This is an automated email from the ASF dual-hosted git repository. marcuse pushed a commit to branch trunk in repository https://gitbox.apache.org/repos/asf/cassandra.git
commit 5eea8bf7b59aa0f2f41383b27aaa19fbe3a6ffe1 Merge: d5e5c45 df4180f Author: Marcus Eriksson <[email protected]> AuthorDate: Fri Nov 29 10:08:43 2019 +0100 Merge branch 'cassandra-3.11' into trunk CHANGES.txt | 1 + src/java/org/apache/cassandra/db/Directories.java | 42 ++++-- .../org/apache/cassandra/db/DiskBoundaries.java | 11 +- .../apache/cassandra/db/DiskBoundaryManager.java | 6 +- .../cassandra/db/compaction/CompactionManager.java | 4 +- .../compaction/writers/CompactionAwareWriter.java | 15 ++- .../org/apache/cassandra/io/util/FileUtils.java | 4 +- .../org/apache/cassandra/db/DirectoriesTest.java | 150 +++++++++++++++++++++ .../compaction/CompactionStrategyManagerTest.java | 4 +- .../apache/cassandra/io/util/FileUtilsTest.java | 11 ++ 10 files changed, 214 insertions(+), 34 deletions(-) diff --cc CHANGES.txt index 3d1991a,3ede6a4..cea03ea --- a/CHANGES.txt +++ b/CHANGES.txt @@@ -1,413 -1,11 +1,414 @@@ -3.11.6 +4.0-alpha3 + * Align record header of FQL and audit binary log (CASSANDRA-15076) + * Shuffle forwarding replica for messages to non-local DC (CASSANDRA-15318) + * Optimise native protocol ASCII string encoding (CASSANDRA-15410) + * Make sure all exceptions are propagated in DebuggableThreadPoolExecutor (CASSANDRA-15332) + * Make it possible to resize concurrent read / write thread pools at runtime (CASSANDRA-15277) + * Close channels on error (CASSANDRA-15407) +Merged from 3.11: * Fix SELECT JSON formatting for the "duration" type (CASSANDRA-15075) - * Fix LegacyLayout to have same behavior as 2.x when handling unknown column names (CASSANDRA-15081) +Merged from 2.2: + * In-JVM DTest: Set correct internode message version for upgrade test (CASSANDRA-15371) + + +4.0-alpha2 + * Fix SASI non-literal string comparisons (range operators) (CASSANDRA-15169) + * Upgrade Guava to 27, and to java-driver 3.6.0 (from 3.4.0-SNAPSHOT) (CASSANDRA-14655) + * Extract an AbstractCompactionController to allow for custom implementations (CASSANDRA-15286) + * Move chronicle-core version from snapshot to stable, and include carrotsearch in generated pom.xml (CASSANDRA-15321) + * Untangle RepairMessage sub-hierarchy of messages, use new messaging (more) correctly (CASSANDRA-15163) + * Add `allocate_tokens_for_local_replication_factor` option for token allocation (CASSANDRA-15260) + * Add Alibaba Cloud Platform snitch (CASSANDRA-15092) Merged from 3.0: + * Fix various data directory prefix matching issues (CASSANDRA-13974) * Minimize clustering values in metadata collector (CASSANDRA-15400) - * Avoid over-trimming of results in mixed mode clusters (CASSANDRA-15405) - * validate value sizes in LegacyLayout (CASSANDRA-15373) + * Make sure index summary redistribution does not start when compactions are paused (CASSANDRA-15265) + * Add ability to cap max negotiable protocol version (CASSANDRA-15193) + * Gossip tokens on startup if available (CASSANDRA-15335) + * Fix resource leak in CompressedSequentialWriter (CASSANDRA-15340) + * Fix bad merge that reverted CASSANDRA-14993 (CASSANDRA-15289) + * Add support for network topology and query tracing for inJVM dtest (CASSANDRA-15319) + + +4.0-alpha1 + * Inaccurate exception message with nodetool snapshot (CASSANDRA-15287) + * Fix InternodeOutboundMetrics overloaded bytes/count mixup (CASSANDRA-15186) + * Enhance & reenable RepairTest with compression=off and compression=on (CASSANDRA-15272) + * Improve readability of Table metrics Virtual tables units (CASSANDRA-15194) + * Fix error with non-existent table for nodetool tablehistograms (CASSANDRA-14410) + * Avoid result truncation in decimal operations (CASSANDRA-15232) + * Catch non-IOException in FileUtils.close to make sure that all resources are closed (CASSANDRA-15225) + * Align load column in nodetool status output (CASSANDRA-14787) + * CassandraNetworkAuthorizer uses cached roles info (CASSANDRA-15089) + * Introduce optional timeouts for idle client sessions (CASSANDRA-11097) + * Fix AlterTableStatement dropped type validation order (CASSANDRA-15203) + * Update Netty dependencies to latest, clean up SocketFactory (CASSANDRA-15195) + * Native Transport - Apply noSpamLogger to ConnectionLimitHandler (CASSANDRA-15167) + * Reduce heap pressure during compactions (CASSANDRA-14654) + * Support building Cassandra with JDK 11 (CASSANDRA-15108) + * Use quilt to patch cassandra.in.sh in Debian packaging (CASSANDRA-14710) + * Take sstable references before calculating approximate key count (CASSANDRA-14647) + * Restore snapshotting of system keyspaces on version change (CASSANDRA-14412) + * Fix AbstractBTreePartition locking in java 11 (CASSANDRA-14607) + * SimpleClient should pass connection properties as options (CASSANDRA-15056) + * Set repaired data tracking flag on range reads if enabled (CASSANDRA-15019) + * Calculate pending ranges for BOOTSTRAP_REPLACE correctly (CASSANDRA-14802) + * Make TableCQLHelper reuse the single quote pattern (CASSANDRA-15033) + * Add Zstd compressor (CASSANDRA-14482) + * Fix IR prepare anti-compaction race (CASSANDRA-15027) + * Fix SimpleStrategy option validation (CASSANDRA-15007) + * Don't try to cancel 2i compactions when starting anticompaction (CASSANDRA-15024) + * Avoid NPE in RepairRunnable.recordFailure (CASSANDRA-15025) + * SSL Cert Hot Reloading should check for sanity of the new keystore/truststore before loading it (CASSANDRA-14991) + * Avoid leaking threads when failing anticompactions and rate limit anticompactions (CASSANDRA-15002) + * Validate token() arguments early instead of throwing NPE at execution (CASSANDRA-14989) + * Add a new tool to dump audit logs (CASSANDRA-14885) + * Fix generating javadoc with Java11 (CASSANDRA-14988) + * Only cancel conflicting compactions when starting anticompactions and sub range compactions (CASSANDRA-14935) + * Use a stub IndexRegistry for non-daemon use cases (CASSANDRA-14938) + * Don't enable client transports when bootstrap is pending (CASSANDRA-14525) + * Make antiCompactGroup throw exception on error and anticompaction non cancellable + again (CASSANDRA-14936) + * Catch empty/invalid bounds in SelectStatement (CASSANDRA-14849) + * Auto-expand replication_factor for NetworkTopologyStrategy (CASSANDRA-14303) + * Transient Replication: support EACH_QUORUM (CASSANDRA-14727) + * BufferPool: allocating thread for new chunks should acquire directly (CASSANDRA-14832) + * Send correct messaging version in internode messaging