By the way, we do caching on the agent level.
Windows agents are not single-shot ones, and download happens only once per 
agent provisioning.
So it is not that bad at one may expect

On Thursday, May 2, 2019 at 3:56:01 PM UTC+2, R. Tyler Croy wrote:
>
> (replies inline) 
>
> On Thu, 02 May 2019, Jesse Glick wrote: 
>
> > On Thu, May 2, 2019 at 9:12 AM Oleg Nenashev <[email protected] 
> <javascript:>> wrote: 
> > > The releases are distributed from GitHub, so they are more stable even 
> if we continue downloading JDKs by URL. 
> > 
> > It ought to be easy to configure the (nginx?) mirror running in Azure 
> > for use by ci.jenkins.io builds to also mirror AdoptOpenJDK releases, 
> > so we are not transferring huge archives from GitHub???s download 
> > servers on every build. 
>
>
> This was discussed a bit during the infra meeting this past Tuesday, and 
> we 
> will not be mirroring GitHub's CDN. I am not worried about hitting their 
> edge 
> caches for releases. As Oleg pointed out in a ticket earlier: if GitHub is 
> down, ci.jenkins.io won't be very useful anyways :) 
>
>
> Of course, the infra team needs less services to babysit, not more. 
>
> -- 
> GitHub:  https://github.com/rtyler 
>
> GPG Key ID: 0F2298A980EE31ACCA0A7825E5C92681BEF6CEA2 
>

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/3c0fe7f1-43d1-4b95-bcc2-e5c07856e336%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to