1
0
Fork 0
Commit Graph

518 Commits

Author SHA1 Message Date
Lars Maier dd07d74d69 [devel] Bug fix/bad leader report current (#7585)
* Bug fix 3.4/bad leader report current (#7574)
* Initialize theLeader non-empty, thus not assuming leadership.
* Correct ClusterInfo to look into Target/CleanedServers.
* Prevent usage of to be cleaned out servers in new collections.
* After a restart, do not assume to be leader for a shard.
* Do nothing in phaseTwo if leader has not been touched. (#7579)
* Drop follower if it refuses to cooperate.

This is important since a dbserver that is follower for a shard will
after a reboot think that it is a leader, at least for a short amount
of time. If it came back quickly enough, the leader might not have
noticed that it was away.
2018-12-03 10:20:30 +01:00
Andrey Abramov 2a0fa4946e
improve logging in ClusterInfo::loadPlan (#7511)
* improve logging in ClusterInfo::loadPlan

* address review comments
2018-11-29 15:56:51 +03:00
Andrey Abramov 6674a4282d
avoid calling cluster related functions while instantiating views on … (#7509)
* avoid calling cluster related functions while instantiating views on a db server

* minor cleanup
2018-11-29 15:43:53 +03:00
Max Neunhöffer ae29e5d2ba
Fix index creation in cluster. (#7440)
* Fix index creation in cluster.

Simplify and correct error handling logic in ensureIndexCoordinator.

* After index creation, wait until index appears.

We wait until the Supervision has removed the isBuilding flag and
the coordinator has reloaded the Plan.

* More index handling fixes.

* Directly remove isBuilding in ensureIndexCoordinator (again).

* Fix catch tests by holding mutex shorter.

* Better mutex handling in ClusterInfo.
2018-11-28 16:58:05 +01:00
Vasiliy 1a0b9b9261 issue 153: ensure views are dropped in Agency when database is dropped in cluster, minor fixes (#7370)
* issue 153: ensure views are dropped in Agency when database is dropped in cluster, minor fixes

* backport: add test to ensure views are dropped when database is dropped from plan, fix some issues in ClusterInfo

* optimize primary key lookups in ArangoSearch

* fix test

* Add JS tests

* temporary comment optimizations
2018-11-21 19:18:34 +03:00
Kaveh Vahedipour 9ec6619b84 Bug fix/index readiness (#6541)
* indexes are marked  while still missing in Current
* index handling getCollection
* supervision gets indexes from isbuilding, when coordinator is gone before finishing
* seems right now
* fixed broken views
* remove junk comments
* cleanup
* node / supervision adjustements
* supervision fixes
* neunhoef remarks part i
* neunhoef remarks part ii
* neunhoef remarks part ii
* neunhoef remarks part iiI
* collection's current version please
* no need to wait for current once again
* no longer necessary code
* clear comments
* delete left overs
* dead code revived
2018-11-21 14:42:58 +01:00
Wilfried Goesgens 05a7d4e96e add alternative to ClusterInfo::getCollection() that doesn't throw (#7339) 2018-11-20 16:05:57 +01:00
Markus Pfeiffer 39bdebf851 Port bug-fix-3.4/timeout-create-coll to devel (#7307)
* Fix loophole in error handling.
* Fix inquiry case of id not found: 404.
* Also handle correctly in AgencyComm.
* Fix agency tests.
* Fix error handling in dropCollectionOnCoordinator.
2018-11-14 10:03:55 +01:00
Jan dbf8d582d5
added missing change to clusterinfo (#7294) 2018-11-13 11:37:48 +01:00
Simon 386fc0e9ad Simplify dropDatabaseCoordinator & fix some bugs (#7211) 2018-11-06 15:26:33 +01:00
Vasiliy 8f44afb6cf issue 496.1: switch scope of responsibility between a TRI_vocbase_t and a LogicalView in respect to view creation/deletion (#7101)
* issue 496.1: switch scope of responsibility between a TRI_vocbase_t and a LogicalView in respect to view creation/deletion

