HADOOP-16371: Option to disable GCM for SSL connections when running on Java 8 #970

sahilTakiar · 2019-06-14T18:06:08Z

HADOOP-16371: Option to disable GCM for SSL connections when running on Java 8

Changes:

Patch is based on the merged patch from HADOOP-16050
Decided a better name for SSLSocketFactoryEx would be DelegatingSSLSocketFactory because the class is not OpenSSL specific (e.g. it is capable of just delegating to the JSSE)
Add a bunch of code comments to DelegatingSSLSocketFactory
Documented fs.s3a.ssl.channel.mode in performance.md and core-default.xml
If a user tries to configure OpenSSL as the mode via fs.s3a.ssl.channel.mode an UnsupportedOperationException is thrown.

Testing Done:

Ran all S3 tests mvn verify and S3 scale tests mvn verify -Dparallel-tests -Dscale -DtestsThreadCount=16 (did not have S3Guard or kms tests setup)
Ran TestDelegatingSSLSocketFactory on Ubuntu and OSX with -Pnative and confirmed the test passes on both systems (on OSX it is skipped, on Ubuntu it actually runs)
Ran the ABFS tests against "East US 2" and the only failure was ITestGetNameSpaceEnabled.testNonXNSAccount (known issue)
Ran mvn package -Pdist -DskipTests -Dmaven.javadoc.skip=true -DskipShade, un-tarred hadoop-dist/target/hadoop-3.3.0-SNAPSHOT.tar.gz and ran ./bin/hadoop fs -ls s3a://[my-bucket-name]/ successfully and that I could upload and read a file to S3 using the CLI without the wildlfy jar on the classpath

hadoop-yetus · 2019-06-14T20:13:30Z

🎊 +1 overall

Vote	Subsystem	Runtime	Comment
0	reexec	92	Docker mode activated.
		_ Prechecks _
+1	dupname	1	No case conflicting files found.
+1	@author	0	The patch does not contain any @author tags.
+1	test4tests	0	The patch appears to include 4 new or modified test files.
		_ trunk Compile Tests _
0	mvndep	30	Maven dependency ordering for branch
+1	mvninstall	1139	trunk passed
+1	compile	991	trunk passed
+1	checkstyle	144	trunk passed
+1	mvnsite	162	trunk passed
+1	shadedclient	1062	branch has no errors when building and testing our client artifacts.
+1	javadoc	124	trunk passed
0	spotbugs	58	Used deprecated FindBugs config; considering switching to SpotBugs.
+1	findbugs	237	trunk passed
		_ Patch Compile Tests _
0	mvndep	20	Maven dependency ordering for patch
+1	mvninstall	103	the patch passed
+1	compile	1087	the patch passed
+1	javac	1087	the patch passed
+1	checkstyle	150	the patch passed
+1	mvnsite	168	the patch passed
+1	whitespace	0	The patch has no whitespace issues.
+1	xml	4	The patch has no ill-formed XML file.
+1	shadedclient	714	patch has no errors when building and testing our client artifacts.
+1	javadoc	122	the patch passed
+1	findbugs	260	the patch passed
		_ Other Tests _
+1	unit	549	hadoop-common in the patch passed.
+1	unit	298	hadoop-aws in the patch passed.
+1	unit	86	hadoop-azure in the patch passed.
+1	asflicense	50	The patch does not generate ASF License warnings.
		7572

