From dylan at mozilla.com Thu Jun 2 16:06:22 2016 From: dylan at mozilla.com (Dylan Hardison) Date: Thu, 2 Jun 2016 12:06:22 -0400 Subject: What's up, Bugzilla? Message-ID: Bugs are good for communicating discrete tasks, but I feel a monthly check-in would be healthy for the project. With that in mind, let this be the first volume of "What's up, Bugzilla?" This will obviously not be exhaustive, and is just from my perspective. :-) I'm getting pretty excited about the hopefully-soon release of 6.0. Having the sandstone skin as default[1] the cleaner REST framework (which will hopefully allow us to have more RESTy APIs in the future), faster memcached (both Cache::Memcached::Fast[2] and the recent patch to remove (un-needed) detainting)[3] a standard-ish perl dependency situation[4], and a pretty friendly out-of-the-box development experience[5], these are just some of my favorite things. :) Things To Do for the 6.0 release. 1) Currently :justdave is reviewing Bug 988971[1], but nothing stops other people from taking that patch and applying it to master, and reporting back their findings. Doing that would definitely be appreciated :-) 2) I'm sure :dkl and/or :LpSolit wouldn't mind someone becoming familiar with the release note process. There is a guide for it [10]. 3) I've been trying to create higher-quality bugs, especially when I can't find time to work on them. Many of these are performance related: - Using Sereal for memcache would yield a good boost for sites using memcached [6]. - Being smarter about creating custom fields in Bugzilla::Bug object. One approach is described in the user story field [7]. - There's no bug open for this, but I would like to have more tests for our Markdown implementation. This is probably the easiest to test component of the whole system, and we've had a lot of formatting bugs [8]. There are things to look for beyond 6.0: The Modal UI, used on BMO, is probably not going to make the 6.0 release. Development of it on the BMO side of things is progressing at a furious rate and it is likely to be defaulted to on at the end of this quarter. The task of making that work for vanilla Bugzilla is straight-forward although tedious. A good first step for this would be getting Needinfo included as a core feature[9]. It's not so much asking for help (we'll get to where we're going eventually) but more to communicate what's going on, and to highlight some things that someone else *could* do without having to invest a significant amount of time. [1]: https://bugzilla.mozilla.org/show_bug.cgi?id=988971 "988971 ? Replace Classic with "Sandstone" skin as Standard skin" [2]: https://bugzilla.mozilla.org/show_bug.cgi?id=1189281 "1189281 ? Use Cache::Memcached::Fast instead of Cache::Memcached" [3]: https://bugzilla.mozilla.org/show_bug.cgi?id=1274764 "1274764 ? Increasing memcached performance is possible because Cache::Memcached::Fast does not need detainting" [4]: https://github.com/bugzilla/bugzilla/blob/master/Makefile.PL "bugzilla/Makefile.PL at master ? bugzilla/bugzilla ? GitHub" [5]: http://dylanwh.tumblr.com/post/144277013247/developing-bugzilla-with-plackup-and-cpanminus "snippets Developing Bugzilla with Plackup and cpanminus " [6]: https://bugzilla.mozilla.org/show_bug.cgi?id=1275545 "1275545 ? Use Sereal for Cache::Memcached::Fast" [7]: https://bugzilla.mozilla.org/show_bug.cgi?id=1275112 "1275112 ? _create_cf_accessors() should be called once at startup, not as part of Bugzilla::Bug->new" [8]: https://github.com/bugzilla/bugzilla/blob/master/t/100markdown.t "bugzilla/100markdown.t at master ? bugzilla/bugzilla ? GitHub" [9]: https://bugzilla.mozilla.org/show_bug.cgi?id=1277592 "1277592 ? Bug Modal will require the Needinfo extension to be a core feature of Bugzilla" [10]: https://www.bugzilla.org/docs/releasenote.html "Release Noting Guide :: Bugzilla :: bugzilla.org" From lpsolit at gmail.com Thu Jun 2 19:17:08 2016 From: lpsolit at gmail.com (=?UTF-8?B?RnLDqWTDqXJpYyBCdWNsaW4=?