handshake's third message (CASSANDRA-14896) + * Make Read and Write Latency columns consistent for proxyhistograms and tablehistograms (CASSANDRA-11939) + * Make protocol checksum type option case insensitive (CASSANDRA-14716) + * Forbid re-adding static columns as regular and vice versa (CASSANDRA-14913) + * Audit log allows system keyspaces to be audited via configuration options (CASSANDRA-14498) + * Lower default chunk_length_in_kb from 64kb to 16kb (CASSANDRA-13241) + * Startup checker should wait for count rather than percentage (CASSANDRA-14297) + * Fix incorrect sorting of replicas in SimpleStrategy.calculateNaturalReplicas (CASSANDRA-14862) + * Partitioned outbound internode TCP connections can occur when nodes restart (CASSANDRA-14358) + * Don't write to system_distributed.repair_history, system_traces.sessions, system_traces.events in mixed version 3.X/4.0 clusters (CASSANDRA-14841) + * Avoid running query to self through messaging service (CASSANDRA-14807) + * Allow using custom script for chronicle queue BinLog archival (CASSANDRA-14373) + * Transient->Full range movements mishandle consistency level upgrade (CASSANDRA-14759) + * ReplicaCollection follow-up (CASSANDRA-14726) + * Transient node receives full data requests (CASSANDRA-14762) + * Enable snapshot artifacts publish (CASSANDRA-12704) + * Introduce RangesAtEndpoint.unwrap to simplify StreamSession.addTransferRanges (CASSANDRA-14770) + * LOCAL_QUORUM may speculate to non-local nodes, resulting in Timeout instead of Unavailable (CASSANDRA-14735) + * Avoid creating empty compaction tasks after truncate (CASSANDRA-14780) + * Fail incremental repair prepare phase if it encounters sstables from un-finalized sessions (CASSANDRA-14763) + * Add a check for receiving digest response from transient node (CASSANDRA-14750) + * Fail query on transient replica if coordinator only expects full data (CASSANDRA-14704) + * Remove mentions of transient replication from repair path (CASSANDRA-14698) + * Fix handleRepairStatusChangedNotification to remove first then add (CASSANDRA-14720) + * Allow transient node to serve as a repair coordinator (CASSANDRA-14693) + * DecayingEstimatedHistogramReservoir.EstimatedHistogramReservoirSnapshot returns wrong value for size() and incorrectly calculates count (CASSANDRA-14696) + * AbstractReplicaCollection equals and hash code should throw due to conflict between order sensitive/insensitive uses (CASSANDRA-14700) + * Detect inconsistencies in repaired data on the read path (CASSANDRA-14145) + * Add checksumming to the native protocol (CASSANDRA-13304) + * Make AuthCache more easily extendable (CASSANDRA-14662) + * Extend RolesCache to include detailed role info (CASSANDRA-14497) + * Add fqltool compare (CASSANDRA-14619) + * Add fqltool replay (CASSANDRA-14618) + * Log keyspace in full query log (CASSANDRA-14656) + * Transient Replication and Cheap Quorums (CASSANDRA-14404) + * Log server-generated timestamp and nowInSeconds used by queries in FQL (CASSANDRA-14675) + * Add diagnostic events for read repairs (CASSANDRA-14668) + * Use consistent nowInSeconds and timestamps values within a request (CASSANDRA-14671) + * Add sampler for query time and expose with nodetool (CASSANDRA-14436) + * Clean up Message.Request implementations (CASSANDRA-14677) + * Disable old native protocol versions on demand (CASANDRA-14659) + * Allow specifying now-in-seconds in native protocol (CASSANDRA-14664) + * Improve BTree build performance by avoiding data copy (CASSANDRA-9989) + * Make monotonic read / read repair configurable (CASSANDRA-14635) + * Refactor CompactionStrategyManager (CASSANDRA-14621) + * Flush netty client messages immediately by default (CASSANDRA-13651) + * Improve read repair blocking behavior (CASSANDRA-10726) + * Add a virtual table to expose settings (CASSANDRA-14573) + * Fix up chunk cache handling of metrics (CASSANDRA-14628) + * Extend IAuthenticator to accept peer SSL certificates (CASSANDRA-14652) + * Incomplete handling of exceptions when decoding incoming messages (CASSANDRA-14574) + * Add diagnostic events for user audit logging (CASSANDRA-13668) + * Allow retrieving diagnostic events via JMX (CASSANDRA-14435) + * Add base classes for diagnostic events (CASSANDRA-13457) + * Clear view system metadata when dropping keyspace (CASSANDRA-14646) + * Allocate ReentrantLock on-demand in java11 AtomicBTreePartitionerBase (CASSANDRA-14637) + * Make all existing virtual tables use LocalPartitioner (CASSANDRA-14640) + * Revert 4.0 GC alg back to CMS (CASANDRA-14636) + * Remove hardcoded java11 jvm args in idea workspace files (CASSANDRA-14627) + * Update netty to 4.1.128 (CASSANDRA-14633) + * Add a virtual table to expose thread pools (CASSANDRA-14523) + * Add a virtual table to expose caches (CASSANDRA-14538, CASSANDRA-14626) + * Fix toDate function for timestamp arguments (CASSANDRA-14502) + * Revert running dtests by default in circleci (CASSANDRA-14614) + * Stream entire SSTables when possible (CASSANDRA-14556) + * Cell reconciliation should not depend on nowInSec (CASSANDRA-14592) + * Add experimental support for Java 11 (CASSANDRA-9608) + * Make PeriodicCommitLogService.blockWhenSyncLagsNanos configurable (CASSANDRA-14580) + * Improve logging in MessageInHandler's constructor (CASSANDRA-14576) + * Set broadcast address in internode messaging handshake (CASSANDRA-14579) + * Wait for schema agreement prior to building MVs (CASSANDRA-14571) + * Make all DDL statements idempotent and not dependent on global state (CASSANDRA-13426) + * Bump the hints messaging version to match the current one (CASSANDRA-14536) + * OffsetAwareConfigurationLoader doesn't set ssl storage port causing bind errors in CircleCI (CASSANDRA-14546) + * Report why native_transport_port fails to bind (CASSANDRA-14544) + * Optimize internode messaging protocol (CASSANDRA-14485) + * Internode messaging handshake sends wrong messaging version number (CASSANDRA-14540) + * Add a virtual table to expose active client connections (CASSANDRA-14458) + * Clean up and refactor client metrics (CASSANDRA-14524) + * Nodetool import row cache invalidation races with adding sstables to tracker (CASSANDRA-14529) + * Fix assertions in LWTs after TableMetadata was made immutable (CASSANDRA-14356) + * Abort compactions quicker (CASSANDRA-14397) + * Support light-weight transactions in cassandra-stress (CASSANDRA-13529) + * Make AsyncOneResponse use the correct timeout (CASSANDRA-14509) + * Add option to sanity check tombstones on reads/compactions (CASSANDRA-14467) + * Add a virtual table to expose all running