On May 3, 3:07 pm, "John Cremona" <[EMAIL PROTECTED]> wrote:
> Two successful builds of rc0 (32 and 64 bit linux). The 64-bit one
> seems to have stalled doing testall at this point:
>
> sage -t devel/sage/sage/dsage/dist_functions/all.py
> [1.3 s]
> sage -t devel/sage/sage/dsage/scripts/nodoctest.py (skipping) --
> nodoctest.py file in directory
> sage -t devel/sage/sage/dsage/scripts/dsage_worker.py (skipping) --
> nodoctest.py file in directory
> sage -t devel/sage/sage/dsage/scripts/__init__.py (skipping) --
> nodoctest.py file in directory
> sage -t devel/sage/sage/dsage/scripts/dsage_setup.py (skipping) --
> nodoctest.py file in directory
> sage -t devel/sage/sage/dsage/web/web_server.py
> [1.3 s]
> sage -t devel/sage/sage/dsage/tests/testdoc.py
>
> which has not moved for at least haf an hour.
>
> Switching do a different window on the same machine, "ps -ax" shows this:
>
<SNIP>
>
> which reveals that as well as the stalled run I knew about, there are
> 8 other sage processes stuck in limbo. Before I kkill these jobs
> (which I certainly will -- it is not even my own machine) -- is there
> anything I can do to help diagnose the problem?
Fortunately wjp figured it out:
[15:04] <wjp> ah, I think I see why dsage is behaving strangely
[15:04] <mabshoff> nice.
[15:04] <mabshoff> if you have a clue that is usually it.
[15:04] <gfurnish|away> why is dsage behaving strangely
[15:04] <wjp> in my alpha1 install, there's no local/dsage, and no
local/bin/dsage_worker.py
[15:04] <mabshoff> ok
[15:04] <wjp> (while in my 3.0.rc0 where the doctest works fine, both
exist)
[15:04] <wjp> somebody who understands dsage may be able to say if
that is relevant :-)
[15:05] <mabshoff> But they were not removed in 3.0 or anywhere else.
[15:05] <gfurnish|away> File a bug report for pbuild.
[15:05] <mabshoff> So, you guys used pbuild both?
[15:05] <mabshoff> That would explain it :)
[15:05] <gfurnish|away> I'm guessing he did.
[15:05] <wjp> I did, yes
[15:05] <mabshoff> I think Cremona did, too.
[15:05] <mabshoff> Excellent.
[15:05] <gfurnish|away> actually I think both of those need to be
fixed in the installer
[15:05] <mabshoff> In the rc0 announcement I had a sentence I took
out:
[15:06] <gfurnish|away> I think you should just mkdir local/dsage and
manually install the local/bin/dsage_worker.py
[15:06] <mabshoff> "If pbuild causes any trouble Gary will be taken to
a damp basement room and beaten with reeds"
[15:06] <gfurnish|away> and add it to the scripts repo
[15:06] <mabshoff> gfurnish|away: please see if that fixes the issue.
[15:06] <gfurnish|away> where is dsage_worker.py?
[15:07] <mabshoff> ./local/bin/dsage_worker.py
[15:07] <mabshoff> ./devel/sage-main/sage/dsage/scripts/
dsage_worker.py
[15:07] <wjp> doing a diff on the contents of local/bin, I see
dsage_setup.py is also missing
[15:07] <mabshoff> ./devel/sage-main/build/sage/dsage/scripts/
dsage_worker.py
[15:08] <mabshoff> wjp: with "hg status" ?
[15:08] <wjp> (and a +SbuildHack.pm, -spkg-debian-maybe, but those may
not be relevant)
[15:08] <mabshoff> ok.
[15:08] <wjp> no, just ls > /tmp/blah in both directories
[15:08] <mabshoff> Both of those are taken care of by setup.py
[15:09] <mabshoff> check out the end of setup.py:
[15:09] <mabshoff> scripts = ['sage/dsage/scripts/
dsage_worker.py',
[15:09] <mabshoff> 'sage/dsage/scripts/
dsage_setup.py',
[15:09] <mabshoff> 'spkg-debian-maybe',
[15:09] <mabshoff> ],
[15:09] <gfurnish|away> local/dsage is the dir thats missing?
[15:10] <wjp> it contains index.html and a static subdirectory
[15:10] <gfurnish|away> yeah I see it but it doesn't seem to fix it...
[15:10] <gfurnish|away> ah so there are files too?
[15:10] <wjp> um, sorry, web/index.html and a web/static subdir
[15:10] <gfurnish|away> where do the files come from?
[15:10] <wjp> devel/sage-main/sage/dsage/
[15:10] <gfurnish|away> well maybe I can convince you to make the
patch ;)
[15:10] <gfurnish|away> we should just symlink the files
> John
Cheers,
Michael
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
