git.iko.soy/distributed-process
on remove-table-api help sign in
history ofsrc/Control/Distributed/Process/Node.hs
all commits
f28ea4bRemove Mx Data Tables. This API isn't used, is easy to replace with various other packages, and contains races that are leading to failing test cases in CI.Tim Watson d8c36b5Avoid loop when tracing received messages.Facundo Domínguez e2bfbebThrow NodeClosedException when the node controller is stopped.Facundo Domínguez 07c5422Stop using the transport for local communication.Facundo Domínguez deadc44Fix race in runProcess when it fails.Facundo Domínguez bd28fbffixup! Avoid racy blocking behavior in closeLocalNode.Alexander Vershilov 537e06dSuppress console messages complaining about closed nodes.Facundo Domínguez 1e1636fAvoid racy blocking behavior in closeLocalNode.Facundo Domínguez bbcbd8eunused bind.Mathieu Boespflug 04ba682Don't hold the localState MVar while calling potentially blocking operations.Facundo Domínguez 5336c01Merge pull request #166 from tweag/informative-register-excFacundo Domínguez a386058Merge pull request #188 from mboes/localnode-closed-stateAlexander Vershilov 07dbb9bMerge pull request #185 from tweag/DP-105Alexander Vershilov 8055a5fWorkaround for throwTo problem that can potentially block.Alexander Vershilov ced6592Revert "Named send is intra-node, so we can use the optimised routing / encoding"Alexander Vershilov f575a46closeLocalNode: extra comment about effect of closeEndPoint.Mathieu Boespflug 14bf51eDon't be pedantic about invalid local node state in modifyValidLocalState_.Mathieu Boespflug 2957cbaProperly set LocalNodeClosed as originally intended.Mathieu Boespflug bae76a7Pass whole node rather than just state to withValidLocalNode.Mathieu Boespflug d57e702fixup! Kill processes on a local node upon `closeLocalNode`.Mathieu Boespflug 2069756Kill processes on a local node upon `closeLocalNode`.Mathieu Boespflug 8f5b623Allow LocalNode's to be in closed state.Mathieu Boespflug d63e025Return size of the queue in ProcessInfo.Alexander Vershilov fe428eeMerge pull request #173 from tweag/usend-hdTim Watson a94af5cBreak feedback loop when tracing messages to the console.Facundo Domínguez d126187Implement unreliable send (usend).Facundo Domínguez 3b1335eMore informative ProcessRegistrationException.Mathieu Boespflug 85068c2Part 2 to making forkProcess exception-safe.Mathieu Boespflug cdf5af4Merge pull request #162 from tweag/forkprocess-async-excTim Watson 770da54Make forkProcess exception-safe.Mathieu Boespflug 426beefMake -Wall clean.Mathieu Boespflug baf2e20Keep "Internal" namespaces visible for now to help contributorsTim Watson e947090switch to explicit per-file LANGUAGE pragmas instead of defaultsTim Watson 3bfe610switch to explicit per-file LANGUAGE pragmas instead of defaultsTavis Rudd ea1b3b8Similar to getProcessInfo, these give developers a means to query node statistics for debugging or monitoring purposes without leaking internal details and process references.Roman Gonzalez fad3a08Merge branch 'mx' into developmentTim Watson dcaeb3dRefactor and introduce management API test suiteTim Watson ad1c63bBack out of the split between tracer and local mx event busTim Watson 4957534Management Agent API ImprovementsTim Watson dd2ddb5Build up an initial execution environment for agentsTim Watson fb085c6Relocate trace infrastructure beneath C.D.Process.ManagementTim Watson f2f5fa5TraceEvent => MxEventTim Watson 28beaa2remove support for disabling tracing (since we now fire events anyway)Tim Watson e2d0cf7Split the node management event bus up from the trace controllerTim Watson 5a3da9cProduce useful exit-from messages where possibleTim Watson 6bd32f8Node.runProcess propagates exceptions to its caller - [DP-58 #resolve]Tim Watson aa41971Explicitly enable the initial tracer processTim Watson 39fb200Additional support for remote tracingTim Watson c600749Do not use (re)register APIs to track trace clientsTim Watson af229cfcosmetic/docTim Watson