Scribd
Upload
48 views2 pages

Turn On Slip-Detector

Slip detector is a tool that monitors tasks for processing overruns called "slips". It generates core files when slips occur without requiring special images. Settings can be configured such as the number of core files before disabling, minimum time threshold, and timer trigger. The status and any slip messages can be checked through the CLI and system logs.

Uploaded by

Sivaraman Alagappan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
48 views2 pages

Turn On Slip-Detector

Slip detector is a tool that monitors tasks for processing overruns called "slips". It generates core files when slips occur without requiring special images. Settings can be configured such as the number of core files before disabling, minimum time threshold, and timer trigger. The status and any slip messages can be checked through the CLI and system logs.

Uploaded by

Sivaraman Alagappan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

Slip-detector

Slip detector is one of the powerful tool for checking if any task is consuming more cycle to
process. It does not require any special image to make it work, simply turning on knob will do
the work. Slip detector can also be turned ON on production routers. Slip detector is much
cheaper than task accounting, only trade off is core file.

Turn on slip-detector
To turn on slip-detector 'set task slip-detector on' this is hidden command. This enables or re-
enables with exiting or current configuration. You can set it off by firing 'set task slip-detector
off' on cli.
If there are any slips then core files generated under /var/tmp/cores.

Other Settings
Other than enabling and disabling, slip detector provides more tweakable settings.
set task slip-detector settings max-cores <n> min-time <sec> trigger <wall-clock-time|process-
cpu-time> backoff <yes|no>
For example set max core by
set task slip-detector settings max-cores 2
Where:

 max-cores: The number of core files the slip detector will generate before disabling
itself, defaults to 4 core files.
 min-time: The initial value of the detector timeout, defaults to 4 seconds.
 trigger: Determines whether accumulated wall-clock-time or process-cpu-time cause the
timer to fire. Defaults to wall clock time.
 backoff: If a single slip is longer than min-time, should the slip detector use exponential
back off when re-enabling the timer?. In exponential backoff time is always doubled.
Defaults to true. The timer is reset to min-time when the process returns to the event
loop.
 ignore-config: If set, ignore slips which occur during configuration processing. Its
recommended not to ignore config because majority time committing config lead to slip.
Defaults to true.
Slip-detector status
To get the status of slip-detector just cli
show task slip-detector
Output:
Settings: max-cores 2, min-time 4, trigger wall-clock-time, backoff
true, ignore-config true
State: enabled, total corefile count 0, current corefile count 0,
maximum expired slip timer 0

Logging

You can also see SLIP messages in /var/log/messages, even if slip detector is not enabled.
cat messages | grep "SCHED_SLIP"
2014-11-26T01:40:15.659-05:00 MAH-H2-TTP136 lldp 1152
JTASK_SCHED_SLIP_KEVENT - 64 sec 766619 usec kevent block
2014-11-26T01:40:15.660-05:00 MAH-H2-TTP136 chas 1157
JTASK_SCHED_SLIP_KEVENT - 64 sec 470385 usec kevent block
2014-11-26T01:40:15.680-05:00 MAH-H2-TTP136 sfid 1194
JTASK_SCHED_SLIP_KEVENT - 64 sec 525330 usec kevent block

You might also like