headers
Hisashi T Fujinaka | 1 Jun 2002 03:00

mozilla vs. kde3

Looks like one needs pth and the other needs pth-syscall. I've been
looking for the dependency lists and I can't find them.

Little help here?

Thanks. :)

--

-- 
Hisashi T Fujinaka - htodd <at> twofifty.com
BSEE (6/86) + BSChem (3/95) + BAEnglish (8/95) + $2.50 = mocha latte
Permalink | Reply | 1 comment |
headers
Jeremy C. Reed | 1 Jun 2002 03:19

Re: mozilla vs. kde3

On Fri, 31 May 2002, Hisashi T Fujinaka wrote:

> Looks like one needs pth and the other needs pth-syscall. I've been
> looking for the dependency lists and I can't find them.

I had similar issue this past day with qt3 (as discussed on tech-pkg). I
was able to make it work by modifying USE_PTHREAD for qt3-tools and
qt3-libs to use pth before pth-syscall.

(I haven't yet used qt3 yet though to see if pth works for it.)

   Jeremy C. Reed
   http://www.reedmedia.net/
Permalink | Reply | 0 comments |
headers
Julio Merino | 1 Jun 2002 15:04

Re: Problems with hard disk

On Sat, Jun 01, 2002 at 02:46:04PM +0200, Manuel Bouyer wrote:
> On Sat, Jun 01, 2002 at 12:41:53PM +0200, Julio Merino wrote:
> > Hello
> > 
> > Is this something dangerous? I mean... maybe my disk is going to die?
> 
> No, it's just data corruption on the IDE bus which has been detected by
> the controller, and the driver redid the transfer. Nothing dangerous here.

I'm not so sure it's not dangerous... I've got another message, saying that
the system has downgraded data transfer to pio mode 4 from udma 2.
I've rebooted the machine to get udma2 again, and when trying to start X,
the system has got stalled while reading from disk... I don't think it
is a -current problem, because I haven't updated it for some days, and it
has also failed "outside" netbsd, as I'll say now.

Ok, another reboot, what I get? Grub (the bootloader) can't start and the
machine reboots itself. And after this, it has booted normal... but fsck
complained, too many errors. I'm now backing up my data, to be safe.
I'll do some more tests with the drive later. it's a Fujitsu 20GB 5400rpm
udma100. Although I think it will die in some days :(

Thanks though.

--

-- 
Of course it runs NetBSD - http://www.netbsd.org
HispaBSD member - http://www.hispabsd.org
Julio Merino <jmmv <at> hispabsd.org>
Permalink | Reply | 6 comments |
headers
Manuel Bouyer | 1 Jun 2002 14:46

Re: Problems with hard disk

On Sat, Jun 01, 2002 at 12:41:53PM +0200, Julio Merino wrote:
> Hello
> 
> I've just noticed the following messages from kernel:
> 
> wd0f: aborted command, interface CRC error reading fsbn 1388128 of 1388128-1388143 (wd0 bn 3641008; cn
3612 tn 1 sn 49), retrying
> wd0: soft error (corrected)
> wd0g: aborted command, interface CRC error writing fsbn 16593056 of 16593056-16593071 (wd0 bn
23966576; cn 23776 tn 5 sn 53), retrying
> wd0: soft error (corrected)
> 
> Is this something dangerous? I mean... maybe my disk is going to die?

No, it's just data corruption on the IDE bus which has been detected by
the controller, and the driver redid the transfer. Nothing dangerous here.

--

-- 
Manuel Bouyer <bouyer <at> antioche.eu.org>
--
Permalink | Reply | 7 comments |
headers
Alan Barrett | 1 Jun 2002 12:54
Gravatar

Re: What am I supposed to do with usr.sbin/config ?

On Fri, May 31, 2002 at 04:14:26PM +0200, Manuel Bouyer wrote:
> Well, it didn't work for me for the 1.6 branch: after a
> cvs co -l -r netbsd-1-6 src
> a cvs update -r netbsd-1-6 -dPA would leave me with an almost-empty
> usr.sbin (only config).

The first "cvs checkout -l" command goes much faster if you
leave out the "-r <tag>".  Like this:

        CVSROOT=:pserver:anoncvs <at> anoncvs.netbsd.org:/cvsroot
        cvs -d ${CVSROOT} login                 # password anoncvs
        mkdir /var/tmp/foo
        cd /var/tmp/foo
        cvs -u -d ${CVSROOT} checkout -l src    # this is reasonably fast
        cd src
        cvs -u update -A -rnetbsd-1-6 -dP       # this is slow

If the first "cvs checkout -l" fails, then the directory structure
needed by the subsequent "cvs update" will probably not be created
correctly.

--apb (Alan Barrett)
Permalink | Reply | 1 comment |
headers
Julio Merino | 1 Jun 2002 12:41

Problems with hard disk

Hello

I've just noticed the following messages from kernel:

wd0f: aborted command, interface CRC error reading fsbn 1388128 of 1388128-1388143 (wd0 bn 3641008; cn
3612 tn 1 sn 49), retrying
wd0: soft error (corrected)
wd0g: aborted command, interface CRC error writing fsbn 16593056 of 16593056-16593071 (wd0 bn 23966576;
cn 23776 tn 5 sn 53), retrying
wd0: soft error (corrected)

Is this something dangerous? I mean... maybe my disk is going to die?

