Found in https://bugs.webkit.org/show_bug.cgi?id=36531 After having attached a patch for review and set the cq to +, the bot tried to land the patch directly and failed due to not finding any reviewer. When posting the second patch, the first patch was still r? but I obsoleted it. The new patch original flags were r? and cq+.
If you mark a patch commit-queue+, the bot will try to land it. If it wasn't reviewed, it doesn't sound like it's ready for commit. Why did you mark it commit-queue+?
Oh posh. I think we should handle this case. We should not try to land patches which still have r? or which are marked r-. Only empty review or r+.
(In reply to comment #1) > If you mark a patch commit-queue+, the bot will try to land it. If it wasn't > reviewed, it doesn't sound like it's ready for commit. Why did you mark it > commit-queue+? Simple, because I wanted the bot to land it for me (I usually land my patches manually to deal with the fallout). This one was special: it was a test fix so it was better to get it landed ASAP. I have barely used the commit bot so my naive thought was that the bot would just wait until was reviewed to land it.
Created attachment 78357 [details] should work, needs testing
Comment on attachment 78357 [details] should work, needs testing Looks right.
Created attachment 78381 [details] Patch
Comment on attachment 78381 [details] Patch Now, this patch, on the other hand, is very nice.
Comment on attachment 78381 [details] Patch Clearing flags on attachment: 78381 Committed r75366: <http://trac.webkit.org/changeset/75366>
All reviewed patches have been landed. Closing bug.