sstable tasks (CASSANDRA-14457) + * Let nodetool import take a list of directories (CASSANDRA-14442) + * Avoid unneeded memory allocations / cpu for disabled log levels (CASSANDRA-14488) + * Implement virtual keyspace interface (CASSANDRA-7622) + * nodetool import cleanup and improvements (CASSANDRA-14417) + * Bump jackson version to >= 2.9.5 (CASSANDRA-14427) + * Allow nodetool toppartitions without specifying table (CASSANDRA-14360) + * Audit logging for database activity (CASSANDRA-12151) + * Clean up build artifacts in docs container (CASSANDRA-14432) + * Minor network authz improvements (Cassandra-14413) + * Automatic sstable upgrades (CASSANDRA-14197) + * Replace deprecated junit.framework.Assert usages with org.junit.Assert (CASSANDRA-14431) + * Cassandra-stress throws NPE if insert section isn't specified in user profile (CASSSANDRA-14426) + * List clients by protocol versions `nodetool clientstats --by-protocol` (CASSANDRA-14335) + * Improve LatencyMetrics performance by reducing write path processing (CASSANDRA-14281) + * Add network authz (CASSANDRA-13985) + * Use the correct IP/Port for Streaming when localAddress is left unbound (CASSANDRA-14389) + * nodetool listsnapshots is missing local system keyspace snapshots (CASSANDRA-14381) + * Remove StreamCoordinator.streamExecutor thread pool (CASSANDRA-14402) + * Rename nodetool --with-port to --print-port to disambiguate from --port (CASSANDRA-14392) + * Client TOPOLOGY_CHANGE messages have wrong port. (CASSANDRA-14398) + * Add ability to load new SSTables from a separate directory (CASSANDRA-6719) + * Eliminate background repair and probablistic read_repair_chance table options + (CASSANDRA-13910) + * Bind to correct local address in 4.0 streaming (CASSANDRA-14362) + * Use standard Amazon naming for datacenter and rack in Ec2Snitch (CASSANDRA-7839) + * Fix junit failure for SSTableReaderTest (CASSANDRA-14387) + * Abstract write path for pluggable storage (CASSANDRA-14118) + * nodetool describecluster should be more informative (CASSANDRA-13853) + * Compaction performance improvements (CASSANDRA-14261) + * Refactor Pair usage to avoid boxing ints/longs (CASSANDRA-14260) + * Add options to nodetool tablestats to sort and limit output (CASSANDRA-13889) + * Rename internals to reflect CQL vocabulary (CASSANDRA-14354) + * Add support for hybrid MIN(), MAX() speculative retry policies + (CASSANDRA-14293, CASSANDRA-14338, CASSANDRA-14352) + * Fix some regressions caused by 14058 (CASSANDRA-14353) + * Abstract repair for pluggable storage (CASSANDRA-14116) + * Add meaningful toString() impls (CASSANDRA-13653) + * Add sstableloader option to accept target keyspace name (CASSANDRA-13884) + * Move processing of EchoMessage response to gossip stage (CASSANDRA-13713) + * Add coordinator write metric per CF (CASSANDRA-14232) + * Correct and clarify SSLFactory.getSslContext method and call sites (CASSANDRA-14314) + * Handle static and partition deletion properly on ThrottledUnfilteredIterator (CASSANDRA-14315) + * NodeTool clientstats should show SSL Cipher (CASSANDRA-14322) + * Add ability to specify driver name and version (CASSANDRA-14275) + * Abstract streaming for pluggable storage (CASSANDRA-14115) + * Forced incremental repairs should promote sstables if they can (CASSANDRA-14294) + * Use Murmur3 for validation compactions (CASSANDRA-14002) + * Comma at the end of the seed list is interpretated as localhost (CASSANDRA-14285) + * Refactor read executor and response resolver, abstract read repair (CASSANDRA-14058) + * Add optional startup delay to wait until peers are ready (CASSANDRA-13993) + * Add a few options to nodetool verify (CASSANDRA-14201) + * CVE-2017-5929 Security vulnerability and redefine default log rotation policy (CASSANDRA-14183) + * Use JVM default SSL validation algorithm instead of custom default (CASSANDRA-13259) + * Better document in code InetAddressAndPort usage post 7544, incorporate port into UUIDGen node (CASSANDRA-14226) + * Fix sstablemetadata date string for minLocalDeletionTime (CASSANDRA-14132) + * Make it possible to change neverPurgeTombstones during runtime (CASSANDRA-14214) + * Remove GossipDigestSynVerbHandler#doSort() (CASSANDRA-14174) + * Add nodetool clientlist (CASSANDRA-13665) + * Revert ProtocolVersion changes from CASSANDRA-7544 (CASSANDRA-14211) + * Non-disruptive seed node list reload (CASSANDRA-14190) + * Nodetool tablehistograms to print statics for all the tables (CASSANDRA-14185) + * Migrate dtests to use pytest and python3 (CASSANDRA-14134) + * Allow storage port to be configurable per node (CASSANDRA-7544) + * Make sub-range selection for non-frozen collections return null instead of empty (CASSANDRA-14182) + * BloomFilter serialization format should not change byte ordering (CASSANDRA-9067) + * Remove unused on-heap BloomFilter implementation (CASSANDRA-14152) + * Delete temp test files on exit (CASSANDRA-14153) + * Make PartitionUpdate and Mutation immutable (CASSANDRA-13867) + * Fix CommitLogReplayer exception for CDC data (CASSANDRA-14066) + * Fix cassandra-stress startup failure (CASSANDRA-14106) + * Remove initialDirectories from CFS (CASSANDRA-13928) + * Fix trivial log format error (CASSANDRA-14015) + * Allow sstabledump to do a json object per partition (CASSANDRA-13848) + * Add option to optimise merkle tree comparison across replicas (CASSANDRA-3200) + * Remove unused and deprecated methods from AbstractCompactionStrategy (CASSANDRA-14081) + * Fix Distribution.average in cassandra-stress (CASSANDRA-14090) + * Support a means of logging all queries as they were invoked (CASSANDRA-13983) + * Presize collections (CASSANDRA-13760) + * Add GroupCommitLogService (CASSANDRA-13530) + * Parallelize initial materialized view build (CASSANDRA-12245) + * Fix flaky SecondaryIndexManagerTest.assert[Not]MarkedAsBuilt (CASSANDRA-13965) + * Make LWTs send resultset metadata on every request (CASSANDRA-13992) + * Fix flaky indexWithFailedInitializationIsNotQueryableAfterPartialRebuild (CASSANDRA-13963) + * Introduce leaf-only iterator (CASSANDRA-9988) + * Upgrade Guava to 23.3 and Airline to 0.8 (CASSANDRA-13997) + * Allow only one concurrent call to StatusLogger (CASSANDRA-12182) + * Refactoring to specialised functional interfaces (CASSANDRA-13982) + * Speculative retry should allow more friendly params (CASSANDRA-13876) + * Throw exception if we send/receive repair messages to incompatible nodes (CASSANDRA-13944) + * Replace usages of MessageDigest with Guava's Hasher (CASSANDRA-13291) + * Add nodetool cmd to print hinted handoff window (CASSANDRA-13728) + * Fix