Please update your mail aliases (if any) from lists.rpath.org
to lists.foresightlinux.org so that when you send mail to
foresight-devel, it goes to the right list and is archived in
the right place.
I wanted to make sure all recipients of the foresight-devel lists
know to use lists.foresightlinux.
On Mon, Jan 07, 2013 at 09:38:14PM -0500, Michael K. Johnson wrote:
> Very much so. In general, being able to use binaries built for RHEL
> (that is, having a compatible set of core libraries available even
> if there are also newer libraries available) will make it more appealing
>
On Sun, Jan 06, 2013 at 03:46:48PM +0100, Rune Morling wrote:
> As I understood it, when I asked about Michael's idea of an ideal FL
> base system, he mentioned how in his view RH is moving too slow, while
> fedora is perhaps sometimes too close to the bleeding/experimental edge
> for comfort. T
Update: the production repositories are back and expected to remain
up. JIRA, Confluence, and build services are currently expected to
be up by some time tomorrow.
We'll continue to work on transitioning to the new infrastructure
quickly.
On Tue, Dec 18, 2012 at 09:51:49AM -0500, Mich
The infrastructure changes that have been posted in the news section
of the web site, on the planet, on G+, and discussed on IRC somehow
never made it to this list, for which I apologize.
We are in the process of deploying a new cluster of systems for both
developer services and public repositorie
On Tue, Sep 11, 2012 at 10:22:09PM +0200, Tomas Forsman wrote:
> We will still be using devel, qa and stable stages?
> Feels like we can go down to devel and stable only. But that's only my
> thought about it.
Well, there are several reasons we're only using two now. Among
them:
* An ancient (
One of the design goals in Conary's "promote" process is to rewrite
all trove references as if they existed on the target label, where
that makes sense, in order to avoid references that expose the
development process. This was really designed for a small number
of packages and proprietary OEM pro
Just to be clear, this is a question that rPath would have to address
as long as the gameway repository continues to be maintained through
the mostly-deactivated rBuilder infrastructure. We can migrate it to
operating with a foresight-hosted repository and mirroring like the
foresight.rpath.org re
ed?
> Maybe that's a bad idea too, as if they change in the future, it wont
> change backif i understands it right.
>
> So, im not sure what's best to do now.
>
>
> 2012-05-14 17:20, Michael K. Johnson skrev:
> > So, if anyone has LXDE installed on their
So, if anyone has LXDE installed on their systems, you want to
arbitrarily push around the osd notifications, even if they also
have GNOME installed? It's not clear to me that this makes sense.
On Mon, May 14, 2012 at 10:26:01AM +0200, Tomas Forsman wrote:
> Hello all,
>
> Need some help to crea
I missed TForsman asking for recipe review on IRC so responding in email:
+r.SetModes('%(destdir)s/opt/dropbox/dropbox', 0755)
+r.SetModes('%(destdir)s/opt/dropbox/dropboxd', 0755)
You don't want %(destdir)s in there.
___
Foresight-devel
On Sat, Feb 25, 2012 at 03:00:22PM +0100, Tomas Forsman wrote:
> Hello, we only unpin the kernels, then they will be uninstalled with
> updateall or sync.
>
> sudo conary unpin kernel=3.1.6-1-1
Actually, with system model, unpin is "unpin and sync" which erases
the kernel as well as unpinning it
http://lists.rpath.org/pipermail/foresight-commits/2011-October/050767.html
is a bad idea. Part of the point of *having* developer contexts is
specifically not to use them as root. Perhaps instead, we could have
a blank developer context in the default config that is overridden
in the developer c
On Fri, Jul 15, 2011 at 06:36:59PM +0100, António Meireles wrote:
> as soon as the last promote to qa (done) finishes mirroring i will
> generate a formal set of 2qa isos and tarballs. (and will post here URLs
> mid tomorrow).
Based on a message António sent me:
group-gnome-dist-devel=foresight.
On Fri, Jul 15, 2011 at 10:59:35AM -0400, Michael K. Johnson wrote:
> With luck you'll have a working Foresight system. If you don't,
> ask for help. Probably I forgot an important step. Sorry.
I think I had to modify /etc/bootloader.d/root.conf to change
the "root&qu
A few weeks ago, I had to install on Sandy Bridge hardware that is not
supported in the kernel in the 2.5.0 media.