* backport: address test failures

* backport: ensure arangosearch links get exported in the dump

* backport: ensure view is created during restore on the coordinator

* Updates for ArangoSearch DDL tests, IResearchView unregistration and known issues

* Add fix for internal issue 483
2018-10-30 12:50:35 +03:00
Simon c72818a9dc Make ensureIndexOnCoordinator more robust (#7110) 2018-10-29 17:45:46 +01:00
Simon 8b7a4099b8 Properly compare velocypack objects in Agency operations (#6921)
* Properly compare velocypack objects in Agency operations

* Add changelog

* added option for VPackDumper
2018-10-17 20:03:53 +02:00
Jan c7cd0262aa
suppress some of these dreaded error messages (#6786) 2018-10-11 10:46:12 +02:00
Dan Larkin-York 4644d2b023 Fix issue with colleciton/view name conflict checking in cluster. (#6796) 2018-10-11 10:45:29 +02:00
Jan c06f2d77da
Feature/velocypack update (#6678) 2018-10-02 14:04:14 +02:00
Simon 0fa7f01c66 Resilience test failure points (#6539) 2018-09-20 01:05:10 +02:00
Jan e18e9158d8
fix cluster selectivity estimates (#6488) 2018-09-14 00:22:01 +02:00
Max Neunhöffer 84735955ea Add advertised endpoints. (#6104) 2018-09-13 16:30:55 +02:00
Jan 3b16913b1b
fix cluster index selectivity (#6467) 2018-09-12 14:35:39 +02:00
Kaveh Vahedipour 6b2733625c Feature/static const strings cleanup (#6352)
* AgentConfiguration cleanup
* static strings in maintenance / agency
* more strings unified
* fix windows build
2018-09-11 13:40:03 +02:00
Simon 0661a4c1fe Hide Links from getIndexes() (#6325) 2018-09-03 15:17:24 +02:00
Lars Maier 63d9cfa081 Maintenance Fixes (#6284)
* Clean up for `FIXMEMAINTENANCE` comments: removed race condition, added errors and `notify()`s.
* Removed dublicated code.
* Added requested changes. Added error reporting for `UpdateCollection`.
* Make it compile. Add missing `notify()`.
* `CreateCollection` generates errors in all code paths.
* Fixed catch test.
2018-08-31 15:24:29 +02:00
Jan 5022ccc24d
Bug fix/fixes 2508 (#6254) 2018-08-27 21:36:39 +02:00
Vasiliy 5d14775de8 issue 459.3: ensure collection permissions are checked before updating/dropping an IResearch view (#6253)
* issue 459.3: ensure collection permissions are checked before updating/dropping an IResearch view

* backport: ensure collection permissions are checked before updating/dropping an IResearch view on cluster

* backport: address test failures

* backport: address more test failures

* reuse existing classes for scoping ExecContext
2018-08-26 18:00:16 +03:00
Kaveh Vahedipour 28754cbf15 Feature/schmutz plus plus (#5972)
- Schmutz now called "Maintenance" and completely implemented in C++
 - Fix index locking bug in mmfiles
 - Fix a bug in mmfiles with silent option and repsert
 - Slightly increase supervision okperiod and graceperiod
2018-08-24 12:15:35 +02:00
Matthew Von-Maszewski 86ea784372 bugfix: establish unique function name & implementation for communication retry status (#6150)
* initial checkin of isRetryOK().  Includes fixes to known code that has previously hung shutdowns by performing infinite retries.

