Changes between Version 25 and Version 26 of LAMP/Tutorial


Ignore:
Timestamp:
09/22/10 01:24:23 (14 years ago)
Author:
fernande@cis.udel.edu
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • LAMP/Tutorial

    v25 v26  
    381381
    382382
    383 === 9. Pushing our changes to UNIS ===
     383=== 9. Pushing our Changes to UNIS ===
    384384
    385385We have now finished the configuration of our I&M services. The host monitoring tool (Ganglia) will be enabled with default parameters (monitor all metrics available by default). The host monitoring daemons will be configured by pSConfig to send their announcements to the host monitoring collector. However, we have only made changes to the configuration stored on our LAMP Portal node. We need to push these changes to UNIS so that the nodes in our slice can fetch it and reconfigure themselves.
     
    388388
    389389[[Image(portal6.png)]]
     390
     391
     392=== 10. Reaping the Benefits ===
     393
     394We have finished configuring our slice with instrumentation & measurement! It will take a couple of minutes until everything is configured and stabilized. After taking a dinner break (and, in my case, fixing the NTP bug mentioned above), we come back to our LAMP Portal to check on our services and see the data we've collected so far.
     395
     396
     397==== Services Registration on UNIS ====
     398
     399Our services should have started and registered themselves with UNIS. We can check this by going to the ''Registered Services'' overview page. (If you don't see any services, click on the ''Pull Registered Services from UNIS'' button.)
     400
     401[[Image(portal7.png)]]
     402
     403[[Color(white, red,This directory has a bug right now that the measurement tools registration data is not being fetched by the portal, but the tools are most likely registered with UNIS by now. The double registration of the PSB_OWAMP services is also a bug.]]
     404
     405We can see that all our perfSONAR (measurement) services have registered! The SNMP MA services is the 'Ganglia MA' we configured before; it is a SNMP MA exporting the data collected by the Ganglia Meta Daemon (our ''host monitoring colletor''). Now lets go over the visualization for each type of data.
     406
     407
     408==== Visualizing Ping Data  ====
     409
     410We can see on the left sidebar the four links for visualizing One-way latency, Ping latency, Throughput and Host Monitoring data. However, clicking on these links brings us back to the same page we were on the previous step, the ''Registered Services'' page. Right now we do not have a dedicated page for each service/type of data. What we can do is click on the 'Query' button next to the service we want to query. So if we click on the Query button for the PingER MA (3rd row), we should a page similar to the following.
     411
     412[[Image(pingerma.png)]]
     413
     414Great! We can see that our dataset exists on the PingER MA. Although the source address appears as the control interface's address (it just reverse resolves the hostname), the tests do go through the virtual topology because the destination address is the address for node2's '''iface0'''. We can now select between 12, 24 and 48 hours graphs. We will take the 12 hours graph since we haven't been running that long. A nifty little graph will popup.
     415
     416[[Image(pingerma-graph.png)]]
     417