Re: Testing code in geode-junit and geode-dunit src/main

r been allowed to happen in the first place. Be patient and > we will get to a good clean end state. > > -Jake > > > > On Aug 10, 2018, at 9:37 AM, Kirk Lund wrote: > > > > In my opinion geode-junit and geode-dunit should be treated as testing > > frameworks tha

Re: Testing code in geode-junit and geode-dunit src/main

anks for finding the tests that were meant to test the framework. I have > refactored them into the appropriate test sources sets. > > More inline below: > > On Thu, Aug 9, 2018 at 3:56 PM Kirk Lund wrote: > > > Here are the tests that are no longer running because they've been

Re: August 2018 Board Report Volunteer

+1 looks good! On Thu, Aug 9, 2018 at 3:46 PM, Anthony Baker wrote: > This is a bit late so please review asap. Thanks! > > ## Description: > Apache Geode provides a database-like consistency model, reliable > transaction processing and a shared-nothing architecture to maintain > very low

Testing code in geode-junit and geode-dunit src/main

We now have geode-junit and geode-dunit modules which contain the testing frameworks and rules that we've been using in Geode tests. These are also being published to maven and have release jars. Yay! There is some code however that was moved from src/test of various modules into the src/main of

Re: Reverting bad commit

This should be fixed now. On Thu, Aug 9, 2018 at 10:35 AM, Kirk Lund wrote: > Looks like I introduced a bad commit. I'm working on reverting it. Sorry > for the inconvenience! >

Reverting bad commit

Looks like I introduced a bad commit. I'm working on reverting it. Sorry for the inconvenience!

Re: [geode-dev] Gfsh Start Server option "disable-exit-when-out-of-memory"

We might want to: 1) introduce disable-default-jvm-arguments and implement it to ensure that GFSH is not adding any -X args when this option is enabled 2) deprecate disable-exit-when-out-of-memory and then remove it in 2.0 On Tue, Aug 7, 2018 at 3:59 PM, Helena Bales wrote: > Hello Geode

AsyncEventListener concurrency bugs

Several AsyncEventListener implementations in our test code are not thread-safe and this is causing tests that use them to be flaky. I discovered this while trying to fix the flakiness in some disabled tests in AsyncEventListenerDUnitTest. The issue here is that AsyncEventListener.processEvents

Re: AsyncEventListenerDUnitTest

to work when I add the annotation: public void testParallelAsyncEventQueueMovePrimaryAndMoveItBackDuringDispatching() { 3) Flakiness which prompted the disabling of tests in #1 caused by concurrency bugs in the various implementations of AsyncEventListener. On Fri, Jul 27, 2018 at 12:10 PM, Kirk Lund wrote: > If you upda

Re: Proposed API for a ConcurrencyTestHelper

implemented it as a Rule, you could throw an exception if any > Runnables are still running when the test finishes, and make sure whatever > Executor you use is cleaned up. I like it! > > As for name, I agree with Kirk that "Helper" is not a particularly > descriptive name. Pe

Re: Proposed API for a ConcurrencyTestHelper

nnables are still running when the test finishes, and make sure whatever > Executor you use is cleaned up. I like it! > > As for name, I agree with Kirk that "Helper" is not a particularly > descriptive name. Perhaps you could call it AsyncTaskAssertionRule or such? > &g

Gradle OutOfMemoryErrors while building Geode

I've helped a couple people who hit OutOfMemoryError while building Geode with Gradle 4.8. If you see OutOfMemoryError while compiling then you've hit the same issue. Geode source contains a gradle.properties file with settings including: org.gradle.daemon = true org.gradle.jvmargs = -Xmx2048m

AsyncEventListenerDUnitTest

If you update to the new version of IntelliJ (or switch to using Eclipse), you may see an error message about AsyncEventListenerDUnitTest. I'm currently reworking this test to fix this issue. Eclipse: The code of method $deserializeLambda$(SerializedLambda) is exceeding the 65535 bytes limit

Re: Assertions in Geode

org.apache.geode.internal.Assert was introduced as a temporary stand-in for Java assertions when GemFire was built using a version of JDK that was pre-assertions. I think we probably should plan to delete it whether we use assertions or not. On Thu, Jul 19, 2018 at 1:07 PM, Udo Kohlmeyer wrote:

Re: Proposed API for a ConcurrencyTestHelper

I'd love to collaborate on improving ExecutorServiceRule. Let me know if you or anyone else is interested. Honestly, I can't remember why I had it default to one thread unless I was favoring very explicit and tight control over what resources the tests use. Unlimited sounds reasonable for this.