* slight help on getting out of a loop faster during shutdown.  not essential.
2018-08-17 14:57:12 +02:00
Dan Larkin-York 5f87f57cd0 Improved sharding algorithms (#6089) 2018-08-09 19:03:32 +02:00
Kaveh Vahedipour fd60b359b6 fixed parallel creation of indexes in cluster (#6088)
* fixed parallel creation of indexes in cluster

* added tests
2018-08-07 10:00:15 +02:00
Simon 42cabb858a Fix dumping of views in the cluster (#6024) 2018-08-02 17:25:49 +02:00
Jan e4d7f1c5f0
Bug fix/wenn der shard mann 2mal klingelt (#5890) 2018-07-26 15:37:40 +02:00
Simon 2dd8593609 View Replication (#5915) 2018-07-26 10:28:46 +02:00
Max Neunhöffer 014c3f7f53
Only load Plan and Current in ClusterInfo when actually needed. (#5649)
* Only update Plan and Current from Agency if not already done.
* Add read protection for getPlanVersion and getCurrentVersion.
* Add a further check to loadPlan and loadCurrent.
* Fix tests to new behaviour.
* Try to increase Plan/Version and Current/Version with every change.
* Add two more increments of Plan/Version
* Add missing increments in tests for Plan/Version.
* Add changelog entry.
2018-07-16 12:20:13 +02:00
Jan 208f1297e1
Bug fix/fixes 1007 (#5815) 2018-07-10 13:47:04 +02:00
Dan Larkin-York 21e16a8a24 Add load balancer awareness for cursor API (#5682) 2018-07-03 14:29:09 +02:00
Vasiliy 7aaeab50fb issue 402.1: share sync thread between IResearchView and IResearchViewDBServer (#5733) 2018-07-02 15:03:00 +03:00
Jan 5e8e1e71c4
don't hang in shutdown while creating a collection or an index (#5710) 2018-06-29 11:21:00 +02:00
Simon ec0d2a1b7b Remove Coordinator DBs (#5661) 2018-06-25 19:18:11 +02:00
Andrey Abramov 5eef6cd618
Feature/test iresearch (#5610)
* start implementing arangosearch cluster tests.

* backport: ensure view lookup is done via collectionNameResover, ensure updateProperties returns current view properties

* first attempt to fix failing tests

* refactor cluster wide view creation logic

* if view is not found in the new plan then check the old plan too

* ensure the cluster-wide view is looked up in vocbase as well on startup/recovery

* do not store cluster-wide IResearchView in vocbase

* move stale view cleanup to the shared pointer deleter, address test failures

* do not print warning

* enable arangosearch tests by default

* fix catch tests

* address icorrect return value for cluster-wide links

* address some issues with test failures due to cluster-view allocated within TRI_vocbase_t

* simplify per-cid view name, address 'catch' test failures

* ensure IResearchViewNode volatility is properly calculated in cluster

* invoke callbacks directly in AgencyMock instead of waiting for timeout

* ensure view updates via JavaScript always use the latest view definition

* pass a list of shards to `IResearchViewDBServer::snapshot`

* extend cluster aql tests

* fixes after merge

* fix class/struct inconsistencies

* comment failing tests

* remove debug logging

* add debug function

* tests cleanup

* simplify upcoming merge: pass resolver from a side

* backport: move all transaction status callback logic to Methods

* add changes missed from previous commit

* fix js and ruby tests

* more tests for IResearchViewNode

* pass transaction to IResearchViewDBServer::snapshot, address IResearchViewDBServer tests segfault

* pass transaction to IResearchView::snapshot instead of transaction state

* temporarily add trace log output to tests to try to find the cause of the core dump on Jenkins

* add more temporary debug output to trace down the segfault on Jenkins

* add even more temporary debug output to trace down the segfault on Jenkins

* ensure Vieew related maps are cleared during shutdown

* reset ClusterInfo::instance() before DatabaseFeature::unprepare()

* remove extraneous debug output

* missed line from previous commit

* uncomment required line

* add nullptr checks to RocksDBIndexFactory::prepareIndexes(...) similar to the ones in MMFilesIndexFactory::prepareIndexes(...)

* attempt to fix deadlock in tests

* add comment as per reviewer request

* fix aql test suite name

* add some debug logging

* address deadlock between ClusterInfo::loadPlan() and CollectionNameResolver::localNameLookup(...)

