debezium/dbz#1089 Add Infinispan connection validator by yuanglili · Pull Request #288 · debezium/debezium-platform

yuanglili · 2026-03-09T03:42:23Z

Summary
Implements InfinispanConnectionValidator for checking connection to Infinispan.

Changes

Add Infinispan connection schema to connection-schemas.json
Add InfinispanConnectionValidator using infinispan-client-hotrod for HotRod protocol validation
Add destinations.infinispan.connection.timeout to application.yml
Add InfinispanConnectionValidatorIT integration test with Testcontainers

Fixes debezium/dbz#1089

...a/io/debezium/platform/environment/connection/destination/InfinispanConnectionValidator.java

mfvitale · 2026-03-11T10:07:22Z

...a/io/debezium/platform/environment/connection/destination/InfinispanConnectionValidator.java

+
+            String serverUri;
+            if (user != null && password != null) {
+                serverUri = String.format("hotrod://%s:%s@%s:%d", user, password, serverHost, serverPort);


Could you create a constant for "hotrod://%s:%s@%s:%d" and the use the "".formatted() method?

Hi @mfvitale, thanks for the review! Here's what I've updated:
Extracted HOTROD_URI_WITH_AUTH / HOTROD_URI_NO_AUTH as constants and switched to .formatted()

mfvitale

hi @yuanglili Overall it LGTM. I just left some minor improvements.

mfvitale · 2026-03-16T08:40:03Z

@yuanglili Could you please rebase on main?

Naros · 2026-03-16T16:21:42Z

...a/io/debezium/platform/environment/connection/destination/InfinispanConnectionValidator.java

+        if (user != null && password != null) {
+            serverUri = HOTROD_URI_WITH_AUTH.formatted(user, password, serverHost, serverPort);
+        }


What about if the supplied user/password is not null but is an empty string? Is the expectation that this should go through the "auth" path or the "no auth" path?

What about if the supplied user/password is not null but is an empty string? Is the expectation that this should go through the "auth" path or the "no auth" path?

@Naros Good catch! But in the case, I think neither auth nor no-auth is the right path here, it should fail validation instead. Because if a user puts in a value, even a blank one, they probably mean to use authentication, so silently ignoring it seems wrong.

Also noticed I was missing another case: if only one of user or password is set, that should be an error too.

My plan to updated the logic to handle all cases:

Both non-blank → auth path

Both absent → no-auth path

One missing, or either is blank → validation error

WDYT?

@Naros @yuanglili I think the correct anwer is to check ISPN and confir if either username or password can be empty. Based on that it will either go through auth path or fail validation.

I checked the Infinispan codebase:

AuthenticationConfigurationBuilder (L165-168): username() just assigns the value with no empty string check

validate() (L231-243): only checks username == null, so empty string passes right through

HotRodURI (L59): parses hotrod://:@host:port and assigns "" directly without converting to null

Test results:

Both blank (hotrod://:@host) → connected successfully (treated as no-auth)

Only one blank (e.g. hotrod://:secret@host or hotrod://admin:@host) → TransportException (auth failure)

So I think the updated logic should be :

Both non-blank → auth path

Both null or both blank → no-auth path

One blank, one not → validation error

Updated the PR：

Fail validation when only one of user/password is provided or is empty

Add parameterized tests to cover partial credential scenarios

mfvitale · 2026-03-18T08:19:20Z

...a/io/debezium/platform/environment/connection/destination/InfinispanConnectionValidator.java

+        boolean hasUser = user != null && !user.isEmpty();
+        boolean hasPassword = password != null && !password.isEmpty();
+
+        if (hasUser && hasPassword) {
+            serverUri = HOTROD_URI_WITH_AUTH.formatted(user, password, serverHost, serverPort);
+        }
+        else {
+            serverUri = HOTROD_URI_NO_AUTH.formatted(serverHost, serverPort);
+        }


Since you have already done the check for the password and user, could you move this block out of this function and then just pass the serverUri as parameter here?