While I've done some installer work recently, it will be at least weeks
before there's a new installer you can use.
In the meantime, here's an alternative, manual workaround process
On Mon, Apr 04, 2011 at 11:37:08AM -0400, Scott Parkerson wrote:
> Looks like fdev is back, but JIRA/Confluence are down. gxti has tried
> to power on the box, but it doesn't stay powered on.
Well, I have months of incremental backups of both jira and
confluence, so if we have to install on the ma
On Thu, Mar 31, 2011 at 03:49:19PM -0400, Elliot Peele wrote:
> There is a named running on fdev, but I'm not sure if it is being used. The
> whois records for foresightlinux.org point at Ken's dns servers.
fdev is the master; ken and I run the two slaves. The glue records
point to the slaves, a
On Thu, Mar 31, 2011 at 11:36:54AM -0400, Og Maciel wrote:
> As the rPath guys will be moving to a new office over this weekend
> (April 1 - 3), be aware that all of the systems (mostly build machines
> afaik) will be shutdown and unavailable. I will try to email whenever
> all systems are back onl
On Wed, Jan 26, 2011 at 05:26:32PM +0100, Martin Bähr wrote:
> i think the current info is already sufficient to help find the low
> hanging fruits which will help bring down the size of the install.
I'm saying that the ordering could be substantially different
when you calculate sizes differently
On Wed, Jan 26, 2011 at 08:58:24PM +0800, Jesse Zhang wrote:
> For example on the x86_64 system, glibc accounts for 12% of the whole
> system. And from big to small, mono>faenza-icon-theme>sun-jdk>kernel.
...
> See it for yourself. On the second sheet, the size of :locale troves are
> shown.
Comp
On Wed, Jan 26, 2011 at 08:57:59AM -0500, Og Maciel wrote:
> mkj is currently working on that and it may even be done already
> (there was an issue around)...
According to my testing, it's all done, so let me know whether
anything fails to work.
What I tested was that both http://foresightlinux.o
Confluence is essentially recovered.
Scott Parkerson managed to recover the confluence database; he was
able to determine that only a referrer log table was damaged and
convince postgres to recover sanity, using careful work with dd and
hexdump.
Thanks, Scott!
Now, we need to get proper backups
I went to build a new version of pdftk, and discovered in gcc.recipe:
conf += ('--enable-languages=c,c++,objc,obj-c++,fortran '
# we have openjdk now, right ?
'--disable-libgcj '
However, pdftk doesn't work with JDK, it's clearly s
On Tue, Dec 07, 2010 at 01:18:17AM +0100, Bertrand Juglas wrote:
> By searching "InfoURL getheader python" on Google as suggested by the
> error page I've found as first results:
> http://docs.python.org/library/urllib.html
> which is linked by another result :
> http://docs.python.org/library/http
On Mon, Dec 06, 2010 at 06:51:52PM +0100, Bertrand Juglas wrote:
> Do you know if conary 2.2 has already been tested using python 2.7 ? Because
> Fedora 14 uses already Python 2.7 but i should still be able to install
> python 2.6 if needed by conary.
Previously, I thought we had looked at it, b
On Mon, Dec 06, 2010 at 09:08:55AM +0100, Bertrand Juglas wrote:
> But how have you tried ?
I tried running two commands, both of which succeeded:
cvc newpkg libelf-lgpl=bertux.rpath@boots:f14-devel
cvc newpkg libelf-lgpl=mkj.rpath@boots:f14-devel
"cvc newpkg" does not require write acce
On Mon, Dec 06, 2010 at 12:47:18AM +0100, Bertrand Juglas wrote:
> When i try to run :
> cvc newpkg libelf-lgpl=bertux.rpath@boots:f14-devel
> it replies :
> error: Error occurred opening repository
> https://:@bertux.rpath.org/conary/: InfoURL instance has no
> attribute 'getheader'
>
> What
On Thu, Dec 02, 2010 at 10:31:19PM -0500, Michael K. Johnson wrote:
> We will need to mirror all of the F14 x86/x86_64 yum repositories
> to fdev. We'll need to do the initial mirror operations in a way
> that does not adversely impact rPath's operations, either by choosing
>
On Fri, Dec 03, 2010 at 06:49:13PM +0100, Bertrand Juglas wrote:
> Le 3 déc. 2010 à 16:15, "Michael K. Johnson" a écrit :
> > You say you documented it -- I don't recall seeing the summary sent to
> > this list.