* eplicitly state which index definition failed in the log message

* use vocbase from shard-view isntead just in case

* explicitly state which index definition failed in the log message

* do not create shard-view instances from cluster-link instances (only register existing ones)

* add some tests
2018-06-21 20:35:04 +03:00
Vasiliy 4253dca6aa issue 381.5: ensure the LogicalView definition that is persisted to the Agency matches the definition that gets created (#5518)
* issue 381.5: ensure the LogicalView definition that is persisted to the Agency matches the definition that gets created

* backport: correct comment
2018-06-02 17:21:55 +03:00
Simon 332a7958f5 Cleanup cluster selectivity (#5440) 2018-05-23 18:00:14 +02:00
Simon 35992ad67b Coordinator storage engine (#5405) 2018-05-22 19:30:27 +02:00
Vasiliy 6a53154160 issue 389.2: use static strings for Index definition json attributes, use TRI_vocbase_t references instead of pointers in V8Context, use TRI_vocbase_t references instead of pointers in DatabaseInitialSyncer (#5344) 2018-05-14 19:06:24 +03:00
Andrey Abramov b69b5bdfdf
Bug fix/issue #5186 (#5269)
* do not persist view on startup

* small refactoring

* ensure view is being opened after creation
2018-05-06 20:38:32 +03:00
Simon 828f1d423c S2 based Geo-Spatial index (#5249) 2018-05-02 23:54:41 +02:00
Andrey Abramov 4649b40b96
Coordinator ArangoSearch view + Execution nodes + AgencyMock (#5160)
* add initial implementation of scatter view rule and node

* add tests for `IResearchViewNode` and `IResearchViewScatterNode`

* add missing check

* modify IResearch execution nodes to use references instead of pointers

* use view id in searialized `ExecutionNode` representation instead of the name

* add cluster mode stubs and checks

* very first attempt to distribute IResearchViewNode

* further implementation of cluster-wide arangosearch views

* fix invalid json format

* add tests for coordinator iresearch view

* allow to retrieve a list of existing views on a coordinator

* more tests for coordinator iresearch view

* some fixes to enable query explanation

* remove Collection dependency from RemoteNode

* remove unnecessary remote ArangoSearch view scatter

* fix explanation appearance

* add some assertions

* minor fixes

* implement IResearchViewCoordinator::updateProperties

* fix view DDL issues

* handle link modifications in DDL operations

* add coordinator implementation of iresearch view links

* fix tests

* further coordinator based view DDL implementation

* further IResearchViewCoordinator implementation

* add initial implementation of AgencyMock

* fix some tests

* code cleanup

* extend test + some fixes

* more tests for IResearchViewCoordinator

* fix tests for IResearchLinkCoordinator

* some fixes after merge

* fix tests

* remove declaration of nonexistent (previously removed) method

* some fixes after review

* remove string duplication

* more tests and fixes

* more fixes and tests

* more tests

* one more test

* fix 'use-after-free' asan error

* fix non-enterprise tests issues
2018-05-02 00:15:11 +03:00
Vasiliy e4368b0991 issue 355.6: remove create() from LogicalView, remove IResearch dependency from IndexFactory, store vocbase reference in LogicalDataSource 2018-04-06 16:38:34 +03:00
Jan a2f8077cd4
fix restoring of smart graph edge collections so it does not time out (#5049) 2018-04-06 14:14:52 +02:00
Vasiliy 99b83ba8c8 issue 355.5: remove more unused methods, move view-related storage engine functionality from vocbase into DBServerLogicalView, address MSVC cmake dependency issue 2018-04-05 16:17:07 +03:00
Vasiliy 635db3b409 issue 355.4: remove redundant methods and code, use 'cp' instead of 'cmake copy_directory' where possible, use vocbase reference instead of pointer 2018-04-04 10:53:48 +03:00