On Jan 5, 2013, at 9:04 AM, Joshua Root wrote:
> I've actually never tried forcing a build on multiple slaves from the
> builders page.
I also always only used a single bb.
___
macports-dev mailing list
macports-dev@lists.macosforge.org
http://lists.maco
On 2013-1-5 13:45 , Ryan Schmidt wrote:
>
> All that happened was that I was redirected to https://build.macports.org .
> No error message was displayed and looking at the console it doesn't look
> like the build was started.
IIRC, that's what used to happen when you entered a portlist containi
On 2013-1-5 13:45 , Ryan Schmidt wrote:
>
> On Jan 4, 2013, at 16:25, mk-macpo...@techno.ms wrote:
>
>> On Jan 4, 2013, at 11:16 PM, Ryan Schmidt wrote:
>>
>>> I am logged in, and I'm looking at this page:
>>>
>>> https://build.macports.org/builders
>>>
>>> If I want to force a particular port to
On 2013-1-5 09:25 , mk-macpo...@techno.ms wrote:
> On Jan 4, 2013, at 11:16 PM, Ryan Schmidt wrote:
>
>> I am logged in, and I'm looking at this page:
>>
>> https://build.macports.org/builders
>>
>> If I want to force a particular port to build* on two of the buildbots, what
>> field do I put the
On Jan 4, 2013, at 16:25, mk-macpo...@techno.ms wrote:
> On Jan 4, 2013, at 11:16 PM, Ryan Schmidt wrote:
>
>> I am logged in, and I'm looking at this page:
>>
>> https://build.macports.org/builders
>>
>> If I want to force a particular port to build* on two of the buildbots, what
>> field do
On Jan 4, 2013, at 11:16 PM, Ryan Schmidt wrote:
> I am logged in, and I'm looking at this page:
>
> https://build.macports.org/builders
>
> If I want to force a particular port to build* on two of the buildbots, what
> field do I put the port name into? We've had the buildbots for so long now,
On Aug 2, 2011, at 11:16, Joshua Root wrote:
> Any or all of the values in
> the form can be blank. The portlist is actually meant to be space
> separated, but a buildbot bug is currently preventing requesting more
> than one port from working.
I am logged in, and I'm looking at this page:
http
On Aug 2, 2011, at 3:30 PM, William Siegrist wrote:
> You need to setup a master and slave buildbot instance locally to test these
> changes. You might find virtualenv helpful for that. See buildbot.net for
> more details.
Something like this?
http://trac.buildbot.net/wiki/RunningBuildbotWith
You need to setup a master and slave buildbot instance locally to test these
changes. You might find virtualenv helpful for that. See buildbot.net for more
details.
-Bill
On Aug 2, 2011, at 1:08 PM, Marko Käning wrote:
> From the lengthy manual example I get the impression that it might be q
I see now that it does work from http://build.macports.org/builders/buildports
but it doesn't from http://build.macports.org/builders
with the buildports bot checkmarked/selected.
___
macports-dev mailing list
macports-dev@lists.macosforge.org
http://list
I just tried to force a build, since the bot was idling around, but it wouldn't
allow me to do so:
...
Authentication Failed
The username or password you entered were not correct. Please go back and try
again.
…
is what I got
As username I used my macports handle m...@macports.org
There is
>From the lengthy manual example I get the impression that it might be quite
>resource-wasting to go through the whole build log and extract the relevant
>lines where an error occurs.
I just wished there was an easy way to test the previously sent script in a
safe environment.
_
Here is my slightly adapted version:
---
from buildbot.status.builder import Results
def messageFormatter(mode, name, build, results, master_status):
result = Results[results]
text = list()
text.append("STATUS: %s" % result.title())
return {
On Aug 2, 2011, at 9:38 PM, Bradley Giesbrecht wrote:
> Having a summarized build result emailed to a mailing list
> (build...@macports.org) would be a nice start.
>
> From http://buildbot.net/buildbot/docs/current/:
>
>> For example, if only short emails are desired (e.g., for delivery to phon
On Aug 2, 2011, at 12:29 PM, William Siegrist wrote:
> On Aug 2, 2011, at 12:26 PM, Marko Käning wrote:
>
>>
>> On Aug 2, 2011, at 6:06 PM, William Siegrist wrote:
>>> The buildbot will eventually email committers and maintainers when
>>> something fails.
>>
>> I'd vote for an (perhaps
> I suggest starting with something simple, like an email that says "step X
> failed. ", send it to the committer (which should be easy
> with buildbot's MailNotifier), and the maintainers(s) of all ports in that
> build run (harder, but possible). Under normal circumstances, each run/commit
>
On Aug 2, 2011, at 12:31 PM, Marko Käning wrote:
>
> On Aug 2, 2011, at 9:29 PM, Jeremy Lavergne wrote:
>> Hopefully just an email containing a link not a 1GB email :-)
> You've got a point! :-)
> One would have to filter out the relevant port and the error message(s) found
> during build, of co
On Aug 2, 2011, at 9:29 PM, Jeremy Lavergne wrote:
> Hopefully just an email containing a link not a 1GB email :-)
You've got a point! :-)
One would have to filter out the relevant port and the error message(s) found
during build, of course. :-)
___
mac
> I dug myself through some compile-stdio right now (I stopped downloading
> one after it had reached a size of 1GB...) and spotted that e.g.
