http://www.datastax.com/support-forums/topic/a-lot-of-disk-space-used-and-unbalanced-nodes Software, Support, and Training for Apache Cassandra en-US Wed, 22 May 2013 14:19:24 +0000 http://bbpress.org/?v=1.0.3 q http://www.datastax.com/support-forums/search.php http://www.datastax.com/support-forums/topic/a-lot-of-disk-space-used-and-unbalanced-nodes#post-151 Fri, 03 Jun 2011 11:23:22 +0000 ithkuil 151@http://www.datastax.com/support-forums/ <p>I was using cassandra 0.7.6, replication factor 3. Feeding data with hadoop.</p> <p>I written a java client to feed the data without hadoop and I never run into this issue again. I'll retry later when I upgrade to cassandra 0.8. </p> http://www.datastax.com/support-forums/topic/a-lot-of-disk-space-used-and-unbalanced-nodes#post-118 Wed, 25 May 2011 22:22:24 +0000 joaquin 118@http://www.datastax.com/support-forums/ <p>What versions of Cassandra and OpsCenter are you running?</p> <p>Having that much data usage to actual data is not normal.</p> <p>Having one node bigger doesn't appear to be normal behavior either. Because random partitoner evenly distributes the data, I don't see how this is happening.</p> <p>What replication factor are you using? Are there any errors in any of your logs? How are you sending the data to Cassandra? </p> http://www.datastax.com/support-forums/topic/a-lot-of-disk-space-used-and-unbalanced-nodes#post-77 Wed, 18 May 2011 17:21:27 +0000 ithkuil 77@http://www.datastax.com/support-forums/ <p>I have a cassandra installation with 3 nodes + OpsCenter.</p> <p>I loaded about 800k rows containing a single value, about 2.6k size each.</p> <p>The content was loaded using a maponly job from hadoop, the input file 2 Gb input file.</p> <p>I started with an empty cassandra cluster (except for OpsCenter data, about 50M)<br /> After the import the cassandra nodes have from 20 to 30 Gb used data.</p> <p>I tried compacting, multiple times, draning, shutting down, restarting, compacting, repairing, etc</p> <p>Usage is still high, far above what is needed to store the data. There are no deleted<br /> rows, and the commit logs are drained and everything compacted.</p> <p>Why on earth does cassandra require so much disk space?</p> <p>(I have to store max 10 Gb of real data, I have to allocate 100Gb now to hold 2-3 Gb and have free space to perform compactions and repairs etc)</p> <p>Furthermore, why one node is so bigger than others? I use random partitioner.</p> <p><a href="http://awesomescreenshot.com/05cd8igb1" rel="nofollow">http://awesomescreenshot.com/05cd8igb1</a></p> <p>Is there a way to see that the 30 Gb of the biggest node are indeed filled also by data<br /> arriving from the another peer; and that part of the data which belongs to this biggest node is actually replicated? I have somehow the feeling that one node gets more data and doesn't replicate. (my payloads are quite homogeneous)</p> <p>Any help is appreciated!</p> <p>Marko </p>