Re: make world fails in usr.sbin/config?

[Christer Solskogen]

On Tue, May 25, 2010 at 1:05 PM, Matthew Seaman
 wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> On 25/05/2010 11:40:23, Stefan Bethke wrote:
>> For the record: I'm now running -stable as of last night, compiled
>> without issue on ZFS filesystems throughout.  No idea what caused the
>> issue in the first place, and what made it disappear though, but
>> updating to the correctly built -stable made the build on ZFS work
>> again.  (It also involved an accidential upgrade and downgrade via
>> -current, since I checked out the wrong tag with csup.  Yikes.)
>
> I've a new machine that's been running 8-STABLE on ZFS for about a week
> now.  Had no problems installing and then upgrading to recent 8-STABLE
> although I did start with installing an 8-STABLE snapshot rather than
> 8.0-RELEASE. (See: http://wiki.freebsd.org/RootOnZFS/GPTZFSBoot/Mirror)
>
> Verb. Sap.  If you're booting from ZFS, beware of updating the zpool
> version without due care and attention.  8.0-RELEASE was on version 13,
> 8-STABLE is now on version 14.  (Use 'zpool update' to see what the
> status is on your machine -- this just gives you a report, and doesn't
> update anything.)  Updating the zpool version is pretty smooth and
> simple, but *remember to immediately rebuild and reinstall gptzfsboot or
> zfsboot bootcode on your drives*.  If you don't do that, your system
> won't be able to find the pool with the root filesystem and so won't be
> able to reboot.
>
>

Old thread, I know. But I encountered this error today and my solution
was that /tmp was tmpfs and it was full. Cleaning that up fixed the
error.


-- 
chs,
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Matthew Seaman]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 25/05/2010 11:40:23, Stefan Bethke wrote:
> For the record: I'm now running -stable as of last night, compiled
> without issue on ZFS filesystems throughout.  No idea what caused the
> issue in the first place, and what made it disappear though, but
> updating to the correctly built -stable made the build on ZFS work
> again.  (It also involved an accidential upgrade and downgrade via
> -current, since I checked out the wrong tag with csup.  Yikes.)

I've a new machine that's been running 8-STABLE on ZFS for about a week
now.  Had no problems installing and then upgrading to recent 8-STABLE
although I did start with installing an 8-STABLE snapshot rather than
8.0-RELEASE. (See: http://wiki.freebsd.org/RootOnZFS/GPTZFSBoot/Mirror)

Verb. Sap.  If you're booting from ZFS, beware of updating the zpool
version without due care and attention.  8.0-RELEASE was on version 13,
8-STABLE is now on version 14.  (Use 'zpool update' to see what the
status is on your machine -- this just gives you a report, and doesn't
update anything.)  Updating the zpool version is pretty smooth and
simple, but *remember to immediately rebuild and reinstall gptzfsboot or
zfsboot bootcode on your drives*.  If you don't do that, your system
won't be able to find the pool with the root filesystem and so won't be
able to reboot.

Cheers,

Matthew 

- -- 
Dr Matthew J Seaman MA, D.Phil.   7 Priory Courtyard
  Flat 3
PGP: http://www.infracaninophile.co.uk/pgpkey Ramsgate
JID: matt...@infracaninophile.co.uk   Kent, CT11 9PW
-BEGIN PGP SIGNATURE-
Version: GnuPG/MacGPG2 v2.0.14 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkv7rvYACgkQ8Mjk52CukIzoEQCeNccarvp+NUrOAvoomJIRiT+H
N6MAn0UbAcoAIdrgyI8CiEvhoiY1mCLh
=ZfJJ
-END PGP SIGNATURE-
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Stefan Bethke]

Am 24.05.2010 um 19:49 schrieb Jeremy Chadwick:

> On Mon, May 24, 2010 at 09:24:00AM -0700, Jeremy Chadwick wrote:
>> Builds are underway now (following /usr/src/Makefile method), I'll
>> report back when those are done.  I'm also adding "time" in front of the
>> "make buildXXX" portions just to see now long things take.
> 
> The build portions finished.  Here are the numbers (quite high due to a
> combination of limited memory constraints (intentional) and the fact
> that VMware Workstation isn't the fastest thing on the planet.  :-) )

For the record: I'm now running -stable as of last night, compiled without 
issue on ZFS filesystems throughout.  No idea what caused the issue in the 
first place, and what made it disappear though, but updating to the correctly 
built -stable made the build on ZFS work again.  (It also involved an 
accidential upgrade and downgrade via -current, since I checked out the wrong 
tag with csup.  Yikes.)

Thanks for all the support to all of you!


Stefan

-- 
Stefan BethkeFon +49 151 14070811

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Jeremy Chadwick]

On Mon, May 24, 2010 at 09:24:00AM -0700, Jeremy Chadwick wrote:
> Builds are underway now (following /usr/src/Makefile method), I'll
> report back when those are done.  I'm also adding "time" in front of the
> "make buildXXX" portions just to see now long things take.

The build portions finished.  Here are the numbers (quite high due to a
combination of limited memory constraints (intentional) and the fact
that VMware Workstation isn't the fastest thing on the planet.  :-) )

--
>>> World build completed on Mon May 24 10:18:55 PDT 2010
--
1950.588u 1791.577s 32:36.44 191.2% 5154+1888k 26+70io 28130pf+0w

--
>>> Kernel build for GENERIC completed on Mon May 24 10:33:58 PDT 2010
--
754.864u 739.046s 14:29.00 171.9%   5258+1816k 1+9io 427pf+0w

These should also act as verification that building world (or at least
most of it, see my src.conf) with /usr/src and /usr/obj as ZFS
filesystems is safe to do.

Installation bits:

- make installkernel completed without issue
- rebooted into single-user
- ran: mount -a ; /etc/rc.d/hostid start ; /etc/rc.d/zfs start
- cd /usr/src
- mergemaster -p-- no changes
- make installworld -- no errors witnessed
- yes | make delete-old -- removal of ipfilter + kerberos
- mergemaster   -- lots of changes...
- reboot
- Once system was back up, ran "make delete-old-libs"