Subsystem	Report/Notes
Docker	Client=18.09.5 Server=18.09.5 base: https://builds.apache.org/job/hadoop-multibranch/job/PR-970/1/artifact/out/Dockerfile
GITHUB PR	#970
JIRA Issue	HADOOP-16371
Optional Tests	dupname asflicense compile javac javadoc mvninstall mvnsite unit shadedclient xml findbugs checkstyle
uname	Linux bc86dcd2f684 4.15.0-48-generic #51-Ubuntu SMP Wed Apr 3 08:28:49 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
Build tool	maven
Personality	personality/hadoop.sh
git revision	trunk / ae4143a
Default Java	1.8.0_212
Test Results	https://builds.apache.org/job/hadoop-multibranch/job/PR-970/1/testReport/
Max. process+thread count	1347 (vs. ulimit of 5500)
modules	C: hadoop-common-project/hadoop-common hadoop-tools/hadoop-aws hadoop-tools/hadoop-azure U: .
Console output	https://builds.apache.org/job/hadoop-multibranch/job/PR-970/1/console
versions	git=2.7.4 maven=3.3.9 findbugs=3.1.0-RC1
Powered by	Apache Yetus 0.10.0 http://yetus.apache.org

This message was automatically generated.

steveloughran · 2019-06-19T09:36:22Z

...t/hadoop-common/src/main/java/org/apache/hadoop/security/ssl/DelegatingSSLSocketFactory.java

use SLF4J {} expansion rather than inline; there's a lot of commons-logging era log statements, and changing a line is the time to upgrade them

steveloughran · 2019-06-19T09:37:40Z

hadoop-common-project/hadoop-common/src/main/resources/core-default.xml

I don't like having to remember the case of all these options, especially as every other s3a config option is all lower case. these should all be case insensitive.

steveloughran · 2019-06-19T09:47:40Z

hadoop-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/s3a/ITestS3ASSL.java

Better: A parameterized junit test (see ITestS3AMetadataPersistenceException) where the example isn't an enum but the string values we expect to see. I know hard-coding strings is not what I normally like, but doing it here would ensure that we'd have a regression test against the values changing.

steveloughran · 2019-06-19T09:55:29Z

hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/S3AUtils.java

I worry about this as it then ensures that the s3a code will only load when the shaded aws JAR is on the CP; people won't be able to switch to the unshaded JAR with the unshaded httpclient code.

I'm also trying to move off adding more stuff to the S3AUtils class as its become a merge conflict point with no real structure: every patch adds something to it. See Refactoring S3A for my thoughts there.

As this is adding new network setup, I'm going propose

putting this into a new class in o.a.h.fs.s3a.impl; -something like "NetworkBinding", where we can add more stuff later on

whatever is done to set up the AWS networking is done through reflection; if the shaded class can't be found on the CP, then just skip trying to configure it.

steveloughran · 2019-06-19T09:56:22Z

hadoop-tools/hadoop-aws/src/site/markdown/tools/hadoop-aws/performance.md

communicate with AWS Services.

hadoop-tools/hadoop-aws/src/site/markdown/tools/hadoop-aws/performance.md

steveloughran · 2019-06-19T10:13:17Z

(did not have S3Guard or kms tests setup)

You are going to have to do S3Guard tests here because that talks to another AWS endpoint, and it's important to know if there are any regressions. I'm not worried about the KMS stuff, or the assumed role code; there's enough test of AWS Session credentials that talking to AWS STS is handled.

FWIW, we directly talk to the following service s in production code

AWS S3
AWS Secure Token Service
DynamoDB

We need to make sure we are going near all of them.

I've just been looking to see where there's already a good parameterized test we could (abuse) to add another option to the config set, so guaranteeing coverage of interaction with these services without adding new tests (and new test delays). Nothing immediately springs to mind, though ITestS3AContractSeek seems like a good choice. Adding an extra column for each of the three options to set the SSL binding would verify that the seek code was happy with it and that, when run with s3guard enabled, we'd be implicitly testing that.

