I think that this issue isn't really about the docker-plugin, but it's more about the core - when there are a lot of "cloud" instances configured, the main config page takes ages to load. We've got the same symptoms, but that's mostly from vSphere and OpenStack clouds - there's a lot of field verification going on all at once, and this takes time. FYI one of the changes I made to the docker plugin was to hide a lot of information within "optional" sections so that the web-page wasn't horribly long by default, allowing the Jenkins admin to only "expand" the sections they were interested in, but this doesn't make much difference to loading speed, only to the visual appearance and navigability of the page. I think the solution here is for the core to automatically split out the "Clouds" bit into multiple pages when it starts to get large. That isn't an enhancement that belongs to the docker-plugin, that's a wider issue. What I would suggest is that the Jenkins main jelly code has some sort of "if clouds.size>=1 then punt Clouds into its own page" logic, e.g. have a separate "Clouds" section that magically appears when clouds.size>=1 and have the "Clouds" section in the main config magically turn from the current menu/list (that can become huge and slow) into a link to that new page whenever clouds.size >= 1. ...or, alternatively, make it unconditional, much like what was done when the Jenkins "thread dump" functionality was moved to its own page. Either way, it's not something that the docker-plugin can do on its own - that's a core issue. I'll amend the "components" field accordingly... |