I had several issues bringing up conifer with docker-compose on NixOS (a linux distribution) - but it was very simple otherwise. All of these issues were in one way or another permissions related. I'm not sure if this may be due to configuration differences in docker on NixOS and whatever the conifer project may be using (ubuntu? debian?). A friend mentioned it may be a lack of SELinux on NixOS.
I am also not familiar with docker, beyond what fiddling I did here.
I will fill out this issue when I have some more capacity, but off the top of my head:
for some reason directories that were volume roots, created in repo/data were created with root:root ownership. This is obviously not the correct solution, but I was able to temporarily work around this by setting the directories chown 777.
groupadd: GID '100' already exists
groupmod: group 'archivist' does not exist
This may have something to do with: for some reason the webrecorder/pywb (and possibly other images) have VOLUME_DIR set to /data despite the Dockerfile setting it to /webarchive, and I have no idea why.
This issue causes several containers to fail to start.
I worked around it temporarily by setting the uid:gid of /data to a nonexistent uid:gid pair and just set the perms so I could access everything.
node in whichever (app?) container wanted access to webrecorder/webrecorder/proxy-certs.pem, as well as webrecorder/webrecorder/static. So I just chmod ou+w the entire directory.
I probably forgot somthing, and would have to go through the process again, but other than these problems, everything went quite painlessly. Congratulations.
I had several issues bringing up conifer with docker-compose on NixOS (a linux distribution) - but it was very simple otherwise. All of these issues were in one way or another permissions related. I'm not sure if this may be due to configuration differences in docker on NixOS and whatever the conifer project may be using (ubuntu? debian?). A friend mentioned it may be a lack of SELinux on NixOS.
I am also not familiar with docker, beyond what fiddling I did here.
I will fill out this issue when I have some more capacity, but off the top of my head:
This may have something to do with: for some reason the webrecorder/pywb (and possibly other images) have VOLUME_DIR set to /data despite the Dockerfile setting it to /webarchive, and I have no idea why. This issue causes several containers to fail to start. I worked around it temporarily by setting the uid:gid of /data to a nonexistent uid:gid pair and just set the perms so I could access everything.
app?) container wanted access to webrecorder/webrecorder/proxy-certs.pem, as well as webrecorder/webrecorder/static. So I just chmod ou+w the entire directory.I probably forgot somthing, and would have to go through the process again, but other than these problems, everything went quite painlessly. Congratulations.