Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [cross-project-issues-dev] Intermittent p2 failures, best practices for "mirroring"

So we've been occasionally getting failures on http://download.eclipse.org/releases/oxygen/201804111000/content.xml.xz and other similar sites over this past week (including this morning EDT).  From the archive I see this has been an issue this past week so I guess you guys are doing something about it but I have some questions relating to setting up a cache that may save you a small amount of bandwidth.


Does tycho support some form of fail-over such that if we set up a cache for p2 and that server goes down we can fail back to the internet?  I'm using maven+tycho and at one point it seemed like adding multiple p2 repos didn't really help for some cases of failure (the other repo wouldn't get tried), but this was forever ago so maybe i'm miss-remembering or the situation was very specific.

Also is there a preference for caching proxies for those of us who want to stay on the FOSS side of things (even if we get support later).  I've used Nexus 2.x (still do) but it's p2 support wasn't great at the time from what i could tell; I'm looking a bit at Artifactory, is it worth switching?  There's nexus 3.x as well

Any advice would be greatly appreciated
Thanks
Kevin

Back to the top