Present state of the system:

testbox# date
Mon May 24 10:45:36 PDT 2010

testbox# uname -a
FreeBSD testbox.home.lan 8.1-PRERELEASE FreeBSD 8.1-PRERELEASE #0: Mon May 24 
10:32:08 PDT 2010 r...@testbox.home.lan:/usr/obj/usr/src/sys/GENERIC  amd64

testbox# df -k
Filesystem 1024-blocksUsedAvail Capacity  Mounted on
/dev/da0s1a1012974  511974   41996455%/
devfs1   10   100%/dev
/dev/da0s1e 507630  12   467008 0%/tmp
/dev/da0s1f   10655298  179240  9623636 2%/usr
/dev/da0s1d2026030   37218  1826730 2%/var
data/usr_obj  15473740 1581700 1389204010%/usr/obj
data/usr_ports14285680  393639 13892040 3%/usr/ports
data/usr_src  14436336  544295 13892040 4%/usr/src

testbox# zfs list
NAME USED  AVAIL  REFER  MOUNTPOINT
data2.40G  13.2G  24.0K  none
data/usr_obj1.51G  13.2G  1.51G  /usr/obj
data/usr_ports   384M  13.2G   384M  /usr/ports
data/usr_src 532M  13.2G   532M  /usr/src

testbox# zpool status
  pool: data
 state: ONLINE
 scrub: none requested
config:

NAMESTATE READ WRITE CKSUM
dataONLINE   0 0 0
  raidz1ONLINE   0 0 0
da1 ONLINE   0 0 0
da2 ONLINE   0 0 0
da3 ONLINE   0 0 0

errors: No known data errors

testbox# zpool scrub data
{...wait a little while...}
testbox# zpool status | grep scrub
 scrub: scrub completed after 0h3m with 0 errors on Mon May 24 10:48:01 2010

Finally, "zfs get all" output, which is fairly long but I figure it
might help in debugging.  Anything else you'd like me to check before I
delete this VM?

testbox# zfs get all
NAMEPROPERTY  VALUE  SOURCE
datatype  filesystem -
datacreation  Mon May 24  8:37 2010  -
dataused  2.40G  -
dataavailable 13.2G  -
datareferenced24.0K  -
datacompressratio 1.00x  -
datamounted   no -
dataquota none   default
datareservation   none   default
datarecordsize128K   default
datamountpointnone   local
datasharenfs  offdefault
datachecksum  on default
datacompression   offdefault
dataatime on default
datadevices   on default
dataexec  on default
datasetuidon default
datareadonly  offdefault
datajailedoffdefault
datasnapdir   hidden default
dataaclmode   groupmask  default
dataaclinheritrestricted default

Re: make world fails in usr.sbin/config?

[Jeremy Chadwick]

On Mon, May 24, 2010 at 09:18:55AM -0700, Kevin Oberman wrote:
> > Date: Mon, 24 May 2010 05:29:44 -0700
> > From: Jeremy Chadwick 
> > Sender: owner-freebsd-sta...@freebsd.org
> > 
> > On Mon, May 24, 2010 at 02:21:44PM +0200, Stefan Bethke wrote:
> > > Am 24.05.2010 um 14:18 schrieb Jeremy Chadwick:
> > > 
> > > > 1) Were you using any "-j" flags during your make?  If so, try without 
> > > > it.
> > > > Sometimes these are known to cause oddities, even if occasionally.
> > > 
> > > Nope.
> > > 
> > > > 2) Make sure your system clock is correct and isn't drifting badly.
> > > > Highly recommend you use ntpdate to set the clock initially, then run
> > > > ntpd at all times.
> > > 
> > > # ntpq -p
> > >  remote   refid  st t when poll reach   delay   offset  
> > > jitter
> > > ==
> > > +lokschuppen.zs6 131.188.3.2222 u   69  512  377   34.1155.313   
> > > 0.153
> > > *jachthafen.hans 131.188.3.2222 u   52  512  377   33.9664.757   
> > > 0.554
> > > -ps.bucuo.de 192.53.103.108   2 u  185  512  377   39.5677.895   
> > > 0.268
> > > -svr02.teleport- 73.120.242.922 u  187  512  377   44.5726.949   
> > > 0.542
> > > -netzwerkteufel. 192.53.103.104   2 u  202  512  377   35.3387.662   
> > > 0.422
> > > +qraftwerk.de192.53.103.108   2 u  141  512  377   52.5055.228   
> > > 0.256
> > > 
> > > I'll try a new checkout next.
> > 
> > Your clock looks OK (worst drift from a stratum 2 comparison is 5.313
> > seconds).
> 
> Minor correction on this. The offset values from 'ntpq -p' are in
> milliseconds, so the worst offset is <6 ms. Not great, but way better
> than is really needed.

Thanks Kevin -- you're absolutely right.  I often forget this fact,
since comparatively "ntpdc -c peers" outputs seconds.

-- 
| Jeremy Chadwick   j...@parodius.com |
| Parodius Networking   http://www.parodius.com/ |
| UNIX Systems Administrator  Mountain View, CA, USA |
| Making life hard for others since 1977.  PGP: 4BD6C0CB |

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Jeremy Chadwick]

On Mon, May 24, 2010 at 08:07:38AM -0700, Jeremy Chadwick wrote:
> On Mon, May 24, 2010 at 03:39:19PM +0200, Stefan Bethke wrote:
> > Am 24.05.2010 um 15:27 schrieb Stefan Bethke:
> > 
> > > I've now checked out via csup, and I've put /usr/obj on UFS.  The error 
> > > has shifted yet again:
> > > cc -O2 -pipe -I. -I/usr/src/usr.bin/lex -std=gnu99   
> > > -I/usr/obj/usr/src/tmp/legacy/usr/include -c parse.c
> > > /usr/src/usr.bin/lex/parse.y: In function 'build_eof_action':
> > > /usr/src/usr.bin/lex/parse.y:786: error: 'MAXLINE' undeclared (first use 
> > > in this function)
> > > 
> > > I would agree that this looks like time problems or similar, but I don't 
> > > see how that could be the case.
> > > 
> > > I'll put the source on UFS as well, just to make sure.
> > 
> > Putting the sources on a separate UFS file system "fixed" the build issue.  
> > Previously, I did have root on UFS, but /usr/src and /usr/obj on ZFS, so I 
> > don't quite understand what the difference is.
> 
> I'll try reproducing your problem on said VM box I have, once I finish
> testing for a possible a 8.0-RELEASE to 8.1-PRERELEASE issue.

VM system has a fresh 8.0-STABLE-201002 installed on it, taken from the
8.0-STABLE snapshots directory.  This is "fairly close" to your
kernel/world build time.

testbox# uname -a
FreeBSD testbox.home.lan 8.0-STABLE-201002 FreeBSD 8.0-STABLE-201002 #0: Tue 
Feb 16 21:05:59 UTC 2010 
r...@mason.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC  amd64

Things to note up front:

1) I chose *not* to install ports nor src from the installation medium,
   so /usr/src and /usr/ports at this point are empty dirs.
2) VM instance is running without ntpdate + ntpd intentionally.

