[no subject]

My cluster of 11 nodes running Casandra 1.1-5 is  pausing too long for ParNew GC, which increases our response latency, Is it a good idea to have a a smaller HEAP_NEWSIZE so that we can collect more often, but not pause that long? INFO [ScheduledTasks:1] 2013-05-10 01:00:17,245 GCInspector.java

Custom 1.2 Authentication plugin will not work unless user is in system_auth.users column family

Hi,     We have a custom  authenticator that works well with Cassandra 1.1.5. When upgrading to C* 1.2.5, authentication failed. Turn out that in ClientState.login, we make a call to Auth.isExistingUser(user.getName()) if the AuthenticatedUser is not Anonymous user. This isExistingUser method does

Re: Custom 1.2 Authentication plugin will not work unless user is in system_auth.users column family

Sorry for not following up on this one in time. I filed a JIRA (5651) and it seems user lookup is here to stay. https://issues.apache.org/jira/browse/CASSANDRA-5651?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel On a related note, that column family is, by default, set up to h

High-read latency for non-existing rows with LCS and 1.2.5

Hi,     We are using Leveled Compaction with Cassandra 1.2.5. Our sstable size is 100M. On each node, we have anywhere from 700+ to 800+ sstables (for all levels). The bloom_filter_fp_chance is set at 0.000744.   For read requests that ask for existing row records, the latency is great, mostly