* bugzilla slowness
@ 2012-09-19 13:50 Richard Guenther
2012-09-19 14:07 ` Paolo Carlini
2012-09-19 15:53 ` Frank Ch. Eigler
0 siblings, 2 replies; 11+ messages in thread
From: Richard Guenther @ 2012-09-19 13:50 UTC (permalink / raw)
To: overseers
When you do a change to a bug that triggers e-mails to be sent
bugzilla is extremely slow (you can hit "stop" in your browser
after a few seconds and continue happily). I think it was tracked
down to synchronous e-mail sending and outgoing spam verification.
Time spent seems to scale with the number of people CCed in a bug.
Can we please white-list gcc.gnu.org accounts?
This really makes bugzilla triage as part of release-manager duties
extremely painful and time consuming.
Thanks,
Richard.
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: bugzilla slowness
2012-09-19 13:50 bugzilla slowness Richard Guenther
@ 2012-09-19 14:07 ` Paolo Carlini
2012-09-19 15:53 ` Frank Ch. Eigler
1 sibling, 0 replies; 11+ messages in thread
From: Paolo Carlini @ 2012-09-19 14:07 UTC (permalink / raw)
To: Richard Guenther; +Cc: overseers
On 09/19/2012 03:46 PM, Richard Guenther wrote:
> This really makes bugzilla triage as part of release-manager duties
> extremely painful and time consuming.
Seconded.
Paolo.
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: bugzilla slowness
2012-09-19 13:50 bugzilla slowness Richard Guenther
2012-09-19 14:07 ` Paolo Carlini
@ 2012-09-19 15:53 ` Frank Ch. Eigler
2012-09-19 16:46 ` Andreas Schwab
2012-09-20 14:55 ` Richard Guenther
1 sibling, 2 replies; 11+ messages in thread
From: Frank Ch. Eigler @ 2012-09-19 15:53 UTC (permalink / raw)
To: Richard Guenther; +Cc: overseers
Hi -
> When you do a change to a bug that triggers e-mails to be sent
> bugzilla is extremely slow [...]
As an experiment, we switched bugzilla to use /usr/lib/sendmail, which
seems to be faster. Please let us know of adverse consequences if you
notice any.
- FChE
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: bugzilla slowness
2012-09-19 15:53 ` Frank Ch. Eigler
@ 2012-09-19 16:46 ` Andreas Schwab
2012-09-19 21:12 ` Frank Ch. Eigler
2012-09-20 14:55 ` Richard Guenther
1 sibling, 1 reply; 11+ messages in thread
From: Andreas Schwab @ 2012-09-19 16:46 UTC (permalink / raw)
To: Frank Ch. Eigler; +Cc: Richard Guenther, overseers
"Frank Ch. Eigler" <fche@redhat.com> writes:
> As an experiment, we switched bugzilla to use /usr/lib/sendmail, which
> seems to be faster. Please let us know of adverse consequences if you
> notice any.
Apparently due to this change bugzilla mails now contain extra newlines.
Andreas.
--
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: bugzilla slowness
2012-09-19 16:46 ` Andreas Schwab
@ 2012-09-19 21:12 ` Frank Ch. Eigler
0 siblings, 0 replies; 11+ messages in thread
From: Frank Ch. Eigler @ 2012-09-19 21:12 UTC (permalink / raw)
To: Andreas Schwab; +Cc: Richard Guenther, overseers, LpSolit
Hi -
> Apparently due to this change [gcc/sourceware params switching to
> sendmail from smtp] bugzilla mails now contain extra newlines.
This appears to be https://bugzilla.mozilla.org/show_bug.cgi?id=571533
I'm game to hack on bugzilla's Mailer.pm, as per lpsolit's last advice.
- FChE
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: bugzilla slowness
2012-09-19 15:53 ` Frank Ch. Eigler
2012-09-19 16:46 ` Andreas Schwab
@ 2012-09-20 14:55 ` Richard Guenther
1 sibling, 0 replies; 11+ messages in thread
From: Richard Guenther @ 2012-09-20 14:55 UTC (permalink / raw)
To: Frank Ch. Eigler; +Cc: overseers
On Wed, 19 Sep 2012, Frank Ch. Eigler wrote:
> Hi -
>
> > When you do a change to a bug that triggers e-mails to be sent
> > bugzilla is extremely slow [...]
>
> As an experiment, we switched bugzilla to use /usr/lib/sendmail, which
> seems to be faster. Please let us know of adverse consequences if you
> notice any.
It feels faster. Let's hope it's not due to a different phase of the
moon ;)
Thanks,
Richard.
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: bugzilla slowness
2012-10-13 17:02 ` Ian Lance Taylor
2012-10-13 17:40 ` Markus Trippelsdorf
@ 2012-10-13 19:36 ` Frank Ch. Eigler
1 sibling, 0 replies; 11+ messages in thread
From: Frank Ch. Eigler @ 2012-10-13 19:36 UTC (permalink / raw)
To: Ian Lance Taylor; +Cc: Markus Trippelsdorf, overseers
Hi -
I believe the issue is a bug in bugzilla, in that for the
/usr/sbin/sendmail transport, it emits a bunch of \r characters into
the output, even on UNIX. It should either not do that, or else
invoke sendmail -ba.
e.g. https://bugzilla.mozilla.org/show_bug.cgi?id=571533#c7
- FChE
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: bugzilla slowness
2012-10-13 17:40 ` Markus Trippelsdorf
@ 2012-10-13 18:14 ` Markus Trippelsdorf
0 siblings, 0 replies; 11+ messages in thread
From: Markus Trippelsdorf @ 2012-10-13 18:14 UTC (permalink / raw)
To: Ian Lance Taylor; +Cc: Frank Ch. Eigler, overseers
On 2012.10.13 at 19:40 +0200, Markus Trippelsdorf wrote:
> On 2012.10.13 at 10:01 -0700, Ian Lance Taylor wrote:
> > Markus Trippelsdorf <markus@trippelsdorf.de> writes:
> >
> > >> This appears to be https://bugzilla.mozilla.org/show_bug.cgi?id=571533
> > >> I'm game to hack on bugzilla's Mailer.pm, as per lpsolit's last
> > >> advice.