Preparation and creation of the zpool (raidz1 w/ 3x 8GB disks) and
filesystems:

testbox# echo 'zfs_enable="yes"' >> /etc/rc.conf
testbox# /etc/rc.d/hostid start
Setting hostuuid: 564de564-be17-6e27-89b6-ae662003f0db.
Setting hostid: 0xfa4c1c12.
testbox# /etc/rc.d/zfs start
testbox# zpool create data raidz1 da1 da2 da3
testbox# zpool list
NAME   SIZE   USED  AVAILCAP  HEALTH  ALTROOT
data  23.9G   141K  23.9G 0%  ONLINE  -
testbox# zfs set mountpoint=none data
testbox# zfs create -o mountpoint=/usr/obj data/usr_obj
testbox# zfs create -o mountpoint=/usr/ports data/usr_ports
testbox# zfs create -o mountpoint=/usr/src data/usr_src
testbox# zfs list
NAME USED  AVAIL  REFER  MOUNTPOINT
data 228K  15.7G  24.0K  none
data/usr_obj24.0K  15.7G  24.0K  /usr/obj
data/usr_ports  24.0K  15.7G  24.0K  /usr/ports
data/usr_src24.0K  15.7G  24.0K  /usr/src

At this point I tuned loader.conf to permit for larger ZFS ARC size (the
VM is only allocated 1GB of RAM; this is intentional, as I wanted to
simulate a memory-tight environment), explicitly disabled prefetching,
and adjusted vfs.zfs.txg.timeout for an increase in responsiveness:

testbox# echo 'vm.kmem_size="768M"' >> /boot/loader.conf
testbox# echo 'vfs.zfs.arc_max="512M"' >> /boot/loader.conf
testbox# echo 'vfs.zfs.prefetch_disable="1"' >> /boot/loader.conf
testbox# echo 'vfs.zfs.txg.timeout="5"' >> /boot/loader.conf
testbox# shutdown -r now

System rebooted to pick up loader.conf changes, and on to csup:

testbox# csup -h cvsup10.freebsd.org -L 2 -4 
/usr/share/examples/cvsup/stable-supfile
{...}
testbox# csup -h cvsup10.freebsd.org -L 2 -4 
/usr/share/examples/cvsup/ports-supfile
{...}

Removing some base system pieces and adding optimisations:

testbox# cat > /etc/make.conf
KERNCONF=GENERIC
CPUTYPE?=nocona
testbox# cat > /etc/src.conf
WITHOUT_IPFILTER=true
WITHOUT_LIB32=true
WITHOUT_KERBEROS=true
WITHOUT_PROFILE=true
WITHOUT_SENDMAIL=true
testbox#

Builds are underway now (following /usr/src/Makefile method), I'll
report back when those are done.  I'm also adding "time" in front of the
"make buildXXX" portions just to see now long things take.

-- 
| Jeremy Chadwick   j...@parodius.com |
| Parodius Networking   http://www.parodius.com/ |
| UNIX Systems Administrator  Mountain View, CA, USA |
| Making life hard for others since 1977.  PGP: 4BD6C0CB |

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Kevin Oberman]

> Date: Mon, 24 May 2010 05:29:44 -0700
> From: Jeremy Chadwick 
> Sender: owner-freebsd-sta...@freebsd.org
> 
> On Mon, May 24, 2010 at 02:21:44PM +0200, Stefan Bethke wrote:
> > Am 24.05.2010 um 14:18 schrieb Jeremy Chadwick:
> > 
> > > 1) Were you using any "-j" flags during your make?  If so, try without it.
> > > Sometimes these are known to cause oddities, even if occasionally.
> > 
> > Nope.
> > 
> > > 2) Make sure your system clock is correct and isn't drifting badly.
> > > Highly recommend you use ntpdate to set the clock initially, then run
> > > ntpd at all times.
> > 
> > # ntpq -p
> >  remote   refid  st t when poll reach   delay   offset  
> > jitter
> > ==
> > +lokschuppen.zs6 131.188.3.2222 u   69  512  377   34.1155.313   
> > 0.153
> > *jachthafen.hans 131.188.3.2222 u   52  512  377   33.9664.757   
> > 0.554
> > -ps.bucuo.de 192.53.103.108   2 u  185  512  377   39.5677.895   
> > 0.268
> > -svr02.teleport- 73.120.242.922 u  187  512  377   44.5726.949   
> > 0.542
> > -netzwerkteufel. 192.53.103.104   2 u  202  512  377   35.3387.662   
> > 0.422
> > +qraftwerk.de192.53.103.108   2 u  141  512  377   52.5055.228   
> > 0.256
> > 
> > I'll try a new checkout next.
> 
> Your clock looks OK (worst drift from a stratum 2 comparison is 5.313
> seconds).

Minor correction on this. The offset values from 'ntpq -p' are in
milliseconds, so the worst offset is <6 ms. Not great, but way better
than is really needed.
-- 
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: ober...@es.net  Phone: +1 510 486-8634
Key fingerprint:059B 2DDF 031C 9BA3 14A4  EADA 927D EBB3 987B 3751
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Jeremy Chadwick]

On Mon, May 24, 2010 at 03:39:19PM +0200, Stefan Bethke wrote:
> Am 24.05.2010 um 15:27 schrieb Stefan Bethke:
> 
> > I've now checked out via csup, and I've put /usr/obj on UFS.  The error has 
> > shifted yet again:
> > cc -O2 -pipe -I. -I/usr/src/usr.bin/lex -std=gnu99   
> > -I/usr/obj/usr/src/tmp/legacy/usr/include -c parse.c
> > /usr/src/usr.bin/lex/parse.y: In function 'build_eof_action':
> > /usr/src/usr.bin/lex/parse.y:786: error: 'MAXLINE' undeclared (first use in 
> > this function)
> > 
> > I would agree that this looks like time problems or similar, but I don't 
> > see how that could be the case.
> > 
> > I'll put the source on UFS as well, just to make sure.
> 
> Putting the sources on a separate UFS file system "fixed" the build issue.  
> Previously, I did have root on UFS, but /usr/src and /usr/obj on ZFS, so I 
> don't quite understand what the difference is.

I'll try reproducing your problem on said VM box I have, once I finish
testing for a possible a 8.0-RELEASE to 8.1-PRERELEASE issue.

-- 
| Jeremy Chadwick   j...@parodius.com |
| Parodius Networking   http://www.parodius.com/ |
| UNIX Systems Administrator  Mountain View, CA, USA |
| Making life hard for others since 1977.  PGP: 4BD6C0CB |

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Stefan Bethke]

Am 24.05.2010 um 15:27 schrieb Stefan Bethke:

> I've now checked out via csup, and I've put /usr/obj on UFS.  The error has 
> shifted yet again:
> cc -O2 -pipe -I. -I/usr/src/usr.bin/lex -std=gnu99   
> -I/usr/obj/usr/src/tmp/legacy/usr/include -c parse.c
> /usr/src/usr.bin/lex/parse.y: In function 'build_eof_action':
> /usr/src/usr.bin/lex/parse.y:786: error: 'MAXLINE' undeclared (first use in 
> this function)
> 
> I would agree that this looks like time problems or similar, but I don't see 
> how that could be the case.
> 
> I'll put the source on UFS as well, just to make sure.

