incubator-cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Frank LoVecchio <>
Subject Cassandra/Pelops error processing get_slice
Date Thu, 14 Oct 2010 23:32:31 GMT
10:10:21,787 ERROR ~ Error getting Sensor

org.apache.thrift.TApplicationException: Internal error processing get_slice


at org.apache.cassandra.thrift.Cassandra$Client.recv_get_slice(

at org.apache.cassandra.thrift.Cassandra$Client.get_slice(

at org.scale7.cassandra.pelops.Selector$6.execute(

at org.scale7.cassandra.pelops.Selector$6.execute(

at org.scale7.cassandra.pelops.Operand.tryOperation(

at org.scale7.cassandra.pelops.Selector.getSuperColumnsFromRow(

at org.scale7.cassandra.pelops.Selector.getSuperColumnsFromRow(

Not sure if this is a pelops thing or thrift.

I spun up a new cluster of 3 nodes a couple of nights ago with the nightly
build 0.7 beta 2's.  When I include all 3 nodes in the Pelops Pool, and run

List<SuperColumn> cols = selector.getSuperColumnsFromRow(
                               getColFamilyName(), key,
                               Selector.newColumnsPredicateAll(true, 1000),

 I get the error above.  When I create a new Pool with only 1 node, I get
different data back then if I create a new Pool with another single node,
but I don't get the error.  Why is this happening?  Here is my
cassandra.yaml file, which I haven't modified on the fly:

# Configuration wiki:
authority: org.apache.cassandra.auth.AllowAllAuthority
authenticator: org.apache.cassandra.auth.AllowAllAuthenticator
auto_bootstrap: true
binary_memtable_throughput_in_mb: 256
cluster_name: DevCluster
column_index_size_in_kb: 64
commitlog_directory: /cassandra/commitlog
saved_caches_directory: /cassandra/saved_caches
commitlog_rotation_threshold_in_mb: 128
commitlog_sync: periodic
commitlog_sync_period_in_ms: 10000
concurrent_reads: 8
concurrent_writes: 50
- /cassandra/data
disk_access_mode: auto
endpoint_snitch: org.apache.cassandra.locator.SimpleSnitch
dynamic_snitch: true
hinted_handoff_enabled: true
in_memory_compaction_limit_in_mb: 256
    - name: Keyspace
      replication_factor: 3

        - name: Sensor
          column_type: Super
          compare_with: TimeUUIDType
          gc_grace_seconds: 864000
          keys_cached: 200000.0
          preload_row_cache: false
          read_repair_chance: 1.0
          rows_cached: 0.0

listen_address: ec2-internal-ip1
memtable_flush_after_mins: 60
memtable_operations_in_millions: 0.3
memtable_throughput_in_mb: 64
partitioner: org.apache.cassandra.dht.RandomPartitioner
phi_convict_threshold: 8
rpc_port: 9160
rpc_timeout_in_ms: 10000
- ec2-internal-ip1
- ec2-internal-ip2
- ec2-internal-ip3
sliced_buffer_size_in_kb: 64
snapshot_before_compaction: false
storage_port: 7000
thrift_framed_transport_size_in_mb: 15
thrift_max_message_length_in_mb: 16

Frank LoVecchio
Software Engineer
Isidorey LLC

View raw message