[jira] [Resolved] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Andrew S resolved CASSANDRA-8089. - Resolution: Not a Problem Thank you for a quick feedback! It was our own mistake we were inserting data in Cassandra setting some columns to null, this is why we were seeing tombstone warnings. Closing ticket. Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. (We have done this on all nodes having the data and all sstables containing the key) {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Do you have any ideas how can we debug this further? Thanks, Andrew -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Andrew S updated CASSANDRA-8089: Description: Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. (We have done this on all nodes having the data and all sstables containing the key) {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Do you have any ideas how can we debug this further? Thanks, Andrew was: Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. (We have done this on all nodes having the data and all sstables containing the key) {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Thanks, Andrew Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM)
[jira] [Comment Edited] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14166563#comment-14166563 ] Andrew S edited comment on CASSANDRA-8089 at 10/10/14 8:46 AM: --- We are also getting the following exception on one of the servers after flushing another table with similar structure. 2014-10-10T07:02:46.402+ su8 ERROR [CompactionExecutor:2] 2014-10-10 09:02:46,396 CassandraDaemon.java:166 - Exception in thread Thread[CompactionExecutor:2,1,main] java.lang.IllegalStateException: Unable to compute ceiling for max when histogram overflowed at org.apache.cassandra.utils.EstimatedHistogram.mean(EstimatedHistogram.java:203) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.io.sstable.metadata.StatsMetadata.getEstimatedDroppableTombstoneRatio(StatsMetadata.java:98) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.io.sstable.SSTableReader.getEstimatedDroppableTombstoneRatio(SSTableReader.java:1805) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.AbstractCompactionStrategy.worthDroppingTombstones(AbstractCompactionStrategy.java:297) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy.getNextBackgroundSSTables(SizeTieredCompactionStrategy.java:106) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy.getNextBackgroundTask(SizeTieredCompactionStrategy.java:267) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionTask.run(CompactionManager.java:229) ~[apache-cassandra-2.1.0.jar:2.1.0] at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471) ~[na:1.7.0_51] at java.util.concurrent.FutureTask.run(FutureTask.java:262) ~[na:1.7.0_51] at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) ~[na:1.7.0_51] at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) [na:1.7.0_51] at java.lang.Thread.run(Thread.java:744) [na:1.7.0_51] was (Author: andrews): We are also getting the following exception on one of the servers after flushing another table that also had one row with TTL a while back. 2014-10-10T07:02:46.402+ su8 ERROR [CompactionExecutor:2] 2014-10-10 09:02:46,396 CassandraDaemon.java:166 - Exception in thread Thread[CompactionExecutor:2,1,main] java.lang.IllegalStateException: Unable to compute ceiling for max when histogram overflowed at org.apache.cassandra.utils.EstimatedHistogram.mean(EstimatedHistogram.java:203) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.io.sstable.metadata.StatsMetadata.getEstimatedDroppableTombstoneRatio(StatsMetadata.java:98) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.io.sstable.SSTableReader.getEstimatedDroppableTombstoneRatio(SSTableReader.java:1805) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.AbstractCompactionStrategy.worthDroppingTombstones(AbstractCompactionStrategy.java:297) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy.getNextBackgroundSSTables(SizeTieredCompactionStrategy.java:106) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy.getNextBackgroundTask(SizeTieredCompactionStrategy.java:267) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionTask.run(CompactionManager.java:229) ~[apache-cassandra-2.1.0.jar:2.1.0] at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471) ~[na:1.7.0_51] at java.util.concurrent.FutureTask.run(FutureTask.java:262) ~[na:1.7.0_51] at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) ~[na:1.7.0_51] at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) [na:1.7.0_51] at java.lang.Thread.run(Thread.java:744) [na:1.7.0_51] Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808,
