How to read the output of the debug flag "-dbg STAT_HISTO"?

If the debug flag "-dbg STAT_HISTO" is activated detailed information for the processed value changes and changes on alert instances is reported.
The information is written to the log file for the Event Manager.
The interval can be defined with the config entry MessageDiagSecs in the [general] section. The default value is 30 seconds.

For every debug two lines of information is written.
The first line contains information for the value changes in the system.
The second line contains information for the processed changes of alert instances.
The information is written in form of a histogram. Each slot of the histogram has a special meaning.


+++ Value changes +++

The first column contains the total number of value changes for the last period.

In the following columns detailed information for the latency is reported. The latency is the difference between the value change timestamp and the processing in the Event Manager.
If the value is changed due to a dpSet/dpSetWait message, the difference is 0 in any case as the time stamp is given by the Event Manager.
If the value is changed in a PLC and the driver provides the timestamp, some latency might be detected. Due to time synchronization there might be negative
latency intervals as well, which are counted in a separate slot.

The histogram contains information for following latency intervals:

// slots:
// total number  [ 0]
//   0 ..  90 ms [ 1] .. [10]
// 100 .. 900 ms [11] .. [19]
//   1 ..   9  s [21] .. [29]
//  10 ..  90  s [31] .. [39]
// <0.0          [102]
// >= 100s       [101]

The example for the log message shows value changes triggered by a Ctrl script running on the same server like the Event Manager, therefore no latency is detected.
There are 34 changes in total, and all are detected with less than 10 ms latency.

WCCILevent   (0), <TIMESTAMP>, STAT_HISTO, "DPEChanges - decades: 34, 34, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, >100s 0, <0 0"


+++ Changes of alert instances +++

The first column contains the total number of changes of alert instances for the last period.

In the following columns detailed information is listed according to the alert type (continuous alarms, multi instance alarms, suppressed alarms) and the alert state.


// ALERTEVENT_CAME (0)
// ALERTEVENT_CAME_IMPULSE (1)
// ALERTEVENT_GONE (2)
// ALERTEVENT_MULTIPLE_ACK (3)
// ALERTEVENT_SINGLE_ACK (4)
// ALERTEVENT_GONE_INACTIVE (5)
// ALERTEVENT_GONE_IMPULSE (6)
// ALERTEVENT_GONE_OBSOLETE (7)
// NOEVENT (8)

The histogram contains the following slots:

// slots:
// total number                              [ 0]
// continuous alarms                        [ 1] .. [10]  sorted by AlertEvent constant
// multi instance alarms                     [11] .. [20]  sorted by AlertEvent constant
// suppressed alarms due to minPrio setting [32]
// _alert_hdl config was set active/inactive [33]

The following sample shows changes of continuous alarms and multi instance alarms.
1036 alarm changes in total
29   continuous alarms (ALERTEVENT_CAME)
7    continuous alarms (ALERTEVENT_GONE)
1000 multi instance alarms (ALERTEVENT_CAME)

WCCILevent   (0), <TIMESTAMP>,    STAT_HISTO,    "AlarmChanges - decades: 1036, 29, 0, 7, 0, 0, 0, 0, 0, 0, 0, 1000, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, >100s 0, <0 0"

Date added:
Last revised:
Hits:
848
Rating:
Rating: 4.3. 34 vote(s). No rating done at all.
Your vote was '' (0 of 5) Your are an anonymous user.
You may log on to do personalized votings
Click the rating bar to rate this item Please log on to do ratings