http://www.datastax.com/support-forums/forum/datastax-enterprise Software, Support, and Training for Apache Cassandra en-US Sat, 25 May 2013 07:22:12 +0000 http://bbpress.org/?v=1.0.3 q http://www.datastax.com/support-forums/search.php http://www.datastax.com/support-forums/topic/changing-the-jetty-port#post-11332 Sat, 25 May 2013 03:01:56 +0000 Ion Olaru 11332@http://www.datastax.com/support-forums/ <p>Hi</p> <p>I am trying to deploy 2 instances of Cassandra 1.2.5 + their respective agents + 1 opscenter on the same box, my local development box, I know this is weird, but I would like to have this configured locally before I move forward.</p> <p><strong>WARN [Jetty] 2013-05-24 23:02:46,794 FAILED org.eclipse.jetty.server.Server@4c18e99: java.net.BindException: Address already in use<br /> </strong></p> <p>My error now is the Jetty server Exception coming form the 2nd agent on start:<br /> I was wondering if I have any control on what port jetty should run.</p> <p>Thanks </p> http://www.datastax.com/support-forums/topic/there-was-a-problem-verifying-an-ssh-login-from-opscenter-ui#post-11330 Fri, 24 May 2013 18:05:52 +0000 hwangj 11330@http://www.datastax.com/support-forums/ <p>Hi,<br /> I am running into the error while creating cluster using opscenter. SSH connection is not an issue if done out side of the opscenter.<br /> Found the following error message in opscenterd.log:</p> <p>2013-05-24 09:48:20-0700 [] There was a problem verifying an ssh login on t1u1d-daasdev1<br /> Traceback (most recent call last):<br /> Failure: opscenterd.SecureShell.SshFailed: ssh to u't1u1d-daasdev1' failed</p> <p>2013-05-24 09:48:21-0700 [] ERROR: Unexpected error when attempting to login to t1u1d-daasdev1: Timed out verifying ssh login for t1u1d-daasdev1. Previous exception was: SshFailed: ssh to u't1u1d-daasdev1' failed<br /> File "/usr/share/opscenter/lib/py-debian/2.7/amd64/twisted/internet/defer.py", line 1018, in _inlineCallbacks<br /> result = result.throwExceptionIntoGenerator(g)</p> <p> File "/usr/share/opscenter/lib/py-debian/2.7/amd64/twisted/python/failure.py", line 349, in throwExceptionIntoGenerator<br /> return g.throw(self.type, self.value, self.tb)</p> <p> File "/usr/lib/python2.7/dist-packages/opscenterd/SecureShell.py", line 315, in verify_ssh_login </p> http://www.datastax.com/support-forums/topic/dse-node-in-analytics-mode#post-11329 Fri, 24 May 2013 16:27:17 +0000 bcoverston 11329@http://www.datastax.com/support-forums/ <p>You can use CFS just like you would use HDFS. Replication factor (default) is two, it can be changed. You can store your long term data there if you desire, but in most cases it's probably better to just run your MR jobs over cassandra CFs. MapReduce actually has a dependency on HDFS, so it has to be there in order to run MapReduce. </p> http://www.datastax.com/support-forums/topic/dse-node-in-analytics-mode#post-11328 Fri, 24 May 2013 07:41:18 +0000 whoisvolos 11328@http://www.datastax.com/support-forums/ <p>Good day!</p> <p>I'm I right that cfs keyspace lives only on analytics-mode nodes? If it's right, then what's the purpose of cfs? Temporary file system? There're nothing to write in this file system (I mean clients for different languages/frameworks/tool) but <code>dse hadoop fs</code> and modeling file system in client application (inodes/sblocks etc) then write with cassandra client. And what with replication factor of this keyspace? Can it be changed? </p> http://www.datastax.com/support-forums/topic/writing-to-cfs-with-flume#post-11327 Fri, 24 May 2013 07:29:23 +0000 whoisvolos 11327@http://www.datastax.com/support-forums/ <p>Hdfs sink uses TCP connection to HDFS's Namenode and communicates via HDFS (may be Thrift, I dunno) protocol. DSE does not use namenodes. It abstracts HDFS in cfs Cassandra keyspace (file system model). As I know now, the only possible way to write to cfs is to model cfs file system in sink and then work with inodes/sblocks in cfs keyspace. </p> http://www.datastax.com/support-forums/topic/adding-new-node-failed-in-datastax-enterprise-30#post-11326 Thu, 23 May 2013 17:29:56 +0000 hwangj 11326@http://www.datastax.com/support-forums/ <p>The issue has been resolved. </p> http://www.datastax.com/support-forums/topic/slow-delete-by-query-on-indexed-fields#post-11325 