> It's on http://www.netinfolution.info/pack
On Fri, Dec 03, 2010 at 11:51:23AM +0100, Bertrand Juglas wrote:
> I've already a public unused server installed with Fedora 14 and I've
> documented how to fetch, compile and install conary 2.2 to make it work on
> Fedora 14.
> So if it would be simpler to start in LXC container hosted on my ser
On Fri, Dec 03, 2010 at 04:40:22AM +0100, Martin Baehr wrote:
> the next think (from my perspective) is to figure out exactly where to
> start. if you could name one thing that someone of us can do now. (even
> if that thing is to prod someone specific to do the next step)
The best answer I have i
On Fri, Dec 03, 2010 at 03:10:21AM +0100, Martin Bähr wrote:
> can you elaborate on the steps needed at this point to get to an
> installable system.
Only a bit. I haven't been involved in mirrorball development for
months, so I'm a bit out of date with regard to recent development
anyway. This
On Thu, Dec 02, 2010 at 08:02:10AM -0800, Adam Chasen wrote:
> I have minimal experience with mirrorball and I am not up to speed with what
> we are attempting to achieve with boots, but I will take on getting
> mirrorball up and running starting the beginning of next year. Do we have
> any wiki or
On Tue, Nov 30, 2010 at 04:13:59AM +0100, Martin Baehr wrote:
> On Mon, Nov 22, 2010 at 07:50:51PM -0500, Michael K. Johnson wrote:
> > I tried in 2.2.0 beta to improve the man page descriptions of system
> > model update operations. Let me know if it's still confusing;
>
On Tue, Nov 30, 2010 at 02:09:05AM +0100, Martin Baehr wrote:
> On Mon, Nov 29, 2010 at 02:04:46PM -0500, Michael K. Johnson wrote:
> > > maybe i need to update more than just the conary package.
> > I assume you have conary, conary-build, and conary-repository in sync?
>
&g
On Mon, Nov 29, 2010 at 07:53:12PM +0100, Martin Bähr wrote:
> genmodel crashes for me.
> i tried to run updateall, which failed, then tried to just update conary
> which also failed. but then i figured my systemi-model is to old
> anyways, so i wiped it and the model cache, at which point i could
On Thu, Nov 25, 2010 at 01:55:42PM -0600, Andres Vargas wrote:
> it is updated on fdev ?
>
> On Tue, Nov 23, 2010 at 10:28 AM, Michael K. Johnson
> wrote:
> > This update has already been applied to the fdev build cluster.
Yes.
_
I have built the 2.2.1 beta release of Conary on fl:2-devel and on
fl:2-qa directly (to make sure it is compatible with the rest of
fl:2-qa right now). I have not rebuilt groups on either label;
I'll leave that to António.
This second beta of the 2.2 series has two significant changes:
* The gen
On Tue, Nov 23, 2010 at 11:27:07PM +0100, Bertrand Juglas wrote:
> Do you think i should only follow
> https://bitbucket.org/johnsonm/mirrorball-boots-config without
> updating for more recent mirrorball ? Or use hg mergre feature to try
> to merge recent mirrorball in your mirrorball-boots-config
I haven't meant to ignore you, I've just been consumed with Conary 2.2
system model work. I have been thinking about what we'd *really* need
to do to make a boots import useful.
I don't think that we need a development stage at all. I think that
we can import packages directly onto the release l
If you are building packages for fl:2-devel, you will first need to
update rmake to 2.0.6. The new glibc on fl:2-devel exposes
https://issues.rpath.com/browse/RMK-1062 which was fixed in rmake
2.0.6 which is available on both fl:2-devel and fl:2-qa.
