The calculation method of checkpoint lag of the Oracle GoldenGate (OGG) Target rep Process

This article is original, reproduced please indicate the source.

The calculation method of checkpoint lag of Oracle GoldenGate (OGG) OGG end rep Process

Background: The following is the display of the process of the rep of Ogg destination, at that time, the transmission process of Ogg source caused by the network congestion, the transmission speed dropped sharply, reduced to 1.5m/minutes. At this point, the source has a backlog of trail file.
That is, the target's rep process is in a state of "not enough to eat"-the transfer speed of the transfer process is not as fast as the rep process.

Ggsci (HOSTB) 5> info repbinfo repbreplicat   repb last      Started 2014-08-01 14:43   Status runningcheckpoint Lag       00:00:00 (updated 00:00:12 ago) Log Read Checkpoint  file/u02/ggs/dirdat/tb001541                     2014-08-01 05:26:32.766558  RBA 32744080GGSCI (hostb) 6>!info repbreplicat   REPB last      Started 2014-08-01 14:43   Status runningcheckpoint Lag       09:21:35 (updated 00:00:00 ago) Log Read Checkpoint  file/u02/ggs/dirdat/ tb001541                     2014-08-01 05:26:34.768028  RBA 33105784

Observe the differences in checkpoint Lag:

The first one is checkpoint lag       00:00:00  The first one is checkpoint lag       09:21:35

The current time is around 2014-08-01 15 o'clock in the afternoon,

Combined with the reduced speed of the trail file transfer, we can speculate on how the checkpoint lag is calculated:

When a new trail file is transferred to the destination (when the trail is not yet applied), Ogg reads the timestamp in the trail file and makes a comparison with the current OS time, resulting in a checkpoint lag of 09:21:35
When all trail files are applied, Checkpoint lag is 00:00:00