some alerts raised by static analysis (CASSANDRA-13799) + * Checksum sstable metadata (CASSANDRA-13321, CASSANDRA-13593) + * Add result set metadata to prepared statement MD5 hash calculation (CASSANDRA-10786) + * Refactor GcCompactionTest to avoid boxing (CASSANDRA-13941) + * Expose recent histograms in JmxHistograms (CASSANDRA-13642) + * Fix buffer length comparison when decompressing in netty-based streaming (CASSANDRA-13899) + * Properly close StreamCompressionInputStream to release any ByteBuf (CASSANDRA-13906) + * Add SERIAL and LOCAL_SERIAL support for cassandra-stress (CASSANDRA-13925) + * LCS needlessly checks for L0 STCS candidates multiple times (CASSANDRA-12961) + * Correctly close netty channels when a stream session ends (CASSANDRA-13905) + * Update lz4 to 1.4.0 (CASSANDRA-13741) + * Optimize Paxos prepare and propose stage for local requests (CASSANDRA-13862) + * Throttle base partitions during MV repair streaming to prevent OOM (CASSANDRA-13299) + * Use compaction threshold for STCS in L0 (CASSANDRA-13861) + * Fix problem with min_compress_ratio: 1 and disallow ratio < 1 (CASSANDRA-13703) + * Add extra information to SASI timeout exception (CASSANDRA-13677) + * Add incremental repair support for --hosts, --force, and subrange repair (CASSANDRA-13818) + * Rework CompactionStrategyManager.getScanners synchronization (CASSANDRA-13786) + * Add additional unit tests for batch behavior, TTLs, Timestamps (CASSANDRA-13846) + * Add keyspace and table name in schema validation exception (CASSANDRA-13845) + * Emit metrics whenever we hit tombstone failures and warn thresholds (CASSANDRA-13771) + * Make netty EventLoopGroups daemon threads (CASSANDRA-13837) + * Race condition when closing stream sessions (CASSANDRA-13852) + * NettyFactoryTest is failing in trunk on macOS (CASSANDRA-13831) + * Allow changing log levels via nodetool for related classes (CASSANDRA-12696) + * Add stress profile yaml with LWT (CASSANDRA-7960) + * Reduce memory copies and object creations when acting on ByteBufs (CASSANDRA-13789) + * Simplify mx4j configuration (Cassandra-13578) + * Fix trigger example on 4.0 (CASSANDRA-13796) + * Force minumum timeout value (CASSANDRA-9375) + * Use netty for streaming (CASSANDRA-12229) + * Use netty for internode messaging (CASSANDRA-8457) + * Add bytes repaired/unrepaired to nodetool tablestats (CASSANDRA-13774) + * Don't delete incremental repair sessions if they still have sstables (CASSANDRA-13758) + * Fix pending repair manager index out of bounds check (CASSANDRA-13769) + * Don't use RangeFetchMapCalculator when RF=1 (CASSANDRA-13576) + * Don't optimise trivial ranges in RangeFetchMapCalculator (CASSANDRA-13664) + * Use an ExecutorService for repair commands instead of new Thread(..).start() (CASSANDRA-13594) + * Fix race / ref leak in anticompaction (CASSANDRA-13688) + * Expose tasks queue length via JMX (CASSANDRA-12758) + * Fix race / ref leak in PendingRepairManager (CASSANDRA-13751) + * Enable ppc64le runtime as unsupported architecture (CASSANDRA-13615) + * Improve sstablemetadata output (CASSANDRA-11483) + * Support for migrating legacy users to roles has been dropped (CASSANDRA-13371) + * Introduce error metrics for repair (CASSANDRA-13387) + * Refactoring to primitive functional interfaces in AuthCache (CASSANDRA-13732) + * Update metrics to 3.1.5 (CASSANDRA-13648) + * batch_size_warn_threshold_in_kb can now be set at runtime (CASSANDRA-13699) + * Avoid always rebuilding secondary indexes at startup (CASSANDRA-13725) + * Upgrade JMH from 1.13 to 1.19 (CASSANDRA-13727) + * Upgrade SLF4J from 1.7.7 to 1.7.25 (CASSANDRA-12996) + * Default for start_native_transport now true if not set in config (CASSANDRA-13656) + * Don't add localhost to the graph when calculating where to stream from (CASSANDRA-13583) + * Make CDC availability more deterministic via hard-linking (CASSANDRA-12148) + * Allow skipping equality-restricted clustering columns in ORDER BY clause (CASSANDRA-10271) + * Use common nowInSec for validation compactions (CASSANDRA-13671) + * Improve handling of IR prepare failures (CASSANDRA-13672) + * Send IR coordinator messages synchronously (CASSANDRA-13673) + * Flush system.repair table before IR finalize promise (CASSANDRA-13660) + * Fix column filter creation for wildcard queries (CASSANDRA-13650) + * Add 'nodetool getbatchlogreplaythrottle' and 'nodetool setbatchlogreplaythrottle' (CASSANDRA-13614) + * fix race condition in PendingRepairManager (CASSANDRA-13659) + * Allow noop incremental repair state transitions (CASSANDRA-13658) + * Run repair with down replicas (CASSANDRA-10446) + * Added started & completed repair metrics (CASSANDRA-13598) + * Added started & completed repair metrics (CASSANDRA-13598) + * Improve secondary index (re)build failure and concurrency handling (CASSANDRA-10130) + * Improve calculation of available disk space for compaction (CASSANDRA-13068) + * Change the accessibility of RowCacheSerializer for third party row cache plugins (CASSANDRA-13579) + * Allow sub-range repairs for a preview of repaired data (CASSANDRA-13570) + * NPE in IR cleanup when columnfamily has no sstables (CASSANDRA-13585) + * Fix Randomness of stress values (CASSANDRA-12744) + * Allow selecting Map values and Set elements (CASSANDRA-7396) + * Fast and garbage-free Streaming Histogram (CASSANDRA-13444) + * Update repairTime for keyspaces on completion (CASSANDRA-13539) + * Add configurable upper bound for validation executor threads (CASSANDRA-13521) + * Bring back maxHintTTL propery (CASSANDRA-12982) + * Add testing guidelines (CASSANDRA-13497) + * Add more repair metrics (CASSANDRA-13531) + * RangeStreamer should be smarter when picking endpoints for streaming (CASSANDRA-4650) + * Avoid rewrapping an exception thrown for cache load functions (CASSANDRA-13367) + * Log time elapsed for each incremental repair phase (CASSANDRA-13498) + * Add multiple table operation support to cassandra-stress (CASSANDRA-8780) + * Fix incorrect cqlsh results when selecting same columns multiple times (CASSANDRA-13262) + * Fix WriteResponseHandlerTest is sensitive to test execution order (CASSANDRA-13421) + * Improve incremental repair logging (CASSANDRA-13468) + * Start compaction when incremental repair finishes (CASSANDRA-13454) + * Add repair streaming preview (CASSANDRA-13257) + * Cleanup isIncremental/repairedAt usage (CASSANDRA-13430) + * Change protocol to allow sending key space independent of query string (CASSANDRA-10145) + * Make gc_log and gc_warn settable at runtime (CASSANDRA-12661) + * Take number of files in