Skip to content

Commit

Permalink
STAR-543 Port remaining guardrails and sync with cndb (#180)
Browse files Browse the repository at this point in the history 
Co-authored-by: Zhao Yang <[email protected]>
Co-authored-by: Paulo Ricardo Motta Gomes <[email protected]>
Co-authored-by: Eduard Tudenhöfner <[email protected]>
Co-authored-by: Andrés de la Peña <[email protected]>
Co-authored-by: Benjamin Lerer <[email protected]>
(cherry picked from commit a83e0e8)
(cherry picked from commit b90b433)
(cherry picked from commit 36600df)
(cherry picked from commit abc39d6)
(cherry picked from commit 7a263dc)
  • Loading branch information
@djatnieks @jacek-lewandowski
djatnieks authored and jacek-lewandowski committed Oct 18, 2022
1 parent d3ce6ca commit 159a1fa
Show file tree
Hide file tree
Showing 47 changed files with 1,297 additions and 876 deletions.
157 changes: 78 additions & 79 deletions conf/cassandra.yaml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -1266,18 +1266,6 @@ transparent_data_encryption_options:
# SAFETY THRESHOLDS #
#####################

# When executing a scan, within or across a partition, we need to keep the
# tombstones seen in memory so we can return them to the coordinator, which
# will use them to make sure other replicas also know about the deleted rows.
# With workloads that generate a lot of tombstones, this can cause performance
# problems and even exaust the server heap.
# (http://www.datastax.com/dev/blog/cassandra-anti-patterns-queues-and-queue-like-datasets)
# Adjust the thresholds here if you understand the dangers and want to
# scan more tombstones anyway. These thresholds may also be adjusted at runtime
# using the StorageService mbean.
tombstone_warn_threshold: 1000
tombstone_failure_threshold: 100000

# Filtering and secondary index queries at read consistency levels above ONE/LOCAL_ONE use a
# mechanism called replica filtering protection to ensure that results from stale replicas do
# not violate consistency. (See CASSANDRA-8272 and CASSANDRA-15907 for more details.) This
Expand All @@ -1298,19 +1286,6 @@ replica_filtering_protection:
cached_rows_warn_threshold: 2000
cached_rows_fail_threshold: 32000

# Log WARN on any multiple-partition batch size exceeding this value. 5kb per batch by default.
# Caution should be taken on increasing the size of this threshold as it can lead to node instability.
batch_size_warn_threshold_in_kb: 5

# Fail any multiple-partition batch exceeding this value. 50kb (10x warn threshold) by default.
batch_size_fail_threshold_in_kb: 50

# Log WARN on any batches not of type LOGGED than span across more partitions than this limit
unlogged_batch_across_partitions_warn_threshold: 10

# Log a warning when compacting partitions larger than this value
compaction_large_partition_warning_threshold_mb: 100

# GC Pauses greater than 200 ms will be logged at INFO level
# This threshold can be adjusted to minimize logging if necessary
# gc_log_threshold_in_ms: 200
Expand Down Expand Up @@ -1426,13 +1401,17 @@ enable_transient_replication: false
# 'ALTER ... DROP COMPACT STORAGE' is considered experimental and is not recommended for production use.
enable_drop_compact_storage: false

# Apply database-as-a-service defaults.
# Emulates DataStax Constellation database-as-a-service defaults.
#
# When enabled, some defaults are modified to match those used by DataStax Constellation (DataStax cloud data
# platform). This includes (but is not limited to) stricter guardrails defaults.
#
# When enabled, some guardrails defaults are modified to values that are appropriate for cloud environments.
# This includes (but is not limited to) stricter guardrails defaults.
# This can be used as an convenience to develop and test applications meant to run on DataStax Constellation.
#
# This can be used as an convenience to develop and test applications meant to run in a cloud environment.
# apply_dbaas_defaults: false
# Warning: when enabled, the updated defaults reflect those of DataStax Constellation _at the time_ of the currently
# used DSE release. This is a best-effort emulation of said defaults. Further, all nodes must use the same
# config value.
# emulate_dbaas_defaults: false

# Guardrails settings.
# guardrails:
Expand All @@ -1446,93 +1425,113 @@ enable_drop_compact_storage: false
# scan more tombstones anyway. These thresholds may also be adjusted at runtime
# using the StorageService mbean.
#
# Default: tombstone_warn_threshold is 1000, may differ if apply_dbaas_defaults is enabled
# Default: tombstone_failure_threshold is 100000, may differ if apply_dbaas_defaults is enabled
# Default tombstone_warn_threshold is 1000, may differ if emulate_dbaas_defaults is enabled
# Default tombstone_failure_threshold is 100000, may differ if emulate_dbaas_defaults is enabled
# tombstone_warn_threshold: 1000
# tombstone_failure_threshold: 100000

