6 2012-02-15T14:32:31+06:00 <pre><code> trait FutureHelpers extends WhenReady { implicit def convertFutureOfAkka[T](futureOfAkka: Future[T]) = new FutureConcept[T] { def value: Option[Either[Throwable, T]] = futureOfAkka.value def isExpired: Boolean = false // Akka Futures don't support the notion of a timeout def isCanceled: Boolean = false // Akka Futures don't support the notion of being canceled } } object FutureHelpers extends FutureHelpers</code></pre> 1 15016563 1822 false dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1822 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

ScalaTest: update to 1.8 and provide FutureHelpers

2012-04-24T22:55:13+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 1 2012-05-24T18:18:02+06:00 Currently % is banned because it must be followed by exactly two hex digits. It should be allowed, properly checked, and used to convert otherwise illegal names. 1 21971413 1984 false dYbsaSchCr4k9jeJe5cbCb 2123 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

ActorPath: allow escape sequences

2012-05-24T18:18:02+06:00 0.0 0.0 Small 1.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCb 2011-12-15T21:04:40+06:00 Error Kernel etc. 1 11504523 1525 false cw1KuKVAKr3OgceJe5afGb,dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr 1525 4 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Screencast for new ActorSystem

2012-04-24T05:12:36+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner dYbsaSchCr4k9jeJe5cbCb 2011-12-15T21:05:05+06:00 1 11504543 1526 false cw1KuKVAKr3OgceJe5afGb,dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr 1526 4 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Screencast for new addressing

2012-04-24T05:13:00+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner alYZpMVA0r3RWMeJe5afGb 2011-12-15T21:05:27+06:00 1 11504573 1527 false cw1KuKVAKr3OgceJe5afGb,alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr 1527 4 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Screencast for new Future API (SIP-14 / Scala 2.10)

2012-05-03T22:58:38+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2011-12-15T21:06:01+06:00 1 11504593 1528 false cw1KuKVAKr3OgceJe5afGb,d3-V0kbESr4l5SeJe5cbLr,da_DgGqT0r35syeJe5aVNr 1528 4 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Screencast for new Routing

2012-04-24T05:13:22+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 19 2012-04-03T18:41:08+06:00 1 18204864 1976 false 1104053 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1976 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

ømq: verify and if necessary backport 2.0 fixes to poll loop

2012-05-03T19:29:09+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn alYZpMVA0r3RWMeJe5afGb 1 2012-05-25T20:35:31+06:00 currently the routees are guarded against BD, but it should be router instead (i.e. check RouterConfig.routerDispatcher instead of Props.dispatcher) 1 22061403 2046 false 1079503 dYbsaSchCr4k9jeJe5cbCb,alYZpMVA0r3RWMeJe5afGb 2136 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 4 Test 0

disallow running “head” Router actor on BalancingDispatcher

2012-05-25T21:45:27+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-04-23T17:18:24+06:00 1 19367643 2009 false 934623 alYZpMVA0r3RWMeJe5afGb 2009 1 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add filterException to Camel tests to avoid stack traces in test output

2012-04-23T17:18:24+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang dP16y-gWmr37KceJe5aVNr 2011-09-21T00:12:13+06:00 (taken from akka-modules) This is the suggested way of doing multi-part requests on Camel's Jetty Component: h1. *Multi-part Form support* From Camel 2.3.0, camel-jetty support to multipart form post out of box. The submitted form-data are mapped into the message header. Camel-jetty creates an attachment for each uploaded file. The file name is mapped to the name of the attachment. The content type is set as the content type of the attachment file name. You can find the example here. Note: getName() functions as shown below in versions 2.5 and higher. In earlier versions you receive the temporary file name for the attachment instead <pre><code>// Set the jetty temp directory which store the file for multi part form // camel-jetty will clean up the file after it handled the request. // The option works rightly from Camel 2.4.0 getContext().getProperties().put("CamelJettyTempDir", "target"); from("jetty://http://localhost:{{port}}/test").process(new Processor() { public void process(Exchange exchange) throws Exception { Message in = exchange.getIn(); assertEquals("Get a wrong attachement size", 1, in.getAttachments().size()); // The file name is attachment id DataHandler data = in.getAttachment("NOTICE.txt"); assertNotNull("Should get the DataHandle NOTICE.txt", data); // This assert is wrong, but the correct content-type (application/octet-stream) // will not be returned until Jetty makes it available - currently the content-type // returned is just the default for FileDataHandler (for the implentation being used) //assertEquals("Get a wrong content type", "text/plain", data.getContentType()); assertEquals("Got the wrong name", "NOTICE.txt", data.getName()); assertTrue("We should get the data from the DataHandle", data.getDataSource() .getInputStream().available() > 0); // The other form date can be get from the message header exchange.getOut().setBody(in.getHeader("comment")); } }); </code></pre> However we're missing the getAttachment and Attachment method family in general in Akka-Camel's Message hence it's impossible to achieve the desired functionality through mentioned component. 0 6780873 1214 false 934623 dP16y-gWmr37KceJe5aVNr,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb,dIWZSyoser4ys0acwqjQXA 1214 3 dP16y-gWmr37KceJe5aVNr ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: Akka-Camel Message is not wrapping Camel Jetty Message's .getAttachment method (for multi-part form support)

2012-02-06T20:24:27+06:00 0.0 0.0 None 0.0 0.0 dP16y-gWmr37KceJe5aVNrkrassermkrassermkrasserm dP16y-gWmr37KceJe5aVNrkrassermkrassermkrasserm 7 2012-01-12T17:35:32+06:00 1 12750843 1634 false 934623 dRo0FKiOKr4yCYacwqjQWU,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1634 3 dRo0FKiOKr4yCYacwqjQWU ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: Fix online documentation to reflect changes after migration

2012-01-19T20:59:50+06:00 0.0 0.0 None 0.0 0.0 dRo0FKiOKr4yCYacwqjQWUpiotrgapiotrgaPiotr Gabryanczyk 7 2012-01-12T20:26:22+06:00 1 12759213 1640 false 934623 dRo0FKiOKr4yCYacwqjQWU,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb,cSuS7M4GKr4l7cacwqjQXA 1640 3 dRo0FKiOKr4yCYacwqjQWU ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: Add examples of all API calls from Java to test directory in order to make sure they are usable from Java

2012-01-19T20:59:51+06:00 0.0 0.0 None 0.0 0.0 dRo0FKiOKr4yCYacwqjQWUpiotrgapiotrgaPiotr Gabryanczyk 2012-01-19T06:32:52+06:00 https://github.com/jboner/akka/pull/217/files#r363042 1 13114033 1672 false 934623 dRo0FKiOKr4yCYacwqjQWU,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1672 3 dRo0FKiOKr4yCYacwqjQWU ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: ConsumerConfig - clarify the docs

2012-01-19T20:59:51+06:00 0.0 0.0 None 0.0 0.0 dRo0FKiOKr4yCYacwqjQWUpiotrgapiotrgaPiotr Gabryanczyk 7 2012-01-19T22:00:39+06:00 1 13163613 1688 false 934623 dRo0FKiOKr4yCYacwqjQWU,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1688 3 dRo0FKiOKr4yCYacwqjQWU ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: Write tests for DefaultCamel covering all methods

2012-01-20T01:25:33+06:00 0.0 0.0 None 0.0 0.0 dRo0FKiOKr4yCYacwqjQWUpiotrgapiotrgaPiotr Gabryanczyk dIWZSyoser4ys0acwqjQXA 7 2012-03-01T19:24:57+06:00 pull the config from the applicaiton config, falling back to akka-camel's reference.conf for ConsumerConfig 1 15982223 1877 false 934623 dIWZSyoser4ys0acwqjQXA,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1877 3 dIWZSyoser4ys0acwqjQXA ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Make ConsumerConfig available through application configuration

2012-03-01T19:24:57+06:00 0.0 0.0 None 0.0 0.0 dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXA 7 2012-03-17T18:13:42+06:00 1 17072053 1923 false 934623 dIWZSyoser4ys0acwqjQXA,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1923 3 dIWZSyoser4ys0acwqjQXA ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: rewrite ActivationTracker using FSM

2012-04-02T15:19:54+06:00 0.0 0.0 None 0.0 0.0 dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXA 7 2012-03-18T01:36:29+06:00 Rewrite register, unregister and activation for both consumers and producers in one solution. Use a deeper hierarchy for the supervising actors. 1 17077403 1924 false 934623 dIWZSyoser4ys0acwqjQXA,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1924 3 dIWZSyoser4ys0acwqjQXA ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: rewrite register, unregister, activation generically in a supervision hierarchy

2012-04-02T15:21:20+06:00 0.0 0.0 None 0.0 0.0 dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXA 7 2012-03-18T02:01:51+06:00 1 17077743 1925 false 934623 dIWZSyoser4ys0acwqjQXA,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1925 3 dIWZSyoser4ys0acwqjQXA ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: add test for concurrent registrations of Producers and Consumers

2012-03-18T02:01:51+06:00 0.0 0.0 None 0.0 0.0 dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXA 7 2012-03-18T02:03:24+06:00 1 17077783 1926 false 934623 dIWZSyoser4ys0acwqjQXA,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1926 3 dIWZSyoser4ys0acwqjQXA ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL:write more tests for synchronous process(exchange) method in ActorProducer

2012-03-18T02:03:24+06:00 0.0 0.0 None 0.0 0.0 dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXA 7 2012-03-18T02:25:17+06:00 1 17077893 1927 false 934623 dIWZSyoser4ys0acwqjQXA,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1927 3 dIWZSyoser4ys0acwqjQXA ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: Write a test that shows onRouteDefinition is called on ConsumerConfig

2012-03-18T02:25:17+06:00 0.0 0.0 None 0.0 0.0 dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXA 7 2012-03-18T15:44:43+06:00 1 17085053 1928 false 934623 dIWZSyoser4ys0acwqjQXA,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1928 3 dIWZSyoser4ys0acwqjQXA ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: move examples to sub project like other akka examples.

