cancel
Showing results for 
Search instead for 
Did you mean: 

VCS probe question

Hello,

IF the resouces didn't probe in VCS startup and if you go and start the resource manually?  would VCS will probe the resouce manually? Or would it show as not probed.

How about the hares  -state?

 

5 Solutions

Accepted Solutions
Accepted Solution!

Hi, If the monitor script of

Hi, If the monitor script of resource is correct, resource would probe automatically at VCS start up. If resource is not probed during startup, there is very slim chance that resource will probe manually without taking any corrective action. You would need to examine issues with monitoring script and fix it G

View solution in original post

Accepted Solution!

VCS probes online resource

VCS probes online resource every MonitorInterval and offline resources every OfflineMonitorInterval. So if you manually online a resource then VCS should prove within OfflineMonitorInterval for that resource type. Example for IP: hatype -display IP | grep MonitorInt IP MonitorInterval 60 IP OfflineMonitorInterval 300 But as Gaurav says if VCS had a problems probing at start up then it may not probe later either. Mike

View solution in original post

Accepted Solution!

Hi.. Two Questions, Where do

Hi..

Two Questions,

Where do i find the monitering script to analyze this?

2.If i want to change the attribute value of a resource i.e wants to make a oracle resource as a critical then how to proceed?

 

Thanks,

Allaboutunix

View solution in original post

Accepted Solution!

Monitoring script is located

Monitoring script is located by default in 2 folders

/opt/VRTSagents/ha/bin/<AgentName>/<AgentNameAgent>

/opt/VRTSvcs/bin/CustomAgents/

however monitor process or program can be kept in any folder with appropriate permissions ... & full path can be defined in main.cf  ... for e.g

Application myapp_app2(
     StartProgram="/usr/sbin/myapp start"
     StopProgram="/usr/sbin/myapp stop"
     CleanProgram="usr/sbin/myapp force stop"
     MonitorProgram="/usr/local/bin/myappMonitor all"
     MonitorProcesses={ "myapp1" , "myapp2" }

Question 2 is completely unrelated to original post however following would be steps

# haconf -makerw

# hares -modify <resource_name> Critical 1

# haconf -dump -makero

 

make sure you are clear with dependencies defined when you are setting critical 1 for any resources. VCS admin guides is the perfect source to understand the concepts.

 

G

 

View solution in original post

Accepted Solution!

To analyse you should also

To analyse you should also look in engine_A.log and agent log for the type of your resource - example IP.log. Also note that MonitorInterval and OfflineMonitorInterval are set per type, not per resource, so to change these you would need to run for example: hatype -modify IP OfflineMonitorInterval 240 Mike

View solution in original post

6 Replies
Accepted Solution!

Hi, If the monitor script of

Hi, If the monitor script of resource is correct, resource would probe automatically at VCS start up. If resource is not probed during startup, there is very slim chance that resource will probe manually without taking any corrective action. You would need to examine issues with monitoring script and fix it G

View solution in original post

Accepted Solution!

VCS probes online resource

VCS probes online resource every MonitorInterval and offline resources every OfflineMonitorInterval. So if you manually online a resource then VCS should prove within OfflineMonitorInterval for that resource type. Example for IP: hatype -display IP | grep MonitorInt IP MonitorInterval 60 IP OfflineMonitorInterval 300 But as Gaurav says if VCS had a problems probing at start up then it may not probe later either. Mike

View solution in original post

Accepted Solution!

Hi.. Two Questions, Where do

Hi..

Two Questions,

Where do i find the monitering script to analyze this?

2.If i want to change the attribute value of a resource i.e wants to make a oracle resource as a critical then how to proceed?

 

Thanks,

Allaboutunix

View solution in original post

Accepted Solution!

Monitoring script is located

Monitoring script is located by default in 2 folders

/opt/VRTSagents/ha/bin/<AgentName>/<AgentNameAgent>

/opt/VRTSvcs/bin/CustomAgents/

however monitor process or program can be kept in any folder with appropriate permissions ... & full path can be defined in main.cf  ... for e.g

Application myapp_app2(
     StartProgram="/usr/sbin/myapp start"
     StopProgram="/usr/sbin/myapp stop"
     CleanProgram="usr/sbin/myapp force stop"
     MonitorProgram="/usr/local/bin/myappMonitor all"
     MonitorProcesses={ "myapp1" , "myapp2" }

Question 2 is completely unrelated to original post however following would be steps

# haconf -makerw

# hares -modify <resource_name> Critical 1

# haconf -dump -makero

 

make sure you are clear with dependencies defined when you are setting critical 1 for any resources. VCS admin guides is the perfect source to understand the concepts.

 

G

 

View solution in original post

Accepted Solution!

To analyse you should also

To analyse you should also look in engine_A.log and agent log for the type of your resource - example IP.log. Also note that MonitorInterval and OfflineMonitorInterval are set per type, not per resource, so to change these you would need to run for example: hatype -modify IP OfflineMonitorInterval 240 Mike

View solution in original post

Thank you alll. It is much

Thank you alll. It is much clear now.