L0 in account when estimating remaining compaction tasks (CASSANDRA-13354) + * Skip building views during base table streams on range movements (CASSANDRA-13065) + * Improve error messages for +/- operations on maps and tuples (CASSANDRA-13197) + * Remove deprecated repair JMX APIs (CASSANDRA-11530) + * Fix version check to enable streaming keep-alive (CASSANDRA-12929) + * Make it possible to monitor an ideal consistency level separate from actual consistency level (CASSANDRA-13289) + * Outbound TCP connections ignore internode authenticator (CASSANDRA-13324) + * Upgrade junit from 4.6 to 4.12 (CASSANDRA-13360) + * Cleanup ParentRepairSession after repairs (CASSANDRA-13359) + * Upgrade snappy-java to 1.1.2.6 (CASSANDRA-13336) + * Incremental repair not streaming correct sstables (CASSANDRA-13328) + * Upgrade the jna version to 4.3.0 (CASSANDRA-13300) + * Add the currentTimestamp, currentDate, currentTime and currentTimeUUID functions (CASSANDRA-13132) + * Remove config option index_interval (CASSANDRA-10671) + * Reduce lock contention for collection types and serializers (CASSANDRA-13271) + * Make it possible to override MessagingService.Verb ids (CASSANDRA-13283) + * Avoid synchronized on prepareForRepair in ActiveRepairService (CASSANDRA-9292) + * Adds the ability to use uncompressed chunks in compressed files (CASSANDRA-10520) + * Don't flush sstables when streaming for incremental repair (CASSANDRA-13226) + * Remove unused method (CASSANDRA-13227) + * Fix minor bugs related to #9143 (CASSANDRA-13217) + * Output warning if user increases RF (CASSANDRA-13079) + * Remove pre-3.0 streaming compatibility code for 4.0 (CASSANDRA-13081) + * Add support for + and - operations on dates (CASSANDRA-11936) + * Fix consistency of incrementally repaired data (CASSANDRA-9143) + * Increase commitlog version (CASSANDRA-13161) + * Make TableMetadata immutable, optimize Schema (CASSANDRA-9425) + * Refactor ColumnCondition (CASSANDRA-12981) + * Parallelize streaming of different keyspaces (CASSANDRA-4663) + * Improved compactions metrics (CASSANDRA-13015) + * Speed-up start-up sequence by avoiding un-needed flushes (CASSANDRA-13031) + * Use Caffeine (W-TinyLFU) for on-heap caches (CASSANDRA-10855) + * Thrift removal (CASSANDRA-11115) + * Remove pre-3.0 compatibility code for 4.0 (CASSANDRA-12716) + * Add column definition kind to dropped columns in schema (CASSANDRA-12705) + * Add (automate) Nodetool Documentation (CASSANDRA-12672) + * Update bundled cqlsh python driver to 3.7.0 (CASSANDRA-12736) + * Reject invalid replication settings when creating or altering a keyspace (CASSANDRA-12681) + * Clean up the SSTableReader#getScanner API wrt removal of RateLimiter (CASSANDRA-12422) + * Use new token allocation for non bootstrap case as well (CASSANDRA-13080) + * Avoid byte-array copy when key cache is disabled (CASSANDRA-13084) + * Require forceful decommission if number of nodes is less than replication factor (CASSANDRA-12510) + * Allow IN restrictions on column families with collections (CASSANDRA-12654) + * Log message size in trace message in OutboundTcpConnection (CASSANDRA-13028) + * Add timeUnit Days for cassandra-stress (CASSANDRA-13029) + * Add mutation size and batch metrics (CASSANDRA-12649) + * Add method to get size of endpoints to TokenMetadata (CASSANDRA-12999) + * Expose time spent waiting in thread pool queue (CASSANDRA-8398) + * Conditionally update index built status to avoid unnecessary flushes (CASSANDRA-12969) + * cqlsh auto completion: refactor definition of compaction strategy options (CASSANDRA-12946) + * Add support for arithmetic operators (CASSANDRA-11935) + * Add histogram for delay to deliver hints (CASSANDRA-13234) + * Fix cqlsh automatic protocol downgrade regression (CASSANDRA-13307) + * Changing `max_hint_window_in_ms` at runtime (CASSANDRA-11720) + * Trivial format error in StorageProxy (CASSANDRA-13551) + * Nodetool repair can hang forever if we lose the notification for the repair completing/failing (CASSANDRA-13480) + * Anticompaction can cause noisy log messages (CASSANDRA-13684) + * Switch to client init for sstabledump (CASSANDRA-13683) + * CQLSH: Don't pause when capturing data (CASSANDRA-13743) + * nodetool clearsnapshot requires --all to clear all snapshots (CASSANDRA-13391) + * Correctly count range tombstones in traces and tombstone thresholds (CASSANDRA-8527) + * cqlshrc.sample uses incorrect option for time formatting (CASSANDRA-14243) + * Multi-version in-JVM dtests (CASSANDRA-14937) + * Allow instance class loaders to be garbage collected for inJVM dtest (CASSANDRA-15170) + + +3.11.6 +Merged from 3.0: * Ensure that tracing doesn't break connections in 3.x/4.0 mixed mode by default (CASSANDRA-15385) * Make sure index summary redistribution does not start when compactions are paused (CASSANDRA-15265) * Ensure legacy rows have primary key livenessinfo when they contain illegal cells (CASSANDRA-15365) diff --cc src/java/org/apache/cassandra/db/Directories.java index afec1a4,4731b3e..de5ab88 --- a/src/java/org/apache/cassandra/db/Directories.java +++ b/src/java/org/apache/cassandra/db/Directories.java @@@ -21,16 -21,17 +21,17 @@@ import java.io.File import java.io.FileFilter; import java.io.IOError; import java.io.IOException; +import java.nio.file.FileStore; import java.nio.file.Files; import java.nio.file.Path; + import java.nio.file.Paths; import java.util.*; import java.util.concurrent.ThreadLocalRandom; -import java.util.function.BiFunction; - +import java.util.function.BiPredicate; import com.google.common.annotations.VisibleForTesting; -import com.google.common.base.Predicate; import com.google.common.collect.ImmutableMap; import com.google.common.collect.Iterables; +import com.google.common.collect.Maps; import org.apache.commons.lang3.StringUtils; import org.slf4j.Logger; @@@ -174,11 -175,12 +175,12 @@@ public class Directorie } } - private final CFMetaData metadata; + private final TableMetadata metadata; private final DataDirectory[] paths; private final File[] dataPaths; + private final ImmutableMap<Path, DataDirectory> canonicalPathToDD; - public Directories(final CFMetaData metadata) + public Directories(final TableMetadata metadata) { this(metadata, dataDirectories); } @@@ -198,11 -200,13 +200,11 @@@ { this.metadata = metadata; this.paths = paths; -- + ImmutableMap.Builder<Path, DataDirectory> canonicalPathsBuilder = ImmutableMap.builder(); - - String cfId = ByteBufferUtil.bytesToHex(ByteBufferUtil.bytes(metadata.cfId)); - int idx = metadata.cfName.indexOf(SECONDARY_INDEX_NAME_SEPARATOR); - String cfName = idx >= 