[jira] [Updated] (CASSANDRA-10964) Startup errors in Docker containers depending on memtable allocation type
[ https://issues.apache.org/jira/browse/CASSANDRA-10964?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jeremy Hanna updated CASSANDRA-10964: - Component/s: (was: Materialized Views) > Startup errors in Docker containers depending on memtable allocation type > - > > Key: CASSANDRA-10964 > URL: https://issues.apache.org/jira/browse/CASSANDRA-10964 > Project: Cassandra > Issue Type: Bug > Components: Configuration > Environment: Docker, Debian Testing, 3.0.1 >Reporter: Jacek Furmankiewicz > > We are creating Docker containers for various versions of Cassandra. All are > based on Debian, Oracle JDK 1.8 and the Cassandra versions are installed > directly from the DataStax Debian repos via apt-get. > We noticed that with 3.0.1 (only that version, 2.1.11 and 2.2.4 work always > fine) the Cassandra process fails to start up randonly (about 50% of the > time) with the following error: > {noformat} > Caused by: java.lang.RuntimeException: > system_distributed:parent_repair_history not found in the schema definitions > keyspace. > at > org.apache.cassandra.schema.SchemaKeyspace.fetchTable(SchemaKeyspace.java:940) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.fetchTables(SchemaKeyspace.java:931) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.fetchKeyspace(SchemaKeyspace.java:894) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.fetchKeyspacesOnly(SchemaKeyspace.java:886) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.mergeSchema(SchemaKeyspace.java:1276) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.mergeSchemaAndAnnounceVersion(SchemaKeyspace.java:1255) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.service.MigrationManager$1.runMayThrow(MigrationManager.java:531) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) > ~[na:1.8.0_45] > at java.util.concurrent.FutureTask.run(FutureTask.java:266) > ~[na:1.8.0_45] > at > java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) > ~[na:1.8.0_45] > at > java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) > ~[na:1.8.0_45] > at java.lang.Thread.run(Thread.java:745) ~[na:1.8.0_45] > {noformat} > Started playing with different configuration parameters and by trial and > error figured out it seems to be related to this configuration parameter: > {noformat} > memtable_allocation_type: offheap_buffers > {noformat} > If we set it to offheap_buffers, this error occurs about 50% of the time > (when starting on a new clean filesystem). > If we set it to heap_buffers, it works always, 100% of the time, never an > issue. > Attaching full stack output to help debug: > {noformat} > INFO 16:11:44 Configuration location: > file:/etc/cassandra/cassandra.yaml > INFO 16:11:44 Node > configuration:[authenticator=PasswordAuthenticator; > authorizer=CassandraAuthorizer; auto_snapshot=true; > batch_size_fail_threshold_in_kb=50; batch_size_warn_threshold_in_kb=5; > batchlog_replay_throttle_in_kb=1024; cas_contention_timeout_in_ms=1000; > client_encryption_options=; cluster_name=TEST_CLUSTER; > column_index_size_in_kb=64; commit_failure_policy=stop; > commitlog_directory=/var/lib/cassandra/commitlog2; > commitlog_segment_size_in_mb=32; commitlog_sync=periodic; > commitlog_sync_period_in_ms=1; > compaction_large_partition_warning_threshold_mb=100; > compaction_throughput_mb_per_sec=16; concurrent_counter_writes=12; > concurrent_materialized_view_writes=7; concurrent_reads=64; > concurrent_writes=10; counter_cache_save_period=17200; > counter_cache_size_in_mb=1027; counter_write_request_timeout_in_ms=5000; > cross_node_timeout=false; data_file_directories=[/var/lib/cassandra/data2]; > disk_failure_policy=stop; disk_optimization_strategy=spinning; > dynamic_snitch_badness_threshold=0.1; > dynamic_snitch_reset_interval_in_ms=60; > dynamic_snitch_update_interval_in_ms=100; > enable_scripted_user_defined_functions=false; > enable_user_defined_functions=false; endpoint_snitch=SimpleSnitch; > gc_warn_threshold_in_ms=1000; hinted_handoff_enabled=true; > hinted_handoff_throttle_in_kb=1024; > hints_directory=/var/lib/cassandra/hints2; hints_flush_period_in_ms=1; >