[jira] [Commented] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14166563#comment-14166563 ] Andrew S commented on CASSANDRA-8089: - We are also getting the following exception on one of the servers after flushing another table that also had one row with TTL a while back. 2014-10-10T07:02:46.402+ su8 ERROR [CompactionExecutor:2] 2014-10-10 09:02:46,396 CassandraDaemon.java:166 - Exception in thread Thread[CompactionExecutor:2,1,main] java.lang.IllegalStateException: Unable to compute ceiling for max when histogram overflowed at org.apache.cassandra.utils.EstimatedHistogram.mean(EstimatedHistogram.java:203) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.io.sstable.metadata.StatsMetadata.getEstimatedDroppableTombstoneRatio(StatsMetadata.java:98) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.io.sstable.SSTableReader.getEstimatedDroppableTombstoneRatio(SSTableReader.java:1805) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.AbstractCompactionStrategy.worthDroppingTombstones(AbstractCompactionStrategy.java:297) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy.getNextBackgroundSSTables(SizeTieredCompactionStrategy.java:106) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy.getNextBackgroundTask(SizeTieredCompactionStrategy.java:267) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionTask.run(CompactionManager.java:229) ~[apache-cassandra-2.1.0.jar:2.1.0] at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471) ~[na:1.7.0_51] at java.util.concurrent.FutureTask.run(FutureTask.java:262) ~[na:1.7.0_51] at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) ~[na:1.7.0_51] at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) [na:1.7.0_51] at java.lang.Thread.run(Thread.java:744) [na:1.7.0_51] Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. (We have done this on all nodes having the data and all sstables containing the key) {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Do you have any ideas how can we debug this further? Thanks, Andrew -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Comment Edited] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14166563#comment-14166563 ] Andrew S edited comment on CASSANDRA-8089 at 10/10/14 9:03 AM: --- We are also getting the following exception on one of the servers after manually flushing (using nodetool) another table with similar structure. 2014-10-10T07:02:46.402+ host9 ERROR [CompactionExecutor:2] 2014-10-10 09:02:46,396 CassandraDaemon.java:166 - Exception in thread Thread[CompactionExecutor:2,1,main] java.lang.IllegalStateException: Unable to compute ceiling for max when histogram overflowed at org.apache.cassandra.utils.EstimatedHistogram.mean(EstimatedHistogram.java:203) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.io.sstable.metadata.StatsMetadata.getEstimatedDroppableTombstoneRatio(StatsMetadata.java:98) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.io.sstable.SSTableReader.getEstimatedDroppableTombstoneRatio(SSTableReader.java:1805) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.AbstractCompactionStrategy.worthDroppingTombstones(AbstractCompactionStrategy.java:297) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy.getNextBackgroundSSTables(SizeTieredCompactionStrategy.java:106) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy.getNextBackgroundTask(SizeTieredCompactionStrategy.java:267) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionTask.run(CompactionManager.java:229) ~[apache-cassandra-2.1.0.jar:2.1.0] at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471) ~[na:1.7.0_51] at java.util.concurrent.FutureTask.run(FutureTask.java:262) ~[na:1.7.0_51] at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) ~[na:1.7.0_51] at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) [na:1.7.0_51] at java.lang.Thread.run(Thread.java:744) [na:1.7.0_51] was (Author: andrews): We are also getting the following exception on one of the servers after flushing another table with similar structure. 2014-10-10T07:02:46.402+ su8 ERROR [CompactionExecutor:2] 2014-10-10 09:02:46,396 CassandraDaemon.java:166 - Exception in thread Thread[CompactionExecutor:2,1,main] java.lang.IllegalStateException: Unable to compute ceiling for max when histogram overflowed at org.apache.cassandra.utils.EstimatedHistogram.mean(EstimatedHistogram.java:203) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.io.sstable.metadata.StatsMetadata.getEstimatedDroppableTombstoneRatio(StatsMetadata.java:98) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.io.sstable.SSTableReader.getEstimatedDroppableTombstoneRatio(SSTableReader.java:1805) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.AbstractCompactionStrategy.worthDroppingTombstones(AbstractCompactionStrategy.java:297) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy.getNextBackgroundSSTables(SizeTieredCompactionStrategy.java:106) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy.getNextBackgroundTask(SizeTieredCompactionStrategy.java:267) ~[apache-cassandra-2.1.0.jar:2.1.0] at org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionTask.run(CompactionManager.java:229) ~[apache-cassandra-2.1.0.jar:2.1.0] at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471) ~[na:1.7.0_51] at java.util.concurrent.FutureTask.run(FutureTask.java:262) ~[na:1.7.0_51] at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) ~[na:1.7.0_51] at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) [na:1.7.0_51] at java.lang.Thread.run(Thread.java:744) [na:1.7.0_51] Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!],
