summaryrefslogtreecommitdiffstats
path: root/searchcore
Commit message (Collapse)AuthorAgeFilesLines
* Use gtestHenning Baldersheim2021-02-047-634/+640
|
* Call std::_Exit(0) if SIGINT or SIGTERM is received while loadingTor Egge2021-02-041-1/+43
| | | | data structures, replaying transaction log or performing startup reprocessing.
* Split to one test file per fixture.Henning Baldersheim2021-02-043-132/+151
|
* Refactor test for easier reuse.Henning Baldersheim2021-02-045-508/+565
|
* Merge pull request #16367 from vespa-engine/toregge/include-thread-as-neededHenning Baldersheim2021-02-033-0/+3
|\ | | | | Include thread as needed.
| * Include thread as needed.Tor Egge2021-02-033-0/+3
| |
* | Count on the right places.Henning Baldersheim2021-02-031-2/+13
| |
* | Remove the sync interface from Bucket Executor. Due to the nature of ↵Henning Baldersheim2021-02-027-32/+36
|/ | | | | | requiring a bucket lock it is very hard to get sync to work in a safe way. Instead the users must do their own accounting as they know their own threading model.
* Merge pull request #16348 from vespa-engine/balder/refactor-bucketmove-jobHenning Baldersheim2021-02-027-94/+98
|\ | | | | Factor out the scan iterator from the BucketMoveJob.
| * Factor out the scan iterator from the BucketMoveJob.Henning Baldersheim2021-02-027-94/+98
| |
* | Extend proton_config_fetcher unit test.Tor Egge2021-02-021-0/+30
| |
* | Extend unit test for ReconfigParams.Tor Egge2021-02-022-0/+8
| |
* | Update class comments.Tor Egge2021-02-022-2/+4
| |
* | More descriptive test name.Tor Egge2021-02-021-1/+1
| |
* | Improve handling of changed alloc config.Tor Egge2021-02-0233-157/+342
|/
* Pretend that compaction strategy config can be changed without needing restart.Tor Egge2021-02-011-2/+2
|
* Wire in config for compaction strategy.Tor Egge2021-02-0114-13/+61
|
* Use correct name for parameter.Henning Baldersheim2021-02-011-1/+1
|
* Add test that verifies that concurrency is upheld during triggerFlush too.Henning Baldersheim2021-01-311-55/+51
|
* Use vespalib::duration for timeoutsHenning Baldersheim2021-01-3120-113/+85
|
* Non-empty _pendingPrune will shortcut throttleing when doing flushAll.Henning Baldersheim2021-01-312-5/+5
| | | | That is done on triggerFlush and prepareForRestart. That is incorrect as it causes a memory surge.
* Use vespalib::duration for time.Henning Baldersheim2021-01-304-26/+26
|
* Various code health.Henning Baldersheim2021-01-3016-98/+64
|
* Wire reporting of attribute resource usage all the way to the cluster ↵Geir Storli2021-01-2910-24/+110
| | | | controller via the host info API.
* Fix forward declarations.Tor Egge2021-01-292-2/+2
|
* Merge pull request #16241 from vespa-engine/balder/some-code-healthHenning Baldersheim2021-01-2814-55/+45
|\ | | | | Some code healt by exposing what is necessary only
| * Some code healt by exposing what is necessary onlyHenning Baldersheim2021-01-2614-55/+45
| |
* | Still test both variantsHenning Baldersheim2021-01-281-1/+1
| |
* | In order to ensure that lid stats are sampled at the right time ensure that ↵Henning Baldersheim2021-01-289-26/+39
| | | | | | | | | | | | | | you do it the *next* time you are in the master thread. This is to ensure that the sync call has taken effect. This also keep the iterator creation logic in one method enhancing readability.
* | Ensure all tasks are queued on the master executor so compaction will pack ↵Henning Baldersheim2021-01-274-2/+9
| | | | | | | | the top too.
* | Copy the maintenance subdb so you can safely use its members without ↵Henning Baldersheim2021-01-271-2/+3
| | | | | | | | worrying that they might change under youor feet.
* | Use shared_ptr to ensure lifetime is sufficient.Henning Baldersheim2021-01-2711-49/+49
| |
* | Call onStop on the jobHenning Baldersheim2021-01-271-2/+5
| |
* | Add a proper way for controlled stop.Henning Baldersheim2021-01-273-1/+6
| |
* | Check _stopped in execute, and guard it by the lock.Henning Baldersheim2021-01-272-7/+13
|/
* gc last usages of closuresHenning Baldersheim2021-01-267-14/+10
|
* Kill jobs when syncing feedview.Henning Baldersheim2021-01-261-0/+1
|
* Remove the syncFeedView upcall as it has no purpose other than forcing it ↵Henning Baldersheim2021-01-267-34/+3
| | | | | | more than necessary. Instead do it in the main documentdb after all subds are reconfigured.
* - Kill maintenance jobs prior to doing reconfig.Henning Baldersheim2021-01-253-17/+25
| | | | - Start if they are stopped.
* - Stop maintenance jobs prior to reconfiguring.Henning Baldersheim2021-01-251-14/+9
| | | | - Always create new LidSpaceCompactionHandlers.
* Merge pull request #16215 from vespa-engine/balder/use-lambda-over-closure-2Henning Baldersheim2021-01-2516-108/+37
|\ | | | | Closure -> lambda
| * Closure -> lambdaHenning Baldersheim2021-01-2516-108/+37
| |
* | Use lambda over closure.Henning Baldersheim2021-01-252-19/+18
|/
* No need to sync the feedviews when we reconfigure searchable.Henning Baldersheim2021-01-251-2/+2
|
* Check gid and bucketid inside the correct master thread to ensure you do the ↵Henning Baldersheim2021-01-253-19/+39
| | | | right thing.
* Sync executor when destructing the jobs.Henning Baldersheim2021-01-252-1/+4
|
* Simplify by avoiding closure.Henning Baldersheim2021-01-251-26/+37
|
* Do not switch default yet.Henning Baldersheim2021-01-241-1/+1
|
* - The metastore must be read in the master thread.Henning Baldersheim2021-01-2412-47/+34
| | | | - The getMetaData belong to the handler, not the iterator.
* Merge pull request #16187 from vespa-engine/balder/use-the-correct-lidHenning Baldersheim2021-01-232-2/+23
|\ | | | | Use the correct lid when looking up the meta data.