# Failure threshold to prevent writing large a column value into Cassandra.
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# Log a warning when compacting partitions larger than this value.
# Default value is 100mb, may differ if emulate_dbaas_defaults is enabled
# partition_size_warn_threshold_in_mb: 100

# Log WARN on any multiple-partition batch size that exceeds this value. 64kb per batch by default.
# Use caution when increasing the size of this threshold as it can lead to node instability.
# Default value is 64kb, may differ if emulate_dbaas_defaults is enabled
# batch_size_warn_threshold_in_kb: 64

# Fail any multiple-partition batch that exceeds this value. The calculated default is 640kb (10x warn threshold).
# Default value is 640kb, may differ if emulate_dbaas_defaults is enabled
# batch_size_fail_threshold_in_kb: 640

# Log WARN on any batches not of type LOGGED than span across more partitions than this limit.
# Default value is 10, may differ if emulate_dbaas_defaults is enabled
# unlogged_batch_across_partitions_warn_threshold: 10

# Failure threshold to prevent writing large column value into Cassandra.
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# column_value_size_failure_threshold_in_kb: -1

# Failure threshold to prevent creating more columns per table than threshold.
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# columns_per_table_failure_threshold: -1

# Failure threshold to prevent creating more secondary indexes per table than threshold (does not apply to CUSTOM INDEX StorageAttachedIndex)
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# Failure threshold to prevent creating more fields in user-defined-type than threshold.
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# fields_per_udt_failure_threshold: -1

# Warning threshold to warn when encountering larger size of collection data than threshold.
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# collection_size_warn_threshold_in_kb: -1

# Warning threshold to warn when encountering more elements in collection than threshold.
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# items_per_collection_warn_threshold: -1

# Whether read-before-write operation is allowed, eg. setting list element by index, removing list element
# by index. Note: LWT is always allowed.
# Default true to allow read before write operation, may differ if emulate_dbaas_defaults is enabled
# read_before_write_list_operations_enabled: true

# Failure threshold to prevent creating more secondary index per table than threshold (does not apply to CUSTOM INDEX StorageAttachedIndex)
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# secondary_index_per_table_failure_threshold: -1

# Failure threshold for number of StorageAttachedIndex per table (only applies to CUSTOM INDEX StorageAttachedIndex)
# Default is 10 (same when apply_dbaas_defaults is enabled)
# Default is 10 (same when emulate_dbaas_defaults is enabled)
# sai_indexes_per_table_failure_threshold: 10
#
# Failure threshold for total number of StorageAttachedIndex across all keyspaces (only applies to CUSTOM INDEX StorageAttachedIndex)
# Default is 10 (same when apply_dbaas_defaults is enabled)
# Default is 10 (same when emulate_dbaas_defaults is enabled)
# sai_indexes_total_failure_threshold: 100

# Failure threshold to prevent creating more materialized views per table than threshold.
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# materialized_view_per_table_failure_threshold: -1

# Warn threshold to warn creating more tables than threshold.
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# tables_warn_threshold: -1

# Failure threshold to prevent creating more tables than threshold.
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# tables_failure_threshold: -1

# Prevents creating tables with provided configurations.
# Default: all properties are allowed, may differ if apply_dbaas_defaults is enabled
# Preventing creating tables with provided configurations.
# Default all properties are allowed, may differ if emulate_dbaas_defaults is enabled
# table_properties_disallowed:

# Whether to allow user-provided timestamps in write requests
# Default: true to allow user-provided timestamps, may differ if apply_dbaas_defaults is enabled
# Whether to allow user-provided timestamp in write request
# Default true to allow user-provided timestamp, may differ if emulate_dbaas_defaults is enabled
# user_timestamps_enabled: true

# Preventing a query with provided consistency levels
# Default: all consistency levels are allowed.
# Preventing query with provided consistency levels
# Default all consistency levels are allowed.
# write_consistency_levels_disallowed:

# Log a warning when compacting partitions larger than this value.
# Default: 100mb, may differ if apply_dbaas_defaults is enabled
# partition_size_warn_threshold_in_mb: 100
# Failure threshold to prevent providing larger paging by bytes than threshold, also served as a hard paging limit
# when paging by rows is used.
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# page_size_failure_threshold_in_kb: -1

# Failure threshold to prevent IN query creating size of cartesian product exceeding threshold, eg.
# "a in (1,2,...10) and b in (1,2...10)" results in cartesian product of 100.
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# in_select_cartesian_product_failure_threshold: -1

# Failure threshold to prevent IN query containing more partition keys than threshold
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# partition_keys_in_select_failure_threshold: -1