Sahil: what would you think about replacing ITestS3ASSL with some extra parameterization of the ITestS3AContractSeek test? I know it's nominally wrong to mix things up this way, but it would (a) not make test time any worse than it is and (b) do a more rigorous test: the seek code is skipping around, aborting connections etc?

steveloughran · 2019-08-09T11:26:49Z

...t/hadoop-common/src/main/java/org/apache/hadoop/security/ssl/DelegatingSSLSocketFactory.java

i dont want a warning in default mode. log at debug unless openssl was explicitly asked for

steveloughran · 2019-08-09T11:28:22Z

...t/hadoop-common/src/main/java/org/apache/hadoop/security/ssl/DelegatingSSLSocketFactory.java

make it 'NoSuchAlgorithmException' so current code can catch it

steveloughran · 2019-08-09T11:31:07Z

hadoop-tools/hadoop-azure/pom.xml

unsure about this as it is an existing dependency.
leave as is

steveloughran · 2019-08-09T11:33:06Z

@sahilTakiar I've left some comments.

it'd be good for @DadanielZ to review too

hadoop-yetus · 2019-09-11T03:55:36Z

💔 -1 overall

Vote	Subsystem	Runtime	Comment
0	reexec	87	Docker mode activated.
		_ Prechecks _
+1	dupname	1	No case conflicting files found.
+1	@author	0	The patch does not contain any @author tags.
+1	test4tests	0	The patch appears to include 4 new or modified test files.
		_ trunk Compile Tests _
0	mvndep	26	Maven dependency ordering for branch
+1	mvninstall	1263	trunk passed
+1	compile	1211	trunk passed
+1	checkstyle	155	trunk passed
+1	mvnsite	172	trunk passed
+1	shadedclient	1122	branch has no errors when building and testing our client artifacts.
+1	javadoc	131	trunk passed
0	spotbugs	64	Used deprecated FindBugs config; considering switching to SpotBugs.
+1	findbugs	274	trunk passed
		_ Patch Compile Tests _
0	mvndep	27	Maven dependency ordering for patch
+1	mvninstall	126	the patch passed
+1	compile	1140	the patch passed
+1	javac	1140	the patch passed
-0	checkstyle	154	root: The patch generated 1 new + 16 unchanged - 0 fixed = 17 total (was 16)
+1	mvnsite	169	the patch passed
+1	whitespace	0	The patch has no whitespace issues.
+1	xml	3	The patch has no ill-formed XML file.
+1	shadedclient	753	patch has no errors when building and testing our client artifacts.
-1	javadoc	34	hadoop-tools_hadoop-aws generated 4 new + 1 unchanged - 0 fixed = 5 total (was 1)
+1	findbugs	310	the patch passed
		_ Other Tests _
+1	unit	615	hadoop-common in the patch passed.
+1	unit	94	hadoop-aws in the patch passed.
+1	unit	83	hadoop-azure in the patch passed.
+1	asflicense	46	The patch does not generate ASF License warnings.
		8051

Subsystem	Report/Notes
Docker	Client=19.03.2 Server=19.03.2 base: https://builds.apache.org/job/hadoop-multibranch/job/PR-970/13/artifact/out/Dockerfile
GITHUB PR	#970
JIRA Issue	HADOOP-16371
Optional Tests	dupname asflicense compile javac javadoc mvninstall mvnsite unit shadedclient xml findbugs checkstyle
uname	Linux 6153829a62f6 4.15.0-58-generic #64-Ubuntu SMP Tue Aug 6 11:12:41 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
Build tool	maven
Personality	personality/hadoop.sh
git revision	trunk / `524b553`
Default Java	1.8.0_212
checkstyle	https://builds.apache.org/job/hadoop-multibranch/job/PR-970/13/artifact/out/diff-checkstyle-root.txt
javadoc	https://builds.apache.org/job/hadoop-multibranch/job/PR-970/13/artifact/out/diff-javadoc-javadoc-hadoop-tools_hadoop-aws.txt
Test Results	https://builds.apache.org/job/hadoop-multibranch/job/PR-970/13/testReport/
Max. process+thread count	1348 (vs. ulimit of 5500)
modules	C: hadoop-common-project/hadoop-common hadoop-tools/hadoop-aws hadoop-tools/hadoop-azure U: .
Console output	https://builds.apache.org/job/hadoop-multibranch/job/PR-970/13/console
versions	git=2.7.4 maven=3.3.9 findbugs=3.1.0-RC1
Powered by	Apache Yetus 0.10.0 http://yetus.apache.org