Thanks.

--

-- 
Of course it runs NetBSD - http://www.netbsd.org
HispaBSD member - http://www.hispabsd.org
Julio Merino <jmmv <at> hispabsd.org>
Permalink | Reply | 8 comments |
headers
Brian A. Seklecki | 1 Jun 2002 13:18

Re: What am I supposed to do with usr.sbin/config ?

AB>
AB>         CVSROOT=:pserver:anoncvs <at> anoncvs.netbsd.org:/cvsroot
AB>         cvs -d ${CVSROOT} login                 # password anoncvs
AB>         mkdir /var/tmp/foo
AB>         cd /var/tmp/foo
AB>         cvs -u -d ${CVSROOT} checkout -l src    # this is reasonably fast
AB>         cd src
AB>         cvs -u update -A -rnetbsd-1-6 -dP       # this is slow
AB>
AB> If the first "cvs checkout -l" fails, then the directory structure

1) Try using a CVS mirror other than anoncvs.*
2) Maybe try doing something more extravgant w/ your update scripts (I've
been thinking about writing one)

_checkOut ()
{

if cvs -d $CVSROOT -t co src
then
	echo "CVS Checkout Complete"
	_update;
else
	echo "CVS Checkout Failed"
	exit 1
fi
}

AB> needed by the subsequent "cvs update" will probably not be created
AB> correctly.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Alan Barrett | 1 Jun 2002 16:25
Gravatar

OpenOffice 1.0

Has anybody been able to build OpenOffice 1.0.0 under NetBSD (any
architecture other than sparc)?

The sources are available via links from
http://www.openoffice.org/dev_docs/source/download.html.

The build procedure for Linux is rather vaguely documented at
http://www.openoffice.org/dev_docs/source/build_linux.html.

Build instructions for NetBSD/sparc can be found at
http://www.fs.tum.de/~mrauch/OpenOffice/source.html.  Despite the
heading at the top of the page which says "Building OpenOffice.org under
NetBSD", this page seems to be specific to NetBSD/sparc.  Also, the
set_soenv.1 script in the config_office directory seems to assume that
NetBSD/sparc is the only NetBSD architecture.

--apb (Alan Barrett)
Permalink | Reply | 1 comment |
headers
Love | 1 Jun 2002 17:10
Picon
Picon
Favicon

sparc64 releng-20010531-NNN...


Hi

I tried to install sparc64 releng-20010531-NNN... on a ultra1 but sysinst
failed with the first text below. Fixing the mode-bits make it failel later
(second text). mknod'ing bpf0 dev-node made dhclient happy, bit sysinst
didn't pick up the dhclient values.

Setting the ip-address & mask enable me at least being able to download the
sets and they seemed to install nice (which the 1.6A snapshot didn't do).

Then it failed when installing the boot blocks (third text). I ran the on
the prompt, again the mode-bits must have been wrong.

With this I got my u1 sparc booting.

Love

[ 1. wrong bits on dhclient-script]

     Status: Failed: command failed
    Command: /sbin/dhclient -pf /tmp/dhclient.pid -lf /tmp/dhclient.leases le0
     Press any key to continue
--------------------------------------------------------------------------------
execve (/sbin/dhclient-script, ...): Permission denied
le0: not found

If you did not get this software from ftp.isc.org, please
get the latest from ftp.isc.org and install that before
requesting help.
(Continue reading)

Permalink | Reply | 3 comments |
headers
Luke Mewburn | 1 Jun 2002 18:38

Re: sparc64 releng-20010531-NNN...

On Sat, Jun 01, 2002 at 05:10:50PM +0200, Love wrote:
  | 
  | Hi
  | 
  | I tried to install sparc64 releng-20010531-NNN... on a ultra1 but sysinst
  | failed with the first text below. Fixing the mode-bits make it failel later
  | (second text). mknod'ing bpf0 dev-node made dhclient happy, bit sysinst
  | didn't pick up the dhclient values.

I've fixed the permissions problems for dhclient-script and binstall,
and added bpf0 to the MAKEDEV devices on the image.

Once a new snapshot is available (in a day or two, or build one
yourself once you have the updated versions of the files in
distrib/sparc64/instfs), can you try this again?

I suspect that sysinst won't find the dhclient values, even in the
newer version.  Could you send-pr a report describing any problems in
a newer snapshot (after it's available)?

Thanks for reporting these problems!

Luke.

  | Setting the ip-address & mask enable me at least being able to download the
  | sets and they seemed to install nice (which the 1.6A snapshot didn't do).
  | 
  | Then it failed when installing the boot blocks (third text). I ran the on
  | the prompt, again the mode-bits must have been wrong.
  |
(Continue reading)

Permalink | Reply | 2 comments |

Gmane