Putting the sources on a separate UFS file system "fixed" the build issue.  
Previously, I did have root on UFS, but /usr/src and /usr/obj on ZFS, so I 
don't quite understand what the difference is.


Thanks for all your help!


Stefan

-- 
Stefan BethkeFon +49 151 14070811



___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Paul Mather]

On May 24, 2010, at 9:27 AM, Stefan Bethke wrote:

> I've just moved from a root on UFS plus data on ZFS setup, to root on ZFS; 
> that's the only real difference I can think of.  Although I don't see how 
> that would affect building world, especially since I've had src and obj on 
> ZFS before.

FWIW, I successfully completed buildworld + buildkernel for 8-STABLE on a 
root-on-ZFS system yesterday.

Cheers,

Paul.

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Stefan Bethke]

Am 24.05.2010 um 15:13 schrieb Jeremy Chadwick:

> So now the problem has moved from usr.sbin/config to usr.sbin/awk?
> Weird.  Usually this sort of thing indicates excessive clock skew (as in
> rapidly skewing multiple seconds in bursts), or very strange filesystem
> problems.
> 
> Is it possible for your /usr/obj to be made a UFS2 filesystem and for
> you to re-try your build?

I've now checked out via csup, and I've put /usr/obj on UFS.  The error has 
shifted yet again:
cc -O2 -pipe -I. -I/usr/src/usr.bin/lex -std=gnu99   
-I/usr/obj/usr/src/tmp/legacy/usr/include -c parse.c
/usr/src/usr.bin/lex/parse.y: In function 'build_eof_action':
/usr/src/usr.bin/lex/parse.y:786: error: 'MAXLINE' undeclared (first use in 
this function)

I would agree that this looks like time problems or similar, but I don't see 
how that could be the case.

I'll put the source on UFS as well, just to make sure.

> By the way, the buildworld + buildkernel I was running on the FreeBSD VM
> box I have just finished -- no issues.  And that's with make -j2.
> That's an 8.0-RELEASE machine which is being built to upgrade to
> RELENG_8.

A separate make buildworld on another machine is chugging along just fine, so 
there's definitly something odd about this box.

I've just moved from a root on UFS plus data on ZFS setup, to root on ZFS; 
that's the only real difference I can think of.  Although I don't see how that 
would affect building world, especially since I've had src and obj on ZFS 
before.


Stefan

-- 
Stefan BethkeFon +49 151 14070811



___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[David Wolfskill]

On Mon, May 24, 2010 at 02:59:01PM +0200, Stefan Bethke wrote:
> ...
> It sure looks like it. Now that I've checked out again, the error has moved 
> to:
> cc -O2 -pipe -DHAS_ISBLANK -I. 
> -I/usr/src/usr.bin/awk/../../contrib/one-true-awk -DFOPEN_MAX=64   
> -I/usr/obj/usr/src/tmp/legacy/usr/include  
> -L/usr/obj/usr/src/tmp/legacy/usr/lib 
> /usr/src/usr.bin/awk/../../contrib/one-true-awk/maketab.c  -o maketab
> In file included from 
> /usr/src/usr.bin/awk/../../contrib/one-true-awk/maketab.c:35:
> ./ytab.h:98: warning: data definition has no type or storage class
> ./ytab.h:99: error: expected '=', ',', ';', 'asm' or '__attribute__' before 
> 'yylval'
> *** Error code 1
> 
> Stop in /usr/src/usr.bin/awk.
> *** Error code 1
> 
> Stop in /usr/src.
> *** Error code 1
> 
> Stop in /usr/src.
> *** Error code 1
> 
> Stop in /usr/src.
> 
> Again, it appears as if YYSTYPE is defined but empty.  ytab.h looks a bit odd 
> to me:
> ...

My laptop has finished byilding the stable/8 world successfully at
r208488 (it's still building the kernel); my build machine completed the
stable/8 build -- also at r208488 -- (and is now building head).

I use a -j factor of $(( 2 \* $( sysctl -n hw.ncpu ) )).

Peace,
david
-- 
David H. Wolfskill  da...@catwhisker.org
Depriving a girl or boy of an opportunity for education is evil.

See http://www.catwhisker.org/~david/publickey.gpg for my public key.


pgpHaQGtUQQW6.pgp
Description: PGP signature

Re: make world fails in usr.sbin/config?

[Jeremy Chadwick]

On Mon, May 24, 2010 at 02:59:01PM +0200, Stefan Bethke wrote:
> Am 24.05.2010 um 14:29 schrieb Jeremy Chadwick:
> 
> > All that said: I *have* seen the compiler error you've mentioned, but
> > usually a 2nd rebuild (after nuking /usr/obj/*) usually works.  Probably
> > some weird race condition.
> 
> It sure looks like it. Now that I've checked out again, the error has moved 
> to:
> cc -O2 -pipe -DHAS_ISBLANK -I. 
> -I/usr/src/usr.bin/awk/../../contrib/one-true-awk -DFOPEN_MAX=64   
> -I/usr/obj/usr/src/tmp/legacy/usr/include  
> -L/usr/obj/usr/src/tmp/legacy/usr/lib 
> /usr/src/usr.bin/awk/../../contrib/one-true-awk/maketab.c  -o maketab
> In file included from 
> /usr/src/usr.bin/awk/../../contrib/one-true-awk/maketab.c:35:
> ./ytab.h:98: warning: data definition has no type or storage class
> ./ytab.h:99: error: expected '=', ',', ';', 'asm' or '__attribute__' before 
> 'yylval'
> *** Error code 1
> 
> Stop in /usr/src/usr.bin/awk.
> *** Error code 1
> 
> Stop in /usr/src.
> *** Error code 1
> 
> Stop in /usr/src.
> *** Error code 1
> 
> Stop in /usr/src.
> 
> Again, it appears as if YYSTYPE is defined but empty.  ytab.h looks a bit odd 
> to me:
> 
> #define DECR 346
> #define INCR 347
> #define INDIRECT 348
> #define LASTTOKEN 349
>  YYSTYPE;
> extern YYSTYPE yylval;
> 
> Line 98 is " YYSTYPE;"

So now the problem has moved from usr.sbin/config to usr.sbin/awk?
Weird.  Usually this sort of thing indicates excessive clock skew (as in
rapidly skewing multiple seconds in bursts), or very strange filesystem
problems.

