[
https://issues.apache.org/jira/browse/HADOOP-13793?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15723516#comment-15723516
]
Mingliang Liu commented on HADOOP-13793:
----------------------------------------
{code}
51 @Override
52 public void teardown() throws Exception {
53 super.teardown();
54 }
55
56 @Override
57 public void nameThread() {
58 super.nameThread();
59 }
60
61 @Override
62 protected int getTestTimeoutMillis() {
63 return super.getTestTimeoutMillis();
64 }
{code}
Can we remove this code lines? Seems unnecessary.
For the inconsistent key delay in ms, I found 500ms are not enough to make the
test failure (when commenting out the {{S3Guard.S3_METADATA_STORE_IMPL}} config
key in test). I tried 1000ms+ and found it works. Do you mind increasing the
default interval to 1~3 seconds?
In the {{ITestS3GuardListConsistency#createContract}}, this is setting it
hard-code. Can we let use the config files so we can run with DDBMetadataStore?
{code}
conf.setClass(S3Guard.S3_METADATA_STORE_IMPL, LocalMetadataStore.class,
MetadataStore.class);
{code}
Thanks,
> s3guard: add inconsistency injection, integration tests
> -------------------------------------------------------
>
> Key: HADOOP-13793
> URL: https://issues.apache.org/jira/browse/HADOOP-13793
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs/s3
> Reporter: Aaron Fabbri
> Assignee: Aaron Fabbri
> Attachments: HADOOP-13793-HADOOP-13345.001.patch,
> HADOOP-13793-HADOOP-13345.002.patch
>
>
> Many of us share concerns that testing the consistency features of S3Guard
> will be difficult if we depend on the rare and unpredictable occurrence of
> actual inconsistency in S3 to exercise those code paths.
> I think we should have a mechanism for injecting failure to force exercising
> of the consistency codepaths in S3Guard.
> Requirements:
> - Integration tests that cause S3A to see the types of inconsistency we
> address with S3Guard.
> - These are deterministic integration tests.
> Unit tests are possible as well, if we were to stub out the S3Client. That
> may be less bang for the buck, though.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