Re: How to setup IntelliJ

ffect the Gradle build > too. > > Please don’t check in any “fixes” to this IJ integration issue. > > > > -Jake > > > > > > > On Jul 18, 2018, at 10:54 AM, Kirk Lund wrote: > > > > > > More details: If I open up BasicDistributedTest, I

Re: Proposed API for a ConcurrencyTestHelper

I've actually been trying to remove most of the Util and Helper classes used by tests. My reasons are two-fold: 1) these classes tend to be very non-OOP, 2) they tend to introduce a new API to the test which then obscures what it's actually doing. When I look at a test, I don't want anything

Moving from gradle to maven

Gradle appears to not play well with IntelliJ unless the project is overly simple. I don't want to spend my days fighting with tools that don't work well together. Here's an interesting blog article about moving from gradle to maven: https://blog.philipphauer.de/moving-back-from-gradle-to-maven/

Re: How to setup IntelliJ

18 10:22:16 2018 -0700 GEODE-5363: Fixes issue with build in IJ IDEA. On Wed, Jul 18, 2018 at 10:29 AM, Kirk Lund wrote: > Yes, my head revision is... > > commit afc8dc8fca846d08581d8027f969ceadec911687 (*HEAD -> **develop*, > *origin/develop*, *origin/HEAD*) > &g

Re: How to setup IntelliJ

oved duplication and compile > fail. > > Anthony > > > > On Jul 18, 2018, at 9:55 AM, Kirk Lund wrote: > > > > Does anyone have any instructions on setting up IntelliJ with the latest > > changes? I’ve tried new project from sources and I’ve tried using > ./gradlew > > idea — neither works for me. Basically all non-unit tests end up not > being > > a source set. > >

How to setup IntelliJ

Does anyone have any instructions on setting up IntelliJ with the latest changes? I’ve tried new project from sources and I’ve tried using ./gradlew idea — neither works for me. Basically all non-unit tests end up not being a source set.

Re: [DISCUSS] Standardize use of awaitility to a higher timeout

we need but at least it'll be ok on very slow hardware. On Thu, Jul 12, 2018 at 10:21 AM, Kirk Lund wrote: > I recommend keeping this Awaitility with default timeout separate from any > Rules or MemberVM. I want to be able to use Awaitility in integration tests > and distributed tests t

Re: [DISCUSS] Standardize use of awaitility to a higher timeout

I recommend keeping this Awaitility with default timeout separate from any Rules or MemberVM. I want to be able to use Awaitility in integration tests and distributed tests that don't use MemberVM or the StarterRules. So from that POV, I'd prefer the GeodeAwaitility that Dan proposes. On Thu, Jul

Article and paper about Flaky tests

Here are two good resources about Flaky tests: How to Deal With and Eliminate Flaky Tests [1] An Empirical Analysis of Flaky Tests [2] [1] https://semaphoreci.com/community/tutorials/how-to-deal-with-and-eliminate-flaky-tests [2] http://mir.cs.illinois.edu/~qluo2/fse14LuoHEM.pdf

Re: [DISCUSS] When is a test not flaky anymore?

gt; > > > > +1 for fixing immediately. > > > > > > Since Dan is already trying to shake out more brittleness this seems to > > be > > > the right time to get rid of the flaky label. Let's just treat all test > > the > > > same and fix

Re: [DISCUSS] When is a test not flaky anymore?

