Mathias Bauer wrote: > [ ... snip ...]
(4) The developer that fixed the error must be able to restart the tinderbox build immediately to get a result in an acceptable time frame.
As long as we do not have an endless stack of tinderbox build boxes available which will never be the case you will have to accept that there is a build queue. A cvs commit on the CWS automatically triggers a restart of the tinderbox build this means that we do not need some kind of button to manually restart a build anywhere (except probably for those cases where there was some other error like network problems etc.) and this is as immediately as it can get without stopping other builds currently running in that queue.
It´s the same as with performance tests and convwatch tests where we currently have just one computer per platform and you also have to wait in the queue.
AFAIK there are some people already discussing the possible work flow. So we don't need to rush things.
Agreed. With that Wiki entry i just wanted to show the current possibilities as background information for the ongoing discussion.
[ ... snip ...]
>
Ciao, Mathias
Ciao, Bernd --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