[jira] [Updated] (CASSANDRA-10964) Startup errors in Docker containers depending on memtable allocation type
[ https://issues.apache.org/jira/browse/CASSANDRA-10964?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jeremy Hanna updated CASSANDRA-10964: - Component/s: Materialized Views > Startup errors in Docker containers depending on memtable allocation type > - > > Key: CASSANDRA-10964 > URL: https://issues.apache.org/jira/browse/CASSANDRA-10964 > Project: Cassandra > Issue Type: Bug > Components: Configuration, Materialized Views > Environment: Docker, Debian Testing, 3.0.1 >Reporter: Jacek Furmankiewicz > > We are creating Docker containers for various versions of Cassandra. All are > based on Debian, Oracle JDK 1.8 and the Cassandra versions are installed > directly from the DataStax Debian repos via apt-get. > We noticed that with 3.0.1 (only that version, 2.1.11 and 2.2.4 work always > fine) the Cassandra process fails to start up randonly (about 50% of the > time) with the following error: > {noformat} > Caused by: java.lang.RuntimeException: > system_distributed:parent_repair_history not found in the schema definitions > keyspace. > at > org.apache.cassandra.schema.SchemaKeyspace.fetchTable(SchemaKeyspace.java:940) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.fetchTables(SchemaKeyspace.java:931) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.fetchKeyspace(SchemaKeyspace.java:894) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.fetchKeyspacesOnly(SchemaKeyspace.java:886) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.mergeSchema(SchemaKeyspace.java:1276) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.mergeSchemaAndAnnounceVersion(SchemaKeyspace.java:1255) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.service.MigrationManager$1.runMayThrow(MigrationManager.java:531) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) > ~[na:1.8.0_45] > at java.util.concurrent.FutureTask.run(FutureTask.java:266) > ~[na:1.8.0_45] > at > java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) > ~[na:1.8.0_45] > at > java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) > ~[na:1.8.0_45] > at java.lang.Thread.run(Thread.java:745) ~[na:1.8.0_45] > {noformat} > Started playing with different configuration parameters and by trial and > error figured out it seems to be related to this configuration parameter: > {noformat} > memtable_allocation_type: offheap_buffers > {noformat} > If we set it to offheap_buffers, this error occurs about 50% of the time > (when starting on a new clean filesystem). > If we set it to heap_buffers, it works always, 100% of the time, never an > issue. > Attaching full stack output to help debug: > {noformat} > INFO 16:11:44 Configuration location: > file:/etc/cassandra/cassandra.yaml > INFO 16:11:44 Node > configuration:[authenticator=PasswordAuthenticator; > authorizer=CassandraAuthorizer; auto_snapshot=true; > batch_size_fail_threshold_in_kb=50; batch_size_warn_threshold_in_kb=5; > batchlog_replay_throttle_in_kb=1024; cas_contention_timeout_in_ms=1000; > client_encryption_options=; cluster_name=TEST_CLUSTER; > column_index_size_in_kb=64; commit_failure_policy=stop; > commitlog_directory=/var/lib/cassandra/commitlog2; > commitlog_segment_size_in_mb=32; commitlog_sync=periodic; > commitlog_sync_period_in_ms=1; > compaction_large_partition_warning_threshold_mb=100; > compaction_throughput_mb_per_sec=16; concurrent_counter_writes=12; > concurrent_materialized_view_writes=7; concurrent_reads=64; > concurrent_writes=10; counter_cache_save_period=17200; > counter_cache_size_in_mb=1027; counter_write_request_timeout_in_ms=5000; > cross_node_timeout=false; data_file_directories=[/var/lib/cassandra/data2]; > disk_failure_policy=stop; disk_optimization_strategy=spinning; > dynamic_snitch_badness_threshold=0.1; > dynamic_snitch_reset_interval_in_ms=60; > dynamic_snitch_update_interval_in_ms=100; > enable_scripted_user_defined_functions=false; > enable_user_defined_functions=false; endpoint_snitch=SimpleSnitch; > gc_warn_threshold_in_ms=1000; hinted_handoff_enabled=true; > hinted_handoff_throttle_in_kb=1024; > hints_directory=/var/lib/cassandra/hints2;
[jira] [Updated] (CASSANDRA-10964) Startup errors in Docker containers depending on memtable allocation type