2012-03-18T15:44:43+06:00 0.0 0.0 None 0.0 0.0 dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dIWZSyoser4ys0acwqjQXARayRoestenburgRayRoestenburgRaymond Roestenburg dRo0FKiOKr4yCYacwqjQWU 7 2012-04-02T15:23:58+06:00 1 18106943 1970 false 934623 dRo0FKiOKr4yCYacwqjQWU,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1970 3 dRo0FKiOKr4yCYacwqjQWU ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Clean up/rename camel producer receiveXXXProduce methods to make them more readable for end users

2012-04-24T05:14:09+06:00 0.0 0.0 None 0.0 0.0 dRo0FKiOKr4yCYacwqjQWUpiotrgapiotrgaPiotr Gabryanczyk dRo0FKiOKr4yCYacwqjQWUpiotrgapiotrgaPiotr Gabryanczyk 7 2012-05-22T14:42:38+06:00 exceptions are logged when running tests Perhaps use TestEventListener, see AkkaSpec 1 21774873 2011 false 934623 d3-V0kbESr4l5SeJe5cbLr 2096 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: Suppress superfluous logging in tests

2012-05-22T14:42:38+06:00 0.0 0.0 Small 1.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall 7 2012-01-19T07:02:45+06:00 Add them to CamelExtension config. 1 13115093 1677 false 934623 dRo0FKiOKr4yCYacwqjQWU,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1677 5 dRo0FKiOKr4yCYacwqjQWU ddEDvgVAKr3QrUeJe5aVNr 0 New 0

CAMEL: Make conversions configurable

2012-02-06T20:24:27+06:00 0.0 0.0 None 0.0 0.0 dRo0FKiOKr4yCYacwqjQWUpiotrgapiotrgaPiotr Gabryanczyk 2011-07-29T16:42:22+06:00 https://github.com/harrah/browse 0 4572238 1065 false 1293373 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1065 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add scala x-ray to Akka docs

2012-05-03T04:48:42+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2012-02-07T15:46:42+06:00 1 14381263 1797 false 1293373 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1797 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Write bench for remote communication/network protocol

2012-05-03T05:11:55+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 1 2012-02-11T01:18:31+06:00 Hello, akka-actor jar don't include osgi headers (but in 1.3.1 it do). Can you include them in the future to allow usage of this great 2.0 version in an OSGI world ? Thank you 1 14760883 1812 false 1293373 bJMRtIvbOr4yzXacwqjQWU,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1812 3 bJMRtIvbOr4yzXacwqjQWU ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Regression ; missing osgi headers in META-INF/manifest.mf

2012-05-25T10:09:23+06:00 0.0 0.0 None 0.0 0.0 bJMRtIvbOr4yzXacwqjQWUmathieu.barcikowskimathieu.barcikowskimathieu.barcikowski 2012-03-20T21:14:41+06:00 There does not seem to be any support for TLS on the client or server side in the current akka API. There was a discussion of this on the mailing list https://groups.google.com/d/topic/akka-user/rbfJAUOAmv4/discussion PS. Please consider that TLS is improving in many ways with work from IETF DANE http://tools.ietf.org/wg/dane/ finishing, it can be used on the client with protocols such as WebID http://webid.info/ ( http://webid.info/spec ), and is on the whole a lot more powerful and useful than generally thought. Facebook and Google have moved to TLS to avoid man in the middle attacks in internet cafés with tools such as Firesheep. Btw. an interesting paper by Google engineers "Overclocking SSL" http://www.imperialviolet.org/2010/06/25/overclocking-ssl.html 1 17310283 1933 false 1293373 cRK_YcCPSr4zFzacwqjQWU,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb,cpvsuyLp4r36UZeJe5cbLr,cawvuADQmr4ArpacwqjQYw,cI7Z84IJGr4zluacwqjQXA,cUR2MkMu8r4ABcacwqjQYw 1933 3 cRK_YcCPSr4zFzacwqjQWU ddEDvgVAKr3QrUeJe5aVNr 0 New 0

missing SSL/TLS ( useful for https:// )

2012-05-09T02:52:58+06:00 0.0 0.0 None 0.0 0.0 cRK_YcCPSr4zFzacwqjQWUhenry.storyhenry.storyhenry.story 19 2012-03-30T13:15:39+06:00 inspired by https://github.com/knoldus/Akka2Bench 1 17995213 1968 false 1293373 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1968 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

zeromq: add performance test

2012-05-09T16:09:50+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-05-03T00:47:05+06:00 1 20296763 1934 false 1293373 alYZpMVA0r3RWMeJe5afGb 2037 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Improve & add code examples in the Akka docs (java & scala)

2012-05-03T00:47:05+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2012-05-25T01:11:13+06:00 * find ancestor with name or starts with name N * find any actor in tree with name N * etc. Brainstorm. 1 22006833 1969 false 1293373 cw1KuKVAKr3OgceJe5afGb 2128 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Research richer querying API for navigating/finding actors in a hierarchy

2012-05-25T01:11:13+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-05-25T14:11:47+06:00 Discuss. 1 22038313 1970 false 1293373 cw1KuKVAKr3OgceJe5afGb 2132 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Investigate in an exponential backoff policy for the supervisor strategies

2012-05-25T14:11:47+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-02-02T21:04:34+06:00 Suggested by Martin: <pre> %A John T. Feo, ed. %Z LLNL %T A Comparative Study of Parallel Programming Languages: The Salishan Problems %S Special Topics in Supercomputing %V 6 %I North Holland %C Amsterdam %D 1992 %K book, text, Hamming's Problem (primes), paraffins problem, skyline matrix, doctor's office problem, languages: Ada, C*, Haskell, Id, OCCAM, PCN, Scheme, SISAL, dynamic dataflow, functional, imperative, </pre> 1. Hamming's Problem --------------------- Given a set of primes {a,b,c..} of unknown size and an integer n, output in increasing order and without duplicates all integers of the form a^i * b^j *c^k ... <= n Observe that if r is in the output stream, then, a*r, b*r, c*r, ... <= n are also in the output stream. The problem tests a language's ability to express recursive stream computations and producer/consumer parallelism, and to support dynamic task creation. 2. Paraffins Problem --------------------- Given an integer n, output the chemical structure of all paraffin molecules for i<=n, without repetitions and in order of increasing size. Include all isomers, but no duplicates. The chemical formula for paraffin molecules is C[i]H[2i+2]. You may choose any representation for the molecules, so long as it clearly distinguishes among isomers. The problem addresses the representation of recursive tree structures, the creations and manipulation of those structures and nested loop parallelism. 3. The Doctor's Office ----------------------- Given a set of patients, a set of doctors, and a receptionist, model the following interactions: initially, all patients are well, and all doctors are in FIFO queue awaiting sick patients. At random times, patients become sick and enter a FIFO queue for treatment by one of the doctors. The receptionist handles the two queues, assigning patients to doctors in a first-in-first-out manner. Once a doctor and patient are paired, the doctor diagnoses the illness and cures the patient in a random amount of time. The patient is then released, and the doctor rejoins the doctor queue to await another patient. The output of the problem is intentionally unspecified. The problem tests the language`s ability to program a set of concurrent, asynchronous processes with circular dependencies. 4. Skyline Matrix Solver ------------------------- Solve the system of linear equations. Ax = b without pivoting where A is ann by n skyline matrix. A skyline matrix has nonzero values in row i in column k through i, 1 <= k <= i, and nonzero values in column k through j, 1 <= k <= j. The problem tests the ability to define array structures that include nonessential elements (i.e. the zeros), and given those structure, efficiency of parallel and iterative array computations. 1 14057023 1778 false 1293373 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1778 4 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Look into some (standard) benchmarks for Akka

2012-05-02T17:20:05+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2011-11-28T16:01:20+06:00 Docs here: http://ls.implicit.ly/#publishing 1 10529853 1403 false 1293373 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr,alYZpMVA0r3RWMeJe5afGb 1403 5 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Publish Akka under ls.implicit.ly

2012-05-03T04:43:45+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-05-02T20:42:13+06:00 * Remove: beanstalk:beanstalk-client:1.4.5 -> akka-beanstalk-mailbox * Remove: zkclient:zkclient:0.3 -> akka-zookeeper-mailbox * Remove: com.mongodb.asynch:bson-driver_2.9.0-1:0.2.9-1 -> akka-mongo-mailbox * Remove: com.mongodb.asynch:monog-driver_2.9.0-1:0.2.9-1 -> akka-mongo-mailbox * Remove: com.twitter:util-core:1.8.1 -> akka-mongo-mailbox * Embed the source into akka-zeromq: org.zeromq:zeromq-scala-binding_2.9.1:0.0.3 -> akka-zeromq * Embed the source of jmxclient (into what?) - needs to be an executable JAR (at least how it is used now in the 'akka-cluster' script) so it needs a MANIFEST.MF. 1 20277643 2026 false 1256893 alYZpMVA0r3RWMeJe5afGb 2033 1 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Akka on Maven Central

2012-05-25T01:16:27+06:00 0.0 0.0 Small 3.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2012-05-15T15:12:37+06:00 1 21290643 2039 false 1256893 alYZpMVA0r3RWMeJe5afGb 2075 1 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Pass an external ExecutionContext into ActorSystem?

2012-05-24T18:08:38+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2012-05-01T17:04:04+06:00 Also, rewrite the dataflow documentation!!! 1 20194843 2025 false 1256893 alYZpMVA0r3RWMeJe5afGb 2030 2 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

SIP-14 / Scala 2.10 - Make the dataflow API a pimped behavior on top of scala.concurrent.Future

2012-05-17T05:03:34+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2012-05-17T00:12:57+06:00 1 21435373 2050 false 1256893 alYZpMVA0r3RWMeJe5afGb 2085 2 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Switch to Scala 2.10 ExecutionContext

2012-05-17T05:03:39+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang a-olHMfbSr37OHeJe5aVNr 2012-05-18T21:53:07+06:00 1 21625053 2054 false 1256893 alYZpMVA0r3RWMeJe5afGb,a-olHMfbSr37OHeJe5aVNr 2090 2 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

ScalaDoc akka.util.cps

2012-05-18T21:53:07+06:00 0.0 0.0 Small 1.0 0.0 a-olHMfbSr37OHeJe5aVNrderekjwderekjwDerek Williams alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang alYZpMVA0r3RWMeJe5afGb 2012-05-25T18:51:16+06:00 Remove DeathWatch and replace it with something built on top of the following ideas: Keep a Set[ActorRef] in every Actor with its watchers Keep a Set[ActorRef] in every Actor the actors it's watching Implement Watch/Unwatch as System messages to the Actor that is going to be watched Change Terminated to: case class Terminated(@BeanProperty actor: ActorRef, knownDeath: Boolean) extends PossiblyHarmful On Actor Death publicize a Terminated(self, knownDeath = true) to all its listeners, and an Unwatch to every Actor that it is watching SystemMessages always end up in a DeadLetterActorRef if undeliverable (also applies to death, see cleanUp) DeadLetterActorRef response to Watch-messages by replying Terminated(wached, knownDeath = false) //Simply because we are not sure if it is dead or just never existed This means that the info about watching is local, and can be freely moved around in a cluster, we just need to make sure that the Watch-messages aren't just dropped. 1 22054623 2066 false 1256893 alYZpMVA0r3RWMeJe5afGb,cw1KuKVAKr3OgceJe5afGb 2134 2 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

