Jump to: navigation, search

Cassandra Maintenance

Summary

The Maintenance function in UCS triggers the following:

  • Periodic anti-entropy tasks (repair) in Cassandra.
  • Periodic snapshots of UCS database tables.

Maintenance Configuration Options

Configure all these options in a section called maintenance in the UCS application configuration object.

Name Default Value Valid Values Example
task.snapshot.enabled false true, false  
task.snapshot.period 86400000 (24 hours) int value (milliseconds) 180000
task.snapshot.tagTemplate     snapshot-%tF—generates a file name ending with the date (snapshot-YY-mm-dd)

Note: On Windows, Cassandra cannot overwrite a previous snapshot. A new name must be generated for each snapshot. If more than one snapshot per day is scheduled, a template like the following example can be used to generate a name with date and time:

snapshot-%1$tY-%1$tm-%1$td-%1$tH-%1$tM

If Cassandra runs on Linux a previous snapshot can be overwritten.

task.repair.enabled false true, false  
task.repair.period If not specified, the period is calculated based on the gc_grace_period parameter defined for the keyspace.

If gc_grace_period is set to 0, a default value of 1 week is used (and expressed in milliseconds)

int value (milliseconds) 180000
task.repair.trace false true, false  

Feedback

Comment on this article:

blog comments powered by Disqus
This page was last modified on May 18, 2018, at 07:01.