Yes. If I ever need a snapshot, I take it an deploy it to my corp repository, at least that way when it works I know it will stay working. It is a pain to keep changing the repo though if you depend on many snapshots and maybe that's the case here.
Here's something I thought of several times that might help the internal deploy issue. What if in the pom you could only specify the repository id and not the url. Then in the settings you could define or override the pom s url. This would allow someone to redefine the repository to point local for deployment and make it easy to get the code and redeploy internally. I originally thought about this to use for continuum, my instance is on the same machine as my repo.... For example I would rather have it pointed to the file folder than use ftp for deploying. You can effectively do this now for "regular" repositories using mirrors, but I'm talking about redirecting distributionManagement repositories. -----Original Message----- From: Mike Perham [mailto:[EMAIL PROTECTED] Sent: Monday, April 17, 2006 8:28 PM To: Maven Developers List Subject: RE: rsync-ability for maven-snapshot-repository on apache? Why on earth are you depending on the network at all? We have our build locked down by using an internal mirror of central + custom snapshots. Anything else is non-deterministic and then management starts asking questions... -----Original Message----- From: Daniel Kulp [mailto:[EMAIL PROTECTED] Sent: Monday, April 17, 2006 8:03 PM To: dev@maven.apache.org Cc: Brett Porter; Carlos Sanchez Subject: Re: rsync-ability for maven-snapshot-repository on apache? Brett, > This is the main concern. We should be cutting releases this week, and > fewer people using the snapshots off of Apache. > > I'm curious why a maven-proxy in the middle doesn't alleviate this > issue for you. The maven-proxy doesn't help much on Monday mornings. The rest of the week it's OK. We have the timeout for snapshots set at 48 hours. If no-one does anything over the weekend (and this was a holiday weekend so no-one did), then the caches are pretty much moot on Monday. It was kind of amusing (not) this morning when we all came in. There were four of us trying to build and each of our builds were failing at different spots. Each would get a little bit further. All while trying to check the snapshot server. By the afternoon, we were pretty much all set, but I really don't want this to happen everyweek (or on Wednesday when the timeouts from today start expiring). We used to have the same problem with ibiblio (and even the mirrors) so we started rsyncing ibiblio local. Now we don't have problems with that, just the apache snapshot server. (the codehaus snapshot server is much better behaved) Thanks! -- J. Daniel Kulp Principal Engineer IONA P: 781-902-8727 C: 508-380-7194 [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