DeathWatch 2.0

2012-05-25T21:47:27+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2011-02-03T15:25:34+06:00 0 3063743 644 false 1256893 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 644 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Java API to the FSM module

2012-05-23T15:56:15+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2011-10-07T15:07:56+06:00 0 7770923 1269 false 1256893 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1269 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Switch to Alex Ctries instead of ConcurrentHashMap

2012-05-24T17:42:11+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 1 2011-12-10T17:27:54+06:00 remote supervision means that if the Terminate() or ChildTerminated() messages are lost, something will get stuck. These messages must probably be retransmitted in that case, possible scheme: - supervisor sends Terminate() - supervisor does not get ChildTerminated() after X time => reparent the child to DavyJones - DavyJones periodically sends Terminate() and subscribes to DeathWatch for all its “children” and releases them once he gets any sort of reply - supervisor forwards all ChildTerminated() he gets for “unknown” children to DavyJones 1 11243803 15 false 1256893 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1478 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

system messages shall “eventually” be delivered

2012-05-23T15:56:15+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-01-23T18:44:40+06:00 might want to have a section in the config where you can specify ActorSelections that will have all their received messages logged. This eliminates the need for LoggingReceive. 1 13323283 1715 false 1256893 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1715 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Fine grained loglevel configuration

2012-05-23T15:56:15+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang dYbsaSchCr4k9jeJe5cbCb 6 2012-03-16T00:25:41+06:00 * improve documentation wrt. async stop and how to avoid IllegalActorNameException * possibly add a dedicated method which ensures proper termination (by blocking) 1 16972523 1919 false 1256893 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1919 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

TestKit: improve synchronous actor stop facilities

2012-05-07T15:01:32+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 6 2012-03-26T15:52:48+06:00 possibly add some API where missing Use the blog post by Akka Essentials. 1 17623413 1952 false 1256893 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1952 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

DOC: describe TestKit usage from Java

2012-05-10T01:30:03+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn ar3pRos-Cr4Bo4acwqjQYw 2012-03-27T13:08:07+06:00 https://github.com/typesafehub/mima-free https://github.com/typesafehub/mima-free/blob/master/reporter/src/main/scala/ssol/tools/mima/cli/Main.scala 1 17697243 1953 false 1256893 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr,ar3pRos-Cr4Bo4acwqjQYw 1953 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 4 Test 0

Start using MiMa SBT plugin in release branches

2012-05-25T17:32:31+06:00 0.0 0.0 None 0.0 0.0 ar3pRos-Cr4Bo4acwqjQYwbjorn.antonsson@typesafe.combjorn.antonsson@typesafe.combjorn.antonsson@typesafe.com cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-04-04T04:36:41+06:00 1 18248633 1978 false 1256893 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb,dYbsaSchCr4k9jeJe5cbCb,cGuvdGNQer4A6GacwqjQXA 1978 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add SSL/TLS encryption support to Netty remote transport

2012-05-25T05:19:36+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2012-05-01T20:09:05+06:00 1 20203393 1976 false 1256893 alYZpMVA0r3RWMeJe5afGb 2031 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Make ActorSystem.actorOf non-blocking

2012-05-25T01:36:39+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2012-05-03T19:28:14+06:00 1 20380373 2030 false 1256893 cw1KuKVAKr3OgceJe5afGb 2041 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Ensure that Akka runs on Java 7

2012-05-03T22:57:39+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner a-olHMfbSr37OHeJe5aVNr 1 2012-05-08T03:42:42+06:00 The IOManager abstracts the usage of a socket, implying only TCP connections are supported. A client would like to use IOManager, but requires UDP DatagramSocket support. 1 20754613 1982 false 1256893 dTBQiWMiWr4ApgacwqjQYw,a-olHMfbSr37OHeJe5aVNr 2047 3 dTBQiWMiWr4ApgacwqjQYw ddEDvgVAKr3QrUeJe5aVNr 1 Accepted 0

Need UDP DatagramSocket support for Akka IOManager

2012-05-10T02:01:08+06:00 0.0 0.0 Small 1.0 0.0 a-olHMfbSr37OHeJe5aVNrderekjwderekjwDerek Williams dTBQiWMiWr4ApgacwqjQYwjamie.allenjamie.allenJamie Allen d3-V0kbESr4l5SeJe5cbLr 14 2012-05-09T13:15:12+06:00 in order not to clash with the real thing, especially since scala-library also has it as an external dependency 1 20897273 2036 false 1256893 dYbsaSchCr4k9jeJe5cbCb,d3-V0kbESr4l5SeJe5cbLr 2051 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Config: remove embedded sources and depend on com.typesafe:config artifact

2012-05-24T17:46:00+06:00 0.0 0.0 Small 1.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn ar3pRos-Cr4Bo4acwqjQYw 2012-05-15T01:09:35+06:00 The documentation is unclear about what guarantees can be given or why they can't. Flesh things out, after we decide what we can support long-term. 1 21254403 1982 false 1256893 ar3pRos-Cr4Bo4acwqjQYw 2071 3 ar3pRos-Cr4Bo4acwqjQYw ddEDvgVAKr3QrUeJe5aVNr 0 New 0

DOC: Describe semantics for watch/unwatch and PoisonPill

2012-05-25T21:49:51+06:00 0.0 0.0 Small 1.0 0.0 ar3pRos-Cr4Bo4acwqjQYwbjorn.antonsson@typesafe.combjorn.antonsson@typesafe.combjorn.antonsson@typesafe.com ar3pRos-Cr4Bo4acwqjQYwbjorn.antonsson@typesafe.combjorn.antonsson@typesafe.combjorn.antonsson@typesafe.com b3bGMs4nur3RAbeJe5aVNr 2012-05-15T15:43:27+06:00 1 21292363 2045 false 1256893 b3bGMs4nur3RAbeJe5aVNr 2076 3 b3bGMs4nur3RAbeJe5aVNr ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Streamline usage of parens on arity-0 methods with the Style Guide (once more)

2012-05-21T14:22:38+06:00 0.0 0.0 Small 1.0 0.0 b3bGMs4nur3RAbeJe5aVNrhseebergerhseebergerHeiko Seeberger b3bGMs4nur3RAbeJe5aVNrhseebergerhseebergerHeiko Seeberger dYbsaSchCr4k9jeJe5cbCb 2012-05-15T16:32:07+06:00 Either embed or ensure that the lib will be on Maven Central the day we push out 2.1. 1 21294983 2046 false 1256893 cw1KuKVAKr3OgceJe5afGb,dYbsaSchCr4k9jeJe5cbCb 2078 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Embed the source into akka-zeromq: org.zeromq:zeromq-scala-binding_2.9.1:0.0.3 in akka-zeromq

2012-05-15T17:38:27+06:00 0.0 0.0 Small 1.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-05-15T19:26:39+06:00 Both in Futures and Actors etc.. Let's brainstorm up a list of potential candidates in this ticket. Perhaps use: http://www.websequencediagrams.com/ Makes it very easy. 1 21305633 2048 false 1256893 cw1KuKVAKr3OgceJe5afGb 2081 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

DOC: create sequence diagrams for the most complicated usage flows

2012-05-20T20:16:00+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 16 2012-05-18T15:07:27+06:00 In DeamonMsgCreate there is both Props and Deploy, and Props holds a Deploy. Investigate if it would be possible to remove the Deploy in DaemonMsgCreate and only use Props. Similar with RouterConfig. It is included both in Props and in Deploy. 1 21589163 1984 false 1256893 d3-V0kbESr4l5SeJe5cbLr 2086 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Improve Deploy, Props and RouterConfig associations

2012-05-21T18:29:57+06:00 0.0 0.0 Small 1.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall dYbsaSchCr4k9jeJe5cbCb 6 2012-05-21T17:57:16+06:00 it’s a bug 1 21709153 2040 false 1256893 dYbsaSchCr4k9jeJe5cbCb 2093 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 1 Accepted 0

TestKit: TestActorRef.receive() should take a sender, defaulting to deadLetters

2012-05-25T21:31:29+06:00 0.0 0.0 Small 1.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-05-22T00:23:10+06:00 1 21738833 2057 false 1256893 alYZpMVA0r3RWMeJe5afGb 2094 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Scala 2.10 - create a MessageDispatcherConfigurator that represents the default global execution context

2012-05-22T01:20:10+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 8 2012-05-22T12:59:30+06:00 We should merge the doc generation changes to get google analytics on the stable docs as well. 1 21769533 2041 false 1256893 ar3pRos-Cr4Bo4acwqjQYw 2095 3 ar3pRos-Cr4Bo4acwqjQYw ddEDvgVAKr3QrUeJe5aVNr 0 New 0

DOC: Google analytics on stable docs as well

2012-05-25T21:45:55+06:00 0.0 0.0 Small 1.0 0.0 ar3pRos-Cr4Bo4acwqjQYwbjorn.antonsson@typesafe.combjorn.antonsson@typesafe.combjorn.antonsson@typesafe.com 2012-05-22T14:55:26+06:00 When there is a sbt-ls for sbt 0.11.3 we should enable it again. Commented out in project/plugins.sbt and ls.sbt 1 21775703 2058 false 1256893 d3-V0kbESr4l5SeJe5cbLr 2097 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 0 New 0

enable ls-sbt for sbt 0.11.3

2012-05-22T18:57:52+06:00 0.0 0.0 Small 1.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall alYZpMVA0r3RWMeJe5afGb 1 2012-05-22T17:17:35+06:00 they are only accessible from ActorContext, i.e. only from within an actor 1 21784623 2060 false 1256893 dYbsaSchCr4k9jeJe5cbCb,alYZpMVA0r3RWMeJe5afGb 2099 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 1 Accepted 0

Actor: make watch/unwatch synchronous

2012-05-24T22:37:53+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-05-22T19:10:28+06:00 Not clear if it works for Java API. If so, document both. If not, document that as well. Current ScalaDoc: <pre> The Stash trait enables an actor to temporarily stash away messages that can not or should not be handled using the actor's current behavior. Example: class ActorWithProtocol extends Actor with Stash { def receive = { case "open" ⇒ unstashAll { case "write" ⇒ // do writing... case "close" ⇒ unstashAll() context.unbecome() case msg ⇒ stash() } case "done" ⇒ // done case msg ⇒ stash() } } Note that the Stash trait can only be used together with actors that have a deque-based mailbox. Actors can be configured to use a deque-based mailbox using a configuration like the following (see the documentation on dispatchers on how to configure a custom dispatcher): akka { actor { my-custom-dispatcher { mailbox-type = "akka.dispatch.UnboundedDequeBasedMailbox" } } } Note that the Stash trait must be mixed into (a subclass of) the Actor trait before any trait/class that overrides the preRestart callback. This means it's not possible to write Actor with MyActor with Stash if MyActor overrides preRestart. </pre> 1 21794403 2062 false 1256893 cw1KuKVAKr3OgceJe5afGb 2101 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 1 Accepted 0