I should add that I'm only in favor of deleting the category if we have a new policy of any failure means we have to fix the test and/or product code. Even if you think that failure is in a test that you or your team is not responsible for. That's no excuse to ignore a failure in your private

Re: [DISCUSS] When is a test not flaky anymore?

+1 go ahead and delete FlakyTest category but please NEVER go back to marking flaky tests with @Ignore or renaming JUnit 3 test methods -- in my opinion that was infinitely worse. The category was meant to be a quarantine for FIXING not for IGNORING. On Fri, Jul 6, 2018 at 7:26 AM, Anthony Baker

Rat failure

Here's the rat failure on develop. I'm not sure what to make of it or what changed... Unapproved licenses: /home/jenkins/jenkins-slave/workspace/Geode-nightly/notifications/4.8/release-features.rendered

Possible dunit RMI connection leak

If I run a dunit test in IntelliJ with Repeat Until Failure, it'll run about 825 times before it fails with java.net.ConnectException: Connection refused. I'm wondering if we have some sort of connection leak in dunit (specifically the RMI connections we use for invoking code on VMs).

Re: DISCUSS: Refactor test source set for integrationTest and distributedTest

Don't forget to add "acceptanceTest/" On Tue, Jun 26, 2018 at 3:36 PM, Jacob Barrett wrote: > I'd like to suggest that we refactor our current test source set, which > contains both unit, integration and distributed tests, into distinct source > sets, test, integrationTest, distributedTest.

Re: DISCUSS: Refactor test source set for integrationTest and distributedTest

+1 although I'm not sure I like "commonTest" On Tue, Jun 26, 2018 at 3:36 PM, Jacob Barrett wrote: > I'd like to suggest that we refactor our current test source set, which > contains both unit, integration and distributed tests, into distinct source > sets, test, integrationTest,

Re: Compilation errors in IntelliJ

-modules-does-not-exist- which didn't clear up until I upgraded from 2018.1 to 2018.1.5. IntelliJ 2018.1.5 plus JDK 1.8.0_172 seems to work together on latest develop. On Mon, Jun 25, 2018 at 2:02 PM, Kirk Lund wrote: > I'm also seeing two problems on the command-line. 1) occasionally a bu

Re: Compilation errors in IntelliJ

FAILED *<**===**--> 84% EXECUTING [2h 23m 18s]* On Mon, Jun 25, 2018 at 1:11 PM, Kirk Lund wrote: > I've pulled, done a clean build, recreated my IntelliJ project and still > see this compilation error in buildSrc: > > /Users/klund/dev/gemfire_CLEAN/open/buildSrc/src/test/java/

Re: Compilation errors in IntelliJ

ngs > as of 226e406. If you followed these instructions to uncheck "Create > separate modules per source" to get things working last week you might want > to reimport the gradle project with the default settings. > > -Dan > > > On Thu, Jun 21, 2018 at 2:40 PM, Kirk Lu

Re: Resolving geode-pulse compilation errors in IntelliJ

EnsureCorrectRunsWithProcessor location: class org.apache.geode.javac.TestCompiler Class EnsureCorrectRunsWithProcessor is in buildSrc. I'm not familiar with buildSrc so I'm not sure if it even needs to be built within IntelliJ for debugging Geode tests. On Thu, Jun 21, 2018 at 2:40 PM, Kirk Lund wrote

Re: OPLOG_MAGIC_SEQ_ID

eve that: > > DiskInitFile.OPLOG_MAGIC_SEQ_ID is used for *.if files > Oplog.OPLOG_MAGIC_SEQ_ID is used for all other oplog file types > > HTH, > Anthony > > > > > > On Jun 21, 2018, at 1:36 PM, Kirk Lund wrote: > > > > Please help me understand OPLOG_MAGIC_SEQ_ID in t

Resolving geode-pulse compilation errors in IntelliJ

After the recent Gradle upgrade, you might see some compilation errors in IntelliJ involving geode-pulse such as the following: /Users/klund/dev/gemfire/open/geode-pulse/src/main/java/org/ apache/geode/tools/pulse/internal/service/MemberDetailsService.java Error:(28, 46) java: package

