> One more case. With concurrent votes, it's possible for the IPMC vote to pass while the PPMC vote fails, either due to -1s or because three PPMC members never turn up. What happens then? > > Worst case, three IPMC members have reviewed an RC the podling rejected, and the IPMC has approved something the community producing it didn’t - slightly awkward.
Based on [1], "in most cases the ultimate decision lies with the individual serving as release manager." [1] https://www.apache.org/foundation/voting.html#ReleaseVotes I imagine several different situations where this happens, and I can find a reasonable solution for any of them. 1. IPMC members miss some compliance issues - the RM confirms, and it must be canceled. 2. No compliance issues, but there is a minor bug - typically can be deferred to the next release if cutting a release is lightweight. 3. No compliance issues, but there is a major bug - very rare. I can have some solutions here but basically it would be canceled due to our requirement on Code Quality [2] [2] https://community.apache.org/apache-way/apache-project-maturity-model.html#quality Let's look at the new policy positively: if the IPMC vote gets enough votes simultaneously, it typically comes from the podling's mentors. That means the podling is running well and its mentors are actively overseeing the podling. The new policy encourages these podlings to release as a TLP (a typical 72-hour time window). Best, tison. <[email protected]> 于2026年9月4日周五 00:22写道： > Hi, > > One more case. With concurrent votes, it's possible for the IPMC vote to > pass while the PPMC vote fails, either due to -1s or because three PPMC > members never turn up. What happens then? > > Worst case, three IPMC members have reviewed an RC the podling rejected, > and the IPMC has approved something the community producing it didn’t - > slightly awkward. > > Kind regards, > Justin > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