> > >
> > > Ping. Can you please implement this? Because the status quo looks
> > > awful, see for example:
> > > http://thread.gmane.org/gmane.comp.gcc.bugs/366661
> >
> > I don't understand the reference to that thread. It doesn't seem to say
> > anything about slowness committing changes to bugzilla. It seems to me,
> > subjectively, that Frank's changes have made it somewhat faster.
>
> Sorry, this isn't about any slowness, but about the ugly extra empty
> lines that inserted when mail from bugzilla is received by Exim. See:
>
> https://bugzilla.mozilla.org/show_bug.cgi?id=571533
> (and the thread quoted above as an example)
Looking deeper, it appears that there are good and bad messages, even
when received by Exim. See for example:
http://thread.gmane.org/gmane.comp.gcc.bugs/366320
The first message displays just fine, all other messages in that thread
have the extra newlines.
The difference is that only the good message has:
"Content-Transfer-Encoding: quoted-printable"
in its header.
Only the bad messages all have the following header:
"X-Old-Lines: x"
(where x changes)
--
Markus
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: bugzilla slowness
2012-10-13 17:02 ` Ian Lance Taylor
@ 2012-10-13 17:40 ` Markus Trippelsdorf
2012-10-13 18:14 ` Markus Trippelsdorf
2012-10-13 19:36 ` Frank Ch. Eigler
1 sibling, 1 reply; 11+ messages in thread
From: Markus Trippelsdorf @ 2012-10-13 17:40 UTC (permalink / raw)
To: Ian Lance Taylor; +Cc: Frank Ch. Eigler, overseers
On 2012.10.13 at 10:01 -0700, Ian Lance Taylor wrote:
> Markus Trippelsdorf <markus@trippelsdorf.de> writes:
>
> >> This appears to be https://bugzilla.mozilla.org/show_bug.cgi?id=571533
> >> I'm game to hack on bugzilla's Mailer.pm, as per lpsolit's last
> >> advice.
> >
> > Ping. Can you please implement this? Because the status quo looks
> > awful, see for example:
> > http://thread.gmane.org/gmane.comp.gcc.bugs/366661
>
> I don't understand the reference to that thread. It doesn't seem to say
> anything about slowness committing changes to bugzilla. It seems to me,
> subjectively, that Frank's changes have made it somewhat faster.
Sorry, this isn't about any slowness, but about the ugly extra empty
lines that inserted when mail from bugzilla is received by Exim. See:
https://bugzilla.mozilla.org/show_bug.cgi?id=571533
(and the thread quoted above as an example)
--
Markus
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: bugzilla slowness
2012-10-13 16:29 Markus Trippelsdorf
@ 2012-10-13 17:02 ` Ian Lance Taylor
2012-10-13 17:40 ` Markus Trippelsdorf
2012-10-13 19:36 ` Frank Ch. Eigler
0 siblings, 2 replies; 11+ messages in thread
From: Ian Lance Taylor @ 2012-10-13 17:02 UTC (permalink / raw)
To: Markus Trippelsdorf; +Cc: Frank Ch. Eigler, overseers
Markus Trippelsdorf <markus@trippelsdorf.de> writes:
>> This appears to be https://bugzilla.mozilla.org/show_bug.cgi?id=571533
>> I'm game to hack on bugzilla's Mailer.pm, as per lpsolit's last
>> advice.
>
> Ping. Can you please implement this? Because the status quo looks
> awful, see for example:
> http://thread.gmane.org/gmane.comp.gcc.bugs/366661
I don't understand the reference to that thread. It doesn't seem to say
anything about slowness committing changes to bugzilla. It seems to me,
subjectively, that Frank's changes have made it somewhat faster.
Ian
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: bugzilla slowness
@ 2012-10-13 16:29 Markus Trippelsdorf
2012-10-13 17:02 ` Ian Lance Taylor
0 siblings, 1 reply; 11+ messages in thread
From: Markus Trippelsdorf @ 2012-10-13 16:29 UTC (permalink / raw)
To: Frank Ch. Eigler; +Cc: overseers
> This appears to be https://bugzilla.mozilla.org/show_bug.cgi?id=571533
> I'm game to hack on bugzilla's Mailer.pm, as per lpsolit's last
> advice.
Ping. Can you please implement this? Because the status quo looks
awful, see for example:
http://thread.gmane.org/gmane.comp.gcc.bugs/366661
Thanks.
--
Markus
^ permalink raw reply [flat|nested] 11+ messages in thread
end of thread, other threads:[~2012-10-13 19:36 UTC | newest]
Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-09-19 13:50 bugzilla slowness Richard Guenther
2012-09-19 14:07 ` Paolo Carlini
2012-09-19 15:53 ` Frank Ch. Eigler
2012-09-19 16:46 ` Andreas Schwab
2012-09-19 21:12 ` Frank Ch. Eigler
2012-09-20 14:55 ` Richard Guenther
2012-10-13 16:29 Markus Trippelsdorf
2012-10-13 17:02 ` Ian Lance Taylor
2012-10-13 17:40 ` Markus Trippelsdorf
2012-10-13 18:14 ` Markus Trippelsdorf
2012-10-13 19:36 ` Frank Ch. Eigler
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).