OPLOG_MAGIC_SEQ_ID

Please help me understand OPLOG_MAGIC_SEQ_ID in the oplog code... There is a constant called OPLOG_MAGIC_SEQ_ID which is a byte representing the beginning of every oplog file. Makes sense so far. There are two different constants called OPLOG_MAGIC_SEQ_ID each with a different value. One is in

Revert recent Gradle upgrade

I think we should revert the recent Gradle upgrade. This change apparently causes IntelliJ to not be able to build Geode which prevents us from running any Tests in the IDE. I'd prefer to fix how IntelliJ works with our project but we cannot be without the ability to debug tests. I don't know

package org.springframework.stereotype does not exist

Last week, I was able to build and execute tests in IntelliJ, but after pulling latest changes on develop (include the Gradle upgrade) I'm now getting errors in every class that uses org.springframework classes. Any ideas how to fix this?

PR #1868: new ThreadMonitoring feature

Can a couple more folks please review PR #1868? My review feedback pertained only to using dependency injection instead of a singleton or static getter and Yossi has updated the code as I requested. My review doesn't really address the usefulness of the feature or the impact on features that were

Re: JIRA and Wiki Access

Hi Nick, I gave you permissions in Jira so you can create/edit/assign tickets. I'm not sure how to give you permissions in the wiki (anyone know?). Thanks, Kirk On Tue, May 8, 2018 at 10:12 AM, Nicholas Vallely wrote: > Update on this request: > > I now have a JIRA user

Re: Requesting Geode Wiki Edit Permissions

Done! Let me know if you need anything else. On Fri, May 4, 2018 at 4:55 PM, Dale Emery wrote: > Hello, > > I am requesting permission to add and edit Geode Wiki pages. > > My user ID on Apache Confluence is: demery > > Cheers, > Dale > > — > Dale Emery > dem...@pivotal.io >

Re: Requesting Jira permissions

Done! Thanks. On Fri, May 4, 2018 at 4:49 PM, Dale Emery wrote: > Hi, > > I am requesting Jira permission to assign Jira tickets to myself. > > My user ID is demery > > Cheers, > Dale > > — > Dale Emery > dem...@pivotal.io > > > >

Re: NoSubscriptionServersAvailableException

; https://github.com/spring-projects/spring-session-data- > geode/blob/master/samples/boot/gemfire/src/main/java/sample/client/ > IntegrationTestConfiguration.java#L112-L137 > > > I hit similar Exceptions in the Spring Session Data Geode client/server > integration test suite. > >

Re: [Proposal] add isBound() to CacheServer API

ideas to anyone that wants to solve the issue (this isn't related to what I'm currently working on so I'm done with it). On Thu, May 3, 2018 at 11:26 AM, Kirk Lund <kl...@apache.org> wrote: > The method CacheServer.isRunning() is actually invoked from within the > Accepto

[Proposal] add isBound() to CacheServer API

on CacheServer called isBound() or something like that: org.apache.geode.cache.server.CacheServer */*** * * Returns true if this cache server is bound to its port and accepting connections.* * */* *boolean isBound();* Any objections or suggestions? On Thu, May 3, 2018 at 11:03 AM, Kirk Lund <

Re: NoSubscriptionServersAvailableException

return !this.shutdownStarted *&& serverSock.isBound();* } Any opinions or alternatives? If I add "*&& serverSock.isBound();*" am I going to break isRunning? On Thu, May 3, 2018 at 9:33 AM, Kirk Lund <kl...@apache.org> wrote: > I have a test which starts a ser

NoSubscriptionServersAvailableException

I have a test which starts a server and then starts a client. But the client intermittently fails with NoSubscriptionServersAvailableException (see full stack below). Seems like there must be something asynchronous in the startup of a CacheServer that I need to wait for. Any ideas what I need to

Re: Better support for JSON gfsh results

+1 to improve gfsh and use jackson consistently for all our json parsing On Wed, May 2, 2018 at 3:49 PM, Patrick Rhomberg wrote: > I was under the impression that gfsh console output was intended as a more > "active" interface with an operator. To that end, improved

Re: Renamed last of Bug*Test classes to descriptive name RegressionTests

Feel free to rename, recategorize or alter any of these tests. Sorry for all the typos in my 1st email! On Mon, Apr 30, 2018 at 5:05 PM, Kirk Lund <kl...@apache.org> wrote: > I just merged in PR #1886 (https://github.com/apache/geode/pull/1886) > which renames the remaining regr

Re: [Proposal] geode-dunit sub module.

