Talend 65535 bytes limit Solution (NOT WORKING)

[kotvir]

Dear Team,

The Solution provided on Talend Help portal of Talend (https://help.talend.com/r/en-US/7.2/code-generation-prevent-error) for Talend  65535 bytes limit is incorrect and does not working.

Also,

It seems there is a patch released by Talend to resolve this issue, but not sure how to download for Talend Open Studio 7.3.

https://help.talend.com/r/YUsadFksckofq8UWy846Ow/XCG67zmR6l5a7yNBpUdR7g

Not sure how to apply this patch to Talend Open Studio 7.3, Please advise.

Regards,

Vir

[Anonymous]

Hi

The patch is only available for enterprise subscription product. For open studio, please download the milestone version 7.4.1 from this page and test if the problem still exists.

If you still have the error, please provide additional information such as job design, I will see if I can help in any other way.

Regards

Shong

[kotvir]

Thanks Shong. Wondering why it's not available to download from Talend ?

[Anonymous]

You can find an answer from this post.

[gjeremy1617088143]

hi @Viral Kothari​ , here is some workaround to prevent the error :

https://community.talend.com/s/article/Exceeding-the-Java-bytes-limit-1Z1UZ

Send me love and kudos

[gjeremy1617088143]

the error could also appear if input component have too much columns

[kotvir]

I wish I could send you Kudos, I have already did this workaround. I am looking for JAVA based solution not the SPLIT work around.

[kotvir]

Hi Shong, 7.4.1 or even version 8 did not solve this issue.

[gjeremy1617088143]

the limit come from method code generated limitation in java, talend autogenerate code so if you made a lot of things in one subjob then it throw the error, you have to reduce the amount of generated code in subjob, maybe you can read the data and generate an id with sequence for example,then you store it in a hash

then you call the hash in differents subjob with a tfiltercolumn to reduce the amount of column then you transform data for each 100 columns for example then you aggregate all the columns with the id generated, you send it to an other hash, and finally you read the hash to send data directly to dboutput