Eventlog entries on resource bottlenecks – Kofax Communication Server 9.1 User Manual

Page 47

Advertising
background image

Environment Guide

Version 3.00.04

© Copyright Kofax, Inc. All information is subject to change without notice.

47

After some time it can be seen that the average Packet-Ack counters are within the limit (<50ms), but there

are too high spikes on the peak Packet-Ack counters (reaching up to 2000ms, see blue and red graph

above).

This means that the network between the TCOSS and LS1 servers does not fulfill requirements (peaks

must be lower than 1000ms). As a consequence, such network spikes may often cause LS1 reboots!

Eventlog Entries on Resource Bottlenecks

During the TCOSS operation, the most crucial Key Performance Indicators (“KPIs”) are being observed and
checked against configurable threshold limits. If these limits are exceeded, corresponding Windows event

log entries are generated.

Two kinds of measures are being computed:

1. Average of the particular value during the last measurement cycle

2. Peak of the particular value that occurred during the last measurement cycle

If any of these values (average or peak) exceeds corresponding threshold, following event log types would

be written:

1.

If average for the particular counter exceeds its threshold limit, the event log of the type “Warning”

would be written (and thus sets the error state of this counter to warning). If the counter being in the

Warning state exceeds its threshold limit during any later measurement cycle, the Warning event log would

be also written.

2. If the average for the particular counter being in the warning fulfills its threshold limit, the event log of
the type “Information” would be written (and thus sets the error state of this counter to Ok).

3. If peak for any counter exceeds its threshold limit, the event log of

the type “Information” would be

written (but no error state of peak counters is being kept).

Advertising