+1 I think I'm in favor of all of these changes as well. I don't know anything about buildship though. I do have a few dunit changes on a feature branch that I'd like to get merged to develop before we move dunit out of geode-core. I'll try to wrap up these changes by Tues or Wed. On Mon, Apr

Re: Index on Region

We should probably target removal of the "deprecated" cache xsd types/elements in Geode 2.0. I'm not sure if we can introduce cache-2.0.xsd before Geode 2.0 or not (anyone know?). On Mon, Apr 16, 2018 at 2:59 PM, Jinmei Liao wrote: > Simply searching "deprecated" in

Re: Requesting access to Geode CI Pipeline

You don't need access unless you need to pause/restart or modify the pipeline. Everyone should have read-only access: https://concourse.apachegeode-ci.info/ On Thu, Apr 12, 2018 at 2:57 PM, Dale Emery wrote: > Hello, > > I am a new hire at Pivotal. I would like access to the

Re: [DISCUSS]: Tests requiring external services

> :geode-assembly:acceptanceTest is a target of the precheckin task. > > > > If we want to expand the scope of the AcceptanceTest tag, we'll need to > go > > update the gradle to make sure these tests get picked up. > > > > On Wed, Apr 4, 2018 at 9:39 AM, Kirk Lund &l

Re: [DISCUSS]: Tests requiring external services

+1 to using AcceptanceTest category for the end-to-end JDBC connector service tests On Wed, Apr 4, 2018 at 8:41 AM, Nick Reich wrote: > Using AcceptanceTest category seems like a good solution at the moment to > me. > > On Tue, Apr 3, 2018 at 4:29 PM, Sean Goller

Re: [PROPOSAL] Deprecating DistributedSystem

We should probably consider moving ServerLauncher and LocatorLauncher from org.apache.geode.distributed to a different package (maybe org.apache.geode?). On Thu, Mar 29, 2018 at 10:32 AM, John Blum wrote: > Yes, framework and tooling. > > I see no reason why the functionality

Re: Membership log message questions

will clear up this problem and will more correctly match the intent of the > tests. > > On 3/22/18 7:58 PM, Kirk Lund wrote: > >> If anyone really wants to look at and run the test directly, it's >> *Bug41733DUnitTest* on develop. >> >> Alternatively, you could pu

Membership log message questions

I'm looking through logs from a DUnit test that takes longer to run than I think it should. It looks to me like membership is chewing up more time than it should, and I have questions about some membership log messages. Specifically, I'm worried about: 1) double-logging by vm1 when shutdown

Re: [Proposal] Gfsh Command Feature Flag Annotation

Adding or removing options from a SpringShell command may be a bit tricky. This might require two versions of the command -- one version with the option, one without -- and we then load and register only one at runtime. On Wed, Mar 21, 2018 at 4:55 PM, Swapnil Bawaskar

Re: Recreate Cache -- is it possible?

The non-daemon thread in a process launched with ServerLauncher is looping in waitOnServer. When you close the Cache, that loop exits and the ServerLauncher process exits. As Bruce pointed you, JUnit and the DUnit VMs have other non-daemon threads. You might need to alter

Re: Recreate Cache -- is it possible?

