;
I.5. "numavg"
- the number for local average value to smooth peaks of
original data, integer number; if shown up the raw data read from
lumberjack are interpreted immidiately and represent the source
for next calculations and comparisons;
I.6. "imin" - the value of the lower limit of the integrated value;
I.7. "imax" - a key word introduced for the upper limit of the integrated value;
I.8. "itu" - time units to calculate the integrated value, the options are
as follows:
nsec - nanaseconds,
usec - microseconds,
msec - milliseconds,
sec - seconds,
min - minutes,
hour - hours,
.
I.9. "skip" - number of points to skip;
I.10. "skipsec" - minimum time between points;
I.11. "oper" - arithmetical operation sign, used just in the case of arithmetical
expression, this keyword is necessary for all memebers of the expression
but last one;
The expression (in many cases read as "device" ) attributes like
lj, min, max, numavg, imin, imax, itu are inherited ones,
so if min|max|imin|imax is not important to you
than you have to write
none
or just
n
in this case the corresponding fields in the output file would be blanked,
and this value would be inherited too !
II. The output file format.
After standart header the output text file consists from several columns
representing different kinds of information. The headers of the columns
are as follows:
Name of Column The kind of information in the column
II.1. "Name" - the name of the first device in the expression;
II.2. "IntAvg" - integrated average;
II.3. "Integrated" - integrated value;
II.4. "IntCmp" - results of comparison of integrated value and limits
for the integrated value ( Hi, OK, Low );
II.5. "Minimum" - minimum value of data in the time interval;
II.6. "Maximum" - maximum value of data in the time interval;
II.7. "Average" - arithmetical average;
II.8. "LowCount" - the number of points which values are less than
the minimum limit;
II.9. "HiCount" - the number of points which values are greater than
the maximum limit;
II.10. "Points" - the total number of points in the expression
in the time interval which should be checkuped;
II.11. "FirstLow" - the time of the first ocurrence when a somewhat point
value violated the minimum limit;
II.12. "LastLow" - the time of the last ocurrence when a somewhat point value
violated the minimum limit;
II.13. "FirstHi" - the time of the first ocurrence when a somewhat point
value violated the maximum limit;
II.14. "LastHi" - the time of the last ocurrence when a somewhat point value
violated the maximum limit;
III.The job file format.
The job file consists from the
III.1.Necessary tags & strings.
III.1.1. "name" - tag and string for a name of an input file;
III.1.2. "email" - tag and string for e-mail address of the output file
to be sent to,
III.1.3. "owner" - tag and string for job owner name;
III.1.4. "schedule" - tag and string for start time of the job in hours
after midnight,
III.1.5. "interval" - tag and string for the next runs interval in hours.
III.2.
Another tags are optional and they control the type of information you
prefer to have in a e-mail message to you (the columns in the output file).
III.2.1.
If you don't need some type of information just use the keyword "omit"
omit= omit= ...
where , - strings in " "-s in I.2.-I.14.;
in this case the corresponding column will be omitted in the output file.
III.2.2.
You can control the accuracy of float point data represented in the
output file. Just use the option with the keyword "accuracy"
accuracy=,
where is a digit number 1, 2, 3, ...
If the case the float point numbers less than 10 will have not more than
1, 2 , 3, ... decimal digits behind the point in the representation,
less than 100 and greater or equal to 10 will have 0, 1, 2, behind the
point, etc. But all digits to the left of the point will be left any way.
The default accuracy is equal to 7. That means 8 true decimal digits.
III.2.3.
If the comparison of the integrated value with imin amd imax gives
that the value is between imin and imax the default text reported is "OK",
if it is greater than imax the default text reported is "Hi", if less than
imin - "Low". But the user can define the texts of his own using tags
"ok", "hi" and "low" as follows:
ok=
hi=
low=
where is a character string of 6 characters long,
if it contains spaces should be enclosed in quotes.
Keeper: Yuri Ivanov
Now: Sun Sep 7 14:29:00 2008