=) Date: Thu, 2 Jun 2016 21:17:08 +0200 Subject: What's up, Bugzilla? In-Reply-To: References: Message-ID: <1aa52f7c-2b3a-aa93-3287-186819a64f97@gmail.com> > Things To Do for the 6.0 release. 1') Replace YUI code by jQuery. Bugzilla 6.0 won't be released till this task is done. https://bugzilla.mozilla.org/show_bug.cgi?id=1090710 From jefffearn at gmail.com Fri Jun 3 10:03:12 2016 From: jefffearn at gmail.com (Jeff Fearn) Date: Fri, 3 Jun 2016 20:03:12 +1000 Subject: Fighting Safe for code coverage Message-ID: Hi, I'm trying to get some code coverage metrics and running in to an issue where Safe is preventing Devel::Cover from loading properly ... aiui anyway :) Has anyone got Devel::Cover working with the upstream tests? I've tried every option to Safe I've been able to find and it's not changed anything :( This is a start to getting some unit tests in to our CI system to run before integration testing. Just prove: $ prove t/013dbschema.t t/013dbschema.t .. ok All tests successful. Files=1, Tests=123, 1 wallclock secs ( 0.03 usr 0.01 sys + 0.76 cusr 0.04 csys = 0.84 CPU) Result: PASS excellent, but: $ HARNESS_PERL_SWITCHES=-MDevel::Cover prove t/013dbschema.t t/013dbschema.t .. Can't locate object method "digest" via package "Devel::Cover::DB::Structure" (perhaps you forgot to load "Devel::Cover::DB::Structure"?) at /usr/lib64/perl5/vendor_perl/Devel/Cover/DB/Digests.pm line 66. No language defines the string 'error_localconfig_read' at Bugzilla/Install/Localconfig.pm line 137. BEGIN failed--compilation aborted at t/013dbschema.t line 65. t/013dbschema.t .. Dubious, test returned 255 (wstat 65280, 0xff00) No subtests run Test Summary Report ------------------- t/013dbschema.t (Wstat: 65280 Tests: 0 Failed: 0) Non-zero exit status: 255 Parse errors: No plan found in TAP output Files=1, Tests=0, 13 wallclock secs ( 0.03 usr 0.00 sys + 12.52 cusr 0.16 csys = 12.71 CPU) I added a warn to get the real error: $ git diff Bugzilla/Install/Localconfig.pm diff --git a/Bugzilla/Install/Localconfig.pm b/Bugzilla/Install/Localconfig.pm index 72c26a4..b6cf8ba 100644 --- a/Bugzilla/Install/Localconfig.pm +++ b/Bugzilla/Install/Localconfig.pm @@ -134,6 +134,7 @@ sub read_localconfig { $s->rdo($filename); if ($@ || $!) { my $err_msg = $@ ? $@ : $!; + warn $err_msg; die install_string('error_localconfig_read', { error => $err_msg, localconfig => $filename }), "\n"; } The rest of the tests work fine. I've also tried ignoring the file, but I couldn't get any of the +/-ignore options to actually ignore this file. Help me Obiwan! Cheers, Jeff. From dylan at mozilla.com Fri Jun 3 12:21:10 2016 From: dylan at mozilla.com (Dylan Hardison) Date: Fri, 3 Jun 2016 08:21:10 -0400 Subject: Fighting Safe for code coverage In-Reply-To: References: Message-ID: What version of Devel::Cover and Safe? I can't reproduce here. perl -MSafe\ 9999 perl -MDevel::Cover\ 9999 On Fri, Jun 3, 2016 at 6:03 AM, Jeff Fearn wrote: > Hi, I'm trying to get some code coverage metrics and running in to an > issue where Safe is preventing Devel::Cover from loading properly ... > aiui anyway :) Has anyone got Devel::Cover working with the upstream > tests? I've tried every option to Safe I've been able to find and it's > not changed anything :( > > > This is a start to getting some unit tests in to our CI system to run > before integration testing. > > Just prove: > > > $ prove t/013dbschema.t > t/013dbschema.t .. ok > All tests successful. > Files=1, Tests=123, 1 wallclock secs ( 0.03 usr 0.01 sys + 0.76 cusr > 0.04 csys = 0.84 CPU) > Result: PASS > > excellent, but: > > $ HARNESS_PERL_SWITCHES=-MDevel::Cover