[jira] [Commented] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14166657#comment-14166657 ] Andrew S commented on CASSANDRA-8089: - Same error message on all nodes and tables mentioned. /opt/apache-cassandra/bin/nodetool --host xxx.xxx.xxx.xxx cfhistograms XXX xxx nodetool: Unable to compute when histogram overflowed It looks from here that this histogram issue may not be related to tombstone warnings we get, as we only get them on one table, but histograms are not working for more tables. Do you think these are two separate issues? Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. (We have done this on all nodes having the data and all sstables containing the key) {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Do you have any ideas how can we debug this further? Thanks, Andrew -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14166698#comment-14166698 ] Andrew S commented on CASSANDRA-8089: - nodetool cfstats Table: table_with_tombstone_issues SSTable count: 8 Space used (live), bytes: 33464296209 Space used (total), bytes: 33464296209 Space used by snapshots (total), bytes: 67910279993 SSTable Compression Ratio: 0.24045657357329253 Memtable cell count: 118368 Memtable data size, bytes: 9541504 Memtable switch count: 543 Local read count: 34239312 Local read latency: 0.092 ms Local write count: 580120 Local write latency: 0.086 ms Pending flushes: 0 Bloom filter false positives: 430017 Bloom filter false ratio: 0.98621 Bloom filter space used, bytes: 23072 Compacted partition minimum bytes: 216 Compacted partition maximum bytes: 30753941057 Compacted partition mean bytes: 6764845 Average live cells per slice (last five minutes): 0.3324931968307692 Average tombstones per slice (last five minutes): 0.6754286242686398 Table: table_unable_to_compute_ceiling_for_max_when_histogram_overflowed SSTable count: 11 Space used (live), bytes: 139688159520 Space used (total), bytes: 139688160740 Space used by snapshots (total), bytes: 53189649117 SSTable Compression Ratio: 0.23556415797023755 Memtable cell count: 439254 Memtable data size, bytes: 0702 Memtable switch count: 817 Local read count: 0 Local read latency: NaN ms Local write count: 2692402 Local write latency: 0.081 ms Pending flushes: 0 Bloom filter false positives: 0 Bloom filter false ratio: 0.0 Bloom filter space used, bytes: 3205776 Compacted partition minimum bytes: 373 Compacted partition maximum bytes: 110196931118 Compacted partition mean bytes: 144738 Average live cells per slice (last five minutes): 0.0 Average tombstones per slice (last five minutes): 0.0 Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. (We have done this on all nodes having the data and all sstables containing the key) {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Do you have any ideas how can we debug this further? Thanks, Andrew -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14166901#comment-14166901 ] Andrew S commented on CASSANDRA-8089: - looks like you have a 110GB partition in that second table, is that expected? Thank you, the largest should be 5GB, this was a partition with an empty key, I have removed it and we no longer get the following exception on flush: java.lang.IllegalStateException: Unable to compute ceiling for max when histogram overflowed Cfhistograms still returns the same error for this and the other tables: nodetool: Unable to compute when histogram overflowed More importantly do you have any ideas on the first issue with tombstones? Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. (We have done this on all nodes having the data and all sstables containing the key) {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Do you have any ideas how can we debug this further? Thanks, Andrew -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Comment Edited] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14166901#comment-14166901 ] Andrew S edited comment on CASSANDRA-8089 at 10/10/14 2:21 PM: --- looks like you have a 110GB partition in that second table, is that expected? Thank you, the largest should be 5GB, this was a partition with an empty key, I have removed it but we still get the exception. java.lang.IllegalStateException: Unable to compute ceiling for max when histogram overflowed Cfhistograms still returns the same error for this and the other tables: nodetool: Unable to compute when histogram overflowed More importantly do you have any ideas on the first issue with tombstones? was (Author: andrews): looks like you have a 110GB partition in that second table, is that expected? Thank you, the largest should be 5GB, this was a partition with an empty key, I have removed it and we no longer get the following exception on flush: java.lang.IllegalStateException: Unable to compute ceiling for max when histogram overflowed Cfhistograms still returns the same error for this and the other tables: nodetool: Unable to compute when histogram overflowed More importantly do you have any ideas on the first issue with tombstones? Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. (We have done this on all nodes having the data and all sstables containing the key) {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Do you have any ideas how can we debug this further? Thanks, Andrew -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Comment Edited] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14166901#comment-14166901 ] Andrew S edited comment on CASSANDRA-8089 at 10/10/14 2:27 PM: --- looks like you have a 110GB partition in that second table, is that expected? Thank you, the largest should be 5GB, this was a partition with an empty key full of ttled values, I have removed the row but we still get the exception. java.lang.IllegalStateException: Unable to compute ceiling for max when histogram overflowed Cfhistograms still returns the same error for this and the other tables: nodetool: Unable to compute when histogram overflowed More importantly do you have any ideas on the first issue with tombstones? was (Author: andrews): looks like you have a 110GB partition in that second table, is that expected? Thank you, the largest should be 5GB, this was a partition with an empty key, I have removed it but we still get the exception. java.lang.IllegalStateException: Unable to compute ceiling for max when histogram overflowed Cfhistograms still returns the same error for this and the other tables: nodetool: Unable to compute when histogram overflowed More importantly do you have any ideas on the first issue with tombstones? Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. (We have done this on all nodes having the data and all sstables containing the key) {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Do you have any ideas how can we debug this further? Thanks, Andrew -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Created] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
Andrew S created CASSANDRA-8089: --- Summary: Invalid tombstone warnings / exceptions Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Casasandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Thanks, Andrew -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Andrew S updated CASSANDRA-8089: Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) was: Casasandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Thanks, Andrew -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Andrew S updated CASSANDRA-8089: Description: Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Thanks, Andrew was: Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Thanks, Andrew Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Casasandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was
[jira] [Updated] (CASSANDRA-8089) Invalid tombstone warnings / exceptions
[ https://issues.apache.org/jira/browse/CASSANDRA-8089?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Andrew S updated CASSANDRA-8089: Description: Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. (We have done this on all nodes having the data and all sstables containing the key) {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Thanks, Andrew was: Hey, We are having a strange issue with tombstone warnings which look like this: {code} WARN 12:28:42 Read 129 live and 4113 tombstoned cells in XXX.xxx (see tombstone_warn_threshold). 500 columns was requested, slices=[31660a4e-4f94-11e4-ac1d-53f244a29642-0a8073aa-4f9f-11e4-87c7-5b3e253389d8:!], delInfo={deletedAt=-9223372036854775808, localDeletion=2147483647} {code} What is strange is that the row requested should not contain any tombstones as we never delete data from that row. (We do delete data from other row in the same column family) To debug the issue we have dumped the data for this row using sstable2json and the result does not contain any tombstones. {code} ./sstable2json /var/lib/cassandra/data/XXX/xxx/XXX-xxx-ka-81524-Data.db -k xxx {code} We are getting the warnings after issuing a simple query: {code} select count(*) from xxx where key = 'x' and aggregate='x'; {code} There are only ~500 cells but it issues a warning about scanning 1700 tombstones. We are very worried about this because for some of the queries we are hitting TombstoneOverwhelmingException for no obvious reason. Here is the table definiion: {code} CREATE TABLE Xxxx.xxx ( key text, aggregate text, t timeuuid, . {date fields } PRIMARY KEY ((key, aggregate), t) ) WITH CLUSTERING ORDER BY (t ASC) AND bloom_filter_fp_chance = 0.01 AND caching = '{keys:ALL, rows_per_partition:NONE}' AND comment = 'we love cassandra' AND compaction = {'min_threshold': '6', 'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32'} AND compression = {'sstable_compression': 'org.apache.cassandra.io.compress.SnappyCompressor'} AND dclocal_read_repair_chance = 0.0 AND default_time_to_live = 0 AND gc_grace_seconds = 3600 AND max_index_interval = 2048 AND memtable_flush_period_in_ms = 0 AND min_index_interval = 128 AND read_repair_chance = 0.1 AND speculative_retry = '99.0PERCENTILE'; {code} Thanks, Andrew Invalid tombstone warnings / exceptions --- Key: CASSANDRA-8089 URL: https://issues.apache.org/jira/browse/CASSANDRA-8089 Project: Cassandra Issue Type: Bug Components: Core Environment: Cassandra 2.1.0 Debian 7.6, 3.2.0-4-amd64 GNU/Linux java version 1.7.0_51 Java(TM) SE Runtime Environment (build 1.7.0_51-b13) Java HotSpot(TM) 64-Bit Server VM (build 24.51-b03, mixed mode) Reporter: Andrew S Hey, We are having a strange issue with tombstone warnings