1
0
Fork 0
Commit Graph

1387 Commits

Author SHA1 Message Date
Kaveh Vahedipour 211812c6f5 Metrics tests (#10578)
* adding tests

* consolidate

* tests working again

* KAVEH!

* more tests

* jsteemann rev pt 1

* minor rewrite

* remove static pointer in metrics feature

* feature api && tests

* fixed tests

* reflective dep

* invalidated iterators on obi watch

* fix test

* back in

* fix counter fetch

* obi1

* added _admin/metrics documentation

* try_emplace and docu touch up
2019-12-02 16:55:29 +01:00
Wilfried Goesgens 589f321ace only anounce coordinator readiness once the agency is convinced we are (#10562)
* only anounce coordinator readiness once the agency is convinced we are

* wait for DB-Servers too

* move test creating the database to the front

* move test to be run first
2019-11-28 18:56:59 +01:00
Jan a50f07e928 obsolete option `--vst.maxsize` requires a value if specified (#10547) 2019-11-26 13:23:20 +01:00
Kaveh Vahedipour 1489afa148 metricsfeature cleanup in unprepare (#10514) 2019-11-22 15:02:36 +01:00
Kaveh Vahedipour b94f40d890 feature/metrics (#10423)
- New internal counters and histograms.
- Prometheus compatible endpoint to expose new metrics, old statistics and RocksDB metrics.
2019-11-22 08:02:48 +01:00
Jan 932f8b5874
undistribute a few more queries (#10459) 2019-11-20 08:21:44 +01:00
Lars Maier ebf3296e9c Backup with view (#10386)
* Commit and store view related data on create backup.
* Fixes to upgrade on env-var.
* Fixed CMakeLists.txt.
* Changes to allow restore from 3.5 to 3.6.
* Remove unnecessary code introduced in merge.
* CHANGELOG:
2019-11-15 17:50:04 +01:00
Jan 1476ddbd0d
Bug fix/fix cluster upgrade (#10411) 2019-11-15 10:42:04 +01:00
Max Neunhöffer b318ca8084
ARANGODB_UPGRADE_DURING_RESTORE env variable. (#10385)
* ARANGODB_UPGRADE_DURING_RESTORE env variable.

If this is set during startup, a database upgrade is performed
and an automatic restart (with the same process ID) is happening.

* Windows compatibility.

* Deactivate code for Windows. Back to unsetenv.
2019-11-13 10:00:48 +01:00
Jan Christoph Uhde 484e2971d6 C++17 clean up (#10287) 2019-11-11 08:49:54 +01:00
Jan 3bafcf2556
try to improve query shutdown in case DB server(s) can't be reached (#10359) 2019-11-05 20:41:51 +01:00
Jan 1a9ebff60a
always create `_analyzers` collection if missing (#10357) 2019-11-04 22:59:46 +01:00
Jan ae818e07d5
Feature/add cluster force one shard option (#10300) 2019-10-28 13:23:33 +01:00
Jan b124113190
abort write transactions (#10248) 2019-10-23 15:49:47 +02:00
jsteemann 1c7a60a587 pacify Valgrind on shutdown 2019-10-20 18:18:54 +02:00
Simon d526805e81 Bug fix/fix suspicous stuff (#10273) 2019-10-17 15:34:22 +02:00
Jan Christoph Uhde 99a5749ab7 more vocbase clean up (#10247) 2019-10-15 12:05:02 +02:00
Tobias Gödderz e2c84acfaf Use explicit default destructors where possible (#10166) 2019-10-04 15:58:30 +02:00
Max Neunhöffer 9b3f911cb5
Use execvp instead of execv and handle error. (#10102)
* Use execvp instead of execv and handle error.
* CHANGELOG.
2019-10-01 16:19:47 +02:00
Jan 2e293b85ca
Bug fix/cppcheck issues (#10105) 2019-09-30 09:27:33 +02:00
Dan Larkin-York a83c2323c9 Refactor ApplicationServer stack (#9965) 2019-09-25 17:31:59 +02:00
Jan Christoph Uhde 0b8c75c7b7 one shard db - devel (#9395) 2019-09-23 15:48:37 +02:00
Max Neunhöffer 71c8314107 Add more strong references to pthread stuff. (#10038)
* Add more strong references to pthread stuff.