The symptom of RMK-1062 is an error that looks
On Fri, Nov 19, 2010 at 10:22:28AM -0500, Michael K. Johnson wrote:
> I don't mean to be a RTFM person, but I do want to make the man page
> useful for this. Therefore, could you please re-read the man page
> description of system model updates, so that we can have this discu
I have built the 2.2.0 beta release of Conary on fl:2-devel and on
fl:2-qa directly (to make sure it is compatible with the rest of
fl:2-qa right now). I have not rebuilt groups on either label;
I'll leave that to António. This first beta of the 2.2 series
contains several changes from the previo
On Fri, Nov 19, 2010 at 05:20:06AM +0100, Martin Baehr wrote:
> On Thu, Nov 18, 2010 at 10:19:02PM -0500, Michael K. Johnson wrote:
> > Each line manipulates the overall state relative to the state in
> > the previous line. This is an ordered list of operations, not an
>
On Fri, Nov 19, 2010 at 03:44:52AM +0100, Martin Bähr wrote:
> seperate lines are ok too?
> update exif=contrib.rpath@rpl:2/0.6.9-7-0.1
> update exiftags=contrib.rpath@rpl:2/1.01-2-0.1
> update jhead=contrib.rpath@rpl:2/2.8-1-0.1
>
> i think it makes the file more read
On Thu, Nov 18, 2010 at 04:03:56PM -0500, Michael K. Johnson wrote:
> When you run the "conary updateall" command, all trove reference that
> include a version (not just a label) will update to the latest version.
Three giant steps forward, one tiny step back.
It turns out
I just built conary=2.2.alpha_a98b4e8eaf6f-1-1 which has several
important changes, several of which were suggested by foresight
developers. Again, I thank you for your testing and suggestions.
Also, I thank you for your patience as you read through this long
email. If you use system model, you d
On Fri, Nov 12, 2010 at 06:28:29PM -0500, Michael K. Johnson wrote:
> Once we're satisfied with the conversion, we can add code to load
> the contents into a new confluence instance as a "restoreBot" user
> to make it clear where the contents came from. By keeping the H
On Thu, Nov 11, 2010 at 10:13:24AM -0500, Michael K. Johnson wrote:
> Looking closer, I think that it's pretty minimal and it will be less
> work to do our own thing with lxml/xpath.
OK, I started work on that. I've checked the scraper (thank John!),
the combined scraped pages,
On Fri, Nov 12, 2010 at 02:28:51PM -0500, Michael K. Johnson wrote:
> In particular, the next conary release (later today) will introduce
> a modelcache format change, so AFTER you do the next update that
> includes a new conary, you'll need to remove the
> /var/lib/conarydb/model
Thanks again for all the great feedback several of you have been
providing on the new system model updates.
I wanted to let you know that as we make changes to system model
code, we occasionally are changing the format of the file
/var/lib/conarydb/modelcache in incompatible ways. After we do
a r
On Mon, Nov 08, 2010 at 11:34:00AM -0500, Michael K. Johnson wrote:
> Thanks! If you would, please open an issue at issues.rpath.com in the
> Conary project against version 2.2(mercurial) with this information,
> and attach two files: /etc/conary/system-model and
> /var/lib/conar
On Thu, Nov 11, 2010 at 10:09:16AM -0500, Michael K. Johnson wrote:
> http://confluence.atlassian.com/display/DISC/HTML+To+Confluence+Converter
> looks like it might be interesting.
Looking closer, I think that it's pretty minimal and it will be less
work to do our own thing with
On Wed, Nov 10, 2010 at 06:23:18PM -0800, John Welborn wrote:
> In my home directory on fdev under confluence I am mirroring all the
> cached pages. Google has already blocked me from fdev so I'm using a
> couple proxies to generate mirrors in parallel for comparison.
>
> Then it will be a matte
On Wed, Nov 10, 2010 at 04:52:32PM -0500, Michael K. Johnson wrote:
> certainly did. So, visit:
> http://www.google.com/search?q=foresight+site%3Awiki.foresightlinux.org
> and start following the "Cached" links and cutting and pasting content.
Be aware that "Cached&quo
OK, Now we need volunteers. No programming experience is required.
There was a problem with our confluence installation: apparently no
one realized that backup dumps were not being made, and then when
data corruption hit on the storage, we ended up with a uselessly
corrupt database. (We will not
Scott Parkerson kindly resurrected the issues system.
issues.foresightlinux.org shares a public IP with
foresight-commits.rpath.com (via proxy). In order to make this
work with SSL, we had to replace the SSL certificate associated
with that IP with a wildcard certificate. It currently has some
b
Thanks! If you would, please open an issue at issues.rpath.com in the
Conary project against version 2.2(mercurial) with this information,
and attach two files: /etc/conary/system-model and
/var/lib/conarydb/manifest
Thanks again!