This message was automatically generated.

steveloughran · 2019-09-11T09:45:55Z

thanks for updating this. Was this just a rebase + push or have you made other changes?

sahilTakiar · 2019-09-11T19:56:42Z

Yes, I rebased this and addressed your comments. Having some trouble running all the tests though. I ran against us-east-1 and all the tests except the follow pass: mvn test -Dtest=ITestS3AContractRename,ITestAuthoritativePath,ITestS3GuardTtl . I think it is because I don't have S3Guard access (working on getting access). Getting a bunch of org.junit.AssumptionViolatedException: FS needs to have a metadatastore exceptions.

I'm working on getting S3Guard access, but wanted to update the PR in the meantime. Once I get access will re-run the tests with mvn verify -Ds3guard -Ddynamo.

Still trying to understand how to test against AWS Secure Token Service as well.

sahilTakiar · 2019-09-11T22:17:33Z

Re-ran tests with:

<property>
  <name>test.fs.s3a.sts.enabled</name>
  <value>true</value>
</property>

In my auth-keys.xml file, and all the tests passed (except for ITestAuthoritativePath and ITestS3GuardTtl).

I re-ran ITestS3AContractRename and it works now, so the maybe the test was just flaky.

steveloughran · 2019-09-16T11:36:02Z

...t/hadoop-common/src/main/java/org/apache/hadoop/security/ssl/DelegatingSSLSocketFactory.java

By calling wildfly methods without using reflection, this class can only be loaded or used if wildfly is on the CP. While this is already a requirement of hadoop-azure, I don't want to add it to hadoop-aws. Somehow reflection is going to be needed here.

The check in NetworkBinding#bindSSLChannelMode explicitly prevents S3A users from setting fs.s3a.ssl.channel.mode to default or OpenSSL, so there should be no way an S3A user can trigger the Wildfly jar from actually being used.

IIUC Java correctly, a class should still be able to load this class without Wildfly on the classpath. Java only looks for the Wildly classes when a Wildly class is initialized (in this case OpenSSLProvider). The import statements are only used during compilation. ref: https://stackoverflow.com/a/12620773/11511572

Thanks.

I always had the belief that integers and other simple constants may just get compiled in but that the class load Will kick off as soon as you reference a string in class. I should do more experiments. Maybe even learn Java assembly language. Anyway, as long as we don't try referencing, things like strings from the class, we should be okay. And as the little wild fly JAR is not currently on the hadoop-aws test CP, we are implicitly verifying this. We probably have to be careful once we had more support for it -we need to make sure we've not accidentally made it mandatory.

hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/Constants.java

steveloughran · 2019-09-16T11:38:42Z

hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/impl/NetworkBinding.java

these should go up into the same import block as com.amazonaws

...p-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/contract/s3a/ITestS3AContractSeek.java

steveloughran · 2019-09-16T12:19:21Z

...p-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/contract/s3a/ITestS3AContractSeek.java

While it's good to see this coverage, we do now have a test which is taking 4 min, and I'm trying to keep execution time of a non-scale test with a runner pool of 12 down to < 20 mins -this isn't going to help.
I propose only having the random and normal as the seek policies for the GCM tests; and remove these from the default jsse. Yes, it's a simpler matrix but as the default_jsse option gets implicitly tested everywhere,we aren't losing coverage.

steveloughran · 2019-09-16T12:23:49Z

Patch is coming together nicely -nearly there. Done CLI tests as well as the -aws suite,

A big fear of mine is that the current patch will, through transitive references, fail if the
wildfly JAR isn't on the CP.

