Skip to:

  1. Skip to Jira Navigation
  2. Skip to Side Navigation
  3. Skip to Main Content

Replace Artifactory with Nexus

Description

Nexus is the main competitor to Artifactory. It has similar features. Lokking at documentation available about it, it looks like the open source version of it supports a much bigger feature set the Artifactory without requiring an upgrade to a payed version:

https://help.sonatype.com/display/NXRM3/Repository+Manager+Feature+Matrix

Of interest to us are the PyPi (Python) support, the NPM (Javascript) and the Docker support.

It seems that Nexus is available as a container, so we may try deploying it on OpenShift.

I think we need to plan and execute a migration to Nexus.

This ticket should server as a tracking ticket for the entire work. We will create blocker tickets to track needed steps.

Activity

Show:

Anton Marchukov February 15, 2019 at 2:44 PM

We should open items for specific problems instead

Eyal Edri June 3, 2018 at 1:25 PM

I don't see this as an immediate need to solve any issue we're facing, let's make sure we don't invest in it if not needed fro a specific task/value.

Barak Korren November 25, 2017 at 2:27 PM
Edited

I actually played with it a little already and have both the V2 and the V3 containers installed on OpenShift ATM (With no persistent storage though)

At the time i was looking at hosting RPMs, and it does not seem to have the features we need for that, but I might look at it again for hosting container images soon.

Having said that, if anyone wants to look at using it as a Maven cache for engine builds, be my guest....

BTW, I've seen some docs discussing how to use Nexus as the container repository for OpenShift itself. Not sure we need to be looking at that though.

Former user November 23, 2017 at 4:39 PM

Setting up nexus is described in detail in official docs (linked).

Need to define the volume size to use for artifacts. We can probably survive without persistent storage as long as mirror configurations are saved in the deployment configs but we should then set up quotas to not let the pod fill up the node's disk.

Won't Fix

Details

Assignee

Reporter

Components

Priority

Created August 26, 2017 at 6:57 AM
Updated August 29, 2019 at 2:12 PM
Resolved February 15, 2019 at 2:44 PM