On Sun, Nov 07, 2010 at 10:06:19AM +0800, Jesse Zhang wrote:
> Ru
On Tue, Nov 02, 2010 at 09:57:27AM +0100, Tomas Forsman wrote:
> If you ask me, i think we should. Can't see any problems to add it.
> We should try to make wine as perfect as we can, as many users use wine
> for games, spotify and so on.
To be clear, I'm not taking responsibility for figuring ou
Wine 1.3.6 has gstreamer support of some sort -- should we be adding
gstreamer to the wine buildreqs?
___
Foresight-devel mailing list
Foresight-devel@lists.rpath.org
http://lists.rpath.org/mailman/listinfo/foresight-devel
On Mon, Nov 01, 2010 at 04:00:31PM +0100, Rune Morling wrote:
> I read that mail (but not the entire thread) and apparently I didn't
> understand its implications.
Sorry not to have been clear enough... :(
> That said, this is how I would guess a minimal system-model for an
> x86_64 sytem with a
On Sun, Oct 31, 2010 at 05:38:18PM +0100, Rune Morling wrote:
> I ran an updateall on my x86_64 system from the 2010-10-18 group-world
> to the 2010-10-26 group-world, which updated conary, restarted and
> promptly 'broke' conary (it appears a 32bit version was installed).
The pain of alpha testin
On Fri, Oct 29, 2010 at 05:03:30PM +0200, Martin Baehr wrote:
> ah, hmm. how about a mode like: i know all my top-level packages are
> mine and not zombies?
That's a thought. Probably not that hard to implement. I'll look
at that.
> yes, that is quite likely. a side effect of the old conary upd
On Fri, Oct 29, 2010 at 06:45:15AM +0200, Martin Baehr wrote:
> tried that, but still found some issues:
> you received my system model email so i won't reproduce it here,
> the points that i am concerned about are the following steps to match
> the model.
>
> Erase foresight-contexts(:config)=0
On Mon, Oct 25, 2010 at 11:50:34AM +0200, Martin Baehr wrote:
> yes, i am hitting that error on my machine in the office.
Sorry -- I've been trying to make genmodel more reliably build
working models with this fix in place -- it has exposed some
weaknesses in the simplification in genmodel.
> is
On Thu, Oct 21, 2010 at 07:22:10PM +0800, Jesse Zhang wrote:
> On Thu, Oct 21, 2010 at 3:10 AM, Michael K. Johnson
> wrote:
> > search
> > group-world=foresight.rpath@fl:2-devel/2.4.9.1+2010.10.18-0.1-2[~!gcc.core]
>
> I wrote it as "search group-world=f
On Thu, Oct 21, 2010 at 06:05:09PM +0200, Martin Baehr wrote:
> i have made some changes to fix a few missing items and get rid of
> unwanted stuff.
>
> i remember you mentioned earlier that testing genmodel would be more
> helpful if done on a fresh system.
I was thinking there mainly of a syste
On Thu, Oct 21, 2010 at 05:01:18AM +0200, Martin Baehr wrote:
> is it helpful (for genmodel testing) if i roll back all changes to the
> point before i started using system model (that's just a handful of
> updates) and run genmodel there?
No, doesn't matter, unless you have afterward made signifi
On Wed, Oct 20, 2010 at 08:25:53PM -0400, Michael K. Johnson wrote:
> I have successfully built conary 2.2.alpha_7e40b066b71c and after
> new groups finish building, your updates will include conary
> 2.2.alpha_7e40b066b71c.
It looks like António and I both built groups at the same time,
On Wed, Oct 20, 2010 at 05:39:02PM -0400, Michael K. Johnson wrote:
> On Wed, Oct 20, 2010 at 03:10:21PM -0400, Michael K. Johnson wrote:
> > I am currently building a new snapshot: conary=2.2.alpha_caf708cb7e9e
> >
> > IMMEDIATELY AFTER you update to this shapshot (
On Wed, Oct 20, 2010 at 03:10:21PM -0400, Michael K. Johnson wrote:
> I am currently building a new snapshot: conary=2.2.alpha_caf708cb7e9e
>
> IMMEDIATELY AFTER you update to this shapshot (or later, if I update
> further)
There was a bug in caf708cb7e9e and I updated further, to
On Mon, Oct 18, 2010 at 05:18:01PM -0400, Michael K. Johnson wrote:
> On Fri, Oct 15, 2010 at 03:23:27AM +0200, Martin Bähr wrote:
> > conary-2.2 is not released yet, so there is still time to change how
> > conary update works.
