utkarsh-1905
commented
6 months ago When manually writing the variable in /opt/couchdb/etc/vm.args the same error is being thrown and container crashes.
To reproduce:
docker exec -it rcdb-3 cat /opt/couchdb/etc/vm.args > vm.args- Add
-name couchdb@127.0.0.1(from docs) at the end of this file. - Copy the file
docker cp <your-local-path>/vm.args rcdb-3:/opt/couchdb/etc/ - Restart the container, it will crash
Other ways to test out:
At port 5985 is the official image of couchdb running where as on 5984, rapidfort/couchdb-official is running, we can see that in the first container, clustering is working fine and setup is done, where as in second container its giving an error due to above mentioned causes
Name and Version
rapidfort/couchdb-official:3.3.3
Which runtime are you using to reproduce this issue?
Is this issue reproducible on the original source image?
Reproducible
Could you please identify the category? Details in TROUBLE_SHOOTING.md
Unknown
What steps will reproduce the bug?
Run the container using the following command
docker run -d --name rcdb-3e -e COUCHDB_SECRET=hello -e COUCHDB_USER=admin -e COUCHDB_PASSWORD=password -e ERL_FLAGS="-setcookie brumbrum" -e NODENAME=couchdb1 rapidfort/couchdb-official:3.3.3The container should exit.Are you using any custom parameters or values?
Yes,
-e NODENAME=couchdb1, the error is coming on passing this variableNo response
What is the expected behavior?
The container should start normally
What do you see instead?
sh: 1: /opt/couchdb/bin/../erts-12.3.2.14/bin/epmd: Exec format erroras it is not able to parse it. I believe it has similar reasons related to env as #610Additional information
This prevents the user to create a cluster setup directly. Other environment variables are parsed only this is one is crashing the container. Docs and requirement of the
NODENAMEvariable.Rapidfort Image => rapidfort/couchdb-official:3.3.3 tested against => couchdb:3.3