DOC: Add docs about Stash

2012-05-23T15:48:48+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner alYZpMVA0r3RWMeJe5afGb 2012-05-23T00:55:39+06:00 https://github.com/netty/netty/issues/356 1 21826263 2063 false 1256893 alYZpMVA0r3RWMeJe5afGb 2102 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Bug in HWT

2012-05-25T14:38:31+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 14 2012-05-24T17:12:12+06:00 The publish tasks should fail if running under Java 1.7. 1 21966473 2042 false 1256893 dYbsaSchCr4k9jeJe5cbCb 2119 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

publish: verify correct java version

2012-05-25T21:13:25+06:00 0.0 0.0 Small 1.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-05-25T01:02:51+06:00 Now the user needs to do this: context.actorOf(Props[MyChild].withRouter(RoundRobinRouter(1).withSupervisorStrategy(…))) Not that nice. 1 22006293 2064 false 1256893 cw1KuKVAKr3OgceJe5afGb 2127 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create a DSL for creating middle man supervisors in an actor hierarchy

2012-05-25T01:02:51+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-05-25T01:16:06+06:00 https://github.com/typesafehub/sbt-nexus-plugin 1 22007203 2065 false 1256893 cw1KuKVAKr3OgceJe5afGb 2129 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add the sbt-nexus-plugin for publishing to Maven Central

2012-05-25T01:16:27+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner dYbsaSchCr4k9jeJe5cbCb 1 2012-05-25T19:16:50+06:00 currently the class name is not logged (which it is for all other exceptions), but a blank line is inserted even for NoCause. 1 22056503 2067 false 1256893 dYbsaSchCr4k9jeJe5cbCb 2135 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 4 Test 0

improve logging of NoStackTrace exceptions

2012-05-25T19:56:46+06:00 0.0 0.0 Small 1.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2011-05-11T09:25:59+06:00 Create a reST :api: directive for linking to api from docs. 0 3823166 849 false 1256893 da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 849 4 da_DgGqT0r35syeJe5aVNr ddEDvgVAKr3QrUeJe5aVNr 1 Accepted 0

reST support for linking to API from docs

2012-05-23T15:56:15+06:00 0.0 0.0 None 0.0 0.0 da_DgGqT0r35syeJe5aVNrpvlugterpvlugterPeter VlugterTypesafehttp://typesafe.com d3-V0kbESr4l5SeJe5cbLr 11 2012-01-27T13:06:05+06:00 The durable mailboxes has error handling that reconnects on every failure. Add circuit breaker that can be used for all durable mailboxes to improve stability. 1 13665023 1734 false 1256893 d3-V0kbESr4l5SeJe5cbLr,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1734 4 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 1 Accepted 0

Circuit Breaker in durable mailboxes

2012-05-24T17:48:25+06:00 0.0 0.0 None 0.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall 2012-03-18T19:00:26+06:00 Needs Scala 2.10 I presume. 1 17087543 1929 false 1256893 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1929 4 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Scala 2.10 - Enable -Xlog-reflective-calls in AkkaBuild

2012-05-10T22:02:44+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2012-05-15T19:40:52+06:00 Add this custom search to the Akka Documentation (perhaps style it at the Google Custom Search creator site): <pre> <div id="cse" style="width: 100%;">Loading</div> <script src="http://www.google.com/jsapi" type="text/javascript"></script> <script type="text/javascript"> google.load('search', '1', {language : 'en', style : google.loader.themes.MINIMALIST}); google.setOnLoadCallback(function() { var customSearchOptions = {}; var customSearchControl = new google.search.CustomSearchControl( '003065520604945464838:izzukx8-qba', customSearchOptions); customSearchControl.setResultSetSize(google.search.Search.FILTERED_CSE_RESULTSET); customSearchControl.draw('cse'); }, true); </script> <style type="text/css"> .gsc-control-cse { font-family: Arial, sans-serif; border-color: #FFFFFF; background-color: #FFFFFF; } .gsc-control-cse .gsc-table-result { font-family: Arial, sans-serif; } input.gsc-input { border-color: #BBBBBB; } input.gsc-search-button { border-color: #000000; background-color: #333333; } .gsc-tabHeader.gsc-tabhInactive { border-color: #777777; background-color: #777777; } .gsc-tabHeader.gsc-tabhActive { border-color: #333333; background-color: #333333; } .gsc-tabsArea { border-color: #333333; } .gsc-webResult.gsc-result, .gsc-results .gsc-imageResult { border-color: #FFFFFF; background-color: #FFFFFF; } .gsc-webResult.gsc-result:hover, .gsc-imageResult:hover { border-color: #000000; background-color: #FFFFFF; } .gsc-webResult.gsc-result.gsc-promotion:hover { border-color: #000000; background-color: #FFFFFF; } .gs-webResult.gs-result a.gs-title:link, .gs-webResult.gs-result a.gs-title:link b, .gs-imageResult a.gs-title:link, .gs-imageResult a.gs-title:link b { color: #444444; } .gs-webResult.gs-result a.gs-title:visited, .gs-webResult.gs-result a.gs-title:visited b, .gs-imageResult a.gs-title:visited, .gs-imageResult a.gs-title:visited b { color: #444444; } .gs-webResult.gs-result a.gs-title:hover, .gs-webResult.gs-result a.gs-title:hover b, .gs-imageResult a.gs-title:hover, .gs-imageResult a.gs-title:hover b { color: #444444; } .gs-webResult.gs-result a.gs-title:active, .gs-webResult.gs-result a.gs-title:active b, .gs-imageResult a.gs-title:active, .gs-imageResult a.gs-title:active b { color: #777777; } .gsc-cursor-page { color: #444444; } a.gsc-trailing-more-results:link { color: #444444; } .gs-webResult .gs-snippet, .gs-imageResult .gs-snippet, .gs-fileFormatType { color: #333333; } .gs-webResult div.gs-visibleUrl, .gs-imageResult div.gs-visibleUrl { color: #000000; } .gs-webResult div.gs-visibleUrl-short { color: #000000; } .gs-webResult div.gs-visibleUrl-short { display: none; } .gs-webResult div.gs-visibleUrl-long { display: block; } .gs-promotion div.gs-visibleUrl-short { display: none; } .gs-promotion div.gs-visibleUrl-long { display: block; } .gsc-cursor-box { border-color: #FFFFFF; } .gsc-results .gsc-cursor-box .gsc-cursor-page { border-color: #777777; background-color: #FFFFFF; color: #444444; } .gsc-results .gsc-cursor-box .gsc-cursor-current-page { border-color: #333333; background-color: #333333; color: #444444; } .gsc-webResult.gsc-result.gsc-promotion { border-color: #CCCCCC; background-color: #E6E6E6; } .gsc-completion-title { color: #444444; } .gsc-completion-snippet { color: #333333; } .gs-promotion a.gs-title:link, .gs-promotion a.gs-title:link *, .gs-promotion .gs-snippet a:link { color: #0000CC; } .gs-promotion a.gs-title:visited, .gs-promotion a.gs-title:visited *, .gs-promotion .gs-snippet a:visited { color: #0000CC; } .gs-promotion a.gs-title:hover, .gs-promotion a.gs-title:hover *, .gs-promotion .gs-snippet a:hover { color: #444444; } .gs-promotion a.gs-title:active, .gs-promotion a.gs-title:active *, .gs-promotion .gs-snippet a:active { color: #00CC00; } .gs-promotion .gs-snippet, .gs-promotion .gs-title .gs-promotion-title-right, .gs-promotion .gs-title .gs-promotion-title-right * { color: #333333; } .gs-promotion .gs-visibleUrl, .gs-promotion .gs-visibleUrl-short { color: #00CC00; }</style> </pre> 1 21306863 2049 false 1256893 cw1KuKVAKr3OgceJe5afGb 2082 5 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add Google Custom Search to the docs

2012-05-17T05:03:16+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 16 2011-12-07T19:05:17+06:00 not only to speed up the process but also not to litter the memory with many representations of the same path, i.e. to canonicalize them 1 11083873 1446 false 765763 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1446 2 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

ActorPath: cache actorFor lookups for RemoteActorRefs

2012-05-03T04:44:03+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2011-11-25T15:49:29+06:00 1 10457623 1387 false 765763 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1387 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Optimize for method inlining (< 35 bytes) in the hot path

2011-12-17T21:58:46+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 1 2011-12-05T21:38:26+06:00 If GC pressure of toString is more relevant than size of the ActorPath objects, a possible improvement would be to add a field to each ChildActorPath which holds the total length of the string up to this point. Then toString would allocate a StringBuilder of the right size and fill it using .replace() from the end, no copying needed. 1 10948423 1434 false 765763 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1434 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

ActorPath: possible toString improvement

2012-05-11T11:53:31+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2011-12-06T16:07:58+06:00 1 11005753 1436 false 765763 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1436 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

In LocalActorRef, make the ActorCell swappable, so we can replace it with a DeadActorCell (GC)

2011-12-06T16:29:29+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2011-12-09T14:42:12+06:00 might just add fields for Class[_<:Actor] and closure and keep the right copy methods, and then have Props BE the factory; since ActorCell holds onto the Props anyway, why waste another object? 1 11195683 1461 false 765763 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1461 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

rework Props.creator to not require a closure when just using a Class or Manifest

2012-05-03T05:11:16+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 1 2011-12-09T20:55:08+06:00 normally, an actor is created from within its parent, which means that the childrenRefs are already correctly set-up and the Supervise() message serves no purpose. I didn’t remove it just yet in case there is a legitimate case (e.g. in clustering) for creating an actor “from the outside” by the provider, in which case Supervise() would really be needed. 1 11214973 1473 false 765763 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1473 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

elide Supervise message when possible

2011-12-09T20:55:08+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 16 2012-01-04T15:16:22+06:00 This is how it works now: "Addresses with the local system’s name, host and port are resolved to local actor refs, meaning that no network transport is used. Any other address will use remoting." Communication between two actor systems in same jvm should be optimized to local sends. 1 12300103 1603 false 765763 d3-V0kbESr4l5SeJe5cbLr,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1603 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Optimize remote send in same jvm

2012-01-04T15:16:22+06:00 0.0 0.0 None 0.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall 2012-02-07T15:48:07+06:00 Same IDL as Protobuf but 1-3 times performance. Should just be a matter of compiling using Protostuff instead of Protobuf and test it through. Need bench first. E.g. depends on #1797 http://code.google.com/p/protostuff/ 1 14381403 1798 false 765763 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1798 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Switch network protocol to use Protostuff