[ https://issues.apache.org/jira/browse/CASSANDRA-10964?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jacek Furmankiewicz updated CASSANDRA-10964: Summary: Startup errors in Docker containers depending on memtable allocation type (was: Starup errors in Docker containers depending on memtable allocation type) > Startup errors in Docker containers depending on memtable allocation type > - > > Key: CASSANDRA-10964 > URL: https://issues.apache.org/jira/browse/CASSANDRA-10964 > Project: Cassandra > Issue Type: Bug > Components: Configuration > Environment: Docker, Debian Testing, 3.0.1 >Reporter: Jacek Furmankiewicz > > We are creating Docker containers for various versions of Cassandra. All are > based on Debian, Oracle JDK 1.8 and the Cassandra versions are installed > directly from the DataStax Debian repos via apt-get. > We noticed that with 3.0.1 (only that version, 2.1.11 and 2.2.4 work always > fine) the Cassandra process fails to start up randonly (about 50% of the > time) with the following error: > {noformat} > Caused by: java.lang.RuntimeException: > system_distributed:parent_repair_history not found in the schema definitions > keyspace. > at > org.apache.cassandra.schema.SchemaKeyspace.fetchTable(SchemaKeyspace.java:940) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.fetchTables(SchemaKeyspace.java:931) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.fetchKeyspace(SchemaKeyspace.java:894) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.fetchKeyspacesOnly(SchemaKeyspace.java:886) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.mergeSchema(SchemaKeyspace.java:1276) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.schema.SchemaKeyspace.mergeSchemaAndAnnounceVersion(SchemaKeyspace.java:1255) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.service.MigrationManager$1.runMayThrow(MigrationManager.java:531) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28) > ~[apache-cassandra-3.0.1.jar:3.0.1] > at > java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) > ~[na:1.8.0_45] > at java.util.concurrent.FutureTask.run(FutureTask.java:266) > ~[na:1.8.0_45] > at > java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) > ~[na:1.8.0_45] > at > java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) > ~[na:1.8.0_45] > at java.lang.Thread.run(Thread.java:745) ~[na:1.8.0_45] > {noformat} > Started playing with different configuration parameters and by trial and > error figured out it seems to be related to this configuration parameter: > {noformat} > memtable_allocation_type: offheap_buffers > {noformat} > If we set it to offheap_buffers, this error occurs about 50% of the time > (when starting on a new clean filesystem). > If we set it to heap_buffers, it works always, 100% of the time, never an > issue. > Attaching full stack output to help debug: > {noformat} > INFO 16:11:44 Configuration location: > file:/etc/cassandra/cassandra.yaml > INFO 16:11:44 Node > configuration:[authenticator=PasswordAuthenticator; > authorizer=CassandraAuthorizer; auto_snapshot=true; > batch_size_fail_threshold_in_kb=50; batch_size_warn_threshold_in_kb=5; > batchlog_replay_throttle_in_kb=1024; cas_contention_timeout_in_ms=1000; > client_encryption_options=; cluster_name=TEST_CLUSTER; > column_index_size_in_kb=64; commit_failure_policy=stop; > commitlog_directory=/var/lib/cassandra/commitlog2; > commitlog_segment_size_in_mb=32; commitlog_sync=periodic; > commitlog_sync_period_in_ms=1; > compaction_large_partition_warning_threshold_mb=100; > compaction_throughput_mb_per_sec=16; concurrent_counter_writes=12; > concurrent_materialized_view_writes=7; concurrent_reads=64; > concurrent_writes=10; counter_cache_save_period=17200; > counter_cache_size_in_mb=1027; counter_write_request_timeout_in_ms=5000; > cross_node_timeout=false; data_file_directories=[/var/lib/cassandra/data2]; > disk_failure_policy=stop; disk_optimization_strategy=spinning; > dynamic_snitch_badness_threshold=0.1; > dynamic_snitch_reset_interval_in_ms=60; > dynamic_snitch_update_interval_in_ms=100; > enable_scripted_user_defined_functions=false; > enable_user_defined_functions=false; endpoint_snitch=SimpleSnitch; > gc_warn_threshold_in_ms=1000;