[
https://issues.apache.org/jira/browse/CASSANDRA-20402?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17940975#comment-17940975
]
Ekaterina Dimitrova edited comment on CASSANDRA-20402 at 4/4/25 1:39 PM:
-------------------------------------------------------------------------
{quote}So I'd say changing testReadOnlyIndex and testWriteOnlyIndex exceptions
is probably the right way to fix this. Caleb Rackliffe, what do you think?
{quote}
{quote}Sounds good to me.
{quote}
and
{quote}Aaaand in the meantime I realized we probably need to check also for the
new exception here -
[https://github.com/apache/cassandra/blob/trunk/src/java/org/apache/cassandra/net/InboundSink.java#L130]
{quote}
Both addressed
[here|https://github.com/ekaterinadimitrova2/cassandra/commit/67eacaca10644cdc95a385c13b3a7d7ea2314d14]
>From
>[CI|https://app.circleci.com/pipelines/github/ekaterinadimitrova2/cassandra/2743/workflows/620f2d2b-b2c1-465b-8db1-9b7289d1d009]:
test_resumable_decommission - known from CASSANDRA-19542
test_size_estimates_multidc - it doesn't seem related - I opened CASSANDRA-20515
largecolumn_test.TestLargeColumn .test_cleanup - CASSANDRA-20509
MultiNodeSAITest - timing out, known, CASSANDRA-20307
testIncompleteState-_jdk11 - CASSANDRA-19524
testIndexExceptionsTwoIndexesOn3NodeCluster- it failed and I noticed it was not
deterministic. I fixed that
[here|https://github.com/ekaterinadimitrova2/cassandra/commit/940214604641e69c294c5c1c22f331043560b80b].
[Here|https://app.circleci.com/pipelines/github/ekaterinadimitrova2/cassandra/2744/workflows/86df43cd-9fa0-4777-a6d7-fdc22d063901/jobs/62689/parallel-runs/1?filterBy=ALL]
are 500 successful runs.
Starting commit soon
was (Author: e.dimitrova):
{quote}So I'd say changing testReadOnlyIndex and testWriteOnlyIndex exceptions
is probably the right way to fix this. Caleb Rackliffe, what do you think?
{quote}
{quote}Sounds good to me.
{quote}
and
{quote}Aaaand in the meantime I realized we probably need to check also for the
new exception here -
[https://github.com/apache/cassandra/blob/trunk/src/java/org/apache/cassandra/net/InboundSink.java#L130]
{quote}
Addressed
>From
>[CI|https://app.circleci.com/pipelines/github/ekaterinadimitrova2/cassandra/2743/workflows/620f2d2b-b2c1-465b-8db1-9b7289d1d009]:
test_resumable_decommission - known from CASSANDRA-19542
test_size_estimates_multidc - it doesn't seem related - I opened CASSANDRA-20515
largecolumn_test.TestLargeColumn .test_cleanup - CASSANDRA-20509
MultiNodeSAITest - timing out, known, CASSANDRA-20307
testIncompleteState-_jdk11 - CASSANDRA-19524
testIndexExceptionsTwoIndexesOn3NodeCluster- it failed and I noticed it was not
deterministic. I fixed that
[here|https://github.com/ekaterinadimitrova2/cassandra/commit/940214604641e69c294c5c1c22f331043560b80b].
[Here|https://app.circleci.com/pipelines/github/ekaterinadimitrova2/cassandra/2744/workflows/86df43cd-9fa0-4777-a6d7-fdc22d063901/jobs/62689/parallel-runs/1?filterBy=ALL]
are 500 successful runs.
Starting commit soon
> Create new RequestFailureReason - INDEX_BUILDING and return it to end users
> while an index is building
> ------------------------------------------------------------------------------------------------------
>
> Key: CASSANDRA-20402
> URL: https://issues.apache.org/jira/browse/CASSANDRA-20402
> Project: Apache Cassandra
> Issue Type: Improvement
> Components: Feature/2i Index, Feature/SAI
> Reporter: Ekaterina Dimitrova
> Assignee: Ekaterina Dimitrova
> Priority: Normal
> Fix For: 5.x
>
>
> Currently, we return INDEX NOT AVAILABLE to the users while an index is
> building, which makes it unclear to users whether the index is building or
> there is an issue.
> Differentiate the two cases by adding INDEX_BUILDING.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