I'm actively working to remove direct usage of GemFireCacheImpl from all of our code. Is there a reason you have to directly use GemFireCacheImpl? This is going to add one more thing that I have to change. Maybe the VM exit is happening because you've forked a new VM with ServerLauncher or

Re: [Discussion] Improving Spotless to be Even More Beautiful

+1 On Mon, Mar 19, 2018 at 2:51 PM, Patrick Rhomberg wrote: > Hello all! > > I'm making another pass at patching up some of our smellier broken > windows, to mix metaphors. To that end, there are a few things I'd like to > add to spotless to more closely enforce

Re: Disabling DistributedRestoreSystemPropertiesTest for now

Thanks Jens! I was searching for the cause on Friday but couldn't find it. Looks like any static usage of ClusterStartupRule ends up launching DUnit VMs. Gradle JUnit class loads (but does not instantiate) *Test.java. The tests that have public static ClusterStartupRules result in

Please use Awaitilty instead of Thread.sleep

I can't believe we're still checking in new tests with Thread.sleep of 10 seconds instead of using Awaitility. org.apache.geode.management.internal.configuration.ClusterConfigLocatorRestartDUnitTest In both test methods: // Wait some time to reconnect Thread.sleep(1); Please don't

Please remember to search JIRA

When you work on something, please remember to search JIRA to see if someone else already filed a JIRA ticket for the work. I'm not sure how many open "old" tickets that we have like this, but here's one example: GEODE-2557: Upgrade jna dependency https://issues.apache.org/jira/browse/GEODE-2557

[PROPOSAL] use default value for validate-serializable-objects in dunit

I want to propose using the default value for validate-serializable-object in dunit tests instead of forcing it on for all dunit tests. I'm sympathetic to the reason why this was done: ensure that all existing code and future code will function properly with this feature enabled. Unfortunately

Re: jdbc schema is missing on apache

t; ./schema/lucene/lucene-1.0.xsd > ./schema/cache/cache-1.0.xsd > > > Anthony > > > > On Mar 7, 2018, at 10:08 AM, Kirk Lund <kl...@apache.org> wrote: > > > > Yep, I believe it should exist. Any ideas where and how to copy it from > > G

Re: Trying to cleanup dunit use of client cache

e. On Fri, Mar 9, 2018 at 3:32 PM, Kirk Lund <kl...@apache.org> wrote: > I keep running into things that the dunit tests are doing in a client > cache that cannot be done with non-deprecated APIs. > > Is there a non-deprecated way to create a PERSISTENT_REPLICATE region in > a

Simple serialize/deserialize test using DataSerializers

Does anyone know what Geode API(s) I should use instead of Apache Geode SerializationUtils to change the following test to use Geode DataSerializers? @Test public void serializesAndDeserializes() throws Exception { PartitionRegionConfig config = new PartitionRegionConfig(prId, path,

Re: jdbc schema is missing on apache

Yep, I believe it should exist. Any ideas where and how to copy it from Geode src so that it appears at http://geode.apache.org/schema/jdbc? On Wed, Mar 7, 2018 at 9:59 AM, Jinmei Liao wrote: > I am looking at some xml that specifies jdbc connector, the namespace is >

Re: Debugging dunit in intellij now fails?

looks like you have already have a process running on the port that > the debugger is trying to connect on. I'd try to netstat -an and kill any > process bound to that port. > > On Tue, Mar 6, 2018 at 1:06 PM, Kirk Lund <kl...@apache.org> wrote: > > > Anyone else see

Debugging dunit in intellij now fails?

Anyone else see this when attempting to use intellij debugger on dunit?? [locator] ERROR: transport error 202: bind failed: Address already in use [locator] ERROR: JDWP Transport dt_socket failed to initialize, TRANSPORT_INIT(510) [locator] JDWP exit error AGENT_ERROR_TRANSPORT_INIT(197): No

EvictionDUnitTest and PRCacheListenerWithInterestPolicyAllInvocationTest

I saw these fail in a precheckin. I'll dig into these and see if I can fix them. :geode-core:distributedTest Do not allow more than 4 test workers org.apache.geode.internal.cache.eviction.EvictionDUnitTest > testDummyInlineNCentralizedEviction FAILED java.lang.AssertionError: expected:<8>