Is it possible for your /usr/obj to be made a UFS2 filesystem and for
you to re-try your build?

By the way, the buildworld + buildkernel I was running on the FreeBSD VM
box I have just finished -- no issues.  And that's with make -j2.
That's an 8.0-RELEASE machine which is being built to upgrade to
RELENG_8.

-- 
| Jeremy Chadwick   j...@parodius.com |
| Parodius Networking   http://www.parodius.com/ |
| UNIX Systems Administrator  Mountain View, CA, USA |
| Making life hard for others since 1977.  PGP: 4BD6C0CB |

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Stefan Bethke]

Am 24.05.2010 um 14:29 schrieb Jeremy Chadwick:

> All that said: I *have* seen the compiler error you've mentioned, but
> usually a 2nd rebuild (after nuking /usr/obj/*) usually works.  Probably
> some weird race condition.

It sure looks like it. Now that I've checked out again, the error has moved to:
cc -O2 -pipe -DHAS_ISBLANK -I. 
-I/usr/src/usr.bin/awk/../../contrib/one-true-awk -DFOPEN_MAX=64   
-I/usr/obj/usr/src/tmp/legacy/usr/include  
-L/usr/obj/usr/src/tmp/legacy/usr/lib 
/usr/src/usr.bin/awk/../../contrib/one-true-awk/maketab.c  -o maketab
In file included from 
/usr/src/usr.bin/awk/../../contrib/one-true-awk/maketab.c:35:
./ytab.h:98: warning: data definition has no type or storage class
./ytab.h:99: error: expected '=', ',', ';', 'asm' or '__attribute__' before 
'yylval'
*** Error code 1

Stop in /usr/src/usr.bin/awk.
*** Error code 1

Stop in /usr/src.
*** Error code 1

Stop in /usr/src.
*** Error code 1

Stop in /usr/src.

Again, it appears as if YYSTYPE is defined but empty.  ytab.h looks a bit odd 
to me:

#define DECR 346
#define INCR 347
#define INDIRECT 348
#define LASTTOKEN 349
 YYSTYPE;
extern YYSTYPE yylval;

Line 98 is " YYSTYPE;"


Stefan

-- 
Stefan BethkeFon +49 151 14070811



___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Stefan Bethke]

Am 24.05.2010 um 14:40 schrieb Paul Mather:

> On May 24, 2010, at 8:29 AM, Jeremy Chadwick wrote:
> 
>> For added posterity, it looks like usr.sbin/config has been mostly
>> untouched for quite some time, sans mkoptions.c and mkmakefile.c:
>> 
>> http://www.freebsd.org/cgi/cvsweb.cgi/src/usr.sbin/config/
> 
> Having said that, there is this entry in /usr/src/UPDATING dating from early 
> May:
> 
> 20100502:
>The config(8) command has been updated to maintain compatibility
>with config files from 8.0-RELEASE.  You will need a new version
>of config to build kernels (this version can be used from 8.0-RELEASE
>forward).  The buildworld target will generate it, so following
>the instructions in this file for updating will work glitch-free.
>Merely doing a make buildkernel without first doing a make buildworld
>(or kernel-toolchain), or attempting to build a kernel using
>traidtional methods will generate a config version warning, indicating
>you should update.
> 
> 
> Stefan's kernel looks to have last been built on 20th February 2010.  It 
> isn't explicit in the first posting of this thread how Stefan is doing his 
> build, so there is a possibility that he's being affected by the above 
> UPDATING entry.

# make buildworld buildkernel


Stefan

-- 
Stefan BethkeFon +49 151 14070811



___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Paul Mather]

On May 24, 2010, at 8:29 AM, Jeremy Chadwick wrote:

> For added posterity, it looks like usr.sbin/config has been mostly
> untouched for quite some time, sans mkoptions.c and mkmakefile.c:
> 
> http://www.freebsd.org/cgi/cvsweb.cgi/src/usr.sbin/config/

Having said that, there is this entry in /usr/src/UPDATING dating from early 
May:

20100502:
The config(8) command has been updated to maintain compatibility
with config files from 8.0-RELEASE.  You will need a new version
of config to build kernels (this version can be used from 8.0-RELEASE
forward).  The buildworld target will generate it, so following
the instructions in this file for updating will work glitch-free.
Merely doing a make buildkernel without first doing a make buildworld
(or kernel-toolchain), or attempting to build a kernel using
traidtional methods will generate a config version warning, indicating
you should update.


Stefan's kernel looks to have last been built on 20th February 2010.  It isn't 
explicit in the first posting of this thread how Stefan is doing his build, so 
there is a possibility that he's being affected by the above UPDATING entry.

Cheers,

Paul.

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Jeremy Chadwick]

On Mon, May 24, 2010 at 02:21:44PM +0200, Stefan Bethke wrote:
> Am 24.05.2010 um 14:18 schrieb Jeremy Chadwick:
> 
> > 1) Were you using any "-j" flags during your make?  If so, try without it.
> > Sometimes these are known to cause oddities, even if occasionally.
> 
> Nope.
> 
> > 2) Make sure your system clock is correct and isn't drifting badly.
> > Highly recommend you use ntpdate to set the clock initially, then run
> > ntpd at all times.
> 
> # ntpq -p
>  remote   refid  st t when poll reach   delay   offset  jitter
> ==
> +lokschuppen.zs6 131.188.3.2222 u   69  512  377   34.1155.313   0.153
> *jachthafen.hans 131.188.3.2222 u   52  512  377   33.9664.757   0.554
> -ps.bucuo.de 192.53.103.108   2 u  185  512  377   39.5677.895   0.268
> -svr02.teleport- 73.120.242.922 u  187  512  377   44.5726.949   0.542
> -netzwerkteufel. 192.53.103.104   2 u  202  512  377   35.3387.662   0.422
> +qraftwerk.de192.53.103.108   2 u  141  512  377   52.5055.228   0.256
> 
> I'll try a new checkout next.

Your clock looks OK (worst drift from a stratum 2 comparison is 5.313
seconds).

The only other thing I can think of would be to try doing this, to
ensure absolutely no corruption or oddities with the csup CVS DB file:

rm -fr /var/db/sup/src-all
rm -fr /usr/src/*
csup -h  -L 2 /usr/share/examples/cvsup/stable-supfile

All that said: I *have* seen the compiler error you've mentioned, but
usually a 2nd rebuild (after nuking /usr/obj/*) usually works.  Probably
some weird race condition.