> gwenhywfar had a missing dep to pkgconfig.
>
> I'd vote for an (perhaps opt-in) email sending possibility to
> maintainers
> already now, since it s
On Aug 2, 2011, at 12:26 PM, Marko Käning wrote:
>
> On Aug 2, 2011, at 6:06 PM, William Siegrist wrote:
>> The buildbot will eventually email committers and maintainers when something
>> fails.
>
> Yes, that feature would be really great! Even now, although buildbot is not
> yet widely anno
On Aug 2, 2011, at 6:06 PM, William Siegrist wrote:
> The buildbot will eventually email committers and maintainers when something
> fails.
Yes, that feature would be really great! Even now, although buildbot is not yet
widely announced.
I dug myself through some compile-stdio right now (I st
On 2011-8-3 02:23 , Marko Käning wrote:
> On Aug 2, 2011, at 6:16 PM, Joshua Root wrote:
>> Nope. Some ports don't fail but just hang, and have to be fixed so the
>> build can continue. It's getting further each time.
> Can you proceed somehow incrementally with these "all" builds, or do you have
On 2011-8-3 02:29 , Arno Hautala wrote:
> On Tue, Aug 2, 2011 at 12:23, Marko Käning wrote:
>>
>> How many ports could you successfully build so far? I mean, are you stuck at
>> 10, 30, or only 80% of all ports?
>
> And how does the build proceed? Seeing as there are ports that
> conflict with e
On Aug 2, 2011, at 6:29 PM, Arno Hautala wrote:
> And how does the build proceed? Seeing as there are ports that
> conflict with each other, I imagine that a straight "port install all"
> would run into problems. Starting each port build without any other
> ports already installed would probably be
On Tue, Aug 2, 2011 at 12:23, Marko Käning wrote:
>
> How many ports could you successfully build so far? I mean, are you stuck at
> 10, 30, or only 80% of all ports?
And how does the build proceed? Seeing as there are ports that
conflict with each other, I imagine that a straight "port install
On Aug 2, 2011, at 6:16 PM, Joshua Root wrote:
> Anyone in the world can initiate it AFAIK.
OK.
> Any or all of the values in the form can be blank.
OK.
> The portlist is actually meant to be space separated,
OK.
> but a buildbot bug is currently preventing requesting more than one port from
>
On 2011-8-3 02:14 , Marko Käning wrote:
>> What about it? Are you seeing packages that are not signed?
> Oh, they are? I thought it's only the various checksums in the portfiles
> which make sure that the source is clean.
> So, you say, the Portfiles themselves are being signed on your end and eve
On 2011-8-3 02:07 , Marko Käning wrote:
> On Aug 2, 2011, at 5:52 PM, Joshua Root wrote:
>> You could also
>> just force a build for the port you're interested in by setting a build
>> property called 'portlist' with the port name as its value.
> Can every registered maintainer initiate this? This,
> What about it? Are you seeing packages that are not signed?
Oh, they are? I thought it's only the various checksums in the portfiles which
make sure that the source is clean.
So, you say, the Portfiles themselves are being signed on your end and every
users port command verifies portfile trustw
On 2011-8-3 02:08 , Marko Käning wrote:
>> Yes, plus catch build, checksum, and base failures.
> What about package signing?
What about it? Are you seeing packages that are not signed?
- Josh
___
macports-dev mailing list
macports-dev@lists.macosforge.
> Yes, plus catch build, checksum, and base failures.
What about package signing?
___
macports-dev mailing list
macports-dev@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macports-dev
On Aug 2, 2011, at 5:52 PM, Joshua Root wrote:
>> not emailing m...@macports.org (not set up yet)
> Same as every other feature, when someone implements it.
So, this is actually then a buildbot configuration issue?
> The buildslave is behind a restrictive firewall, so it can't fetch from
> sv
On Aug 2, 2011, at 8:52 AM, Joshua Root wrote:
> On 2011-8-3 01:37 , Marko Käning wrote:
>> Hi,
>>
>> only a few days ago I learnt about MacPorts' buildbot. And I have a few
>> questions.
>>
I don't think it has been widely announced since it is still a work in
progress.
>> for ports which
On 2011-8-3 01:37 , Marko Käning wrote:
> Hi,
>
> only a few days ago I learnt about MacPorts' buildbot. And I have a few
> questions.
>
>
> 1) When or is it scheduled at all that port maintainers get an email when
> their port(s) fail building? I just saw in the Waterfall graph for
> status-
Hi,
only a few days ago I learnt about MacPorts' buildbot. And I have a few
questions.
1) When or is it scheduled at all that port maintainers get an email when their
port(s) fail building? I just saw in the Waterfall graph for status-stdio
something like
not emailing m...@macports.o
35 matches
Mail list logo