# Warning threshold to warn when local disk usage exceeding threshold. Valid values: (1, 100]
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# disk_usage_percentage_warn_threshold: -1

# Failure threshold to reject write requests if replica disk usage exceeding threshold. Valid values: (1, 100]
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# Default -1 to disable, may differ if emulate_dbaas_defaults is enabled
# disk_usage_percentage_failure_threshold: -1

# Failure threshold to prevent IN query creating size of cartesian product exceeding threshold, eg.
# "a IN (1,2,...10) AND b IN (1,2...10)" results in cartesian product of 100.
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# in_select_cartesian_product_failure_threshold: -1

# Whether to allow user-provided timestamps in write request (USING TIMESTAMP ...)
# Default: true to allow user-provided timestamp, may differ if apply_dbaas_defaults is enabled
# user_timestamps_enabled: true

# Whether read-before-write operation is allowed on lists, eg. setting list element by index, removing list element
# by index. Note: LWT is always allowed.
# Default: true to allow read before write operation on lists, may differ if apply_dbaas_defaults is enabled
# read_before_write_list_operations_enabled: true

# Failure threshold to prevent creating more fields in user-defined-type than threshold.
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# fields_per_udt_failure_threshold: -1

# Warning threshold to warn when encountering larger size of collection data than threshold.
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# collection_size_warn_threshold_in_kb: -1

# Warning threshold to warn when encountering more elements in collection than threshold.
# Default: -1 to disable, may differ if apply_dbaas_defaults is enabled
# items_per_collection_warn_threshold: -1
# Allows configuring max disk size of data directories when calculating thresholds for disk_usage_percentage_warn_threshold
# and disk_usage_percentage_failure_threshold. Valid values: (1, max available disk size of all data directories]
# Default -1 to disable and use the physically available disk size of data directories during calculations.
# may differ if emulate_dbaas_defaults is enabled
# disk_usage_max_disk_size_in_gb: -1
39 changes: 32 additions & 7 deletions src/java/org/apache/cassandra/config/Config.java
View file
Original file line number Diff line number Diff line change
Expand Up @@ -220,12 +220,28 @@ public class Config
/* if the size of columns or super-columns are more than this, indexing will kick in */
public int column_index_size_in_kb = 64;
public volatile int column_index_cache_size_in_kb = 2;
public volatile int batch_size_warn_threshold_in_kb = 5;
public volatile int batch_size_fail_threshold_in_kb = 50;
public Integer unlogged_batch_across_partitions_warn_threshold = 10;
/**
* @deprecated Migrated to 'guardrails.batch_size_warn_threshold_in_kb'
*/
@Deprecated
public int batch_size_warn_threshold_in_kb = 0;
/**
* @deprecated Migrated to 'guardrails.batch_size_fail_threshold_in_kb'
*/
@Deprecated
public int batch_size_fail_threshold_in_kb = 0;
/**
* @deprecated Migrated to 'guardrails.unlogged_batch_across_partitions_warn_threshold'
*/
@Deprecated
public Integer unlogged_batch_across_partitions_warn_threshold = 0;
public volatile Integer concurrent_compactors;
public volatile int compaction_throughput_mb_per_sec = 64;
public volatile int compaction_large_partition_warning_threshold_mb = 100;
/**
* @deprecated Migrated to 'guardrails.compaction_large_partition_warning_threshold_mb'
*/
@Deprecated
public int compaction_large_partition_warning_threshold_mb = 0;
public int min_free_space_per_drive_in_mb = 50;

public volatile int concurrent_materialized_view_builders = 1;
Expand Down Expand Up @@ -352,8 +368,16 @@ public class Config

public MemtableAllocationType memtable_allocation_type = MemtableAllocationType.offheap_objects;

public volatile int tombstone_warn_threshold = 1000;
public volatile int tombstone_failure_threshold = 100000;
/**
* @deprecated Migrated to 'guardrails.tombstone_warn_threshold'
*/
@Deprecated
public int tombstone_warn_threshold = 0;
/**
* @deprecated Migrated to 'guardrails.tombstone_failure_threshold'
*/
@Deprecated
public int tombstone_failure_threshold = 0;

public final ReplicaFilteringProtectionOptions replica_filtering_protection = new ReplicaFilteringProtectionOptions();

Expand Down Expand Up @@ -525,7 +549,8 @@ public class Config
*/
public volatile int validation_preview_purge_head_start_in_sec = 60 * 60;

public boolean apply_dbaas_defaults = false;
public boolean emulate_dbaas_defaults = false;

public GuardrailsConfig guardrails = new GuardrailsConfig();

/**
Expand Down
Loading

0 comments on commit 159a1fa

Please sign in to comment.