For added posterity, it looks like usr.sbin/config has been mostly
untouched for quite some time, sans mkoptions.c and mkmakefile.c:

http://www.freebsd.org/cgi/cvsweb.cgi/src/usr.sbin/config/

I've rebuilt world a couple times in the past few days (on different
boxes, including one under a VMware Workstation VM which is going as I
write this), without issue.

-- 
| Jeremy Chadwick   j...@parodius.com |
| Parodius Networking   http://www.parodius.com/ |
| UNIX Systems Administrator  Mountain View, CA, USA |
| Making life hard for others since 1977.  PGP: 4BD6C0CB |

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Stefan Bethke]

Am 24.05.2010 um 14:18 schrieb Jeremy Chadwick:

> 1) Were you using any "-j" flags during your make?  If so, try without it.
> Sometimes these are known to cause oddities, even if occasionally.

Nope.

> 2) Make sure your system clock is correct and isn't drifting badly.
> Highly recommend you use ntpdate to set the clock initially, then run
> ntpd at all times.

# ntpq -p
 remote   refid  st t when poll reach   delay   offset  jitter
==
+lokschuppen.zs6 131.188.3.2222 u   69  512  377   34.1155.313   0.153
*jachthafen.hans 131.188.3.2222 u   52  512  377   33.9664.757   0.554
-ps.bucuo.de 192.53.103.108   2 u  185  512  377   39.5677.895   0.268
-svr02.teleport- 73.120.242.922 u  187  512  377   44.5726.949   0.542
-netzwerkteufel. 192.53.103.104   2 u  202  512  377   35.3387.662   0.422
+qraftwerk.de192.53.103.108   2 u  141  512  377   52.5055.228   0.256

I'll try a new checkout next.


Stefan

-- 
Stefan BethkeFon +49 151 14070811



___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Jeremy Chadwick]

On Mon, May 24, 2010 at 02:12:13PM +0200, Stefan Bethke wrote:
> Am 24.05.2010 um 14:09 schrieb Jeremy Chadwick:
> 
> > On Mon, May 24, 2010 at 01:59:14PM +0200, Stefan Bethke wrote:
> >> I have a feeling I screwed something up, but I can't find anything wrong 
> >> locally.
> >> ...
> >> ===> usr.sbin/config (obj,depend,all,install)
> >> /usr/obj/usr/src/tmp/usr/src/usr.sbin/config created for 
> >> /usr/src/usr.sbin/config
> >> yacc -d /usr/src/usr.sbin/config/config.y
> >> cp y.tab.c config.c
> >> lex -t  /usr/src/usr.sbin/config/lang.l > lang.c
> >> file2c 'char kernconfstr[] = {' ',0};' < 
> >> /usr/src/usr.sbin/config/kernconf.tmpl > kernconf.c
> >> rm -f .depend
> >> mkdep -f .depend -a-I. -I/usr/src/usr.sbin/config 
> >> -I/usr/obj/usr/src/tmp/legacy/usr/include config.c 
> >> /usr/src/usr.sbin/config/main.c lang.c 
> >> /usr/src/usr.sbin/config/mkmakefile.c /usr/src/usr.sbin/config/mkheaders.c 
> >> /usr/src/usr.sbin/config/mkoptions.c kernconf.c
> >> echo config: /usr/lib/libc.a /usr/lib/libl.a /usr/lib/libsbuf.a 
> >> /usr/obj/usr/src/tmp/legacy/usr/lib/libegacy.a >> .depend
> >> cc -O2 -pipe -I. -I/usr/src/usr.sbin/config   
> >> -I/usr/obj/usr/src/tmp/legacy/usr/include -c config.c
> >> config.c:214: error: expected '=', ',', ';', 'asm' or '__attribute__' 
> >> before '*' token
> >> config.c:215: error: expected '=', ',', ';', 'asm' or '__attribute__' 
> >> before 'yyval'
> >> config.c:216: error: expected '=', ',', ';', 'asm' or '__attribute__' 
> >> before 'yylval'
> >> config.c:219: error: expected '=', ',', ';', 'asm' or '__attribute__' 
> >> before '*' token
> >> /usr/src/usr.sbin/config/config.y: In function 'yyerror':
> >> /usr/src/usr.sbin/config/config.y:312: error: 'yyfile' undeclared (first 
> >> use in this function)
> >> /usr/src/usr.sbin/config/config.y:312: error: (Each undeclared identifier 
> >> is reported only once
> >> /usr/src/usr.sbin/config/config.y:312: error: for each function it appears 
> >> in.)
> >> /usr/src/usr.sbin/config/config.y:312: error: 'yyline' undeclared (first 
> >> use in this function)
> > 
> > 1) Have you tried rm -fr /usr/obj/* prior to building world?
> 
> /usr/obj is a fresh filesystem (zfs).
> 
> > 2) If you already tried that, can you provide your /etc/make.conf and
> > /etc/src.conf contents?
> 
> I have no src.conf, and this is make.conf, unchanged from previous make 
> worlds.
> 
> #
> # make world etc.
> #
> KERNCONF?=DIESEL
> #MODULES_WITH_WORLD=  true
> 
> BOOT_PXELDR_ALWAYS_SERIAL?=   true
> BOOT_PXELDR_PROBE_KEYBOARD?=  true
> 
> # added by use.perl 2009-07-26 23:56:06
> PERL_VERSION=5.8.9

1) Were you using any "-j" flags during your make?  If so, try without it.
Sometimes these are known to cause oddities, even if occasionally.

2) Make sure your system clock is correct and isn't drifting badly.
Highly recommend you use ntpdate to set the clock initially, then run
ntpd at all times.

-- 
| Jeremy Chadwick   j...@parodius.com |
| Parodius Networking   http://www.parodius.com/ |
| UNIX Systems Administrator  Mountain View, CA, USA |
| Making life hard for others since 1977.  PGP: 4BD6C0CB |

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Stefan Bethke]

Am 24.05.2010 um 14:09 schrieb Jeremy Chadwick:

> On Mon, May 24, 2010 at 01:59:14PM +0200, Stefan Bethke wrote:
>> I have a feeling I screwed something up, but I can't find anything wrong 
>> locally.
>> ...
>> ===> usr.sbin/config (obj,depend,all,install)
>> /usr/obj/usr/src/tmp/usr/src/usr.sbin/config created for 
>> /usr/src/usr.sbin/config
>> yacc -d /usr/src/usr.sbin/config/config.y
>> cp y.tab.c config.c
>> lex -t  /usr/src/usr.sbin/config/lang.l > lang.c
>> file2c 'char kernconfstr[] = {' ',0};' < 
>> /usr/src/usr.sbin/config/kernconf.tmpl > kernconf.c
>> rm -f .depend
>> mkdep -f .depend -a-I. -I/usr/src/usr.sbin/config 
>> -I/usr/obj/usr/src/tmp/legacy/usr/include config.c 
>> /usr/src/usr.sbin/config/main.c lang.c /usr/src/usr.sbin/config/mkmakefile.c 
>> /usr/src/usr.sbin/config/mkheaders.c /usr/src/usr.sbin/config/mkoptions.c 
>> kernconf.c
>> echo config: /usr/lib/libc.a /usr/lib/libl.a /usr/lib/libsbuf.a 
>> /usr/obj/usr/src/tmp/legacy/usr/lib/libegacy.a >> .depend
>> cc -O2 -pipe -I. -I/usr/src/usr.sbin/config   
>> -I/usr/obj/usr/src/tmp/legacy/usr/include -c config.c
>> config.c:214: error: expected '=', ',', ';', 'asm' or '__attribute__' before 
>> '*' token
>> config.c:215: error: expected '=', ',', ';', 'asm' or '__attribute__' before 
>> 'yyval'
>> config.c:216: error: expected '=', ',', ';', 'asm' or '__attribute__' before 
>> 'yylval'
>> config.c:219: error: expected '=', ',', ';', 'asm' or '__attribute__' before 
>> '*' token
>> /usr/src/usr.sbin/config/config.y: In function 'yyerror':
>> /usr/src/usr.sbin/config/config.y:312: error: 'yyfile' undeclared (first use 
>> in this function)
>> /usr/src/usr.sbin/config/config.y:312: error: (Each undeclared identifier is 
>> reported only once
>> /usr/src/usr.sbin/config/config.y:312: error: for each function it appears 
>> in.)
>> /usr/src/usr.sbin/config/config.y:312: error: 'yyline' undeclared (first use 
>> in this function)
> 
> 1) Have you tried rm -fr /usr/obj/* prior to building world?

/usr/obj is a fresh filesystem (zfs).

> 2) If you already tried that, can you provide your /etc/make.conf and
> /etc/src.conf contents?

I have no src.conf, and this is make.conf, unchanged from previous make worlds.

#
# make world etc.
#
KERNCONF?=  DIESEL
#MODULES_WITH_WORLD=true

BOOT_PXELDR_ALWAYS_SERIAL?= true
BOOT_PXELDR_PROBE_KEYBOARD?=true

# added by use.perl 2009-07-26 23:56:06
PERL_VERSION=5.8.9


-- 
Stefan BethkeFon +49 151 14070811



___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world fails in usr.sbin/config?

[Jeremy Chadwick]

On Mon, May 24, 2010 at 01:59:14PM +0200, Stefan Bethke wrote:
> I have a feeling I screwed something up, but I can't find anything wrong 
> locally.
> ...
> ===> usr.sbin/config (obj,depend,all,install)
> /usr/obj/usr/src/tmp/usr/src/usr.sbin/config created for 
> /usr/src/usr.sbin/config
> yacc -d /usr/src/usr.sbin/config/config.y
> cp y.tab.c config.c
> lex -t  /usr/src/usr.sbin/config/lang.l > lang.c
> file2c 'char kernconfstr[] = {' ',0};' < 
> /usr/src/usr.sbin/config/kernconf.tmpl > kernconf.c
> rm -f .depend
> mkdep -f .depend -a-I. -I/usr/src/usr.sbin/config 
> -I/usr/obj/usr/src/tmp/legacy/usr/include config.c 
> /usr/src/usr.sbin/config/main.c lang.c /usr/src/usr.sbin/config/mkmakefile.c 
> /usr/src/usr.sbin/config/mkheaders.c /usr/src/usr.sbin/config/mkoptions.c 
> kernconf.c
> echo config: /usr/lib/libc.a /usr/lib/libl.a /usr/lib/libsbuf.a 
> /usr/obj/usr/src/tmp/legacy/usr/lib/libegacy.a >> .depend
> cc -O2 -pipe -I. -I/usr/src/usr.sbin/config   
> -I/usr/obj/usr/src/tmp/legacy/usr/include -c config.c
> config.c:214: error: expected '=', ',', ';', 'asm' or '__attribute__' before 
> '*' token
> config.c:215: error: expected '=', ',', ';', 'asm' or '__attribute__' before 
> 'yyval'
> config.c:216: error: expected '=', ',', ';', 'asm' or '__attribute__' before 
> 'yylval'
> config.c:219: error: expected '=', ',', ';', 'asm' or '__attribute__' before 
> '*' token
> /usr/src/usr.sbin/config/config.y: In function 'yyerror':
> /usr/src/usr.sbin/config/config.y:312: error: 'yyfile' undeclared (first use 
> in this function)
> /usr/src/usr.sbin/config/config.y:312: error: (Each undeclared identifier is 
> reported only once
> /usr/src/usr.sbin/config/config.y:312: error: for each function it appears 
> in.)
> /usr/src/usr.sbin/config/config.y:312: error: 'yyline' undeclared (first use 
> in this function)

1) Have you tried rm -fr /usr/obj/* prior to building world?

2) If you already tried that, can you provide your /etc/make.conf and
/etc/src.conf contents?

-- 
| Jeremy Chadwick   j...@parodius.com |
| Parodius Networking   http://www.parodius.com/ |
| UNIX Systems Administrator  Mountain View, CA, USA |
| Making life hard for others since 1977.  PGP: 4BD6C0CB |

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: Make world builderror on 7.1-BETA2 with latest cvsup

[Mattias Björk]

Dominik Żyła wrote:

On Sun, Jan 04, 2009 at 11:39:28AM +0100, Mattias Björk wrote:

Hello again,


Garrett Cooper wrote:

On Fri, Jan 2, 2009 at 11:52 AM, Mattias Björk  wrote:

Hello Garrett,


Sorry for top posting, but here is the attachment of the log.


Garrett Cooper wrote:

On Jan 1, 2009, at 19:23, Mattias Björk  wrote:


Hello everybody,

First of all, my bad if this get sent two times.

I have a compile error/problem when building world.

My uname -a output is:

FreeBSD barabolaptop 7.1-BETA2 FreeBSD 7.1-BETA2 #0: Thu Dec  4 20:52:35
CET 2008 r...@barabolaptop:/usr/obj/usr/src/sys/BARABOLAPTOP  i386

My make.conf looks like:

# added by use.perl 2008-12-03 00:58:09
PERL_VER=5.8.8
PERL_VERSION=5.8.8
WRKDIRPREFIX=/var/tmp


The options.h file in /usr/obj/usr/src/gnu/usr.bin/cc/cc_tools/options.h

Here are the output of options.h from pastebin:

http://pastebin.com/m254dfedf


When I run "make -j1 buildworld" i get the error is as following:

http://pastebin.com/m57738677

I don't know much of programming (if any), but I have tried to change
things in (remove OPT_w from both places) options.h and runned
"make -j1 -DNOCLEAN buildworld"

But that have not solve anything of this, perhaps some can shine some
light on this or have I missed something trivial?
I can build ports and so on without any problems.

