I just added a depending bug as well as a blocker on bug #321043... The information Mail sent out only contains the change on the dependent Bug... Noting about the new blocker!? Everything went as expected, it's just the mail that is a bit incomplete! ;-) Reproducible: Didn't try Steps to Reproduce:
Can you please show me the complete mail including the header?
Created attachment 233883 [details] the whole mail as requested
I've just seen that i have the tab still open with the process-messages above the bug... This is the relevant/missing part (i also haven't got a mail for that bug-change [i'm on the x86 alias]) Checking for dependency changes on bug 304057 Email sent to: ruby@gentoo.org, gentoolists@vectrocomputer.com, srinclan@gmail.com, ia64@gentoo.org, graaff@gentoo.org, myckel@sdf.lonestar.org, amd64@gentoo.org, x86@gentoo.org, pandaswarm@gmail.com, oehme.markus@gmx.de, xarthisius@gentoo.org, will@teknofire.net, voxus@gentoo.org, me@neurogeek.org.ve Excluding: mholzer@gentoo.org, diox@gentoo.org, Florian.Steinel@googlemail.com, bugbot@landfill.bugzilla.org If you wish to tweak the kinds of mail Bugzilla sends you, you can change your preferences.
Created attachment 233887 [details] bug-304057.txt I suppose that's the "missing" mail? ;)
Jep, this is the one! :-) Strange!? Do we have actually two bugs together!? -The first initial mail has a missing part (or do you have this info in the mail?) -The corresponding mail was only sent to amd64 and not to all cc's
(In reply to comment #5) > Jep, this is the one! :-) > > Strange!? Do we have actually two bugs together!? Each bug get a mail in this case (separate) and only for the changes related to the "own" bug. > -The first initial mail has a missing part (or do you have this info in the > mail?) > -The corresponding mail was only sent to amd64 and not to all cc's > Note the "X-Bugzilla-CC: amd64@gentoo.org, ia64@gentoo.org, x86@gentoo.org" each CC will get its own mail. So both mails seems to be complete so far, do you agree?
(In reply to comment #6) > So both mails seems to be complete so far, do you agree? > Yes, the headers are complete! :-) But it seems to have forgotten that x86 is on the cc of #304057 while actually producing the mails!?! -That was the last bug in the list of "checked dependency changes", according to the process-messages after sending... Maybe it was just too much in one take! ;-) On the other hand the "source bug" had the missing text in the mail and that was the first action according to the list!
I just had a look in bugzilla-3.6 and i personally would bet for a "my" in line 570 at process_bug.cgi! Just as a wild guess! ;-) # To get a list of all changed dependencies, convert the "changes" arrays # into a long string, then collapse that string into unique numbers in # a hash. my $all_changed_deps = join(', ', @{ $changes->{'dependson'} || [] }); -> $all_changed_deps = join(', ', @{ $changes->{'blocked'} || [] }, $all_changed_deps); my %changed_deps = map { $_ => 1 } split(', ', $all_changed_deps); # When clearning one field (say, blocks) and filling in the other # (say, dependson), an empty string can get into the hash and cause # an error later. delete $changed_deps{''};
Ok, it seems that i lost my bet! What do you like me to do now!? i switched the bug numbers around... sorry, I also got that email that you've attached! i miss the one for bug #312547 furthermore i once again found out, that computers are smarter than me after a beer! ;-) This dependency was already reported a week ago and bugzilla just ignored that i entered the same bug twice! ;-) Sorry for the noise, it's probably a bit late for me... lol ;-)
No problem ;)