* Circumvent libgcc/libmusl mistake in multi-threaded detection.

* CHANGELOG.
2019-09-19 12:36:15 +03:00
Max Neunhöffer 7d5313611f
Fix a shutdown busy loop after main if two exceptions collide. (#9978) 2019-09-11 14:47:49 +02:00
Jan aada04e75b
don't assert/crash when using an unknown collection/shard (#9959) 2019-09-11 12:03:13 +02:00
KVS85 4fc39dd4b3
Debug segfault reimplementation (#9940)
* Changed debugSegfault to debugTerminate

* Fix *nix compilation

* More data for broken reconnect

* Remove circumventCores completely

* Fix forgotten calls
2019-09-09 23:07:45 +03:00
Jan 9f078f363d
Bug fix/harden database creation against duplicate name (#9951) 2019-09-09 11:00:33 +02:00
Tobias Gödderz b7f8f01a22 Feature/split libarangoserver (#9670)
* Made usage of named variables for targets a little more consistent

* Bumped minimum cmake version

- removed policies superseded by minimum cmake version
- fixed static linking due to new policy CMP0060

* Fixed library suffixes for windows

* Extracted libs arango_mmfiles and _rocksdb from arangoserver

* Replaced target variables by constants

* Extracted arango_cluster_engine from arangoserver

* Extracted llhttp from arangoserver

* First successful split of arangoserver

* Moved enterprise files to enterprise

* Again only optionally include RestTestHandler and AcceptorUnixDomain

* Cleaned source files from other libraries

* Removed old commented sources

* Split off a small third library

* Fixed boost dependency for cluster engine

* Added some missing dependencies

* Added arango_geo dep, and -J on windows

* Began to split off an arango_graph lib

* Moved more files to arango_graph

* Do not set /J globally for ATL

* Moved more files to arango_graph

* Moved graph-RestHandlers to arango_graph

* Added arango_geo dependency to _mmfiles and _rocksdb

* Updated graph dependencies

* Split off arango_pregel

* Split off arango_aql

* Added missing boost_system dependency to pregel

* Split off arango_vocbase

* Cleanup

* Added missing boost_system dependency for arango_vocbase

* Split off arango_v8server

* Split off arango_utils

* Minor cleanup

* Split of arango_storage_engine

* Split off arango_indexes and arango_cache

* Fixed some dependencies

* Split off arango_replication

* Resolved two todos

* Split off arango_agency

* Reordered some statements

* Ordered dependency definitions alphabetically

* Cleaned some deps

* Break one cycle, comment on another

* Merge the remaining arangoserver_part[123] sources

* Moved some utils to vocbase to break cycles

* Added missing backtrace dependency to iresearch-s

* Added missing boost dependency

* Added dependency arango_indexes -> arango_geo

* Added deps to arango_cluster_engine, cleaned duplicate deps

* Broke remaining dependency cycles

* Actually, missed one cycle...

* Re-added include for Mac

* arango_cache needs SharedPRNG
2019-09-05 09:37:12 +02:00
Simon 0ee0cebb11 Non-Blocking inserts (#9823) 2019-08-30 09:17:58 +02:00
Simon 9a43b28f8f Improve ExecContext usability (#9806) 2019-08-28 19:05:23 +02:00
Dronplane 9559c8a80e Bug fix/internal issue #622 (#9781)
* Added analyzer cache invalidation for dropped database

* Fixed jslint reported errors
2019-08-21 23:34:24 +03:00
Frank Celler aa3d3f8e40
Feature/cleanup ccpcheck (#9665) 2019-08-12 11:11:49 +02:00
Jan 5576e44547 miscellaneous adjustments to pacify cppcheck (#9677) 2019-08-09 14:29:21 +02:00
Dan Larkin-York 3d0246cb18 Decentralize includes (#9623) 2019-08-06 15:32:09 +02:00
Jan 5bca5c1920
fix yet another undefined behavior (#9630) 2019-08-02 16:26:01 +02:00
Lars Maier ed496fe5dd Feature/hotbackup devel (#9495)
Hotbackup
2019-08-02 11:39:46 +02:00
KVS85 7945e01037
Fix ArangoSearch view creation errors (#9600) 2019-07-29 20:43:08 +03:00
Jan 973f00a997
Added startup error for bad temporary directory setting. (#9552)
* Added startup error for bad temporary directory setting.

