Versions Compared

Version Old Version 27 New Version 28
Changes made by

JuneL (Unlicensed)

정은진

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Section
Column

It is classified resources according the types and has different attributes for each type.

Column
width350px
Panel

Table of Contents

Table of Contents
maxLevel4

Resource Type

Image RemovedImage Added

[Figure] Resource Types

...

when you add two NIC resource, these will be classified under the NIC type as shown below.
If you edit an attribute of resource type, this will affect all of the resources classified this type.

Image RemovedImage Added

[Figure] Classifying of resources under the type

...

Resource type has specified attributes for each type.
For an example, when compare to NIC and mirror disk, each service has different characteristics and has specific monitor interval, online/offline timeout and managed policies. 
In addition, even in the environment where high availability clusters are configured, if the network condition is not good, you need to adjust the monitoring timeout. Depending on the requests, you can change the attribute of each type.
Resource type attribute can be checked in the resource type attribute menu of the MCCS web console.

Image RemovedImage Added

[Figure] Resource Type Attribute View


AttributeDescriptionTypeDefaultEditNote
Actions

Define the resource operation type.
OnOff value is the command resource. None is the resource command only for monitoring.

STRING X 
ArgList

It represents the necessary arguments for each type of resource.
Based on these arguments, you can create variables in an agent and characterize the types.
You can check it in the type.json file but the value cannot be modified by a user.

STRING X 
CleanTimeout

It represents the time out until execution of CleanProgram is finished.
When the limit is exceeded, it will be forcefully terminated.

INTEGER60 secO 
FaultOnMonitorTimeouts

If the timeout specified by this value occurs, it is regarded as trouble.
If it is '0', it is not regarded as trouble.

INTEGER4 timesO 
MonitorInterval

It represents the time interval until the next monitoring of online resource.
If the value is too small, it could mistakenly start monitoring before receiving the proper monitoring results.
If the value is too big, then it will take too long until it detects the status changes.

INTEGER10 secO 
MonitorTimeoutIt represents the time elapsed until monitoring is finished or terminated.INTEGER10 secO 
OfflineMonitorInterval

It is the time interval until the next monitoring of offline resource.
If the time interval is '0', the offline resources will not be monitored.

INTEGER60 secO 
OfflineTimeout It represents the max time until the final decision is made on the success/failure of resource going offline.INTEGER300 secO 
OnlineRetryLimitSet the number of retries if the resource fails to restart.INTEGER0 timeO 
OnlineTimeoutIt represents the max time until the final decision is made on the success/failure of resource going online.INTEGER300 secO 
OnlineTrustTime

It is the time to reset the restart number of resource.
Initialize the number of retries if the normal condition is maintained for the duration specified after the resource is started.

INTEGER600 secO 
OnlineWaitLimit

If the time our occurs while determining whether the resource goes online properly, it will retry as many times as designated.
It is the time interval of retry is defined in 'MonitorInterval'.

NTEGER2 timesO 
*ToleranceLimit

When the monitoring results cannot be trusted , monitoring will be retried as many times as designated.
For instance, if the network condition is not good, and you cannot trust the ping test results, then you need to perform monitoring several more times until you get a good result.

INTEGER0 timeO 

...