DUnit tests failing in intellij

If I randomly pick a dunit test that was passing in my intellij a week ago, I find it failing now and eventually hit a hang. For example, try running PRBasicIndexCreationDUnitTest. I git two GREEN tests, three RED tests (caused by SerializationException) and then it hangs in

Re: Starter tickets

+1 for using starter and starter++ On Fri, Mar 2, 2018 at 1:45 PM, Joey McAllister wrote: > +1 to "starter" over "newbie." I think the ++ plan makes sense, too. > > Thanks for the initiative on this, Alexander. It seems like a very valuable > effort for the community. >

Re: [Discuss] Member Launcher API

You might want to also consider moving these Launcher classes from org.apache.geode.distributed to some other package such as one of the following: 1) org.apache.geode 2) org.apache.geode.api 3) org.apache.geode.launcher 4) org.apache.geode.management 5) org.apache.geode.process The existing

Re: Broken build?

Nevermind... On Thu, Feb 22, 2018 at 4:07 PM, Kirk Lund <kl...@apache.org> wrote: > My latest precheckin failed to compile with the following. This is on a > branch that I rebased on latest develop HEAD. Has this already been fixed? > > :gemfire-test:compileTestJava/tmp/buil

Broken build?

My latest precheckin failed to compile with the following. This is on a branch that I rebased on latest develop HEAD. Has this already been fixed? :gemfire-test:compileTestJava/tmp/build/ae3c03f4/gemfire/closed/gemfire-test/src/test/java/parReg/execute/FunctionServiceTest.java:500: error: cannot

Re: Naming unit tests and integration tests

examples (see the TBD someone else previously added to this wiki page) and then I was planning to move on to updating https://cwiki.apache.org/confluence/display/GEODE/Distributed-Unit-Tests Thanks, Kirk On Tue, Feb 6, 2018 at 9:24 AM, Kirk Lund <kl...@apache.org> wrote: > +1 for renaming

Re: [PROPOSAL] Deprecate running JMX Manager on non-locator members

his.. > > > > > > > I may even launch and connect additional servers using Gfsh within my > > IDE > > > (or using *Gfsh*) that are connected to my *Boot* server, particularly > if > > > that *Boot* server also embeds a *Locator*. > > > > &

Re: Logging in to concourse.apachegeode-ci.info

> > > > I don't think you can log in to that pipeline. What are you wanting to do > > to it? > > > > -Dan > > > > On Wed, Feb 21, 2018 at 9:15 AM, Kirk Lund <kl...@apache.org> wrote: > > > >> How are we supposed to login to >

Re: [PROPOSAL] Deprecate running JMX Manager on non-locator members

Most of the geode developers are working to get rid of singleton calls (and next up static code) with the goal of being able to have multiple Cache instances in one JVM. This is to facilitate writing geode and geode application tests in one JVM. This fits in with what John is saying as well.

Re: Logging in to concourse.apachegeode-ci.info

ub itself working for you as expected? Using GitHub > for auth on the pipeline worked for me without any issues. > > On Wed, Feb 21, 2018 at 9:15 AM, Kirk Lund <kl...@apache.org> wrote: > > > How are we supposed to login to > > https://concourse.apachegeode-ci.info/team

Logging in to concourse.apachegeode-ci.info

How are we supposed to login to https://concourse.apachegeode-ci.info/teams/main/login? I'm trying to use the "login with GitHub" option (I think this worked for me before) but it keeps failing in a couple different ways... latest failure was "failed to verify token", previous one was

Intermittent join-response hang

I'm intermittently seeing dunit tests hang (actually fails after timing out) at this point in joining a cluster: [locator] [info 2018/02/15 12:56:25.583 PST tid=71] This member is becoming coordinator [vm0] [info 2018/02/15 12:56:25.584 PST tid=20] Probable coordinator is still

Re: [PROPOSAL] Add mod and arithmetic functionality to OQL

+1 for adding these ops to OQL On Tue, Feb 6, 2018 at 11:45 AM, Anilkumar Gingade wrote: > +1 like the idea... > > On Tue, Feb 6, 2018 at 10:22 AM, Jason Huynh wrote: > > > This is a proposal to add the ability to execute arithmetic operations on > >