2012-05-03T05:11:55+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 19 2012-04-03T18:11:07+06:00 1 18202944 1975 false 765763 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1975 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

ømq: make polling scheme more intelligent for REQ/REP sockets

2012-04-03T18:11:07+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-05-03T00:23:47+06:00 https://github.com/peter-lawrey/Java-Thread-Affinity 1 20295013 1977 false 765763 alYZpMVA0r3RWMeJe5afGb 2035 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Investigate possibility to use Thread Affinity for dispatchers

2012-05-03T00:23:47+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 16 2012-03-29T13:14:50+06:00 The upstream handler will simply invoke tell() on the ActorRef; if that blocks, that TCP connection is blocked. Cannot be solved in general because of possibly misbehaving custom mailbox impls, but we can abort early in case of our own BoundedMailbox types. 1 17899473 1961 false 765763 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1961 4 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Sending to an Actor with BoundedMailbox via the network blocks Netty thread

2012-05-02T22:25:18+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2010-10-31T23:38:44+06:00 0 2465795 503 false 441778 alYZpMVA0r3RWMeJe5afGb 503 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add exponential back-off to restart fail of immortal actor

2012-05-02T17:18:45+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 1 2012-01-03T02:36:42+06:00 The question about message converters made me think: given an actor A which transforms requests, sends to B, receives replies, transforms them and sends them back to the original sender. This breaks down when there are multiple senders and the sender cannot be embedded within the message (fixed protocol). In that case the closest solution is to spawn one actor per sender, keep them in a map and use these for transforming and routing the replies. This could in principle be avoided if the sender reference would carry some meta-information. Attaching it to the sender and not to the Envelope means that forwarding messages works, even to multiple destinations, while retaining the information at the “right” spot. We might use the query part of the URI for this (currently simply disallowed): <code>akka://sys@host:port/user/service/forwarder?orig=akka://other@host2:port/user/client</code> When a reply is sent to this address, the “forwarder” actor could retrieve the meta-info and route the reply to the “client”. *IMPACT* This is of course not free: we would need to store the recipient address in the Envelope, which would typically be the same object as the actor’s path, but could potentially differ. In practice we would probably just extract the meta-info and store that or null, but the extra field is needed. And of course there would need to be an accessor method and an underlying field which is potentially set upon each message processed. 1 12224513 1602 false 441778 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1602 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

allow attaching of meta-information to sender reference

2012-05-03T04:49:45+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 8 2012-04-11T17:02:40+06:00 1 18618823 1994 false 441778 dYbsaSchCr4k9jeJe5cbCb 1994 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

DOCS: investigate JavaDoc-like output format for Java API

2012-04-23T22:06:57+06:00 0.0 0.0 Medium 3.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 16 2012-05-09T18:58:22+06:00 1 20915913 2037 false 441778 dYbsaSchCr4k9jeJe5cbCb 2053 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

create actor-based remote transport

2012-05-10T17:13:04+06:00 0.0 0.0 Small 1.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-05-10T22:02:17+06:00 1 21028383 2038 false 441778 alYZpMVA0r3RWMeJe5afGb 2060 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Investigate possibility to use Thread.stop(reason) to kill the processing of an actor

2012-05-10T22:02:17+06:00 0.0 0.0 Small 1.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 14 2012-03-21T18:27:11+06:00 * “normal” tests on fast and slow system * timing tests on unloaded system * performance tests on 48-core box * … 1 17383223 1939 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1939 2 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

setup Jenkins with swarm plugin to run various different tests

2012-03-21T19:08:12+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-03-21T18:35:24+06:00 1 17383643 1941 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1941 2 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

send all Jenkins failures to dedicated mailing list

2012-03-21T19:08:23+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-03-21T18:36:51+06:00 should contains the logs and the JVM’s gcore dump, stored on some network-accessible large disk and linked from the ticket 1 17383723 1942 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1942 2 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

have Jenkins automatically file a ticket when the build fails

2012-03-21T19:08:30+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-04-23T17:48:18+06:00 Split up scalable2 so we can do cluster testing locally without interfering with the other VMs. 1 19369373 2 false 1120813 ar3pRos-Cr4Bo4acwqjQYw 2012 2 ar3pRos-Cr4Bo4acwqjQYw ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Split scalable2 into multiple virtual machines

2012-04-23T17:51:23+06:00 0.0 0.0 Medium 3.0 0.0 ar3pRos-Cr4Bo4acwqjQYwbjorn.antonsson@typesafe.combjorn.antonsson@typesafe.combjorn.antonsson@typesafe.com 2011-02-03T02:47:57+06:00 Purpose: To introduce nondeterministic lag to provoke races and thread unsafe code Something like this: object BugShaker { val random = DynamicVariable(new Random(System.currentTimeMillis)) val on = false def shake() { random.nextInt(100) match { case 10 => Thread.sleepNanos(10) case 20 => Thread.yield } } } usage: if (BugShaker.on) BugShaker.shake //Will be optimized away by compiler if off 0 3061011 641 false 1120813 alYZpMVA0r3RWMeJe5afGb 641 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add a bugshaker that's run when testing

2012-04-27T06:06:06+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2012-02-03T16:46:48+06:00 should run tests with <code>-Dakka.test.tags.exclude=</code> 1 14145023 1781 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb,ar3pRos-Cr4Bo4acwqjQYw 1781 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 1 Accepted 0

Setup Jenkins on dedicated box for timing-sensitive tests

2012-04-23T17:52:50+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 18 2012-03-21T18:20:38+06:00 This should be trivially done by unifying both concepts and reusing the same actual test code. 1 17382803 1935 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1935 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Port all MultiJvm tests to MultiNode

2012-03-21T18:23:13+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 18 2012-03-21T18:23:03+06:00 * severed network connection * silently dropped network connection (i.e. no ChannelClosed generated) * spuriously unavailable network connection: hiccups * slowing down network connections * corrupted byte streams 1 17382973 1936 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1936 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add failure simulator for MultiNode tests

2012-03-21T18:23:03+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-03-21T18:46:18+06:00 50 to 500 nodes, on-demand 1 17384363 1945 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1945 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

automatize deployment of large test runs on EC2

2012-03-21T18:46:18+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 14 2012-03-30T13:14:58+06:00 * all durable mailboxes * zeromq * amqp 1 17995163 1967 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1967 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

add Jenkins installation which runs ALL tests

2012-03-30T13:14:58+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCb 2012-04-23T17:19:43+06:00 1 19367693 2010 false 1120813 dYbsaSchCr4k9jeJe5cbCb 2010 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Test: clean up test output (filter out expected log messages)

2012-04-23T17:19:43+06:00 0.0 0.0 Small 1.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 8 2012-05-08T19:21:48+06:00 1 20827113 1983 false 1120813 ar3pRos-Cr4Bo4acwqjQYw 2049 3 ar3pRos-Cr4Bo4acwqjQYw ddEDvgVAKr3QrUeJe5aVNr 0 New 0

DOC: Add documentation for Test Conductor Extension

2012-05-10T18:42:30+06:00 0.0 0.0 Small 1.0 0.0 ar3pRos-Cr4Bo4acwqjQYwbjorn.antonsson@typesafe.combjorn.antonsson@typesafe.combjorn.antonsson@typesafe.com 12 2012-05-09T15:00:57+06:00 Right now the file transfer only have a single chunk per file and client in transit at a time. The transfer rate is suffering. 1 20902573 2011 false 1120813 ar3pRos-Cr4Bo4acwqjQYw 2052 3 ar3pRos-Cr4Bo4acwqjQYw ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Enhancement: multi-node-test should have multiple file chunks in transit for performance reasons

2012-05-09T15:00:58+06:00 0.0 0.0 Small 1.0 0.0 ar3pRos-Cr4Bo4acwqjQYwbjorn.antonsson@typesafe.combjorn.antonsson@typesafe.combjorn.antonsson@typesafe.com ar3pRos-Cr4Bo4acwqjQYw 14 2012-05-10T21:22:41+06:00 the plugin created in #2057 shall be able to support the following: * start up N nodes, giving each the list of nodes and their own index * the first node fires up the TestConductor and puts the port of that one into this store * all others wait for that value to appear, retrieve it and start their TestConductor slaves with it All TestConductor usage is in client code, i.e. within the JARs under test; this might make the provision of the interface a bit more interesting given the need to segregate the class loaders. 1 21024703 2014 false 1120813 dYbsaSchCr4k9jeJe5cbCb,ar3pRos-Cr4Bo4acwqjQYw 2058 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

add simple key-value store to SBT plugin

2012-05-25T16:56:22+06:00 0.0 0.0 Small 1.0 0.0 ar3pRos-Cr4Bo4acwqjQYwbjorn.antonsson@typesafe.combjorn.antonsson@typesafe.combjorn.antonsson@typesafe.com dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn cw1KuKVAKr3OgceJe5afGb 3 2012-05-24T12:39:54+06:00 1 21944993 2018 false 1120813 d3-V0kbESr4l5SeJe5cbLr,cw1KuKVAKr3OgceJe5afGb 2109 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 4 Test 0

ClientDowningSpec

2012-05-25T21:47:33+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall 3 2012-05-24T12:40:33+06:00 1 21945033 2019 false 1120813 d3-V0kbESr4l5SeJe5cbLr 2110 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 0 New 0

GossipingAccrualFailureDetectorSpec

2012-05-24T16:02:59+06:00 0.0 0.0 Small 1.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall d3-V0kbESr4l5SeJe5cbLr 3 2012-05-24T12:41:03+06:00 1 21945053 2020 false 1120813 d3-V0kbESr4l5SeJe5cbLr 2111 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 4 Test 0

JoinTwoClusterSpec

2012-05-25T13:02:37+06:00 0.0 0.0 Small 1.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall cw1KuKVAKr3OgceJe5afGb 3 2012-05-24T12:41:30+06:00 1 21945103 2021 false 1120813 d3-V0kbESr4l5SeJe5cbLr,cw1KuKVAKr3OgceJe5afGb 2112 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 0 New 0

LeaderDowningSpec

2012-05-26T02:22:49+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall d3-V0kbESr4l5SeJe5cbLr 3 2012-05-24T12:42:11+06:00 1 21945123 2022 false 1120813 d3-V0kbESr4l5SeJe5cbLr,ar3pRos-Cr4Bo4acwqjQYw 2113 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 1 Accepted 0

LeaderElectionSpec

2012-05-25T13:22:16+06:00 0.0 0.0 Small 1.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall d3-V0kbESr4l5SeJe5cbLr 3 2012-05-24T12:42:56+06:00 1 21945203 2023 false 1120813 d3-V0kbESr4l5SeJe5cbLr 2114 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 4 Test 0

