Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 9 Feb 2004 18:45:17 +0200
From:      Alexey Zelkin <phantom@FreeBSD.org>
To:        Dan Langille <dan@langille.org>
Cc:        ports@FreeBSD.org
Subject:   Re: ports/java/jdk14 looping on install?
Message-ID:  <20040209164517.GC23768@phantom.cris.net>
In-Reply-To: <40273DAD.14286.14410189@localhost>
References:  <40264324.19153.106E4782@localhost> <40273DAD.14286.14410189@localhost>

next in thread | previous in thread | raw e-mail | index | archive | help
hi,

On Mon, Feb 09, 2004 at 07:58:37AM -0500, Dan Langille wrote:
> On 9 Feb 2004 at 0:20, Alexey Zelkin wrote:
> 
> > On Sun, Feb 08, 2004 at 02:09:40PM -0500, Dan Langille wrote:
> > > On 8 Feb 2004 at 18:23, Alexey Zelkin wrote:
> > > 
> > > > hi,
> > > > 
> > > > is linuxprocfs mounted ?
> > > 
> > > It appears not.
> > 
> > It appears to be strange that you even passed pre-build: checks then.
> > If you are using linux jdk as bootstrap jdk then linprocfs should
> > be mounted and active.
> 
> After more thatn 36 hours of CPU time, I terminated the build.  The 
> java process continued to run so I terminated that manually.  A kill -
> TERM did not kill it so I resorted to a kill -KILL.

Yep.  It was reported many times before.  Most usual reason is hard
dependancy linux_base on linprocfs (in linux jdk case).  Second most
reported case with such behaviour is mixing threading libraries.  Since
you are building jdk from scratch I assume linux_base is your problem.

NOTE: Actually, I am talking about this issue (linux_base related) only by
reports of other people.  I never was able to reproduce such behavior
on my build machine. :(

> This process needs to be cleaner if we expect people to start using
> Open Office.
> 
> I am pleased to report this:

[linprocfs warning skiped]

Did you changed something ?  Why it did not appear before ?

> After doing the above:
> 
> # mount
> /dev/ad0s1a on / (ufs, local)
> /dev/ad0s1f on /tmp (ufs, local, soft-updates)
> /dev/ad0s1g on /usr (ufs, NFS exported, local, soft-updates)
> /dev/ad0s1e on /var (ufs, local, soft-updates)
> procfs on /proc (procfs, local)
> xeon:/usr/ports/distfiles on /usr/ports/distfiles (nfs)
> linprocfs on /usr/compat/linux/proc (linprocfs, local)
> 
> Now why can't the port just do that?

Sorry.  This is too intrusive change as for me.  I prefer to force
people do it by hands.  At least they'll know how to revert this
change in runtime.

-- 
/* Alexey Zelkin             && Independent Contractor      */
/* phantom(at)FreeBSD.org    && http://www.FreeBSD.org/java */
/* phantom(at)cris.net       && http://www.FreeBSD.org.ua/  */



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040209164517.GC23768>