prove t/013dbschema.t > t/013dbschema.t .. Can't locate object method "digest" via package > "Devel::Cover::DB::Structure" (perhaps you forgot to load > "Devel::Cover::DB::Structure"?) at > /usr/lib64/perl5/vendor_perl/Devel/Cover/DB/Digests.pm line 66. > No language defines the string 'error_localconfig_read' at > Bugzilla/Install/Localconfig.pm line 137. > BEGIN failed--compilation aborted at t/013dbschema.t line 65. > t/013dbschema.t .. Dubious, test returned 255 (wstat 65280, 0xff00) > No subtests run > > Test Summary Report > ------------------- > t/013dbschema.t (Wstat: 65280 Tests: 0 Failed: 0) > Non-zero exit status: 255 > Parse errors: No plan found in TAP output > Files=1, Tests=0, 13 wallclock secs ( 0.03 usr 0.00 sys + 12.52 cusr > 0.16 csys = 12.71 CPU) > > > I added a warn to get the real error: > > > $ git diff Bugzilla/Install/Localconfig.pm > diff --git a/Bugzilla/Install/Localconfig.pm > b/Bugzilla/Install/Localconfig.pm > index 72c26a4..b6cf8ba 100644 > --- a/Bugzilla/Install/Localconfig.pm > +++ b/Bugzilla/Install/Localconfig.pm > @@ -134,6 +134,7 @@ sub read_localconfig { > $s->rdo($filename); > if ($@ || $!) { > my $err_msg = $@ ? $@ : $!; > + warn $err_msg; > die install_string('error_localconfig_read', > { error => $err_msg, localconfig => $filename }), "\n"; > } > > The rest of the tests work fine. I've also tried ignoring the file, but > I couldn't get any of the +/-ignore options to actually ignore this file. > > Help me Obiwan! > > > Cheers, Jeff. > > > - > To view or change your list settings, click here: > From jefffearn at gmail.com Fri Jun 3 23:42:58 2016 From: jefffearn at gmail.com (Jeff Fearn) Date: Sat, 4 Jun 2016 09:42:58 +1000 Subject: Fighting Safe for code coverage In-Reply-To: References: Message-ID: On 3/06/2016 10:21 PM, Dylan Hardison wrote: > What version of Devel::Cover and Safe? I can't reproduce here. > > perl -MSafe\ 9999 > perl -MDevel::Cover\ 9999 Safe 2.37 Devel::Cover 1.17 perl 5.20.3 Oh nasty, I ran this on a clean upstream and it works fine, I guess I need to go diving in to our patches and extensions >_< Thanks, Jeff. > On Fri, Jun 3, 2016 at 6:03 AM, Jeff Fearn wrote: >> Hi, I'm trying to get some code coverage metrics and running in to an >> issue where Safe is preventing Devel::Cover from loading properly ... >> aiui anyway :) Has anyone got Devel::Cover working with the upstream >> tests? I've tried every option to Safe I've been able to find and it's >> not changed anything :( >> >> >> This is a start to getting some unit tests in to our CI system to run >> before integration testing. >> >> Just prove: >> >> >> $ prove t/013dbschema.t >> t/013dbschema.t .. ok >> All tests successful. >> Files=1, Tests=123, 1 wallclock secs ( 0.03 usr 0.01 sys + 0.76 cusr >> 0.04 csys = 0.84 CPU) >> Result: PASS >> >> excellent, but: >> >> $ HARNESS_PERL_SWITCHES=-MDevel::Cover prove t/013dbschema.t >> t/013dbschema.t .. Can't locate object method "digest" via package >> "Devel::Cover::DB::Structure" (perhaps you forgot to load >> "Devel::Cover::DB::Structure"?) at >> /usr/lib64/perl5/vendor_perl/Devel/Cover/DB/Digests.pm line 66. >> No language defines the string 'error_localconfig_read' at >> Bugzilla/Install/Localconfig.pm line 137. >> BEGIN failed--compilation aborted at t/013dbschema.t line 65. >> t/013dbschema.t .. Dubious, test returned 255 (wstat 65280, 0xff00) >> No subtests run >> >> Test Summary Report >> ------------------- >> t/013dbschema.t (Wstat: 65280 Tests: 0 Failed: 0) >> Non-zero exit status: 255 >> Parse errors: No plan found in TAP output >> Files=1, Tests=0, 13 wallclock secs ( 0.03 usr 0.00 sys + 12.52 cusr >> 0.16 csys = 12.71 CPU) >> >> >> I added a warn to get the real error: >> >> >> $ git diff Bugzilla/Install/Localconfig.pm >> diff --git