But I couldn't actually create that failure condition when I tried on the CLI.

first, extended patched cloudstore s3a diagnostics to look for the new class

class: org.wildfly.openssl.OpenSSLProvider
       Not found on classpath: org.wildfly.openssl.OpenSSLProvider

tested IO against a store -all good.

and when I switch to an unsupported mode I get the expected stack trace

2019-09-16 13:06:11,124 [main] INFO  diag.StoreDiag (DurationInfo.java:<init>(53)) - Starting: Creating filesystem s3a://hwdev-steve-ireland-new/
2019-09-16 13:06:11,683 [main] INFO  diag.StoreDiag (DurationInfo.java:close(100)) - Creating filesystem s3a://hwdev-steve-ireland-new/: duration 0:00:561
java.lang.UnsupportedOperationException: S3A does not support setting fs.s3a.ssl.channel.mode OpenSSL or Default
	at org.apache.hadoop.fs.s3a.impl.NetworkBinding.bindSSLChannelMode(NetworkBinding.java:86)
	at org.apache.hadoop.fs.s3a.S3AUtils.initProtocolSettings(S3AUtils.java:1266)
	at org.apache.hadoop.fs.s3a.S3AUtils.initConnectionSettings(S3AUtils.java:1230)
	at org.apache.hadoop.fs.s3a.S3AUtils.createAwsConf(S3AUtils.java:1211)
	at org.apache.hadoop.fs.s3a.DefaultS3ClientFactory.createS3Client(DefaultS3ClientFactory.java:58)
	at org.apache.hadoop.fs.s3a.S3AFileSystem.bindAWSClient(S3AFileSystem.java:543)
	at org.apache.hadoop.fs.s3a.S3AFileSystem.initialize(S3AFileSystem.java:364)
	at org.apache.hadoop.fs.FileSystem.createFileSystem(FileSystem.java:3370)
	at org.apache.hadoop.fs.FileSystem.access$200(FileSystem.java:136)
	at org.apache.hadoop.fs.FileSystem$Cache.getInternal(FileSystem.java:3419)
	at org.apache.hadoop.fs.FileSystem$Cache.get(FileSystem.java:3387)
	at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:502)
	at org.apache.hadoop.fs.Path.getFileSystem(Path.java:365)
	at org.apache.hadoop.fs.store.diag.StoreDiag.executeFileSystemOperations(StoreDiag.java:860)
	at org.apache.hadoop.fs.store.diag.StoreDiag.run(StoreDiag.java:409)
	at org.apache.hadoop.fs.store.diag.StoreDiag.run(StoreDiag.java:353)
	at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:76)
	at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:90)
	at org.apache.hadoop.fs.store.diag.StoreDiag.exec(StoreDiag.java:1163)
	at org.apache.hadoop.fs.store.diag.StoreDiag.main(StoreDiag.java:1172)
	at storediag.main(storediag.java:25)
	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
	at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
	at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
	at java.lang.reflect.Method.invoke(Method.java:498)
	at org.apache.hadoop.util.RunJar.run(RunJar.java:323)
	at org.apache.hadoop.util.RunJar.main(RunJar.java:236)
2019-09-16 13:06:11,685 [main] INFO  util.ExitUtil (ExitUtil.java:t

which is telling me that my fears are misguided?

what do others say?

BTW, @bgaborg been having problems with STS tests too -try setting a region for the endpoint. Starting to suspect the latest SDK needs this now.

steveloughran · 2019-09-16T14:29:28Z

Did a full test run -Ds3guard -Ddynamodb -Dscale -Dauth; all failures DDB table delete timeouts and the (known) prune one. All good otherwise.

sahilTakiar · 2019-09-16T19:26:21Z

Thanks for the feedback and running all the tests Steve! I left a comment above about why I think everything will still work without wildfly on the classpath.

Working on addressing the other comments.