0 ? metadata.cfName.substring(0, idx) : metadata.cfName; - String indexNameWithDot = idx >= 0 ? metadata.cfName.substring(idx) : null; + String tableId = metadata.id.toHexString(); + int idx = metadata.name.indexOf(SECONDARY_INDEX_NAME_SEPARATOR); + String cfName = idx >= 0 ? metadata.name.substring(0, idx) : metadata.name; + String indexNameWithDot = idx >= 0 ? metadata.name.substring(idx) : null; this.dataPaths = new File[paths.length]; // If upgraded from version less than 2.1, use existing directories @@@ -215,10 -221,15 +219,15 @@@ boolean olderDirectoryExists = Iterables.any(Arrays.asList(dataPaths), File::exists); if (!olderDirectoryExists) { + canonicalPathsBuilder = ImmutableMap.builder(); // use 2.1+ style - String newSSTableRelativePath = join(metadata.ksName, cfName + '-' + cfId); + String newSSTableRelativePath = join(metadata.keyspace, cfName + '-' + tableId); for (int i = 0; i < paths.length; ++i) - dataPaths[i] = new File(paths[i].location, newSSTableRelativePath); + { - File dataPath = new File(paths[i].location, newSSTableRelativePath);; ++ File dataPath = new File(paths[i].location, newSSTableRelativePath); + dataPaths[i] = dataPath; + canonicalPathsBuilder.put(Paths.get(FileUtils.getCanonicalPath(dataPath)), paths[i]); + } } // if index, then move to its own directory if (indexNameWithDot != null) diff --cc src/java/org/apache/cassandra/db/DiskBoundaries.java index 22f17b0,90af893..6c69007 --- a/src/java/org/apache/cassandra/db/DiskBoundaries.java +++ b/src/java/org/apache/cassandra/db/DiskBoundaries.java @@@ -114,8 -115,9 +116,9 @@@ public class DiskBoundarie /** * Try to figure out location based on sstable directory */ - private int getBoundariesFromSSTableDirectory(SSTableReader sstable) + public int getBoundariesFromSSTableDirectory(Descriptor descriptor) { - Directories.DataDirectory actualDirectory = cfs.getDirectories().getDataDirectoryForFile(sstable.descriptor); ++ Directories.DataDirectory actualDirectory = cfs.getDirectories().getDataDirectoryForFile(descriptor); for (int i = 0; i < directories.size(); i++) { Directories.DataDirectory directory = directories.get(i); diff --cc src/java/org/apache/cassandra/db/DiskBoundaryManager.java index 69aabfd,61febe9..2628af8 --- a/src/java/org/apache/cassandra/db/DiskBoundaryManager.java +++ b/src/java/org/apache/cassandra/db/DiskBoundaryManager.java @@@ -107,11 -104,12 +107,11 @@@ public class DiskBoundaryManage while (directoriesVersion != BlacklistedDirectories.getDirectoriesVersion()); // if directoriesVersion has changed we need to recalculate if (localRanges == null || localRanges.isEmpty()) - return new DiskBoundaries(dirs, null, ringVersion, directoriesVersion); + return new DiskBoundaries(cfs, dirs, null, ringVersion, directoriesVersion); - List<Range<Token>> sortedLocalRanges = Range.sort(localRanges); + List<PartitionPosition> positions = getDiskBoundaries(localRanges, cfs.getPartitioner(), dirs); - return new DiskBoundaries(dirs, positions, ringVersion, directoriesVersion); - List<PartitionPosition> positions = getDiskBoundaries(sortedLocalRanges, cfs.getPartitioner(), dirs); + return new DiskBoundaries(cfs, dirs, positions, ringVersion, directoriesVersion); } /** diff --cc test/unit/org/apache/cassandra/db/DirectoriesTest.java index 7629a54,24e03f6..ec1b34c --- a/test/unit/org/apache/cassandra/db/DirectoriesTest.java +++ b/test/unit/org/apache/cassandra/db/DirectoriesTest.java @@@ -46,6 -49,8 +50,7 @@@ import org.apache.cassandra.io.sstable. import org.apache.cassandra.io.util.FileUtils; import org.apache.cassandra.schema.IndexMetadata; import org.apache.cassandra.service.DefaultFSErrorHandler; + import org.apache.cassandra.utils.ByteBufferUtil; -import org.apache.cassandra.utils.Pair; import static org.junit.Assert.assertEquals; import static org.junit.Assert.assertFalse; @@@ -469,6 -475,150 +474,150 @@@ public class DirectoriesTes } } + @Test + public void testGetLocationForDisk() + { + Collection<DataDirectory> paths = new ArrayList<>(); + paths.add(new DataDirectory(new File("/tmp/aaa"))); + paths.add(new DataDirectory(new File("/tmp/aa"))); + paths.add(new DataDirectory(new File("/tmp/a"))); + - for (CFMetaData cfm : CFM) ++ for (TableMetadata cfm : CFM) + { + Directories dirs = new Directories(cfm, paths); + for (DataDirectory dir : paths) + { + String p = dirs.getLocationForDisk(dir).getAbsolutePath() + File.separator; + assertTrue(p.startsWith(dir.location.getAbsolutePath() + File.separator)); + } + } + } + + @Test + public void testGetLocationWithSymlinks() throws IOException + { + Path p = Files.createTempDirectory("something"); + Path symlinktarget = Files.createDirectories(p.resolve("symlinktarget")); + Path ddir = Files.createDirectories(p.resolve("datadir1")); + + Path p1 = Files.createDirectories(ddir.resolve("p1").resolve("ks")).getParent(); // the data dir does not include the keyspace dir + Path p2 = Files.createDirectories(ddir.resolve("p2")); + Path l1 = Files.createSymbolicLink(p2.resolve("ks"), symlinktarget); + + DataDirectory path1 = new DataDirectory(p1.toFile()); + DataDirectory path2 = new DataDirectory(p2.toFile()); + Directories dirs = new Directories(CFM.iterator().next(), new DataDirectory[] {path1, path2}); + dirs.getLocationForDisk(new DataDirectory(p1.toFile())); + dirs.getLocationForDisk(new DataDirectory(p2.toFile())); + + assertTrue(dirs.getLocationForDisk(path2).toPath().startsWith(l1)); + assertTrue(dirs.getLocationForDisk(path1).toPath().startsWith(p1)); + } + + @Test + public void getDataDirectoryForFile() + { + Collection<DataDirectory> paths = new ArrayList<>(); + paths.add(new DataDirectory(new File("/tmp/a"))); + paths.add(new DataDirectory(new File("/tmp/aa"))); + paths.add(new DataDirectory(new File("/tmp/aaa"))); + - for (CFMetaData cfm : CFM) ++ for (TableMetadata cfm : CFM) + { + Directories dirs = new Directories(cfm, paths); + for (DataDirectory dir : paths) + { + Descriptor d = Descriptor.fromFilename(new File(dir.location, getNewFilename(cfm, false)).toString()); + String p = dirs.getDataDirectoryForFile(d).location.getAbsolutePath() + File.separator; + assertTrue(p.startsWith(dir.location.getAbsolutePath() + File.separator)); + } + } + } + + /** + * Makes sure we can find the data directory when it is a symlink + * + * creates the following data directories: + * <tempdir something>/datadir1 + * <tempdir something>/datadir11 (symlink to <tempdir something>/symlinktarget) + * + * and then makes sure that we get the correct directory back. + */ + @Test + public void testDirectoriesSymlinks() throws IOException + { + Path