>
> And, on those lines, I wanted t
On Wed, Oct 20, 2010 at 05:00:27PM +0800, Jesse Zhang wrote:
> Hi,
> On 10/15/2010 12:31 AM, Michael K. Johnson wrote:
> > When you run conary install/update/erase commands, they modify that
> > file, and then apply it to the system.
>
> It seems 'erase -i'
On Wed, Oct 20, 2010 at 04:41:21AM +0200, Martin Baehr wrote:
> is there a way to watch what conary sends over the wire?
Yes! The CONARY_CLIENT_LOG environment variable specifies a
file to which to write a binary log of XML-RPC traffic, and it
works with both conary and cvc.
CONARY_CLIENT_LOG=/t
On Tue, Oct 19, 2010 at 02:39:28AM +0200, Martin Baehr wrote:
> > The simpler query is whether a trove exists. It does.
>
> the simpler query could tell me whether a trove is installable.
> that is the question i would expect most people have.
It *is* installable. It installs by directing to an
On Fri, Oct 15, 2010 at 03:23:27AM +0200, Martin Bähr wrote:
> conary-2.2 is not released yet, so there is still time to change how
> conary update works.
And, on those lines, I wanted to say that this whole conversation
has been very helpful, and we're looking at changing semantics
as a result of
On Mon, Oct 18, 2010 at 07:36:38PM +0200, Martin Baehr wrote:
> On Mon, Oct 18, 2010 at 01:31:27PM -0400, Michael K. Johnson wrote:
> > $ conary rq --components --all-troves --trove-flags
> > uvcvideo-kernel=/foresight.rpath@fl:devel//2-qa-kernel/0-0.2-1
> > uvcvideo-ker
On Mon, Oct 18, 2010 at 07:08:59PM +0200, Martin Baehr wrote:
> ok, so how do i install it?
> even the most explicit:
> # conary update
> uvcvideo-kernel=/foresight.rpath@fl:devel//2-qa-kernel/0-0.2-1
> or
> # conary install
> uvcvideo-kernel=/foresight.rpath@fl:devel//2-qa-kernel/0-0.2-
On Mon, Oct 18, 2010 at 05:44:45PM +0200, Martin Baehr wrote:
> the question is, which behaviour do we want for foresight?
> do we want to add the installLabelPath to a default foresight install?
For normal users, I feel fairly strongly that sticking to the
groups will give them a better experienc
On Mon, Oct 18, 2010 at 05:02:12PM +0200, Martin Baehr wrote:
> here is another puzzling behaviour:
>
> # conary rq uvcvideo-kernel
> uvcvideo-kernel=0-0.2-1
> # conary update uvcvideo-kernel
> No troves found matching: uvcvideo-kernel
> # conary update uvcvideo-kernel=0-0.2-1
> No troves found ma
On Mon, Oct 18, 2010 at 04:42:28PM +0200, Martin Baehr wrote:
> On Mon, Oct 18, 2010 at 10:02:05AM -0400, Michael K. Johnson wrote:
> > I don't follow this. You said, "it doesn't work" and then, as
> > far as I can tell, demonstrated that "conary insta
On Mon, Oct 18, 2010 at 11:09:47AM +0200, Martin Baehr wrote:
> On Fri, Oct 15, 2010 at 10:18:05AM -0400, Michael K. Johnson wrote:
> > > one point i hope to find is that conary gives feedback explaining why
> > > there is nothing to update.
> > That's a flaw, I th
On Fri, Oct 15, 2010 at 05:57:39AM +0200, Martin Baehr wrote:
> i guess, what i don't understand is, why does bypassing the search path
> imply bypassing the system model?
I misunderstood you to have been saying that you wanted conary update
to bypass the system model while install used the system
On Fri, Oct 15, 2010 at 03:23:27AM +0200, Martin Bähr wrote:
> On Thu, Oct 14, 2010 at 12:31:02PM -0400, Michael K. Johnson wrote:
> > You can use vi to edit this file, and then run the "conary sync"
> > command to apply the system model to your system.