Thu, 23 May 2013 15:05:56 +0000 rockbrain 11325@http://www.datastax.com/support-forums/ <p>Each record is deleted from C*, we could probably parallelize that however the time does not look off. </p> http://www.datastax.com/support-forums/topic/question-about-message-in-cassandra-logs-duplicate-core-plugin-detected-waiting-for-it-to-finish#post-11324 Thu, 23 May 2013 14:31:32 +0000 sam 11324@http://www.datastax.com/support-forums/ <p>Hi, that message is safe to ignore - it occurs at a synchronization point when multiple concurrent request threads cause an index to be initialised or reloaded at the same time. In version 3.0, Solr core management in DSE was reworked and the conditions that caused this contention no longer occur.. </p> http://www.datastax.com/support-forums/topic/slow-delete-by-query-on-indexed-fields#post-11323 Thu, 23 May 2013 12:29:36 +0000 jpayne97 11323@http://www.datastax.com/support-forums/ <p>Not sure how long a delete by a query should be taking, but for deleting 50961 records it took 73 seconds to delete. Both the fields in the query are indexed in the schema. Is this normal?</p> <p><a href="http://u060dc2d82:8983/solr/item.location/update?stream.body=%3Cdelete%3E%3Cquery%3E(div:014)AND(store:00305)%3C/query%3E%3C/delete%3E" rel="nofollow">http://u060dc2d82:8983/solr/item.location/update?stream.body=%3Cdelete%3E%3Cquery%3E(div:014)AND(store:00305)%3C/query%3E%3C/delete%3E</a></p> <p>&lt;response&gt;<br /> &lt;lst name="responseHeader"&gt;<br /> &lt;int name="status"&gt;0&lt;/int&gt;<br /> &lt;int name="QTime"&gt;73342&lt;/int&gt;<br /> &lt;/lst&gt;<br /> &lt;/response&gt; </p> http://www.datastax.com/support-forums/topic/question-about-message-in-cassandra-logs-duplicate-core-plugin-detected-waiting-for-it-to-finish#post-11322 Thu, 23 May 2013 12:26:28 +0000 jpayne97 11322@http://www.datastax.com/support-forums/ <p>Using SolrJ to connect to solr and we are constantly seeing these two messages when load testing. Is there something configured wrong? We are just doing GETs to query data when this is happening.</p> <p>INFO [http-8983-82] 2013-05-22 16:04:40,549 CassandraCoreContainer.java (line 265) Creating new core: item.location<br /> INFO [http-8983-82] 2013-05-22 16:04:40,549 CassandraCoreContainerManager.java (line 41) Duplicate core plugin detected, waiting for it to finish... </p> http://www.datastax.com/support-forums/topic/byte-array-from-solr-to-cassandra-cf-dse-301#post-11319 Wed, 22 May 2013 22:27:35 +0000 Jack Krupansky 11319@http://www.datastax.com/support-forums/ <p>Sorry for the delayed response. Use the "binary" field type in Solr:</p> <p> &lt;fieldType name="binary" class="solr.BinaryField"/&gt;</p> <p>Note that a solr.ByteField is simply a single 8-bit signed integer.</p> <p>The value for a Solr binary field is a Base64 string. For example, "Hello World" encodes as "SGVsbG8gV29ybGQK".</p> <p>Also, since there is no way to query the content of a binary field in Solr, set it to indexed="false" and stored="true":</p> <p> &lt;field name="bin" type="binary" indexed="false" stored="true"/&gt;</p> <p>A Solr "BinaryField" will correspond to a CQL "blob" or C* BytesType. </p> http://www.datastax.com/support-forums/topic/dse-30-with-internal-authentication-and-hadoop#post-11318 Wed, 22 May 2013 16:55:20 +0000 felipe.abezerra 11318@http://www.datastax.com/support-forums/ <p>Hi Sam!</p> <p>Thanks a lot for replying.<br /> I'm thinking on change somethings to solve this. </p> http://www.datastax.com/support-forums/topic/writing-to-cfs-with-flume#post-11302 Wed, 22 May 2013 01:15:02 +0000 bcoverston 11302@http://www.datastax.com/support-forums/ <p>If I were you I would try the HDFS sink. It should work. If it doesn't let us know and we can look at it. </p> http://www.datastax.com/support-forums/topic/sstableloader-error#post-11301 Tue, 21 May 2013 21:13:00 +0000 upant 11301@http://www.datastax.com/support-forums/ <p>Finally it worked by putting hadoop core jar, apache commons IO jar, dse.yml and some other jars. </p> http://www.datastax.com/support-forums/topic/how-do-you-change-the-data-location#post-11299 Tue, 21 May 2013 17:47:04 +0000 cb55555 11299@http://www.datastax.com/support-forums/ <p>OK. It turns out that the problem was an illegal character in the yaml file. </p>