305   Reporting with BufTm=0

Created: 31 Mar 2006

Status: In Force (green)

Part: Part 7-2 (2003)

Links:

Page: 080,85

Clause: 14.2.2.9, 14.2.3.2.2.9

Paragraph: Multiple

Category: Issue may impact interoperability of implementations of Edition 1

Issue

7-2 does not properly detail the exepected behavior if two members of a DataSet change at the same time while BufTm=0.

7-2 should be changed to reflect that an "internal event" is the detection of changes to any member of a DatSet(e.g. the event is that 1 or more members have changed). It must be possible for 2 or more members change that generate a single report with BufTm=0 otherwise large numbers of reports may be generated.



Proposal

The recommended change in 14.2.2.9 is to change:

"Upon receipt of the first internal event notification of the referenced..."

to

Upon receipt of the first internal event notification, that may correspond to the change detection of several members, of the referenced....

The recommended change in 14.2.3.2.2.9 is to change:

"In case of (BufTm = 0) only the value of the member of a DATA-SET shall be included that produced the internal event."

to

In case of (BufTm = 0) only the value(s) of the member(s) of a DATA-SET shall be included that produced the internal event.

Discussion Created Status
changed to green. The 7-2 clauses for reporting and logging ( Edition 1) can be found at ftp.sisconet.com. Username=wg10revision pw: iec61850. Passive FTP must be used for retrieval. 16 Feb 07 In Force (green)
changed to green. The 7-2 clauses for reporting and logging ( Edition 1) can be found at ftp.sisconet.com. Username=wg10revision pw: iec61850. Passive FTP must be used for retrieval. 16 Feb 07 In Force (green)
See Tissue 453 for revised chapter.

The text has been changed to :
"The DATA-SET (referencing DATA) represent the real data values. The real data values are conceptually monitored by the event monitors. An event monitor determines, on the basis of the state of the real data and the attributes of the control class, when to generate a notification to the appropriate handlers (e.g. Log or Report Handler). This notification includes the data values and reasons for data inclusion.
Note: The contents of the event notification are determined by a combination of I/O scan and event monitoring. These are conceptually two asynchronous processes. Therefore, the number of data values included within a single notification is a local issue.
The number of data values, within a notification, is a local implementation issue. The report handler assigns EntryID(s) and TimeOfEntry(s) to the values contained within a set of notifications. The number of notifications combined into a single EntryID is determined by the RCB control parameters (e.g. BufTim). The value of the EntryID is a local issue but it shall be a unique arbitrary OCTETSTRING whose value is unique within the scope of entries for a specific RCB. The value of the TimeOfEntry shall be the timestamp representing the time at which the report handler received the first notification that is used to form an EntryID. The report handler decides when and how to send a report to the subscribed client. The log handler stores a log entry to the log."
13 Dec 06 Ballot Period
Agreed in pricipal.

Changed internal notification event to set "notification of an internal set of events." This was done to clarify that an internal notification may contain more than one event.

Otherwise accepted.
17 Oct 06 Ballot Period
accepted; even for bufTm=0 all events available when report assembly starts should be put into one report, even if the have different time stamps - just the additional wait for more events should be skipped. 05 May 06 Discussion (red)

 

Privacy | Contact | Disclaimer

Tissue DB v. 23.12.13.1