>
> can i
The latest groups on fl:2-devel contain a snapshot of the conary-2.2
branch. This is the branch of conary that contains the "system model"
code that you may have heard mentions of on #foresight-devel
First: This new version of Conary will not automatically switch to
"system model" update mode.
On Wed, Oct 06, 2010 at 09:24:46PM -0400, Michael K. Johnson wrote:
> * Enable developer accounts on foresight-commits DONE
Note that I may easily have made mistakes in this process and
failed to turn on developer accounts that I should have. Please,
if you used to be able to commit and
On Wed, Oct 06, 2010 at 09:23:33PM -0400, Michael K. Johnson wrote:
> Foresight developers will need to adjust, very slightly.
>
> * Please DO NOT use foresight-commits for system updates. Your
> system conary configuration should not include a repositoryMap
> entry for
The last (I hope) chapter of the saga of the repository conversion:
* Reverse direction of mirror DONE
* Run mirror DONE
* Enable commitmail from foresight-commits DONE
* Test commit DONE:
http://lists.rpath.org/pipermail/foresight-commits/2010-October/044258.html
* Test build DONE:
http
I have finished the process of splitting the foresight repository into
two repositories.
All commits will be made to the new foresight-commits repository.
Those commits will then be immediately mirrored out to the main
foresight repository. This makes foresight development and system
updates most
This is an update to my previous post with more information. At
this point, this information should be stable, and if someone would
volunteer to make a wiki page from this email, I would appreciate
your effort.
For those of you with fdev accounts, fdev is now fully operational.
fdev is now a for
As discussed on foresight-devel, this plan is now underway.
On Fri, Sep 03, 2010 at 10:28:02AM -0400, Michael K. Johnson wrote:
> I've changed the plan a bit. The remaining plan, in case anyone
> is interested in what I'll be doing:
...
> * Test email from fdev [DONE]
On Thu, Sep 23, 2010 at 03:43:32AM +0200, Martin Bähr wrote:
> actually, after comparing old versions of the code, at least for system
> updates, it aparently never did that.
>
> http://github.com/sgp/PackageKit/blob/6679c05980b5aec8ac8a6f6454c2bddce68c3c14/backends/conary/helpers/conaryBackend.py
On Wed, Sep 22, 2010 at 05:14:29AM +0200, Martin Bähr wrote:
> On Tue, Sep 21, 2010 at 09:59:15PM -0400, Michael K. Johnson wrote:
> > What packagekit does should continue to work on conary 2.2; the
> > old interfaces remain.
>
> ok. i sort of expected that, but it is m
On Wed, Sep 22, 2010 at 02:27:50AM +0200, Martin Bähr wrote:
> unfortunately, we'll have to rewrite the pckagekit-conary backend
> completely for conary 2.2 which has been redesigned to work quite
> differently on how it does updates.
Well...
What packagekit does should continue to work on conary
On Tue, Sep 14, 2010 at 03:39:41PM +0800, Jesse Zhang wrote:
> It can also be put in /etc/bashrc, but I think /etc/skel/.bashrc is better.
I agree. This way there's no question of whether they slow down
starting system bash scripts, and it's easy for each user to control
whether to use the comple
I've changed the plan a bit. The remaining plan, in case anyone
is interested in what I'll be doing:
* Set up database dumps on foresight-commits (in addition to full
backup from streaming mirror) [DONE]
* Convert the filesystem that will hold the local foresight-commits
content store from ex
On Fri, Sep 03, 2010 at 02:01:53PM +0200, Bertrand Juglas wrote:
> As all the software components installed seems publicly available,
> would it be possible for me to setup such a build cluster also ?
Yes. In fact, one of the things I'm doing as part of setting up
this cluster is writing some doc
This is an update to my previous post with more information.
For those of you with fdev accounts, it is now OK to use fdev.
There are a few things to keep in mind.
* fdev is now a foresight-based system. This means, among other
things, that the packaged conary/rmake contexts are available.
R
Since António's build cluster is down, I'm delaying additional fdev
maintenance, because the fdev cluster is working now. Be aware of
two temporary things -- both of these issues are temporary because we
are halfway through the transition to foresight-commits. We will be
able to remove these work
1 - 100 of 215 matches
Mail list logo
