[RESULT][VOTE] Apache Accumulo 1.10.3-rc1

2023-04-13 Thread Christopher
This vote passes with: 5 +1s, and no other votes I will track the release tasks on https://github.com/apache/accumulo/issues/3295 On Wed, Apr 12, 2023 at 1:44 PM Christopher wrote: > > +1 to this release > > I checked: > > * Hashes/signatures > * Source and javadoc jars exist for jars in stagi

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-12 Thread Christopher
+1 to this release I checked: * Hashes/signatures * Source and javadoc jars exist for jars in staging repo * The jars in the binary tarball match the jars in the staging repo, and were built with the specified commit * The source tarball matches the contents of the git branch * Unit tests and ITs

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-12 Thread Christopher
https://github.com/apache/accumulo/issues/3287 was identified as a known issue in the documentation. This can be noted in the release notes and fixed along with any 1.10.4 fixes. I'm wrapping up my own testing, and will close the vote soon. On Wed, Apr 12, 2023 at 7:36 AM Dave Marion wrote: > > T

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-12 Thread Dave Marion
Thanks Christopher for setting me straight. This information did enable the build to move beyond the CredentialProviderFactoryShimTest. I'm unable to get past MiniAccumuloClusterTest though. Two tests timeout for me every time even when increasing the timeout factor (they both specify a timeout in

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-11 Thread Christopher
I think a lot of you are getting bugs because you're trying to build with both the "hadoop-default" / "hadoop2" and the "hadoop3" profile active at the same time, and that's causing dependency resolution issues because the classpath is polluted with a bunch of extra stuff (which you can see by addi

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-11 Thread Mark Owens
If I update the pom's with the commons-configuration changes suggested by Dave in a previous post, I can successfully run the Sunny profile tests with both Hadoop 2 and Hadoop 3. On Tue, Apr 11, 2023 at 1:16 PM Daniel Roberts wrote: > Yes, I changed the memory settings last week and was able to

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-11 Thread Daniel Roberts
Yes, I changed the memory settings last week and was able to pass the ExamplesIT.testScansWithInterference and ExamplesIT.testIsolatedScansWithInterference without issue. Given these environmental tweaks, do we have a documented minimum machine spec that we expect the release tests to pass on? On

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-11 Thread Dave Marion
Chris Shannon pointed out that this test was failing and the resolution was to give it more memory (see https://github.com/apache/accumulo/issues/3281). On Mon, Apr 10, 2023 at 6:04 PM Dave Marion wrote: > *mvn clean verify -Phadoop3,sunny* failed in ExamplesIT with the > aforementioned changes

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-11 Thread Christopher Shannon
The test failure for ExamplesIT.testScansWithInterference is likely from out of memory. See https://github.com/apache/accumulo/issues/3281 On Tue, Apr 11, 2023 at 12:21 PM Dominic Garguilo < dominic.gargu...@gmail.com> wrote: > After running > mvn clean verify -Phadoop3,sunny > I am consistently

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-11 Thread Dominic Garguilo
After running mvn clean verify -Phadoop3,sunny I am consistently seeing the following: [ERROR] > org.apache.accumulo.core.conf.CredentialProviderFactoryShimTest.extractFromHdfs > Time elapsed: 3.605 s <<< ERROR! > java.lang.NoClassDefFoundError: > org/apache/commons/beanutils/BeanIntrospector >

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-10 Thread Dave Marion
*mvn clean verify -Phadoop3,sunny* failed in ExamplesIT with the aforementioned changes. Specifically, ExamplesIT.testScansWithInterference failed with a server error when closing the batch writer. No further information in the logs. On Mon, Apr 10, 2023 at 5:06 PM Dave Marion wrote: > Updating

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-10 Thread Dave Marion
Updating commons-configuration from 1.6 to 1.10 did the trick I think. Versions 1.7, 1.8, and 1.9 had compilation issues. I'm not veto-ing because other people have built with the hadoop3 profile even though more than one of us have had an issue. Here are the changes I made to get me past the Crede

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-10 Thread Dominic Garguilo
+1 I ran the tests with no errors via mvn clean verify package -Psunny and mvn clean verify package -Dhadoop.profile=3 -Psunny Also ran ingest and verify on 195M entries. On Mon, Apr 10, 2023 at 1:31 PM Daniel Roberts wrote: > Dave & Mark, > > Running the sunny tests with the hadoop 3 profile

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-10 Thread Daniel Roberts
Dave & Mark, Running the sunny tests with the hadoop 3 profile resulted in no local errors for me. `mvn clean verify package -Dhadoop.profile=3 -Psunny` So I checked my local maven repo and while I do have commons-configuration 1.6, the pom lists commons-beanutils-core:1.8.0 as a dependency. comm

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-10 Thread Mark Owens
Running the hadoop3 profile, I also had an issue with the CredentialProviderFactoryShimTest as did Dave. I was running on CentOS7. [ERROR] Tests run: 10, Failures: 0, Errors: 1, Skipped: 0, Time elapsed: 7.559 s <<< FAILURE! - in org.apache.accumulo.core.conf.CredentialProviderFactoryShimTest [ERR

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-10 Thread Dave Marion
>From what I can tell: 1. commons-configuration:1.6 has a commons-beanutils:1.7.0 dependency 2. hadoop-common:3.0.3 has a commons-beanutils:1.9.3 dependency, hadoop-common:2.6.5 has no such dependency Hadoop 3.0.3 is failing because the commons-beanutils:1.9.3 dependency must not be on the classp

RE: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-10 Thread dev1
oleman -Original Message- From: Dave Marion Sent: Monday, April 10, 2023 8:24 AM To: dev@accumulo.apache.org Subject: Re: [VOTE] Apache Accumulo 1.10.3-rc1 I built the source tarball using the hadoop3 profile again and encountered the issue below. The version of hadoop (3.0.3) has not chang

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-10 Thread Dave Marion
I built the source tarball using the hadoop3 profile again and encountered the issue below. The version of hadoop (3.0.3) has not changed, so I assume that this is related to a change in our dependencies. [ERROR] Tests run: 10, Failures: 0, Errors: 1, Skipped: 0, Time elapsed: 3.856 s <<< FAILURE!

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-09 Thread Jeffrey Manno
+1 I did run into a few flaky tests but none seem to warrant a stoppage to this release candidate. Two common ones are SuspendTabletsIT and DeleteTableDuringSplitIT. Other checks I did: - Ran the entire test suite, along with sunny day tests. - Verified checksums. - Did some small inges

Re: [VOTE] Apache Accumulo 1.10.3-rc1

2023-04-08 Thread Christopher Shannon
+1 (binding), LGTM Some of the things I did for verification/testing locally: * Validated signatures and checksums * Verified license and notice files in archives * Verified source license headers with 'mvn apache-rat:check' * Built and ran all the sunny integration tests * Started up using Uno t

[VOTE] Apache Accumulo 1.10.3-rc1

2023-04-07 Thread Christopher
Accumulo Developers, Please consider the following candidate for Apache Accumulo 1.10.3. Git Commit: 733863638d85d0109d217da7ea5f36f5e483e207 Branch: 1.10.3-rc1 If this vote passes, a gpg-signed tag will be created using: git tag -f -s -m 'Apache Accumulo 1.10.3' rel/1.10.3 \ 733