devbv commented on a change in pull request #2744: Deploy kafka & zookeeper
cluster with ansible
URL:
https://github.com/apache/incubator-openwhisk/pull/2744#discussion_r152815994
##########
File path: tests/src/test/scala/whisk/core/limits/ActionLimitsTests.scala
##########
@@ -269,13 +269,13 @@ class ActionLimitsTests extends TestHelpers with
WskTestHelpers {
it should "be aborted when exceeding its memory limits" in
withAssetCleaner(wskprops) { (wp, assetHelper) =>
val name = "TestNodeJsMemoryExceeding"
assetHelper.withCleaner(wsk.action, name, confirmDelete = true) {
- val allowedMemory = 256.megabytes
+ val allowedMemory = 128.megabytes
val actionName = TestUtils.getTestActionFilename("memoryWithGC.js")
(action, _) =>
action.create(name, Some(actionName), memory = Some(allowedMemory))
}
- val run = wsk.action.invoke(name, Map("payload" -> 512.toJson))
+ val run = wsk.action.invoke(name, Map("payload" -> 256.toJson))
Review comment:
Travis failed ?Action limits should be aborted when exceeding its memory
limits? after I increased kafka instance number to 2. (All good in my local
environment & dist environment)
```
whisk.core.limits.ActionLimitsTests > Action limits should be aborted when
exceeding its memory limits STANDARD_OUT
check failed for activation 106409d09700435da409d09700835d69:
{"duration":315,"name":"TestNodeJsMemoryExceeding","subject":"guest","activationId":"106409d09700435da409d09700835d69","publish":false,"annotations":[{"key":"limits","value":{"timeout":60000,"memory":256,"logs":10}},{"key":"path","value":"guest/TestNodeJsMemoryExceeding"},{"key":"kind","value":"nodejs:6"}],"version":"0.0.1","response":{"result":{"error":"An
error has occurred: RangeError: Array buffer allocation
failed"},"success":false,"status":"action developer
error"},"end":1511323308912,"logs":["2017-11-22T04:01:48.83040658Z stdout:
helloEatMemory memory 512MB","2017-11-22T04:01:48.899278892Z stderr:
RangeError: Array buffer allocation failed","2017-11-22T04:01:48.899342652Z
stderr: at Buffer.Uint8Array (native)","2017-11-22T04:01:48.899350793Z stderr:
at FastBuffer (buffer.js:11:5)","2017-11-22T04:01:48.89935681Z stderr: at
createUnsafeBuffer (buffer.js:38:12)","2017-11-22T04:01:48.899362782Z stderr:
at new Buffer.alloc (buffer.js:130:12)","2017-11-22T04:01:48.899368458Z stderr:
at eat (eval at <anonymous> (/nodejsAction/runner.js:76:109),
<anonymous>:3:18)","2017-11-22T04:01:48.899379285Z stderr: at
NodeActionRunner.main [as userScriptMain] (eval at <anonymous>
(/nodejsAction/runner.js:76:109),
<anonymous>:11:5)","2017-11-22T04:01:48.899386376Z stderr: at
/nodejsAction/runner.js:95:45","2017-11-22T04:01:48.899391497Z stderr: at
NodeActionRunner.run
(/nodejsAction/runner.js:89:16)","2017-11-22T04:01:48.899397412Z stderr: at
doRun (/nodejsAction/src/service.js:155:31)","2017-11-22T04:01:48.899402993Z
stderr: at runCode
(/nodejsAction/src/service.js:111:20)"],"start":1511323308597,"namespace":"guest"}
Exception occurred during test execution:
org.scalatest.exceptions.TestFailedException: "An error has occurred:
RangeError: Array buffer allocation failed" was not equal to "The action
exhausted its memory and was aborted."
```
I doubt that travis suffers low free memory problem caused by memory
consumption of additional kafka&zookeeper instance. I also reduced kafka
default heap size to 0.5g with same reason.
But it had been changed before I changed kafka image, so I try rollback and
run Travis again.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
