topic Re: Databricks supported LTS version in Qlik Replicate

Databricks supported LTS version

MoeE — Wed, 25 Sep 2024 04:23:34 GMT

Hi,

I am looking through the QR versions from November 2022 up until May 2024. I noticed that none of these versions are explicitly listed to support an All-purpose cluster on runtime version 14.1 LTS.

Is this a typo or is it actually not supported? Context is we have a client that is having issues and they are on 14.1. Cause of the issue is not clear after investigation. I won't get into too much detail here.

Regards,

Mohammed

Re: Databricks supported LTS version

john_wang — Wed, 25 Sep 2024 05:01:14 GMT

Hello Mohammed, @MoeE

Thanks for opening the article.

In Support Matrix, the supported versions of All-purposes (Interactive) cluster are:

9.1 LTS
10.4 LTS
11.3 LTS
12.2 LTS
13.3 LTS
14.3 LTS

Version 14.1 LTS is not explicitly listed, and I wasn't able to find any specific information regarding this version. However, based on my understanding, it should work as expected. Could you please provide more detailed information about the specific issue you're encountering so that I can assist you more effectively?

Thanks,

John.

Re: Databricks supported LTS version

MoeE — Wed, 25 Sep 2024 23:15:06 GMT

Hi John,

Thanks. QR is working mostly as expected, we receive some 503 response errors from the target every now and then. Reason for that is not quite clear yet but I suspect its due to high target load/overuse. Plus some tables are suspended due to conflicts intermittently due to commit conflicts, yet the conflicting operations still can't be identified following long investigation.

Our next move is to enable a setting called row-level concurrency on the target side which may decrease commit conflicts as it allows operations to edit two different rows in the same file/partition without conflicts being raised.

Regards,

Mohammed

Re: Databricks supported LTS version

DesmondWOO — Thu, 26 Sep 2024 01:20:34 GMT

Hi @MoeE ,

Regarding 503 error, please check following article if it helps.

Databricks cluster resizing causes 503 error in Replicate

Regards,
Desmond