MembershipChangeListenerSpec

2012-05-25T13:01:55+06:00 0.0 0.0 Small 1.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall d3-V0kbESr4l5SeJe5cbLr 3 2012-05-24T12:43:30+06:00 1 21945223 2024 false 1120813 d3-V0kbESr4l5SeJe5cbLr 2115 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 4 Test 0

NodeMembershipSpec

2012-05-25T13:01:16+06:00 0.0 0.0 Small 1.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall dYbsaSchCr4k9jeJe5cbCb 18 2012-05-24T16:40:11+06:00 addition to the DSL: <pre><code> deployOn(master, """/blah.remote = "@slave@" """) deployOnAll("""/blah.remote = "@slave@" """) </code></pre> Role names are substitued if appearing between '@' signs, and the config section is injected into the Deployer before yielding control to the actual test procedure. 1 21961833 2026 false 1120813 dYbsaSchCr4k9jeJe5cbCb 2118 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 4 Test 0

MultiNodeSpec: allow deployment to nodes without hard-coding their address

2012-05-24T19:12:40+06:00 0.0 0.0 Small 1.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-05-24T17:19:48+06:00 1 21966893 2027 false 1120813 ar3pRos-Cr4Bo4acwqjQYw 2120 3 ar3pRos-Cr4Bo4acwqjQYw ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Make Jenkins build and test with both JDK 6 and JDK 7

2012-05-24T17:23:01+06:00 0.0 0.0 Small 2.0 0.0 ar3pRos-Cr4Bo4acwqjQYwbjorn.antonsson@typesafe.combjorn.antonsson@typesafe.combjorn.antonsson@typesafe.com 2012-05-24T17:23:01+06:00 Right now the multi-node version of the multi-jvm plugin only runs "java" and gets whatever java is the default. This should be configurable, maybe even per remote host. 1 21967023 2028 false 1120813 ar3pRos-Cr4Bo4acwqjQYw 2121 3 ar3pRos-Cr4Bo4acwqjQYw ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Make multi-jvm:multi-node-test use different JDKs

2012-05-24T17:23:01+06:00 0.0 0.0 Small 1.0 0.0 ar3pRos-Cr4Bo4acwqjQYwbjorn.antonsson@typesafe.combjorn.antonsson@typesafe.combjorn.antonsson@typesafe.com dYbsaSchCr4k9jeJe5cbCb 6 2012-05-24T17:55:03+06:00 MultiNodeSpec-based tests would all have actor systems named “MultiNodeSpec”. Fix AkkaSpec.getCallerName to drop more stuff and THROW IT TO THE GROUND! 1 21970343 2029 false 1120813 dYbsaSchCr4k9jeJe5cbCb 2122 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 4 Test 0

fix auto-generated actor system names

2012-05-24T17:57:06+06:00 0.0 0.0 Small 1.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCb 18 2012-05-24T23:11:15+06:00 should have better name and less parentheses 1 21998463 2030 false 1120813 dYbsaSchCr4k9jeJe5cbCb 2126 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

improve MultiNodeSpec.ifNode syntax

2012-05-25T12:30:07+06:00 0.0 0.0 Small 1.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-02-06T18:13:50+06:00 1. Set up Jenkins. 2. Create a "stress-test-suite" 3. Should run continuously. 4. Report when errors occur. 5. We might need to create some more stress tests. 6. Also run the multi-jvm tests as part of the "stress-test-suite". 1 14306083 1792 false 1120813 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1792 4 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Set up stress test on the 12-core box at Akka HQ

2012-04-23T17:41:54+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-14T17:48:05+06:00 1 16811973 1909 false 1120813 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1909 4 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

If 'long-running' tests are enabled then the normal cluster tests are not run, they should

2012-03-21T19:10:14+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-21T18:34:07+06:00 setup Jenkins job which is triggered by pushing certain branch name patterns, which will then run all tests on that branch, including verifying binary compatibility agains 2.0, and finally merge onto release-2.0. 1 17383563 1940 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1940 4 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

add MigrationManager to automated merge process for 2.0.x series

2012-03-21T19:10:01+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-03-21T18:41:44+06:00 Test applications shall be run continuously, with monitoring console being displayed at Akka HQ at all times (to notice performance degradation, etc.). 1 17384033 1943 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1943 4 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

long-running tests with Typesafe Console

2012-03-21T19:10:50+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn d3-V0kbESr4l5SeJe5cbLr 2012-03-22T13:37:15+06:00 1 17446403 1948 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb,d3-V0kbESr4l5SeJe5cbLr 1948 4 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 1 Accepted 0

port cluster tests to multiJvm/multiNode

2012-05-26T02:22:49+06:00 0.0 0.0 None 7.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2011-08-15T15:34:20+06:00 Also: Rewrite the long-running tests so that they only take as long as they need - Use 'awaitCond' instead of arbitrary sleeps 0 4866742 1112 false 1120813 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1112 5 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Identify slow tests and improve their speed

2012-05-03T05:13:46+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 18 2012-03-21T18:24:36+06:00 verifying correct cluster membership determination under all conceivable conditions 1 17383053 1937 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1937 5 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Write a boat-load of cluster tests

2012-03-21T19:09:25+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-03-21T18:25:39+06:00 1 17383113 1938 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1938 5 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

write long-running load and performance tests

2012-03-21T19:09:47+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-03-21T18:45:07+06:00 and possibly apachebench, plus other “DoS” style checkers. 1 17384283 1944 false 1120813 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1944 5 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

add continuous tests with Apache JMeter

2012-03-21T19:10:34+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-05-23T16:15:27+06:00 Which tests to write? Should cover all cases and corner cases. 1 21869103 2026 false 418132 cw1KuKVAKr3OgceJe5afGb 2108 1 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create a test specification (compliance matrix) for the cluster membership/failure detector implementation

2012-05-23T16:15:27+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2011-08-17T18:39:03+06:00 ActorSerialization has been removed and SerializationProtocol has been pruned, need a new solution for serialization of actors. 0 4984802 1120 false 418132 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1120 2 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create a serialization strategy for Akka Actors

2012-02-06T20:32:36+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 1 2011-12-30T00:30:09+06:00 although it is not clear to me what should happen: if we signal unreachable actors as Terminated, they must also terminate before the node becomes reachable again, otherwise everything will be quite confusing … 1 12095753 1588 false 418132 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1588 2 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

DeathWatch: hook up with failure detector

2012-02-06T20:32:36+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 1 2012-02-06T01:07:23+06:00 1 14278103 1786 false 418132 d3-V0kbESr4l5SeJe5cbLr,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1786 2 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Define SerialVersionUID for Props and related

2012-03-16T22:29:47+06:00 0.0 0.0 None 0.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall 3 2011-06-16T16:35:35+06:00 Needs to be retrieved from Git history and adapted to the new clustering. 0 4179868 939 false 418132 cw1KuKVAKr3OgceJe5afGb,aUIV_Mzpir35KLeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cT24wEOxOr355MeJe5cbCb,cSxu3cXXKr4khHeJe5cbLr,d3-V0kbESr4l5SeJe5cbLr 939 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add internal API for metrics in Akka OSS cluster

2012-04-24T18:36:44+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 3 2011-06-16T16:36:20+06:00 Dependent on #939 * Push or pull? * ZK based? * P2P based? 0 4179884 940 false 418132 cw1KuKVAKr3OgceJe5afGb,aUIV_Mzpir35KLeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cT24wEOxOr355MeJe5cbCb,cSxu3cXXKr4khHeJe5cbLr 940 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add a way of transporting out relevant metrics info to the different metrics dependent routers

2012-04-24T18:35:12+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 3 2011-06-16T16:37:14+06:00 Dependent on #940 * Base it on internal API for metrics * Use Hyperic's Sigar Plus write tests. 0 4179898 941 false 418132 cw1KuKVAKr3OgceJe5afGb,aUIV_Mzpir35KLeJe5cbCb,da_DgGqT0r35syeJe5aVNr 941 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create LeastCPU router

2012-04-24T18:38:06+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 3 2011-06-16T16:38:07+06:00 Dependent on #940 * Base it on internal API for metrics * Perhaps use Hyperic's Sigar or do we have what we need in the System MBeans? Plus write tests. 0 4179908 942 false 418132 cw1KuKVAKr3OgceJe5afGb,aUIV_Mzpir35KLeJe5cbCb,da_DgGqT0r35syeJe5aVNr 942 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create LeastMem router

2012-04-24T18:38:27+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 3 2011-06-16T16:38:48+06:00 Dependent on #940 Plus write tests. 0 4179916 943 false 418132 cw1KuKVAKr3OgceJe5afGb,aUIV_Mzpir35KLeJe5cbCb,da_DgGqT0r35syeJe5aVNr 943 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create LeastMessages router

2012-04-24T18:38:46+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 3 2011-06-22T15:29:08+06:00 Either: 1. Compare the akka.conf for all nodes that are joining the cluster to ensure that they are equal 2. Feed each Akka node with a URI where to fetch the akka.conf, download it and then boot up with it Not the whole config, but the parts of it that makes sense. 0 4225870 962 false 418132 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr,alYZpMVA0r3RWMeJe5afGb,aUIV_Mzpir35KLeJe5cbCb 962 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Ensure that the akka.conf is semantically equivalent on all nodes

2012-05-23T16:02:19+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner dYbsaSchCr4k9jeJe5cbCb 2011-09-05T13:09:38+06:00 With the old remoting gone, there needs to be an easy way for two Akka applications to talk to each other, preferably reusing all the serialization and connection facilities. One idea is to create endpoint actors which talk to each other. 0 5991861 1165 false 418132 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1165 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

create remote client for decoupling Akka application (clusters)

2012-02-06T20:31:15+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 1 2011-12-09T22:56:10+06:00 1 11221993 1476 false 418132 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1476 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create /alias path with configured aliases

2012-01-17T17:02:36+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 16 2011-12-12T00:14:05+06:00 It shall be possible to specify a list of transports in the configuration file. All remote addresses will be parsed according to transport, which is identified by its protocol name. 1 11261443 1481 false 418132 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1481 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Remote: make transport fully pluggable

2011-12-19T17:22:56+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dT87-qXpar36xpeJe5cbLA 2011-12-20T06:20:16+06:00 Create a section of the conf that the Netty remoting will use to filter inbound messages for NAT:ed machines. 1 11716683 1552 false 418132 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb,dT87-qXpar36xpeJe5cbLA 1552 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 1 Accepted 0

NAT Firewall for remote actors

2012-05-10T23:07:50+06:00 0.0 0.0 None 0.0 0.0 dT87-qXpar36xpeJe5cbLAscott.clasenscott.clasenscott.clasen alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2012-02-08T00:44:53+06:00 Like the title says. 1 14428483 1802 false 418132 bd4mIEM_er4kUReJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1802 3 bd4mIEM_er4kUReJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

