Ask is moving to Stack Overflow and Serverfault.com! Please use the "opendaylight" tag on either of these sites. This site is now in Read-Only mode

sahil27's profile - activity

2016-08-12 06:38:16 -0700 received badge  Taxonomist
2016-08-04 11:48:27 -0700 received badge  Popular Question (source)
2016-08-04 11:48:27 -0700 received badge  Notable Question (source)
2016-07-21 00:43:45 -0700 received badge  Famous Question (source)
2016-04-25 20:23:05 -0700 received badge  Popular Question (source)
2016-04-25 20:23:05 -0700 received badge  Notable Question (source)
2015-11-10 00:19:24 -0700 answered a question Single Node clustering - Operational Data Persistence issue

Thanks Tom from the controller-dev list for helping me with this. Just posting the answer here for the community.

There is a cfg file in the etc directory, org.opendaylight.controller.cluster.datastore.cfg, that was added in Lithium for configuring the clustered datastore in lieu of changing the 05-clustering.xml file to make preserving changes on upgrade easier. The cfg file overrides settings in 05-clustering.xml. You'll need to set "operational.persistent=true".

Tried with this and i worked.

2015-11-06 05:20:26 -0700 answered a question Clustering data persistence doesnot work, does it?

Hi Sepehr, Are you able to figure out what was the issue? I am also facing the same issue.

2015-11-06 05:16:32 -0700 answered a question Building ODL with minimum installed features

If you already have the tarball, it means you are trying to run and not build. It again depends on what features are loaded by default. If you want to load lesser features, you can change the same also.

What do you intend to do? That will clarify further,, however, I always have a minimum of 2 GB RAM when I run the distribution or build any project.

2015-11-06 04:55:22 -0700 received badge  Organizer (source)
2015-11-06 03:29:56 -0700 asked a question Single Node clustering - Operational Data Persistence issue

Hi All, I followed the single node clustering tutorial and it was working for the Lithium SR2, I tried to see the operational data getting persisted for netconf - inventory and topology.

Now, when I tried to verify the same with adding a small data model/rpc to the coretutorials/hello, the same is not working fine. MY operational data goes whenever I restart the controller.

Question 1: Is writing to the distributed datastore is different than how we write to the inmemory datastore.

Following are my changes:

  1. hello-karaf-1.1.0-SNAPSHOT/etc/opendaylight/karaf/05-clustering.xml: changed <persistent>true</persistent> for <name>distributed-operational-store-module</name>.

  2. In file: hello-karaf-1.1.0-SNAPSHOT/configuration/initial/modules.conf Added the module: { name = "hello" namespace = "urn:opendaylight:params:xml:ns:yang:hello" shard-strategy = "module" }

  3. In file: hello-karaf-1.1.0-SNAPSHOT/configuration/initial/module-shards.conf { name = "hello" shards = [ { name="hello" replicas = [ "member-1" ] } ] }

Then the following steps:

-> Start the ODL, bin/karaf -> Verify hello features installed successfully. -> Install the clustering feature: feature:install odl-mdsal-clustering -> Install the Jolokia bundle: install -s mvn:org.jolokia/jolokia-osgi/1.1.5

Then called my RPC to put the data to the operational datastore. I am getting the data from the REST. But when I restart the controller the data is gone. Needless to mention, configuration datastore has it.

I want this data to persist in the same way as it is for the configuration database.

Any pointer will be of a great help. It is kind of urgent.

2015-10-07 06:19:40 -0700 asked a question USC demo setup for call home functionality

I am setting up the USC demo with the steps given. USC is bring up and I am able to see the features loaded.

Just wanted to confirm, what can we use as a netconf servee with USC agent running so that we can demo the call home functionality.