  If the temporary directory (--temp.path) setting is identical to the database
  directory (`--database.directory`) this can eventually lead to data loss, as
  temporary files may be created inside the temporary directory, causing overwrites of
  existing database files/directories with the same names.
  Additionally the temporary directory may be cleaned at some point, and this would lead
  to an unintended cleanup of the database files/directories as well.
  Now, if the database directory and temporary directory are set to the same path, there
  will be a startup warning about potential data loss (though in ArangoDB 3.4 allowing to
  continue the startup - in 3.5 and higher we will abort the startup).
2019-07-24 12:47:42 +02:00
Jan b5932c2edf
make TTL indexes behave like other indexes on creation (#9546)
* make TTL indexes behave like other indexes on creation

if a TTL index is already present on a collection, the previous behavior
was to make subsequent calls to `ensureIndex` fail unconditionally with
the error "there can only be one ttl index per collection".

now, we are comparing the attributes of the to-be-created index with the
attributes of the existing TTL index and make it only fail when the
attributes differ. if the attributes are identical, the `ensureIndex`
call succeeds and returns the existing index.
2019-07-23 13:53:58 +02:00
Wilfried Goesgens ca0f2b8b86 All hail to the SI (#9445) 2019-07-19 13:52:12 +02:00
Andrey Abramov 5bc6eac636
bug-fix/internal-issue-#609 (#9464)
* allow to access last operation tick of the transaction

* modify IResearchLink to rely on last commtted tick

* get rid of writing arangosearch markers to WAL

* further implementation

* local changes in iresearch

* add recovery states

* properly handle link creation during recovery

* adjust test cases

* properly handle nested transactions

* ungreylist recovery tests

* add more recovery tests

* do not use transaction to pass recovery tick

* do not store recoveryTick in MMFilesRecoveryState

* adjust tests

* fix mmfiles

* cleanup

* add context validity check

* fix tests

* fix more tests

* ensure subscription is not being released during commit

* address review comment

* final cleanup

* fix crash

* fix tests

* address test failures

* address review comments

* address review comments

* properly set recovery tick even if no recovery happened
2019-07-19 11:10:56 +03:00
Jan 300b8e58f4
Bug fix/fix duplicate actions (#9452) 2019-07-19 09:17:19 +02:00
Simon e5507d840f Feature/comm task refactor (#9426) 2019-07-16 09:43:25 +02:00
Andrey Abramov f2ff365206
bug-fix/fix-arangosearch-recovery-mmfiles (#9448)
* start maintenance arangosearch threads after recovery is done

* ensure flush subscription is not being deallocated while in use

* add some tests

* properly determine storage engine

* adjust default view options

* stick to old index meta format in 3.5

* address test failures

* and cluster tests
2019-07-11 11:42:08 +03:00
Tobias Gödderz f501e00e9d Bug fix/add shard id to replication client identifier (#9366) 2019-07-08 14:03:42 +02:00
Andrey Abramov 174b54b036
Bug fix/internal issue #605 (#9421)
* release tick before committing changes to arangosearch

* fix tests

* address review comments

* address review comments

* add comment
2019-07-08 14:07:11 +03:00
Andrey Abramov 595728ec1b
Bug fix/internal issue #605 (#9405)
* fix recovery

* update iresearch

* more fixes

* address review comments
2019-07-05 13:14:37 +03:00
Jan 1a58cc2213
add VelocyPackHelper::equal method (#9389) 2019-07-03 12:15:11 +02:00
jsteemann b175853bae fix invalid logId 2019-07-02 16:48:06 +02:00
Jan 9cb08ded92
make the comparison functions unambiguous (#9349)
* make the comparison functions unambiguous

* added @kaveh's suggestion
2019-07-01 16:35:28 +02:00