p = Files.createTempDirectory("something"); + Path symlinktarget = Files.createDirectories(p.resolve("symlinktarget")); + Path ddir1 = Files.createDirectories(p.resolve("datadir1")); + Path ddir2 = Files.createSymbolicLink(p.resolve("datadir11"), symlinktarget); + DataDirectory dd1 = new DataDirectory(ddir1.toFile()); + DataDirectory dd2 = new DataDirectory(ddir2.toFile()); + - for (CFMetaData tm : CFM) ++ for (TableMetadata tm : CFM) + { + Directories dirs = new Directories(tm, Sets.newHashSet(dd1, dd2)); - Descriptor desc = Descriptor.fromFilename(ddir1.resolve(getNewFilename(tm, false)).toString()); ++ Descriptor desc = Descriptor.fromFilename(ddir1.resolve(getNewFilename(tm, false)).toFile()); + assertEquals(ddir1.toFile(), dirs.getDataDirectoryForFile(desc).location); - desc = Descriptor.fromFilename(ddir2.resolve(getNewFilename(tm, false)).toString()); ++ desc = Descriptor.fromFilename(ddir2.resolve(getNewFilename(tm, false)).toFile()); + assertEquals(ddir2.toFile(), dirs.getDataDirectoryForFile(desc).location); + } + } + + @Test + public void testDirectoriesOldTableSymlink() throws IOException + { + testDirectoriesSymlinksHelper(true); + } + + @Test + public void testDirectoriesTableSymlink() throws IOException + { + testDirectoriesSymlinksHelper(false); + } + + /** + * Makes sure we can find the data directory for a file when the table directory is a symlink + * + * if oldStyle is false we append the table id to the table directory + * + * creates the following structure + * <tempdir>/datadir1/<ks>/<table> + * <tempdir>/datadir11/<ks>/<table symlink to <tempdir>/symlinktarget> + * + * and then we create a fake descriptor to a file in the table directory and make sure we get the correct + * data directory back. + */ + private void testDirectoriesSymlinksHelper(boolean oldStyle) throws IOException + { + Path p = Files.createTempDirectory("something"); + Path symlinktarget = Files.createDirectories(p.resolve("symlinktarget")); + Path ddir1 = Files.createDirectories(p.resolve("datadir1")); + Path ddir2 = Files.createDirectories(p.resolve("datadir11")); + - for (CFMetaData metadata : CFM) ++ for (TableMetadata tm : CFM) + { - Path keyspacedir = Files.createDirectories(ddir2.resolve(metadata.ksName)); - String tabledir = metadata.cfName + (oldStyle ? "" : Component.separator + ByteBufferUtil.bytesToHex(ByteBufferUtil.bytes(metadata.cfId))); ++ Path keyspacedir = Files.createDirectories(ddir2.resolve(tm.keyspace)); ++ String tabledir = tm.name + (oldStyle ? "" : Component.separator + tm.id.toHexString()); + Files.createSymbolicLink(keyspacedir.resolve(tabledir), symlinktarget); + } + + DataDirectory dd1 = new DataDirectory(ddir1.toFile()); + DataDirectory dd2 = new DataDirectory(ddir2.toFile()); - for (CFMetaData tm : CFM) ++ for (TableMetadata tm : CFM) + { + Directories dirs = new Directories(tm, Sets.newHashSet(dd1, dd2)); - Descriptor desc = Descriptor.fromFilename(ddir1.resolve(getNewFilename(tm, oldStyle)).toFile().toString()); ++ Descriptor desc = Descriptor.fromFilename(ddir1.resolve(getNewFilename(tm, oldStyle)).toFile()); + assertEquals(ddir1.toFile(), dirs.getDataDirectoryForFile(desc).location); - desc = Descriptor.fromFilename(ddir2.resolve(getNewFilename(tm, oldStyle)).toFile().toString()); ++ desc = Descriptor.fromFilename(ddir2.resolve(getNewFilename(tm, oldStyle)).toFile()); + assertEquals(ddir2.toFile(), dirs.getDataDirectoryForFile(desc).location); + } + } + - private String getNewFilename(CFMetaData metadata, boolean oldStyle) ++ private String getNewFilename(TableMetadata tm, boolean oldStyle) + { - return metadata.ksName + File.separator + metadata.cfName + (oldStyle ? "" : Component.separator + ByteBufferUtil.bytesToHex(ByteBufferUtil.bytes(metadata.cfId))) + "/na-1-big-Data.db"; ++ return tm.keyspace + File.separator + tm.name + (oldStyle ? "" : Component.separator + tm.id.toHexString()) + "/na-1-big-Data.db"; + } + private List<Directories.DataDirectoryCandidate> getWriteableDirectories(DataDirectory[] dataDirectories, long writeSize) { // copied from Directories.getWriteableLocation(long) diff --cc test/unit/org/apache/cassandra/db/compaction/CompactionStrategyManagerTest.java index 73e6852,2120757..d29ab52 --- a/test/unit/org/apache/cassandra/db/compaction/CompactionStrategyManagerTest.java +++ b/test/unit/org/apache/cassandra/db/compaction/CompactionStrategyManagerTest.java @@@ -200,203 -158,6 +200,203 @@@ public class CompactionStrategyManagerT } } + @Test + public void testAutomaticUpgradeConcurrency() throws Exception + { + ColumnFamilyStore cfs = Keyspace.open(KS_PREFIX).getColumnFamilyStore(TABLE_PREFIX); + DatabaseDescriptor.setAutomaticSSTableUpgradeEnabled(true); + DatabaseDescriptor.setMaxConcurrentAutoUpgradeTasks(1); + + // latch to block CompactionManager.BackgroundCompactionCandidate#maybeRunUpgradeTask + // inside the currentlyBackgroundUpgrading check - with max_concurrent_auto_upgrade_tasks = 1 this will make + // sure that BackgroundCompactionCandidate#maybeRunUpgradeTask returns false until the latch has been counted down + CountDownLatch latch = new CountDownLatch(1); + AtomicInteger upgradeTaskCount = new AtomicInteger(0); + MockCFSForCSM mock = new MockCFSForCSM(cfs, latch, upgradeTaskCount); + + CompactionManager.BackgroundCompactionCandidate r = CompactionManager.instance.getBackgroundCompactionCandidate(mock); + CompactionStrategyManager mgr = mock.getCompactionStrategyManager(); + // basic idea is that we start a thread which will be able to get in to the currentlyBackgroundUpgrading-guarded + // code in CompactionManager, then we try to run a bunch more of the upgrade tasks which should return false + // due to the currentlyBackgroundUpgrading count being >= max_concurrent_auto_upgrade_tasks + Thread t = new Thread(() -> r.maybeRunUpgradeTask(mgr)); + t.start(); + Thread.sleep(100); // let the thread start and grab the task + assertEquals(1, CompactionManager.instance.currentlyBackgroundUpgrading.get()); + assertFalse(r.maybeRunUpgradeTask(mgr)); + assertFalse(r.maybeRunUpgradeTask(mgr)); + latch.countDown(); + t.join(); + assertEquals(1, upgradeTaskCount.get()); // we should only call findUpgradeSSTableTask once when concurrency = 1 + assertEquals(0, CompactionManager.instance.currentlyBackgroundUpgrading.get()); + + DatabaseDescriptor.setAutomaticSSTableUpgradeEnabled(false); + } + + @Test + public void testAutomaticUpgradeConcurrency2() throws Exception + { + ColumnFamilyStore cfs = Keyspace.open(KS_PREFIX).getColumnFamilyStore(TABLE_PREFIX); + DatabaseDescriptor.setAutomaticSSTableUpgradeEnabled(true); + DatabaseDescriptor.setMaxConcurrentAutoUpgradeTasks(2); + // latch