Naros · 2026-03-18T09:27:41Z

...a/io/debezium/platform/environment/connection/destination/InfinispanConnectionValidator.java

+        boolean hasUser = user != null && !user.isEmpty();
+        boolean hasPassword = password != null && !password.isEmpty();
+
+        if (hasUser && hasPassword) {


Suggested change

if (hasUser && hasPassword) {

if (!Strings.isNullOrEmpty(user) && !Strings.isNullOrEmpty(password)) {

Then you can also remove the hasUser and hasPassword local vars.

mfvitale · 2026-03-24T08:50:53Z

Waiting this #313 for test stabilization

mfvitale · 2026-03-24T10:04:25Z

@yuanglili Could you please rebase with latest changes on main?

…n tests Signed-off-by: nonononoonononon <yuangli971@gmail.com>

Signed-off-by: nonononoonononon <yuangli971@gmail.com>

yuanglili · 2026-03-25T03:08:38Z

@yuanglili Could you please rebase with latest changes on main?

Rebased , thanks！

mfvitale · 2026-03-26T08:54:08Z

@yuanglili Applied, thanks!

Naros reviewed Mar 10, 2026

View reviewed changes

...a/io/debezium/platform/environment/connection/destination/InfinispanConnectionValidator.java Outdated Show resolved Hide resolved

...a/io/debezium/platform/environment/connection/destination/InfinispanConnectionValidator.java Outdated Show resolved Hide resolved

mfvitale reviewed Mar 11, 2026

View reviewed changes

mfvitale mentioned this pull request Mar 12, 2026

Adds NATS Streaming Connection Validator (DBZ-9433) #287

Merged

yuanglili force-pushed the dbz#1089-infinispan-conn-validator branch from 8a501d9 to d62bbb0 Compare March 12, 2026 23:58

mfvitale approved these changes Mar 13, 2026

View reviewed changes

yuanglili requested a review from Naros March 13, 2026 20:01

Naros reviewed Mar 16, 2026

View reviewed changes

yuanglili force-pushed the dbz#1089-infinispan-conn-validator branch from 8438ff8 to cf51b63 Compare March 16, 2026 23:14

yuanglili requested review from Naros, jpechane and mfvitale March 18, 2026 01:22

yuanglili marked this pull request as draft March 18, 2026 01:25

yuanglili marked this pull request as ready for review March 18, 2026 01:26

mfvitale reviewed Mar 18, 2026

View reviewed changes

Naros reviewed Mar 18, 2026

View reviewed changes

yuanglili force-pushed the dbz#1089-infinispan-conn-validator branch from f64c2c5 to 869bab5 Compare March 23, 2026 23:45

Naros approved these changes Mar 24, 2026

View reviewed changes

yuanglili added 5 commits March 24, 2026 23:06

debezium/dbz#1089 Add Infinispan connection validator with integratio…

56e993b

…n tests Signed-off-by: nonononoonononon <yuangli971@gmail.com>

debezium/dbz#1089 Address review comments

b98b31c

Signed-off-by: nonononoonononon <yuangli971@gmail.com>

debezium/dbz#1089 handle empty string credentials for authentication

eed456f

Signed-off-by: nonononoonononon <yuangli971@gmail.com>

debezium/dbz#1089 fix checkstyle

63d201b

Signed-off-by: nonononoonononon <yuangli971@gmail.com>

debezium/dbz#1089 Refactor URI building

94b7a6a

Signed-off-by: nonononoonononon <yuangli971@gmail.com>

yuanglili force-pushed the dbz#1089-infinispan-conn-validator branch from bdc9cca to 94b7a6a Compare March 25, 2026 03:07

mfvitale merged commit e853f27 into debezium:main Mar 26, 2026
7 checks passed

	if (hasUser && hasPassword) {
	if (!Strings.isNullOrEmpty(user) && !Strings.isNullOrEmpty(password)) {

Uh oh!

Conversation

yuanglili commented Mar 9, 2026

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mfvitale left a comment

Choose a reason for hiding this comment

Uh oh!

mfvitale commented Mar 16, 2026

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!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mfvitale commented Mar 24, 2026

Uh oh!

mfvitale commented Mar 24, 2026

Uh oh!

yuanglili commented Mar 25, 2026

Uh oh!

mfvitale commented Mar 26, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants