Closed kript closed 2 years ago
note to self, following UGM hackathon, a suggestion is that it's failing because of two blank lines (line 8 and 9) in the README.md.
https://app.codacy.com/gh/irods/irods_demo/pullRequest?prid=7485349
I feel that the iRODS testing environment is a better fit for this than the demo. We are also about to rework this project (see #29) and I have demonstrated on that branch that the iRODS service can run standalone. I think we can close this in light of this. Thoughts?
I agree. This PR can be closed.
Agreed.
Maybe we put some instructions in the README about how to stand up certain services...
That makes sense to me. A line or two explaining the how/why is good enough IMO. We can direct them to the docker-compose docs for additional info.
I added a line to the README explaining how to stand up individual services using docker-compose
.
I ran into a couple of issues when trying to use the irods container on its own, rather than as part of a compose.
The main issue was that without reserving more shared memory for the docker container, it wouldn't restart. This can be seen by a boost error in the rodsServerLog when the server was restarted after the
pkill
.I also found that it can take some time for the processes to finish, otherwise, the restart would fail as the port was still held open.
In addition, since 4.2.9 was released, there is no requirement for the
irods-rule-engine-plugin-update-collection-mtime
package.Currently, this MR pins the packages to 4.2.8, the last release supported by the package until it was subsumed by 4.2.9