Getting /bin/bash^M: bad interpreter: No such file or directory after running initialisation script on Dataproc

Question

I am getting the below error when I am trying to create a Dataproc cluster in GCP. I am using a windows machine to make changes to GCS bucket which will be used to initialize the dataproc cluster.

/tmp/dataproc-agent1435981490491006254/initialize-env.sh: /etc/google-dataproc/startup-scripts/dataproc-initialization-script-0: /bin/bash^M: bad interpreter: No such file or directory

score 3 · Accepted Answer · answered Aug 24 '22 at 14:29

3

I could understand that when I use Windows machine to push changes to GCS bucket the EOL conversion was not done. Windows and Linux use different characters for declaring new lines or the end of file. So to fix this I used Notepad ++ and from edit menu I used Unix EOL conversion. This was the cause of the error which i was getting while creating a Dataproc cluster.

Reference : How to fix EOL conversion issue

answered Aug 24 '22 at 14:29

sidharth vijayakumar

1,190
5
29

1

Exactly! CRLF usually causes this problem. Running the utility [dos2unix](https://linux.die.net/man/1/dos2unix) should be able to fix this error too. – Bruno Volpato Aug 24 '22 at 16:17
1

Yea. I had saw that too but this was more easier way in my case. – sidharth vijayakumar Aug 24 '22 at 16:34

Getting /bin/bash^M: bad interpreter: No such file or directory after running initialisation script on Dataproc

1 Answers1