Re: Naming unit tests and integration tests

over all of the tests and rename them to follow the > conventions we come up with? It would take some time, and maybe access to > Pivotal's old Trac issue tracker in order to rename files with "Bug" and > numbers for a name. > > > > On 2/5/18 12:28 PM, Kirk L

Re: Naming unit tests and integration tests

name that it's using the > dunit infrastructure is often useful to me. Having to open a file to see > that it's a DistributedTest vs a regular unit test could make life more > difficult. Putting them in their own source tree would help with that. > > > On 2/5/18 11:49 AM, Kirk L

Re: Naming unit tests and integration tests

is sometimes better -- otherwise just add new tests to existing tests). On Mon, Feb 5, 2018 at 11:42 AM, Kirk Lund <kl...@apache.org> wrote: > The industry standard names for tests are FooTest and FooIntegrationTest > and I think we should stick to that for UnitTest and Int

MigrationClient and MigrationServer

I just ran across two classes in org.apache.geode.internal package: * MigrationClient * MigrationServer They have the javadoc @since GemFire 6.0.1 so I suspect these are no longer useful or relevant. Can I delete these classes? Thanks, Kirk

jacocoTestReport

gradle/code-analysis.gradle contains configuration for the jacoco plugin, but if I attempt to execute the standard jacoco task jacocoTestReport, our build just fails and says it doesn't know what that is. How do we run jacoco on the command line? I tried this: ./gradlew geode-core:test --tests

Re: build is broken

ld be discussed on dev-list > >> BEFORE adding them to Geode, 2) we shouldn't need to talk to MS Access > for > >> Cluster Configuration. > >> > >> On Tue, Jan 23, 2018 at 9:30 AM, Kirk Lund <kl...@pivotal.io> wrote: >

Re: build is broken

there are two problems with adding Jackcess: 1) all new dependencies should be discussed on dev-list BEFORE adding them to Geode, 2) we shouldn't need to talk to MS Access for Cluster Configuration. On Tue, Jan 23, 2018 at 9:30 AM, Kirk Lund <kl...@pivotal.io> wrote: > I just tried to comp

build is broken

I just tried to compile Geode develop. ClusterConfigurationService has two broken imports that don't exist: import com.healthmarketscience.rmiio.RemoteInputStream; import com.healthmarketscience.rmiio.RemoteInputStreamClient; Was there a new dependency added to Geode?

Re: Handling files and user.dir in tests

that's a much better strategy for end-to-end cluster tests. Thanks, Kirk On Tue, Jan 9, 2018 at 8:02 AM, Kirk Lund <kl...@apache.org> wrote: > Looks like everyone supports this change. I’m going to change the Jira > ticket into an epic and create a few sub-tasks. I’ll also specify

ExecutorServiceRule

GEODE-4239: Create JUnit Rule to facilitate reuse of ExecutorService pattern for concurrent integration tests I just merged GEODE-4239 to develop. This introduces a new JUnit rule called ExecutorServiceRule. The commit also updates about 8 tests to use this rule. ExecutorServiceRule provides a

Full diffs in Jira comments

I noticed that we are once again getting full diffs added to Jira comments for Pull Requests. (see https://issues.apache.org/jira/browse/GEODE-3965). I've reopened https://issues.apache.org/jira/browse/INFRA-15727.

Re: Handling files and user.dir in tests

8, 2018 at 10:23 AM, Jinmei Liao <jil...@pivotal.io> wrote: > > > > > +1 for always using absolute path in our product code. > > > > > > Also the server/locator launchers should be able to take a working-dir > as > > > parameter to store all the sta

Re: [DISCUSS] Benchmarks module package structure

> in a good level of proximity between the benchmarks and the code they are > testing. > > On Mon, Jan 8, 2018 at 3:15 PM, Kirk Lund <kl...@apache.org> wrote: > > > We'll probably end up writing benchmarks for classes or methods that are > > package-private,

<    1   2   3   4   5   6   7   8   9   10   >