a/Bugzilla/Install/Localconfig.pm >> b/Bugzilla/Install/Localconfig.pm >> index 72c26a4..b6cf8ba 100644 >> --- a/Bugzilla/Install/Localconfig.pm >> +++ b/Bugzilla/Install/Localconfig.pm >> @@ -134,6 +134,7 @@ sub read_localconfig { >> $s->rdo($filename); >> if ($@ || $!) { >> my $err_msg = $@ ? $@ : $!; >> + warn $err_msg; >> die install_string('error_localconfig_read', >> { error => $err_msg, localconfig => $filename }), "\n"; >> } >> >> The rest of the tests work fine. I've also tried ignoring the file, but >> I couldn't get any of the +/-ignore options to actually ignore this file. >> >> Help me Obiwan! >> >> >> Cheers, Jeff. >> >> >> - >> To view or change your list settings, click here: >> > - > To view or change your list settings, click here: > > From jefffearn at gmail.com Fri Jun 3 23:55:38 2016 From: jefffearn at gmail.com (Jeff Fearn) Date: Sat, 4 Jun 2016 09:55:38 +1000 Subject: Fighting Safe for code coverage In-Reply-To: References: Message-ID: <61cc9903-6b8e-a418-cb05-21f7fc196b67@gmail.com> On 4/06/2016 9:42 AM, Jeff Fearn wrote: > On 3/06/2016 10:21 PM, Dylan Hardison wrote: >> What version of Devel::Cover and Safe? I can't reproduce here. >> >> perl -MSafe\ 9999 >> perl -MDevel::Cover\ 9999 > > Safe 2.37 > Devel::Cover 1.17 > perl 5.20.3 > > Oh nasty, I ran this on a clean upstream and it works fine, I guess I > need to go diving in to our patches and extensions >_< Thanks for the pointer, I've isolated it to the ComponentWatching extension we lifted from BMO. Haven't determined the exact cause yet, but touch `extension/ComponentWatching/disabled` allows Devel::Cover to run again. Cheers, Jeff. From jefffearn at gmail.com Sat Jun 4 00:10:22 2016 From: jefffearn at gmail.com (Jeff Fearn) Date: Sat, 4 Jun 2016 10:10:22 +1000 Subject: Fighting Safe for code coverage In-Reply-To: <61cc9903-6b8e-a418-cb05-21f7fc196b67@gmail.com> References: <61cc9903-6b8e-a418-cb05-21f7fc196b67@gmail.com> Message-ID: <327570f2-2946-5bb1-1bdf-ab5c9bf41234@gmail.com> On 4/06/2016 9:55 AM, Jeff Fearn wrote: > On 4/06/2016 9:42 AM, Jeff Fearn wrote: >> On 3/06/2016 10:21 PM, Dylan Hardison wrote: >>> What version of Devel::Cover and Safe? I can't reproduce here. >>> >>> perl -MSafe\ 9999 >>> perl -MDevel::Cover\ 9999 >> >> Safe 2.37 >> Devel::Cover 1.17 >> perl 5.20.3 >> >> Oh nasty, I ran this on a clean upstream and it works fine, I guess I >> need to go diving in to our patches and extensions >_< > > Thanks for the pointer, I've isolated it to the ComponentWatching > extension we lifted from BMO. Haven't determined the exact cause yet, > but touch `extension/ComponentWatching/disabled` allows Devel::Cover to > run again. And solved, an unneeded reference to Bugzilla->bdh was added to the top of db_schema_abstract_schema which causes some insanity. Someone owes me a cider, Matt! Cheers, Jeff. From jefffearn at gmail.com Sun Jun 5 23:51:02 2016 From: jefffearn at gmail.com (Jeff Fearn) Date: Mon, 6 Jun 2016 09:51:02 +1000 Subject: Fighting Safe for code coverage In-Reply-To: <327570f2-2946-5bb1-1bdf-ab5c9bf41234@gmail.com> References: <61cc9903-6b8e-a418-cb05-21f7fc196b67@gmail.com> <327570f2-2946-5bb1-1bdf-ab5c9bf41234@gmail.com> Message-ID: On 4/06/2016 10:10, Jeff Fearn wrote: > On 4/06/2016 9:55 AM, Jeff Fearn wrote: >> On 4/06/2016 9:42 AM, Jeff Fearn wrote: >>> On 3/06/2016 10:21 PM, Dylan Hardison wrote: >>>> What version of Devel::Cover and Safe? I can't reproduce here. >>>> >>>> perl -MSafe\ 9999 >>>> perl -MDevel::Cover\ 9999 >>> >>> Safe 2.37 >>> Devel::Cover 1.17 >>> perl 5.20.3 >>> >>> Oh nasty, I ran this on a clean upstream and it works fine, I guess I >>> need to go diving in to our patches and extensions >_< >> >> Thanks for the pointer, I've isolated it to the ComponentWatching >> extension we lifted from BMO. Haven't