…on Java 8

sahilTakiar · 2019-09-16T22:08:24Z

Addressed comments, re-ran tests, the only additional failure is ITestS3AFileOperationCost, which is failing on trunk for me as well.

hadoop-yetus · 2019-09-16T23:56:08Z

💔 -1 overall

Vote	Subsystem	Runtime	Comment
0	reexec	147	Docker mode activated.
		_ Prechecks _
+1	dupname	1	No case conflicting files found.
+1	@author	0	The patch does not contain any @author tags.
+1	test4tests	0	The patch appears to include 4 new or modified test files.
		_ trunk Compile Tests _
0	mvndep	100	Maven dependency ordering for branch
+1	mvninstall	1532	trunk passed
+1	compile	1361	trunk passed
+1	checkstyle	167	trunk passed
+1	mvnsite	205	trunk passed
-1	shadedclient	1289	branch has errors when building and testing our client artifacts.
+1	javadoc	149	trunk passed
0	spotbugs	71	Used deprecated FindBugs config; considering switching to SpotBugs.
+1	findbugs	270	trunk passed
		_ Patch Compile Tests _
0	mvndep	22	Maven dependency ordering for patch
+1	mvninstall	129	the patch passed
+1	compile	1130	the patch passed
+1	javac	1130	the patch passed
-0	checkstyle	181	root: The patch generated 1 new + 16 unchanged - 0 fixed = 17 total (was 16)
+1	mvnsite	169	the patch passed
+1	whitespace	1	The patch has no whitespace issues.
+1	xml	4	The patch has no ill-formed XML file.
+1	shadedclient	801	patch has no errors when building and testing our client artifacts.
-1	javadoc	33	hadoop-tools_hadoop-aws generated 4 new + 1 unchanged - 0 fixed = 5 total (was 1)
+1	findbugs	266	the patch passed
		_ Other Tests _
+1	unit	545	hadoop-common in the patch passed.
+1	unit	87	hadoop-aws in the patch passed.
+1	unit	92	hadoop-azure in the patch passed.
+1	asflicense	48	The patch does not generate ASF License warnings.
		8761

Subsystem	Report/Notes
Docker	Client=18.09.7 Server=18.09.7 base: https://builds.apache.org/job/hadoop-multibranch/job/PR-970/14/artifact/out/Dockerfile
GITHUB PR	#970
JIRA Issue	HADOOP-16371
Optional Tests	dupname asflicense compile javac javadoc mvninstall mvnsite unit shadedclient xml findbugs checkstyle
uname	Linux 3eeeb1eb687e 4.15.0-58-generic #64-Ubuntu SMP Tue Aug 6 11:12:41 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
Build tool	maven
Personality	personality/hadoop.sh
git revision	trunk / `66bd168`
Default Java	1.8.0_222
checkstyle	https://builds.apache.org/job/hadoop-multibranch/job/PR-970/14/artifact/out/diff-checkstyle-root.txt
javadoc	https://builds.apache.org/job/hadoop-multibranch/job/PR-970/14/artifact/out/diff-javadoc-javadoc-hadoop-tools_hadoop-aws.txt
Test Results	https://builds.apache.org/job/hadoop-multibranch/job/PR-970/14/testReport/
Max. process+thread count	1341 (vs. ulimit of 5500)
modules	C: hadoop-common-project/hadoop-common hadoop-tools/hadoop-aws hadoop-tools/hadoop-azure U: .
Console output	https://builds.apache.org/job/hadoop-multibranch/job/PR-970/14/console
versions	git=2.7.4 maven=3.3.9 findbugs=3.1.0-RC1
Powered by	Apache Yetus 0.10.0 http://yetus.apache.org

This message was automatically generated.

steveloughran · 2019-09-17T10:14:25Z

the only additional failure is ITestS3AFileOperationCost, which is failing on trunk for me as well.

Is it? That's not good. That test uses the file system metrics to make assertions about how many operations actually go to S3. If it's failing either there is something wrong with assertions in your environment, or the connector is potentially making to many or too few calls to S3.

What are your test S3Guard settings?

steveloughran · 2019-09-17T10:33:40Z

thanks for the final revision and explanation of why classloading wasn't going to be a problem

+1 -committed to trunk. Thanks!

