cvs commit: ports/databases/p5-BerkeleyDB Makefile distinfo

[Ion-Mihai Tetcu]

On Wed, 23 Nov 2005 20:22:19 +0100
Kirill Ponomarew <krion at voodoo.bawue.com> wrote:

> On Wed, Nov 23, 2005 at 09:31:37AM +0000, Cheng-Lung Sung wrote:
> > clsung      2005-11-23 09:31:37 UTC
> > 
> >   FreeBSD ports repository
> > 
> >   Modified files:
> >     databases/p5-BerkeleyDB Makefile distinfo 
> >   Log:
> >   * Add SHA256.
> >   * Use bsd.database.mk to decide BerkeleyDB version.
> 
>
> --->  Upgrading 'p5-BerkeleyDB-0.27' to
> 'p5-BerkeleyDB-0.27_1' (databases/p5-BerkeleyDB)

This seems to fail for any BDB version. But the nicest thing was that I
didn't look very careful in the Makefile, I changed WITH_BDB_VER=42 to
WITH_BDB_VER=43 in pkgtools.conf and added USE_BDB=43; as result I
nearly locked my machine because of swapping: 

 # portupgrade -Rp p5-BerkeleyDB-0.27
--->  Upgrading 'p5-BerkeleyDB-0.27' to 'p5-BerkeleyDB-0.27_1' (databases/p5-BerkeleyDB)
--->  Building '/usr/ports/databases/p5-BerkeleyDB' with make flags: WITH_BDB_VER=43 USE_BDB=43
make: Max recursion level (500) exceeded.: Resource temporarily unavailable
make: Max recursion level (500) exceeded.: Resource temporarily unavailable
^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C
C^C^C^C^C^C^C^C

Had to ssh from an other, rtprio 31 csh, use top, attached to the
screen(1) session and try to kill it with CTRL+C w/o success and resort
to detaching screen session and killall make.

Related or not  when this happened the first time I got:
Nov 23 21:28:57 it kernel: kbdc: TEST_AUX_PORT status:0000
Nov 23 21:28:57 it kernel: kbdc: RESET_AUX return code:00fa
Nov 23 21:28:57 it kernel: kbdc: RESET_AUX status:00aa
Nov 23 21:28:57 it kernel: kbdc: RESET_AUX ID:0000

and I'm very interested to know if it's heavy swapping related; the
first time when I tried portupgrade and got the above the swapping was
really hard (I have 1GB memory and had about 300MB swap used at the
moment). I didn't got other kdbc msgs on the next runs but while the
swap used was about 540MB there were about 300 MB free and I killed
portupgrade with CRTL+C when free mem got around 0.

This is on a 6.0-STABLE #7: Mon Nov  7 14:40:46 EET 2005 with 4BSD on
VIA KT600/8237 -based motherboard.


Thanks,

-- 
IOnut
Unregistered ;) FreeBSD "user"