Our scenario is to get .pdf files uploaded in AWS S3 storage and process it later. We want to move to AWS Lambda. However, Lambda requires that the entire package (along with all node_modules) be uploaded as a zip file (i.e. it wont run npm install). This means that tika picks up whatever java path that the local machine happened to have and save it in jvm_dll_path.json. The path to libjvm.so is different on the Lambda machine, and loading the module fails with "libjvm.so: cannot open shared object file: No such file or directory".
I tried just replacing the string in jvm_dll_path.json with the correct AWS path, but no dice.
Really appreciate any help to make this work on Lambda.
Thanks!
Our scenario is to get .pdf files uploaded in AWS S3 storage and process it later. We want to move to AWS Lambda. However, Lambda requires that the entire package (along with all node_modules) be uploaded as a zip file (i.e. it wont run npm install). This means that tika picks up whatever java path that the local machine happened to have and save it in jvm_dll_path.json. The path to libjvm.so is different on the Lambda machine, and loading the module fails with "libjvm.so: cannot open shared object file: No such file or directory". I tried just replacing the string in jvm_dll_path.json with the correct AWS path, but no dice. Really appreciate any help to make this work on Lambda. Thanks!