…to CoordinatorStreamSystemConsumer (apache#970)

steveloughran reviewed Jun 19, 2019

View reviewed changes

hadoop-tools/hadoop-aws/src/site/markdown/tools/hadoop-aws/performance.md Outdated

Copy link

Contributor

steveloughran Jun 19, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

communicate with AWS Services.

steveloughran reviewed Jun 19, 2019

View reviewed changes

hadoop-tools/hadoop-aws/src/site/markdown/tools/hadoop-aws/performance.md Outdated Show resolved Hide resolved

steveloughran reviewed Aug 9, 2019

View reviewed changes

hadoop-tools/hadoop-azure/pom.xml Outdated

Copy link

Contributor

steveloughran Aug 9, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

unsure about this as it is an existing dependency.
leave as is

steveloughran mentioned this pull request Sep 3, 2019

Hadoop-16438 ADLS Gen1 OpenSSL config control #1394

Closed

sahilTakiar force-pushed the HADOOP-16371 branch 2 times, most recently from f4b8f9d to 7fcdf14 Compare September 11, 2019 01:40

apache deleted a comment from hadoop-yetus Sep 11, 2019

steveloughran self-assigned this Sep 16, 2019

steveloughran reviewed Sep 16, 2019

View reviewed changes

hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/Constants.java Outdated Show resolved Hide resolved

steveloughran reviewed Sep 16, 2019

View reviewed changes

...p-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/contract/s3a/ITestS3AContractSeek.java Outdated Show resolved Hide resolved

steveloughran reviewed Sep 16, 2019

View reviewed changes

apache deleted a comment from hadoop-yetus Sep 16, 2019

steveloughran mentioned this pull request Sep 16, 2019

HADOOP-16445. Allow separate custom signing algorithms for S3 and DDB #1332

Merged

HADOOP-16371: Option to disable GCM for SSL connections when running …

f9a550e

…on Java 8

sahilTakiar force-pushed the HADOOP-16371 branch from 7fcdf14 to f9a550e Compare September 16, 2019 20:51

steveloughran closed this Sep 17, 2019

steveloughran mentioned this pull request Sep 27, 2019

[SPARK-29089][SQL] Parallelize blocking FileSystem calls in DataSource#checkAndGlobPathIfNecessary apache/spark#25899

Closed

shanthoosh pushed a commit to shanthoosh/hadoop that referenced this pull request Oct 15, 2019

SAMZA-2143: NPE in CoordinatorStreamMessage#equals and some clean-up …

4aaa589

…to CoordinatorStreamSystemConsumer (apache#970)

HADOOP-16371: Option to disable GCM for SSL connections when running on Java 8 #970

HADOOP-16371: Option to disable GCM for SSL connections when running on Java 8 #970

Uh oh!

Conversation

sahilTakiar commented Jun 14, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

hadoop-yetus commented Jun 14, 2019

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

steveloughran commented Jun 19, 2019

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

steveloughran commented Aug 9, 2019

Uh oh!

hadoop-yetus commented Sep 11, 2019

Uh oh!

steveloughran commented Sep 11, 2019

Uh oh!

sahilTakiar commented Sep 11, 2019

Uh oh!

sahilTakiar commented Sep 11, 2019

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

steveloughran commented Sep 16, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

steveloughran commented Sep 16, 2019

Uh oh!

sahilTakiar commented Sep 16, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

sahilTakiar commented Sep 16, 2019

Uh oh!

hadoop-yetus commented Sep 16, 2019

Uh oh!

steveloughran commented Sep 17, 2019

Uh oh!

steveloughran commented Sep 17, 2019

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

sahilTakiar commented Jun 14, 2019 •

edited

Loading

steveloughran commented Sep 16, 2019 •

edited

Loading

sahilTakiar commented Sep 16, 2019 •

edited

Loading