Skip to Content.
Sympa Menu

perfsonar-user - Re: [perfsonar-user] Using personal to monitor failover testing

Subject: perfSONAR User Q&A and Other Discussion

List archive

Re: [perfsonar-user] Using personal to monitor failover testing


Chronological Thread 
  • From: Andrew Lake <>
  • To: Saravanaraj Ayyampalayam <>, "" <>
  • Subject: Re: [perfsonar-user] Using personal to monitor failover testing
  • Date: Fri, 23 Oct 2020 05:41:36 -0700

Th latency and packet loss tests actually run continuously (at a rate of 10 packets per second by default). Every 600 packets (~60 seconds) it generates a report on what happened to those packets, so you see a data point every minute but the tests is actually running all the time. You can change the packet rate and/or the number of packets before generating the report if you would like one more frequently. Assuming you have a pSConfig JSON file you use to configure the tests, you can set the packet-interval (e.g. 0.1 means 10 packets per second, 0.01 means 100 packets per second, etc) and packet-count (e.g. 300 means every 300 packets). The example below sends packets at 10 per second (0.1) and sends a report every 300 packets which results in reports every 30 seconds:

tests" : {
      "latencybg_30secs" : {
         "spec" : {
            "dest" : "{% address[1] %}",
            "dest-node" : "{% pscheduler_address[1] %}",
            "flip" : "{% flip %}",
            "packet-count" : 300,
            "packet-interval" : 0.1,
            "packet-padding" : 0,
            "source" : "{% address[0] %}",
            "source-node" : "{% pscheduler_address[0] %}"
         },
         "type" : "latencybg"
      }
}

Hope that helps,
Andy

On October 23, 2020 at 12:42:44 AM, Saravanaraj Ayyampalayam () wrote:

Hello,

 

Our network engineering team is going to test different failover scenarios this Saturday. They want to use our 12 perfsonar nodes around campus to monitor latency and packet loss during this testing.

They want to continuously monitor latency and packetloss when they perform manual failovers and record it in the esmond archiver for later analysis and record keeping.

It looks like the latency and packet loss task is run about once every minute. If there a way to run it more frequently so the visibility blind spot is minimized.

 

Thanks,

Raj

 

Raj Ayyampalayam
EITS-GACRC, University of Georgia, Athens GA
phone: (706) 542-0188
email:

--
To unsubscribe from this list: https://lists.internet2.edu/sympa/signoff/perfsonar-user

Attachment:
Description: Binary data




Archive powered by MHonArc 2.6.19.

Top of Page