MultiJVM test harnesses need to be published

2012-02-08T02:44:13+06:00 0.0 0.0 None 0.0 0.0 bd4mIEM_er4kUReJe5cbCbDerek WyattDerek WyattDerek Wyatt 2012-03-14T17:47:19+06:00 for maintenance etc. tricky part is that he will still be able to gossip, but not receive gossip so his failure detector will detect all nodes as 'unreachable' how to force him to go through JOINING phase again? 1 16811903 1907 false 418132 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1907 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Implement support for DOWNING a node that is reachable

2012-03-14T17:47:19+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-14T17:47:46+06:00 1 16811953 1908 false 418132 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1908 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Implement support for re-JOINING a node that have been DOWN

2012-03-14T17:47:46+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-14T17:49:24+06:00 We need to wait until the protocol is fully stabilized. Should we use ScalaBuf? Or ProtoStuff? 1 16812103 1910 false 418132 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1910 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create Protobuf messages out of all the Gossip protocol classes

2012-03-14T17:49:24+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-14T17:50:31+06:00 <pre> trait MembershipChangeListener { def notify(members: SortedSet[Member]): Unit } </pre> 1 16812213 1911 false 418132 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1911 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Cluster MembershipChangeListener needs to have Java friendly API

2012-03-14T17:50:31+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-14T17:51:48+06:00 <pre> trait MetaDataChangeListener { def notify(meta: Map[String, Array[Byte]]): Unit } </pre> 1 16812283 1912 false 418132 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1912 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Cluster MetaDataChangeListener needs Java-friendly API

2012-03-14T17:51:48+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-14T17:52:27+06:00 1 16812323 1913 false 418132 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1913 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Implement management and notification of MetaDataChangeListener

2012-03-14T17:52:27+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-14T17:52:59+06:00 Can't just shut down. Node must first gossip an Leave command, wait for Leader to do proper Handoff and then await an Exit command before switching to Removed. 1 16812353 1914 false 418132 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1914 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Node#shutdown() is cheating

2012-03-14T17:52:59+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-14T17:53:45+06:00 So we can wait for user to invoke explicit DOWN. 1 16812393 1915 false 418132 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1915 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Should we let the leader auto-down every run (as it is now) or just every X seconds?

2012-03-14T17:53:45+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-14T17:55:05+06:00 Some stuff you can do: <pre> val node = Cluster(system) node.registerListener(new MembershipChangeListener { def notify(currentMemberRing: SortedSet[Member]) { .. } }) val membershipStatus = node.status // JOINING, UP, EXITING, LEAVING, REMOVED, DOWN if (node.convergence.isDefined) { .. } // do we have cluster ring convergence? if (node.isLeader) { .. } // is this node the leader? if (node.isAvailable) { .. } // is this node currently considered available? if (node.failureDetector.isAvailable(otherNodeAddress)) { .. } // is the other node available? val likelyhoodBeingDown = node.failureDetector.phi(otherNodeAddress) // how likely is it that he is down? </pre> 1 16812483 1916 false 418132 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1916 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

DOC: document Cluster's user API

2012-05-03T20:13:52+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner cw1KuKVAKr3OgceJe5afGb 2012-04-24T17:31:33+06:00 Document what is done so far. 1 19587233 2014 false 418132 cw1KuKVAKr3OgceJe5afGb 2014 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Document Cluster API and 'akka-cluster' admin tool

2012-04-25T23:04:46+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-04-24T18:46:34+06:00 Now only bash script. Do a straight port. 1 19596113 2017 false 418132 cw1KuKVAKr3OgceJe5afGb 2017 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create Windows (.bat) version of the 'akka-cluster' admin tool

2012-04-24T18:46:34+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-04-24T18:47:38+06:00 Finish the ClusterMBeanSpec.scala - now just a mock up. 1 19596193 2018 false 418132 cw1KuKVAKr3OgceJe5afGb 2018 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create tests for the Cluster JMX API

2012-04-24T18:47:38+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 3 2012-05-14T14:22:38+06:00 Current implementation is an simplification, using only mean value of the intervals. Investigate how it should be implemented according to the paper. If we decide that mean is good enough we should remove current deviation calculation. Jonas comment: "Another thing is how we use the FD. Ideally we want to use the actual phi value and react differently depending on how likely the node is down. That's the whole idea, that it is not a matter of true/false. It is ok as it is now, but we have the option of being more fine-grained in the decisions later. " 1 21209443 2020 false 418132 d3-V0kbESr4l5SeJe5cbLr 2066 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Improve phi in AccrualFailureDetector

2012-05-14T14:22:38+06:00 0.0 0.0 Small 1.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall 2012-05-15T15:48:33+06:00 Needs discussion and some more thinking. Is it what we want? Code: <pre> /** * Merges two Gossip instances including membership tables, meta-data tables and the VectorClock histories. */ def merge(that: Gossip): Gossip = { import Member.ordering // 1. merge vector clocks val mergedVClock = this.version merge that.version // 2. group all members by Address => Vector[Member] var membersGroupedByAddress = Map.empty[Address, Vector[Member]] (this.members ++ that.members) foreach { m ⇒ val ms = membersGroupedByAddress.get(m.address).getOrElse(Vector.empty[Member]) membersGroupedByAddress += (m.address -> (ms :+ m)) } // 3. merge members by selecting the single Member with highest MemberStatus out of the Member groups val mergedMembers = SortedSet.empty[Member] ++ membersGroupedByAddress.values.foldLeft(Vector.empty[Member]) { (acc, members) ⇒ acc :+ members.reduceLeft(Member.highestPriorityOf(_, _)) } // 4. merge meta-data val mergedMeta = this.meta ++ that.meta // 5. merge gossip overview val mergedOverview = GossipOverview( this.overview.seen ++ that.overview.seen, this.overview.unreachable ++ that.overview.unreachable) Gossip(mergedOverview, mergedMembers, mergedMeta, mergedVClock) } /** * Picks the Member with the highest "priority" MemberStatus. */ def highestPriorityOf(m1: Member, m2: Member): Member = (m1.status, m2.status) match { case (Removed, _) ⇒ m1 case (_, Removed) ⇒ m2 case (Down, _) ⇒ m1 case (_, Down) ⇒ m2 case (Exiting, _) ⇒ m1 case (_, Exiting) ⇒ m2 case (Leaving, _) ⇒ m1 case (_, Leaving) ⇒ m2 case (Up, Joining) ⇒ m1 case (Joining, Up) ⇒ m2 case (Joining, Joining) ⇒ m1 case (Up, Up) ⇒ m1 } </code> 1 21292583 2021 false 418132 cw1KuKVAKr3OgceJe5afGb 2077 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Challenge/Verify the current Cluster merge strategy

2012-05-15T15:48:33+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-05-23T15:59:15+06:00 Make the current routers cluster-aware through the use of a ClusterActorRefProvider. Meaning that they should hook in into the cluster membership and be able to use new nodes in the node ring automatically (without hardcoding the actor URL). These routers should all be supported: * round-robin * random * broadcast * direct * scatter-gather 1 21868053 2022 false 418132 cw1KuKVAKr3OgceJe5afGb 2103 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create clustered routers/routees

2012-05-23T15:59:15+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-05-23T16:08:09+06:00 If a node wants to connect to a cluster: cluster://system@host:port If a node wants to lookup an actor in a cluster: cluster://system@/path/to/actor 1 21868633 2023 false 418132 cw1KuKVAKr3OgceJe5afGb 2104 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create a cluster protocol for remoting

2012-05-23T16:19:10+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-05-23T16:10:38+06:00 Should be used together with a RemoteClient that listens in to the cluster membership to be able to interact with the clustered actor system from the outside without being part of it. 1 21868823 2025 false 418132 cw1KuKVAKr3OgceJe5afGb 2106 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add support for 'Watcher' nodes that can join cluster but the clustered actor system

2012-05-23T16:10:38+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner d3-V0kbESr4l5SeJe5cbLr 3 2012-05-24T12:49:44+06:00 singetonCluster (only one node) doesn't reach convergence. Best if it does since then there are no corner cases. It should have one member - itself. FIXME assert in NodeStartupSpec 1 21945413 2027 false 418132 d3-V0kbESr4l5SeJe5cbLr 2117 3 d3-V0kbESr4l5SeJe5cbLr ddEDvgVAKr3QrUeJe5aVNr 0 New 0

singletonCluster convergence

2012-05-24T16:02:56+06:00 0.0 0.0 Small 1.0 0.0 d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall d3-V0kbESr4l5SeJe5cbLrpatriknwpatriknwPatrik Nordwall 2011-06-16T16:40:48+06:00 Use this class: https://github.com/jboner/akka/blob/master/akka-actor/src/main/scala/akka/routing/ConsistentHash.scala Plus write tests. 0 4179926 944 false 418132 cw1KuKVAKr3OgceJe5afGb,aUIV_Mzpir35KLeJe5cbCb,da_DgGqT0r35syeJe5aVNr 944 4 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create ConsistentHashing router

2011-10-28T16:44:10+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2011-09-27T13:51:38+06:00 and implement wildcarding using underscores and return the deployment config with the longest match against the address. 0 7152763 1231 false 418132 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb,d3-V0kbESr4l5SeJe5cbLr 1231 4 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Create a Trie with the deployment ids

2012-01-31T14:35:21+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2011-10-04T19:27:24+06:00 <pre> akka { actor { deployment { my-actor { cluster { local-to-parent = on ... </pre> 0 7564283 1259 false 418132 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr,aR_tjSkIir4ipneJe5cbLA 1259 4 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add 'local-to-parent' config option to force actor to be instantiated on the same node as its parent

2012-02-06T20:31:00+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2011-10-06T14:14:57+06:00 Impl and test. See comments. 0 7697313 1267 false 600323 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1267 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Implement support 'nr-of-instances = "auto"' setting

2012-05-23T16:05:38+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 16 2011-10-17T15:11:05+06:00 Implement vector clock pruning using solution outlined in: http://wiki.basho.com/Vector-Clocks.html#Vector-Clock-Pruning 0 8382863 1289 false 600323 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1289 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add pruning of vector clocks

2012-05-23T16:05:38+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-27T19:39:35+06:00 1 17725883 1955 false 600323 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1955 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Look into how to support the upcoming JCache final API for persistence

