Cube Cloud stopped working with databricks (Multiple entries with same key error)

[Klimmy]

Describe the bug
Starting from the version 1.2.28 it seems that Cube Cloud stopped working with Databricks. At least in our setup.
It works fine with 1.2.27.

We've got this error when trying to run something in the Playground (I replaced actual token with <databricks_token>):

Error running static method java.lang.IllegalArgumentException: Multiple entries with same key: pwd=<databricks_token> and pwd=<databricks_token> at com.databricks.internal.google.common.collect.ImmutableMap.conflictException(ImmutableMap.java:382) at com.databricks.internal.google.common.collect.ImmutableMap.checkNoConflict(ImmutableMap.java:376) at com.databricks.internal.google.common.collect.RegularImmutableMap.checkNoConflictInKeyBucket(RegularImmutableMap.java:246) at com.databricks.internal.google.common.collect.RegularImmutableMap.fromEntryArrayCheckingBucketOverflow(RegularImmutableMap.java:133) at com.databricks.internal.google.common.collect.RegularImmutableMap.fromEntryArray(RegularImmutableMap.java:95) at com.databricks.internal.google.common.collect.ImmutableMap$Builder.build(ImmutableMap.java:576) at com.databricks.internal.google.common.collect.ImmutableMap$Builder.buildOrThrow(ImmutableMap.java:604) at com.databricks.internal.google.common.collect.ImmutableMap$Builder.build(ImmutableMap.java:591) at com.databricks.jdbc.api.impl.DatabricksConnectionContext.buildPropertiesMap(DatabricksConnectionContext.java:88) at com.databricks.jdbc.api.impl.DatabricksConnectionContext.parse(DatabricksConnectionContext.java:130) at com.databricks.jdbc.api.impl.DatabricksConnectionContextFactory.create(DatabricksConnectionContextFactory.java:20) at com.databricks.client.jdbc.Driver.connect(Driver.java:59) at java.sql/java.sql.DriverManager.getConnection(DriverManager.java:681) at java.sql/java.sql.DriverManager.getConnection(DriverManager.java:190) at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:77) at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.base/java.lang.reflect.Method.invoke(Method.java:569) at node.MethodCallBaton.invokeMethod9(MethodCallBaton.java:25)

The PWD in URL is deprecated, so when we remove it from the CUBEJS_DB_DATABRICKS_URL environment variable we've got similar error, but with uid key duplication problem:

Error running static method java.lang.IllegalArgumentException: Multiple entries with same key: uid=token and uid=token at
...

Removing further uid from the CUBEJS_DB_DATABRICKS_URL leads to this error:

Error running static method com.databricks.jdbc.exception.DatabricksSQLException: Error initializing the Java Util Logger (JUL) with error: 
...

Version:
1.2.28

Additional context
Databricks-related envrionment variables we use:
CUBEJS_DB_TYPE=databricks-jdbc
CUBEJS_DB_DATABRICKS_URL
CUBEJS_DB_DATABRICKS_TOKEN

[KSDaemon]

Hi @Klimmy!

Could you please post what your CUBEJS_DB_DATABRICKS_URL param looks like? (you can hide the real url and warehouse id).

Because I've just checked with the latest version - it works fine:

2025-04-03T08:47:29.983Z
🔗 Cube SQL (pg) is listening on 0.0.0.0:5432
2025-04-03T08:47:30.163Z
🚀 Cube API server (1.2.29) is listening on 4000
2025-04-03T08:47:49.292Z
Apr 03, 2025 8:47:48 AM com.databricks.jdbc.common.util.LoggingUtil setupLogger
2025-04-03T08:47:49.292Z
INFO: Setting up JUL logger
2025-04-03T08:47:50.058Z
Apr 03, 2025 8:47:49 AM com.databricks.client.jdbc.Driver connect
2025-04-03T08:47:50.058Z
INFO: DriverSystemConfiguration[driverName='oss-jdbc', driverVersion='1.0.2-oss', osName='Linux', osVersion='6.6.72+', osArch='amd64', runtimeName='OpenJDK 64-Bit Server VM', runtimeVersion='17.0.14', runtimeVendor='Eclipse Adoptium', localeName='en_US', defaultCharsetEncoding='UTF-8']
2025-04-03T08:47:50.505Z
SLF4J(W): No SLF4J providers were found.
2025-04-03T08:47:50.506Z
SLF4J(W): Defaulting to no-operation (NOP) logger implementation
2025-04-03T08:47:50.506Z
SLF4J(W): See https://www.slf4j.org/codes.html#noProviders for further details.
2025-04-03T08:48:10.464Z
Apr 03, 2025 8:48:10 AM com.databricks.client.jdbc.Driver connect
2025-04-03T08:48:10.464Z
INFO: DriverSystemConfiguration[driverName='oss-jdbc', driverVersion='1.0.2-oss', osName='Linux', osVersion='6.6.72+', osArch='amd64', runtimeName='OpenJDK 64-Bit Server VM', runtimeVersion='17.0.14', runtimeVendor='Eclipse Adoptium', localeName='en_US', defaultCharsetEncoding='UTF-8']

[Klimmy]

Hi @KSDaemon, thanks for your quick response!
Sure, here you go:

jdbc:databricks://XXXXX.cloud.databricks.com:443/default;transportMode=http;ssl=1;AuthMech=3;httpPath=/sql/1.0/warehouses/XXXXX;UID=token

[KSDaemon]

It's really interesting. I've just checked mine - and it's literally the same (only difference - no UID).

jdbc:databricks://XXXXX.cloud.databricks.com:443/default;transportMode=http;ssl=1;AuthMech=3;httpPath=/sql/1.0/warehouses/xxxxxxxxx

[KSDaemon]

@Klimmy do you have any other settings in your deployment related to Java? Cause the error is somehow related to Java Logger. And I my case - there is no problem. So it's something on the Java Side on deployment or maybe in Databricks...
Sorry, I'm guessing, I'm not an expert in Databricks Java driver %)

[Klimmy]

Hey @KSDaemon, sorry for the late reply — I was struggling to upgrade our Cube instance back to 1.2.29 to test it out once again. Support team helped with an upgrade and now it's working. I'm not sure what was the reason, as I don't think we have any Java-related settings.

To sum up, if anyone happens to face this issue:

1. Remove PWD=... and UID=token from your CUBEJS_DB_DATABRICKS_URL
2. Don't forget to keep your token in a separate variable CUBEJS_DB_DATABRICKS_TOKEN
3. If you face Error initializing the Java Util Logger (JUL), try running the query once again. From my experience, it happened for the first (of first several) query, and then go away.

Thanks for your help @KSDaemon! I believe we can close the issue, unless you want me to monitor it for this week and report back.

[KSDaemon]

@Klimmy thnx for the updates! Let's close it for now. If you see this once again or get any other related info - feel free to reopen!