hk-lrzy commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1609878354
##
flink-cdc-runtime/src/main/java/org/apache/flink/cdc/runtime/operators/schema/SchemaOperator.java:
##
@@ -127,11 +142,23 @@ public Schema load(TableId tableId) {
hk-lrzy commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1609878354
##
flink-cdc-runtime/src/main/java/org/apache/flink/cdc/runtime/operators/schema/SchemaOperator.java:
##
@@ -127,11 +142,23 @@ public Schema load(TableId tableId) {
hk-lrzy commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1609878354
##
flink-cdc-runtime/src/main/java/org/apache/flink/cdc/runtime/operators/schema/SchemaOperator.java:
##
@@ -127,11 +142,23 @@ public Schema load(TableId tableId) {
leonardBang merged PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
lvyanquan commented on PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#issuecomment-2072115980
> There is an error in the scenario of using route from CI, and I am trying
to fix it.
Done fix.
--
This is an automated message from the Apache Git Service.
To respond
lvyanquan commented on PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#issuecomment-2072051290
There is an error in the scenario of using route from CI, and I am trying to
fix it.
--
This is an automated message from the Apache Git Service.
To respond to the message, please
lvyanquan commented on PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#issuecomment-2071458830
@leonardBang Can you help to start a CI workflow?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL
lvyanquan commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1568374817
##
flink-cdc-common/src/main/java/org/apache/flink/cdc/common/pipeline/PipelineOptions.java:
##
@@ -85,5 +89,12 @@ public class PipelineOptions {
leonardBang commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1567483688
##
flink-cdc-common/src/main/java/org/apache/flink/cdc/common/pipeline/PipelineOptions.java:
##
@@ -85,5 +89,12 @@ public class PipelineOptions {
lvyanquan commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1554820774
##
flink-cdc-common/src/main/java/org/apache/flink/cdc/common/pipeline/PipelineOptions.java:
##
@@ -85,5 +89,12 @@ public class PipelineOptions {
leonardBang commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1554795627
##
flink-cdc-common/src/main/java/org/apache/flink/cdc/common/pipeline/PipelineOptions.java:
##
@@ -85,5 +89,12 @@ public class PipelineOptions {
leonardBang commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1554795438
##
flink-cdc-runtime/src/main/java/org/apache/flink/cdc/runtime/operators/schema/SchemaOperator.java:
##
@@ -127,4 +136,13 @@ RESPONSE
lvyanquan commented on PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#issuecomment-2041034979
> Thanks @lvyanquan for the fix, I left some comments. And I think tests is
necessary, e.g. a test mocked RPC timeout when processing schema change event
is welcome.
Add new
lvyanquan commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1554559194
##
flink-cdc-runtime/src/main/java/org/apache/flink/cdc/runtime/operators/schema/SchemaOperator.java:
##
@@ -127,4 +136,13 @@ RESPONSE
lvyanquan commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1554559016
##
flink-cdc-runtime/src/main/java/org/apache/flink/cdc/runtime/operators/schema/SchemaOperator.java:
##
@@ -111,8 +115,13 @@ private SchemaChangeResponse
leonardBang commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1549367497
##
flink-cdc-runtime/src/main/java/org/apache/flink/cdc/runtime/operators/schema/SchemaOperator.java:
##
@@ -111,8 +115,13 @@ private SchemaChangeResponse
LYanquan commented on PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#issuecomment-2003238562
> Should we clear the flushedSinkWriters as well?
Thanks for this suggestion, address it.
--
This is an automated message from the Apache Git Service.
To respond to the
LYanquan commented on PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#issuecomment-2003214710
@fanqiejiang8 Hi, I have added `getRuntimeContext().getIndexOfThisSubtask()
== 0` check to make sure that it won't send many times.
--
This is an automated message from the Apache
BaoPiao commented on PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#issuecomment-2003029727
> @BaoPiao Yeah, this is because SchemaRegistryRequestHandler doesn't clear
pendingSchemaChanges when SchemaOperator restarted, I've added one commit
yanghuaiGit commented on PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#issuecomment-2003023705
如果schemaRegister的pending里有多个待执行schema事件，isSchemaChangeApplying代表的是正在执行schema变更事件，但是无法得知是哪个schema事件执行，isSchemaChangeApplying无法和客户端的事件绑定，导致schemaoperator可能需要等待所有的schema事件执行完才会转为非阻塞
fanqiejiang8 commented on PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#issuecomment-2002855730
@lvyanquan @BaoPiao When task parallelism is multiple, task retry all sent
SchemaRegistry task more the RefreshPendingListsRequest request, that each
request will not cause data
LYanquan commented on PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#issuecomment-2002529227
@BaoPiao Yeah, this is because SchemaRegistryRequestHandler doesn't clear
pendingSchemaChanges when SchemaOperator restarted, I've added one commit
BaoPiao commented on PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#issuecomment-1996731322
If the SchemaRegistryRequestHandler is in the midst of executing a scheme
change, and the task manager where the SchemaOperator resides undergoes a
restart, could this potentially
lvyanquan commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1524312973
##
flink-cdc-runtime/src/main/java/org/apache/flink/cdc/runtime/operators/schema/coordinator/SchemaRegistryRequestHandler.java:
##
@@ -70,6 +81,8 @@ public
ruanhang1993 commented on code in PR #3128:
URL: https://github.com/apache/flink-cdc/pull/3128#discussion_r1524276055
##
flink-cdc-runtime/src/main/java/org/apache/flink/cdc/runtime/operators/schema/coordinator/SchemaRegistryRequestHandler.java:
##
@@ -70,6 +81,8 @@ public
25 matches
Mail list logo