2012-05-23T16:05:38+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-03-27T19:41:01+06:00 Both 'Interval Tree Clocks' and 'Dotted Version Vectors' keep the size of the history down a lot. So we would not need to implement VectorClock pruning, which is both hard to impl and can never be fully correct. h1. Interval Tree Clocks Paper: http://thor.info.uaic.ro/~vidrascu/MasterSD2/papers/cap3/itc2008.pdf Two different Java implementations that we could use or port to Scala: * https://github.com/haf/Interval-Tree-Clocks * http://code.google.com/p/itc4j/ h1. Dotted Version Vectors Paper: http://gsd.di.uminho.pt/members/vff/dotted-version-vectors-2012.pdf How Riak was improved: http://gsd.di.uminho.pt/members/tome/papers/inforum2011.pdf Erlang implementation we could port to Scala: https://github.com/ricardobcl/Dotted-Version-Vectors 1 17726043 1956 false 600323 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1956 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Research 'Interval Tree Clocks' and 'Dotted Version Vectors' as a replacement for our current VectorClock implementation

2012-05-23T16:05:38+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-05-09T19:28:28+06:00 Mentioned in the Cluster Spec to some extent. Also - see attachment for some interesting ideas how to evolve it. 1 20918343 2019 false 600323 cw1KuKVAKr3OgceJe5afGb 2054 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Design cluster rebalancing of actor partitions

2012-05-23T16:05:38+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2012-05-23T16:13:12+06:00 Instead of only current Push. To minimize data sent across the wire. See the cluster spec. 1 21868953 2020 false 600323 cw1KuKVAKr3OgceJe5afGb 2107 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Change cluster protocol to use a Push-Pull mechanism

2012-05-23T16:13:12+06:00 0.0 0.0 Small 1.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2009-11-24T15:49:07+06:00 To further enhance documentation and API/framework, it would be beneficial to identify and solve common tasks. i.e: * "I need to have a pool of actors to load-balance a stream of messages" * "I have a distributed transaction I need to perform" * "I need to schedule an event on a remote node" * "I need to solve a problem concurrently" Suggested patterns to implement: 1. Pattern: FilterDispatch Does: Forwards messages to other actors depending on the type of the messages 2. Pattern: LoadBalancer Does: Forwards messages to a set of other actors of type T given some algorithm (round-robin, whathaveyou) 3. Pattern: EventSourcing Does: Create a generic event-sourcing "framework". Something like: [[url:http://jonasboner.com/2009/02/12/event-sourcing-using-actors.htmlExampleName]] 4. Pattern: Replicator. Does: Mediator that sends state-changing messages to both original destination as well as a remote replica 5. Pattern: Master/Worker. Does: Sends out functions (Workers) to the data residing on different nodes to do data processing and then aggregates the data back to the original Master. Something like: [[url:http://jonasboner.com/2007/01/29/how-to-build-a-pojo-based-data-grid-using-open-terracotta.htmlExampleName]] 6. Pattern: Map/Reduce. Variation of Master/Worker. Useful? 7. Pattern: Audit/Logging Does: Logs messages it receives before forwarding them. 8. Pattern: Security policy enforcement Does: Actor that acts as a PEP/PDP given a configured policy, either forwards or doesn't Responsible for: * Authentication * Authorization * User attribute lookup * Confidentiality & Integrity 9: Pattern: Transformation Does: Matches incoming messages, and forwards them after optionally transforming the message (e.g. to remove sensitive information, provide interface versioning), etc. 0 665589 3 false 117390 bU2hegTuur4liieJe5cbLr 67 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Patterns for distributed systems

2011-12-25T03:19:53+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 2011-02-10T13:41:57+06:00 https://github.com/remeniuk/akka-cm-dispatcher 0 3114097 657 false 117390 657 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Bring in container-managed dispatcher impl into akka-dispatcher-contrib

2011-10-19T15:42:37+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner dYbsaSchCr4k9jeJe5cbCb 2011-03-20T19:24:32+06:00 It should be optionally possible to accumulate the stack trace even for messages which required queueing. There should be an upper limit, though. 0 3411310 11 false 117390 dYbsaSchCr4k9jeJe5cbCb 713 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

add stack accumulation to CallingThreadDispatcher

2011-10-19T15:42:56+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCb 2011-05-04T11:27:22+06:00 People keep forgetting to put initialize() last in the FSM constructor, so use DelayedInit to make that implicit. Calling it twice will not have bad effects. 0 3771516 835 false 117390 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 835 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Use DelayedInit to remove need for explicit initialize in FSM

2011-12-08T05:33:53+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2011-05-19T15:56:21+06:00 To allow multiple clients, started in sequential order but still maintain fair scheduling between them, so that not one clients "takes over" the service 0 3934787 876 false 117390 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 876 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Add concept of Mediator between remote client and server

2011-10-03T19:03:31+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2011-08-23T13:51:36+06:00 https://github.com/SandroGrzicic/ScalaBuff 0 5240021 1130 false 117390 alYZpMVA0r3RWMeJe5afGb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1130 3 alYZpMVA0r3RWMeJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Switch to ScalaBuff for Protobuf

2011-12-16T22:05:16+06:00 0.0 0.0 None 0.0 0.0 alYZpMVA0r3RWMeJe5afGbviktorklangviktorklangviktorklang 3 2011-10-04T19:30:04+06:00 From user: "Another important aspect of this is how actors should be redistributed in the cluster over time. In the example of the big tournament above the tables should be spread-out across the cluster initially, but as players are eliminated (and the number of table actors decrease) more and more tables should be moved closer to the tournament actor. This, IMO, describes some kind of `dynamic migration strategy` which would need to be defined using application-specific rules (or, once again, by setting up dynamic rules for how to allocate actors based on the hierarchy)." Need to discuss if it makes sense and if so how to do it. 0 7564453 1260 false 117390 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr,aR_tjSkIir4ipneJe5cbLA 1260 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Allocation and Migration Strategy for clustered actors

2012-05-25T14:57:29+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 1 2011-10-18T14:02:31+06:00 In order to cut down on (and keep low) object allocations when creating actors, there should be two varieties: * the basic Actor, which works using overriding methods and even does not allocate a closure for “receive” * the fancy Actor, which provides the low-level methods internally and fills in behavior a DSL (à la “uponBirth { ... }”, “uponRestart { ... }”, “become { ... }”, etc.) The reason is that we will spawn many basic actors when re-implementing ask(), and those don’t need a fancy DSL, not even the current receive(). We could keep the current receive method supported (defaulting to null) and provide a default receive(Any):Boolean which delegates to that. Then everyone could live on happily in peace. 0 8446293 1293 false 117390 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1293 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

ACTOR: split into bare-bones and DSL

2011-12-20T16:51:52+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCb 2011-12-31T18:06:24+06:00 use-case (outside actor): <pre><code> val results = actorRef askStream KickOffComputation for (r <- results) { doSomething } </code></pre> 1 12179833 1596 false 117390 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1596 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Patterns: create askStream(target, message): Iterator[Any]

2012-05-24T16:44:16+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 16 2012-01-18T02:04:26+06:00 The protocol is basically prepared, what is needed is a registry for actor systems and to enable reuse of RemoteSupport. 1 13023063 8 false 117390 dYbsaSchCr4k9jeJe5cbCb,da_DgGqT0r35syeJe5aVNr,cw1KuKVAKr3OgceJe5afGb 1658 3 dYbsaSchCr4k9jeJe5cbCb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

allow multiple ActorSystems to share a remote endpoint

2012-01-25T04:58:12+06:00 0.0 0.0 None 0.0 0.0 dYbsaSchCr4k9jeJe5cbCbrkuhnrkuhnrkuhn 2012-02-27T14:03:18+06:00 From Scott Clasen: <pre> Some food for thought for you... certainly on Heroku and Im guessing for many other cloud platforms, HTTP is definitely the first class transport, and using straight TCP is either more difficult or unsupported. Putting the added overhead of wrapping up/unwrapping the protobuf RPC in an HTTP wrapper, would there be any blockers you can think of that would prevent Akka from running on top of a remoting and clustering layer that did just that and wrapped the protobuf in HTTP and had netty pipelines that did the proper encoding/decoding such that everything else would function normally??? </pre> <pre> For TCP, we have an internal private alpha version of what we call TCPRouter, which we could use to test akka remoting/clustering via TCP. There are both command line tools and soon a java api to control the routing...so here is pseudocode of how it works. https://github.com/JacobVorreuter/heroku-routing Lets say we have an app deployed to heroku, that has one process type called 'akka' and we wanted to do a 32 node test. Pseudocode for(1 to 32) { i => val route = heroku routes:create // route is something like tcp://routes.heroku.com:12345 heroku routes:attach route "akka." + i } heroku scale akka=32 The akka app should then use the api to read attached routes for the app, and use that to seed the gossip protocol. Note that we arent very close to actually releasing this feature, but it will do for akka internal testing. </pre> <pre> There is a REST api sitting behind all of the commands in the command line tooling. We also have a java client wrapper for it, https://github.com/heroku/heroku.jar . (BTW we will include a Spray-can based connector once spray supports SSL clients). So for instance, an akka app could use the api to self scale based on load. There isnt a java wrapper for the routing api yet, since it isnt public, but it would be trivial to hit the REST api directly from an akka app to get the gossip seed. So without TCP routing we still couldn't even do a true HTTP based cluster app on Heroku, because without TCP routing, the individual nodes are not directly addressable, so all HTTP traffic flows through our routing layer and is non-sticky load balanced across nodes. However, with TCP routing, we can also use the tcp-route to talk over HTTP. So for instance....take an app with 2 web nodes (play 2 with akka 'http-remoting' enabled) and 2 akka nodes. We could do something like attach tcp routes to the 2 web processes and the 2 akka processes. Lets assume they are all speaking akka "http-remoting". they also grab the gossip seed via api. Now all the nodes are directly accessible for http based gossip/remoting/clustering, and the play web frontend is publicly accessible through http://my-app-name.herokuapp.com with traffic flowing through the normal path of our erlang routing/load balancing mesh. </pre> 1 15716183 1866 false 117390 cw1KuKVAKr3OgceJe5afGb,da_DgGqT0r35syeJe5aVNr 1866 3 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Investigate running Akka cluster on Heroku

2012-02-27T14:03:18+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner 2009-11-25T19:43:01+06:00 Would work nice with the: actor ! HotSwap(PF) Can perhaps even wrap it in a more high-level manager for actors. 0 670761 17 false 117390 cw1KuKVAKr3OgceJe5afGb 72 4 cw1KuKVAKr3OgceJe5afGb ddEDvgVAKr3QrUeJe5aVNr 0 New 0

Remote REPL (remote management of actors)

2011-10-19T15:43:01+06:00 0.0 0.0 None 0.0 0.0 cw1KuKVAKr3OgceJe5afGbjbonerjbonerJonas BonérTypesafe, Inc.http://typesafe.com+46733777123none@jboner