Commit graph

10 commits

Author SHA1 Message Date
Robert Nishihara
dad57e3b62 Convert actor data structures to C++. (#454) 2017-04-12 01:18:16 -07:00
Robert Nishihara
fb4525f833 Convert some local scheduler data structures to C++ STL. (#445)
* Convert more local scheduler data structures to C++ STL.

* Convert vector pointer to vector.

* Convert some of the UT_arrays to std::vector.

* Simplify worker vectors.

* Simplify remote_object and local_object containers.

* Change some unnecessary checks to DCHECK.
2017-04-10 21:02:36 -07:00
Robert Nishihara
05fd4c2c37 Changes to local scheduler client protocol. (#435)
* Make local scheduler clients receive reply upon registration.

* Fix tests and linting.
2017-04-07 23:03:37 -07:00
Robert Nishihara
fa363a5a3a Notify driver when a worker dies while executing a task. (#419)
* Notify driver when a worker dies while executing a task.

* Fix linting.

* Don't push error when local scheduler is cleaning up.
2017-04-06 00:02:39 -07:00
Stephanie Wang
93679df724 Stopped nodes can rejoin immediately (#428)
* Ignore deleted clients when reading address info from Redis

* Remove self from db_client table when exiting cleanly

* Fix valgrind test

* Do not call plasma_perform_release when disconnecting
2017-04-05 23:50:38 -07:00
Stephanie Wang
083e7a28ad Push an error to the driver when the workload hangs on ray.put reconstruction (#382)
* Fix worker blocked bug

* tmp

* Push an error to the driver on ray.put for non-driver tasks

* Fix result table tests

* Fix test, logging

* Address comments

* Fix suppression bug

* Fix redis module test

* Edit error message

* Get values in chunks during reconstruction

* Test case for driver ray.put errors

* Error for evicting ray.put objects from the driver

* Fix tests

* Reduce verbosity

* Documentation
2017-03-21 00:16:48 -07:00
Stephanie Wang
12c9618c0c Plasma and worker node failure. (#373)
* Failing test case

* Local scheduler exits cleanly after plasma store dies

* Tolerate one plasma store failure

* Tolerate plasma store failures on all nodes except head node

* Plasma manager heartbeats

* Component failure tests

* Don't run the helper for Python testing

* Fix C test

* Fix hanging plasma transfer test

* Fix python3

* Consolidate ClientConnection code

* Fix valgrind test

* fix c test

* We can restart worker nodes!

* Fix flatbuffers bug

* Address comments

* Only register actual workers with the local scheduler

* Fix bug

* Fix segfaults

* Add test case that tests for driver liveness, fix local scheduler bug

* Clean up after tests

* Allocate retry info on the stack

* Send SIGKILL before waiting

* Relax unit test conditions

* Driver liveness test case and documentation
2017-03-17 17:03:58 -07:00
Philipp Moritz
068429ffd8 Convert local scheduler messages to flatbuffers (#340)
* use flatbuffer messages for local scheduler

* make sure constructor gets called for C++ object ObjectInfoT

* fix typo

* fix Robert's comments

* Small change to actor test.

* fix valgrind error

* linting

* free notification

* fix

* valgrind

* fix valgrind

* fix other bugs

* valgrind fix

* fixes

* more fixes

* Small changes to comments.
2017-03-15 16:27:52 -07:00
Philipp Moritz
0b8d279ef2 Convert task_spec to flatbuffers (#255)
* convert Ray to C++

* convert task_spec to flatbuffers

* fix

* it compiles

* latest

* tests are passing

* task2 -> task

* fix

* fix

* fix

* fix

* fix

* linting

* fix valgrind

* upgrade flatbuffers

* use debug mode for valgrind

* fix naming and comments

* downgrade flatbuffers

* fix linting

* reintroduce TaskSpec_free

* rename TaskSpec -> TaskInfo

* refactoring

* linting
2017-03-05 02:05:02 -08:00
Philipp Moritz
793a102846 Make Ray code C++ compatible (#321)
* convert Ray to C++
* const correctness
2017-03-01 01:17:24 -08:00
Renamed from src/local_scheduler/local_scheduler.c (Browse further)