determined the exact cause yet, >> but touch `extension/ComponentWatching/disabled` allows Devel::Cover to >> run again. > > And solved, an unneeded reference to Bugzilla->bdh was added to the > top of db_schema_abstract_schema which causes some insanity. > > Someone owes me a cider, Matt! Turns out it's not Matt, it's DKL! https://github.com/mozilla/webtools-bmo-bugzilla/commit/634f31d6c64894bcf2f0788525e9a9286e9de314 Cheers, Jeff. From gerv at mozilla.org Tue Jun 14 22:17:02 2016 From: gerv at mozilla.org (Gervase Markham) Date: Tue, 14 Jun 2016 23:17:02 +0100 Subject: Landfill slowness Message-ID: Why is Landfill so slow, does anyone know? It just took me 30 seconds to load https://landfill.bugzilla.org/bugzilla-5.0-branch/ , and I'm fairly sure it's not my wifi. Gerv _______________________________________________ dev-apps-bugzilla mailing list dev-apps-bugzilla at lists.mozilla.org https://lists.mozilla.org/listinfo/dev-apps-bugzilla From jfearn at redhat.com Tue Jun 14 22:24:42 2016 From: jfearn at redhat.com (Jeff Fearn) Date: Wed, 15 Jun 2016 08:24:42 +1000 Subject: Landfill slowness In-Reply-To: References: Message-ID: On 15/06/16 08:17, Gervase Markham wrote: > Why is Landfill so slow, does anyone know? It just took me 30 seconds to > load https://landfill.bugzilla.org/bugzilla-5.0-branch/ , and I'm fairly > sure it's not my wifi. Database was not in cache? 2.66s (onload: 2.76s) Cheers, Jeff. -- Jeff Fearn Senior Software Engineer PnT - DevOps - Development Red Hat Asia Pacific Pty Ltd http://dilbert.com/fast/2004-08-17/ PGP Fingerprint: B61A DC52 3E0E B17C 94D7 945C BB37 478C F119 9BCA From gerv at mozilla.org Wed Jun 15 08:05:54 2016 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 15 Jun 2016 09:05:54 +0100 Subject: Landfill slowness In-Reply-To: References: Message-ID: <2d6ea167-2d56-ba15-4275-cafb66e5a97e@mozilla.org> On 14/06/16 23:24, Jeff Fearn wrote: > On 15/06/16 08:17, Gervase Markham wrote: >> Why is Landfill so slow, does anyone know? It just took me 30 seconds to >> load https://landfill.bugzilla.org/bugzilla-5.0-branch/ , and I'm fairly >> sure it's not my wifi. > > Database was not in cache? > > 2.66s (onload: 2.76s) OK, it's faster now, but 30s to get it into cache? Wow. Gerv From jfearn at redhat.com Thu Jun 16 00:34:53 2016 From: jfearn at redhat.com (Jeff Fearn) Date: Thu, 16 Jun 2016 10:34:53 +1000 Subject: Landfill slowness In-Reply-To: <2d6ea167-2d56-ba15-4275-cafb66e5a97e@mozilla.org> References: <2d6ea167-2d56-ba15-4275-cafb66e5a97e@mozilla.org> Message-ID: On 15/06/16 18:05, Gervase Markham wrote: > On 14/06/16 23:24, Jeff Fearn wrote: >> On 15/06/16 08:17, Gervase Markham wrote: >>> Why is Landfill so slow, does anyone know? It just took me 30 seconds to >>> load https://landfill.bugzilla.org/bugzilla-5.0-branch/ , and I'm fairly >>> sure it's not my wifi. >> >> Database was not in cache? >> >> 2.66s (onload: 2.76s) > > OK, it's faster now, but 30s to get it into cache? Wow. It might be hosted in the Sydney AWS center and had to swim to a dry server! Cheers, Jeff From gerv at mozilla.org Wed Jun 22 13:31:06 2016 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 22 Jun 2016 14:31:06 +0100 Subject: Bugzilla meeting for June: cancelled Message-ID: <_5WdnbNx-dWHDvfKnZ2dnUU7-cfNnZ2d@mozilla.org> Hi everyone, Neither Dylan (YAPC) nor I (church group meeting) am available at the relevant time this month, and so we are cancelling this month's Bugzilla meeting. If you have things you wanted to say, you'll need to say them on this list :-) The next meeting is on the day I move house, but I'm still hoping to be there! Gerv _______________________________________________ dev-apps-bugzilla mailing list dev-apps-bugzilla at lists.mozilla.org https://lists.mozilla.org/listinfo/dev-apps-bugzilla