Hmmm, not sure as once it has tried to transition to the new config it saves the new config so that it can be applied. I'll have a look through the code again. Most likely it will be Tuesday before I can take a look though
On 3 May 2013 22:34, Les Mikesell <[email protected]> wrote: > On Fri, May 3, 2013 at 2:19 PM, Les Mikesell <[email protected]> > wrote: > > >> If it doesn't then that's a bug, but nobody has alerted me to issues > with > >> that upgrade path. > > > > Can you describe what is "supposed" to happen in this upgrade? I > > just did an rpm update from a 1.5.00-1.1 (non LTS) to LTS 1.509.1-1 > > and none of the ssh-started slaves are working. > > Part of this may have involved the peculiar upgrade path here. The > initial error (which I can't reproduce now) had to do with some method > not being found. Updating the Credentials and SSH Credentials plugins > fixed that error, but only one slave started with the auto-converted > credentials. I had to re-enter them for the others. Some of the > production slaves are maintained by others so I don't have the > passwords or keys. Is there something I can do to fix things up from > the existing file contents if the production upgrade has similar > problems? > > -- > Les Mikesell > [email protected] > > -- > You received this message because you are subscribed to the Google Groups > "Jenkins Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/groups/opt_out. > > > -- You received this message because you are subscribed to the Google Groups "Jenkins Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
