cassandra-cli has alarming output for "show keyspaces"

(5 posts) (3 voices)

• Started 2 years ago by matt
• Latest reply from matt
• This topic is resolved

Tags:

No tags yet.

1. 

   matt
   Member

   I went poking into the cli to try to figure out what the replication factor is the the keyspace backing cassandraFS, but "show keyspaces did the following... (brief session transcript)

   [default@unknown] connect localhost/9160;
   Connected to: "Brisk One Node Cluster" on localhost/9160
   [default@unknown] show keyspaces;
   Required field 'replication_factor' was not found in serialized data! Struct: KsDef(name:system, .... etc

   I invoked it from the Brisk home directory. Is this indicative of a problem, or is this expected.

   Also, how do I find out what the keyspace definition is for the cassandraFS keyspace, or am I incorrect in assuming that cassandraFS is backed by a regular keyspace?

   Posted 2 years ago #
2. 

   nickmbailey
   Key Master

   Yes that appears to be a problem. How did you install/run brisk?

   The system keyspace should be using LocalStrategy and a replication_factor of 1.

   Posted 2 years ago #
3. 

   matt
   Member

   I did a manual install on one node. I have a cassandra user, the binary release was simply untarred in that directory and I made changes to $BRISK_HOME/resources/cassandra/conf/cassandra.yaml to fit my environment. Nothing major here, I just named the cluster, set the initial_token to zero and changed the paths for data, commitlog, saved_caches etc.

   I'm able to start up the cluster and JNA appears to be functioning correctly. I was able to successfully run some Hadoop TestDFSIO jobs with no problems.

   I didn't really do anything out of the ordinary. I can try the same procedure on another node and see if I get similar results.

   Thanks for the quick response,
   Matt

   Posted 2 years ago #
4. 

   thobbs
   Administrator

   This sounds like you're using an 0.7 (or early 0.8) CLI. I think an up-to-date 0.8 CLI should fix the problem.

   Posted 2 years ago #
5. 

   matt
   Member

   Resolved.

   I'm was invoking the cli with:
   ./resources/cassandra/bin/cassandra-cli

   from $BRISK_HOME, but I had forgotten that puppet was coming and overwriting the .bash_profile that I had edited to comment out the CASSANDRA_HOME setting. It looks like if CASSANDRA_HOME is set to an older cassandra distro, the cli invocation picks that up instead of the brisk environment. Since puppet restored my .bash_profile to the original that set CASSANDRA_HOME to point to my 0.7.5 install, I guess that code was getting invoked.

   Thanks for the help.

   sudo /etc/init.d/puppet stop -- woops, wrong window.

   Posted 2 years ago #

RSS feed for this topic

Reply

You must log in to post.