to block CompactionManager.BackgroundCompactionCandidate#maybeRunUpgradeTask + // inside the currentlyBackgroundUpgrading check - with max_concurrent_auto_upgrade_tasks = 1 this will make + // sure that BackgroundCompactionCandidate#maybeRunUpgradeTask returns false until the latch has been counted down + CountDownLatch latch = new CountDownLatch(1); + AtomicInteger upgradeTaskCount = new AtomicInteger(); + MockCFSForCSM mock = new MockCFSForCSM(cfs, latch, upgradeTaskCount); + + CompactionManager.BackgroundCompactionCandidate r = CompactionManager.instance.getBackgroundCompactionCandidate(mock); + CompactionStrategyManager mgr = mock.getCompactionStrategyManager(); + + // basic idea is that we start 2 threads who will be able to get in to the currentlyBackgroundUpgrading-guarded + // code in CompactionManager, then we try to run a bunch more of the upgrade task which should return false + // due to the currentlyBackgroundUpgrading count being >= max_concurrent_auto_upgrade_tasks + Thread t = new Thread(() -> r.maybeRunUpgradeTask(mgr)); + t.start(); + Thread t2 = new Thread(() -> r.maybeRunUpgradeTask(mgr)); + t2.start(); + Thread.sleep(100); // let the threads start and grab the task + assertEquals(2, CompactionManager.instance.currentlyBackgroundUpgrading.get()); + assertFalse(r.maybeRunUpgradeTask(mgr)); + assertFalse(r.maybeRunUpgradeTask(mgr)); + assertFalse(r.maybeRunUpgradeTask(mgr)); + assertEquals(2, CompactionManager.instance.currentlyBackgroundUpgrading.get()); + latch.countDown(); + t.join(); + t2.join(); + assertEquals(2, upgradeTaskCount.get()); + assertEquals(0, CompactionManager.instance.currentlyBackgroundUpgrading.get()); + + DatabaseDescriptor.setMaxConcurrentAutoUpgradeTasks(1); + DatabaseDescriptor.setAutomaticSSTableUpgradeEnabled(false); + } + + private static void assertHolderExclusivity(boolean isRepaired, boolean isPendingRepair, boolean isTransient, Class<? extends AbstractStrategyHolder> expectedType) + { + ColumnFamilyStore cfs = Keyspace.open(KS_PREFIX).getColumnFamilyStore(TABLE_PREFIX); + CompactionStrategyManager csm = cfs.getCompactionStrategyManager(); + + AbstractStrategyHolder holder = csm.getHolder(isRepaired, isPendingRepair, isTransient); + assertNotNull(holder); + assertSame(expectedType, holder.getClass()); + + int matches = 0; + for (AbstractStrategyHolder other : csm.getHolders()) + { + if (other.managesRepairedGroup(isRepaired, isPendingRepair, isTransient)) + { + assertSame("holder assignment should be mutually exclusive", holder, other); + matches++; + } + } + assertEquals(1, matches); + } + + private static void assertInvalieHolderConfig(boolean isRepaired, boolean isPendingRepair, boolean isTransient) + { + ColumnFamilyStore cfs = Keyspace.open(KS_PREFIX).getColumnFamilyStore(TABLE_PREFIX); + CompactionStrategyManager csm = cfs.getCompactionStrategyManager(); + try + { + csm.getHolder(isRepaired, isPendingRepair, isTransient); + fail("Expected IllegalArgumentException"); + } + catch (IllegalArgumentException e) + { + // expected + } + } + + /** + * If an sstable can be be assigned to a strategy holder, it shouldn't be possibly to + * assign it to any of the other holders. + */ + @Test + public void testMutualExclusiveHolderClassification() throws Exception + { + assertHolderExclusivity(false, false, false, CompactionStrategyHolder.class); + assertHolderExclusivity(true, false, false, CompactionStrategyHolder.class); + assertHolderExclusivity(false, true, false, PendingRepairHolder.class); + assertHolderExclusivity(false, true, true, PendingRepairHolder.class); + assertInvalieHolderConfig(true, true, false); + assertInvalieHolderConfig(true, true, true); + assertInvalieHolderConfig(false, false, true); + assertInvalieHolderConfig(true, false, true); + } + + PartitionPosition forKey(int key) + { + DecoratedKey dk = Util.dk(String.format("%04d", key)); + return dk.getToken().minKeyBound(); + } + + /** + * Test that csm.groupSSTables correctly groups sstables by repaired status and directory + */ + @Test + public void groupSSTables() throws Exception + { + final int numDir = 4; + ColumnFamilyStore cfs = createJBODMockCFS(numDir); + Keyspace.open(cfs.keyspace.getName()).getColumnFamilyStore(cfs.name).disableAutoCompaction(); + assertTrue(cfs.getLiveSSTables().isEmpty()); + List<SSTableReader> transientRepairs = new ArrayList<>(); + List<SSTableReader> pendingRepair = new ArrayList<>(); + List<SSTableReader> unrepaired = new ArrayList<>(); + List<SSTableReader> repaired = new ArrayList<>(); + + for (int i = 0; i < numDir; i++) + { + int key = 100 * i; + transientRepairs.add(createSSTableWithKey(cfs.keyspace.getName(), cfs.name, key++)); + pendingRepair.add(createSSTableWithKey(cfs.keyspace.getName(), cfs.name, key++)); + unrepaired.add(createSSTableWithKey(cfs.keyspace.getName(), cfs.name, key++)); + repaired.add(createSSTableWithKey(cfs.keyspace.getName(), cfs.name, key++)); + } + + cfs.getCompactionStrategyManager().mutateRepaired(transientRepairs, 0, UUID.randomUUID(), true); + cfs.getCompactionStrategyManager().mutateRepaired(pendingRepair, 0, UUID.randomUUID(), false); + cfs.getCompactionStrategyManager().mutateRepaired(repaired, 1000, null, false); + - DiskBoundaries boundaries = new DiskBoundaries(cfs.getDirectories().getWriteableLocations(), ++ DiskBoundaries boundaries = new DiskBoundaries(cfs, cfs.getDirectories().getWriteableLocations(), + Lists.newArrayList(forKey(100), forKey(200), forKey(300)), + 10, 10); + + CompactionStrategyManager csm = new CompactionStrategyManager(cfs, () -> boundaries, true); + + List<GroupedSSTableContainer> grouped = csm.groupSSTables(Iterables.concat( transientRepairs, pendingRepair, repaired, unrepaired)); + + for (int x=0; x<grouped.size(); x++) + { + GroupedSSTableContainer group = grouped.get(x); + AbstractStrategyHolder holder = csm.getHolders().get(x); + for (int y=0; y<numDir; y++) + { + SSTableReader sstable = Iterables.getOnlyElement(group.getGroup(y)); + assertTrue(holder.managesSSTable(sstable)); + SSTableReader expected; + if (sstable.isRepaired()) + expected = repaired.get(y); + else if (sstable.isPendingRepair()) + { + if (sstable.isTransient()) + { + expected = transientRepairs.get(y); + } + else + { + expected = pendingRepair.get(y); + } + } + else + expected = unrepaired.get(y); + + assertSame(expected, sstable); + } + } + } + private MockCFS createJBODMockCFS(int disks) { // Create #disks data directories to simulate JBOD --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