Thank you.

Not enough data. Please send a full compressed version of the log to me.
Thanks!
-Garrett

That's really odd why the symbol was redefined. Here're the relevant
sections of the log:

c_tools/../../../../contrib/gcc/gencheck.c
In file included from ./tm.h:4,
from
/usr/src/gnu/usr.bin/cc/cc_tools/../../../../contrib/gcc/gencheck.c:25:
./options.h:901: error: redeclaration of enumerator 'OPT_w'
./options.h:899: error: previous definition of 'OPT_w' was here
*** Error code 1
1 error
*** Error code 2
1 error
*** Error code 2
1 error
*** Error code 2

Looking back, I'm not sure why OPT_w is printed into the file twice.
I wonder if you met a rare race condition where options.sh printed out
that line twice; then again that's unlikely, given the
reproducibility... I could be wrong however.
Is your source tree based off of RELENG_7 and what cvsup server is it
synced against?
-Garrett


I'm using cvsup.se.freebsd.org, cvsup.dk.freebsd.org, 
cvsup.no.freebsd.org I have tried to remove /usr/src/* right now and 
downloaded it again but it did not help after the build.


So I have no idea what to do from here.

But I guess It would be solved sometime in the future.


Did you clean your /usr/obj/ directory?



Yes I have also done that, still no luck there.

Thanks anyway.
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: Make world builderror on 7.1-BETA2 with latest cvsup

[Dominik Żyła]

On Sun, Jan 04, 2009 at 11:39:28AM +0100, Mattias Björk wrote:
> Hello again,
> 
> 
> Garrett Cooper wrote:
> > On Fri, Jan 2, 2009 at 11:52 AM, Mattias Björk  
> > wrote:
> >> Hello Garrett,
> >>
> >>
> >> Sorry for top posting, but here is the attachment of the log.
> >>
> >>
> >> Garrett Cooper wrote:
> >>> On Jan 1, 2009, at 19:23, Mattias Björk  wrote:
> >>>
>  Hello everybody,
> 
>  First of all, my bad if this get sent two times.
> 
>  I have a compile error/problem when building world.
> 
>  My uname -a output is:
> 
>  FreeBSD barabolaptop 7.1-BETA2 FreeBSD 7.1-BETA2 #0: Thu Dec  4 20:52:35
>  CET 2008 r...@barabolaptop:/usr/obj/usr/src/sys/BARABOLAPTOP  i386
> 
>  My make.conf looks like:
> 
>  # added by use.perl 2008-12-03 00:58:09
>  PERL_VER=5.8.8
>  PERL_VERSION=5.8.8
>  WRKDIRPREFIX=/var/tmp
> 
> 
>  The options.h file in /usr/obj/usr/src/gnu/usr.bin/cc/cc_tools/options.h
> 
>  Here are the output of options.h from pastebin:
> 
>  http://pastebin.com/m254dfedf
> 
> 
>  When I run "make -j1 buildworld" i get the error is as following:
> 
>  http://pastebin.com/m57738677
> 
>  I don't know much of programming (if any), but I have tried to change
>  things in (remove OPT_w from both places) options.h and runned
>  "make -j1 -DNOCLEAN buildworld"
> 
>  But that have not solve anything of this, perhaps some can shine some
>  light on this or have I missed something trivial?
>  I can build ports and so on without any problems.
> 
>  Thank you.
> >>> Not enough data. Please send a full compressed version of the log to me.
> >>> Thanks!
> >>> -Garrett
> > 
> > That's really odd why the symbol was redefined. Here're the relevant
> > sections of the log:
> > 
> > c_tools/../../../../contrib/gcc/gencheck.c
> > In file included from ./tm.h:4,
> > from
> > /usr/src/gnu/usr.bin/cc/cc_tools/../../../../contrib/gcc/gencheck.c:25:
> > ./options.h:901: error: redeclaration of enumerator 'OPT_w'
> > ./options.h:899: error: previous definition of 'OPT_w' was here
> > *** Error code 1
> > 1 error
> > *** Error code 2
> > 1 error
> > *** Error code 2
> > 1 error
> > *** Error code 2
> > 
> > Looking back, I'm not sure why OPT_w is printed into the file twice.
> > I wonder if you met a rare race condition where options.sh printed out
> > that line twice; then again that's unlikely, given the
> > reproducibility... I could be wrong however.
> > Is your source tree based off of RELENG_7 and what cvsup server is it
> > synced against?
> > -Garrett
> 
> 
> I'm using cvsup.se.freebsd.org, cvsup.dk.freebsd.org, 
> cvsup.no.freebsd.org I have tried to remove /usr/src/* right now and 
> downloaded it again but it did not help after the build.
> 
> So I have no idea what to do from here.
> 
> But I guess It would be solved sometime in the future.

Did you clean your /usr/obj/ directory?

-- 
Dominik Żyła

If you want a picture of the future, imagine a boot stamping on a human 
face - for ever..
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: Make world builderror on 7.1-BETA2 with latest cvsup

[Mattias Björk]

Hello again,


Garrett Cooper wrote:

On Fri, Jan 2, 2009 at 11:52 AM, Mattias Björk  wrote:

Hello Garrett,


Sorry for top posting, but here is the attachment of the log.


Garrett Cooper wrote:

On Jan 1, 2009, at 19:23, Mattias Björk  wrote:


Hello everybody,

First of all, my bad if this get sent two times.

I have a compile error/problem when building world.

My uname -a output is:

FreeBSD barabolaptop 7.1-BETA2 FreeBSD 7.1-BETA2 #0: Thu Dec  4 20:52:35
CET 2008 r...@barabolaptop:/usr/obj/usr/src/sys/BARABOLAPTOP  i386

My make.conf looks like:

# added by use.perl 2008-12-03 00:58:09
PERL_VER=5.8.8
PERL_VERSION=5.8.8
WRKDIRPREFIX=/var/tmp


The options.h file in /usr/obj/usr/src/gnu/usr.bin/cc/cc_tools/options.h

Here are the output of options.h from pastebin:

http://pastebin.com/m254dfedf


When I run "make -j1 buildworld" i get the error is as following:

http://pastebin.com/m57738677

I don't know much of programming (if any), but I have tried to change
things in (remove OPT_w from both places) options.h and runned
"make -j1 -DNOCLEAN buildworld"

But that have not solve anything of this, perhaps some can shine some
light on this or have I missed something trivial?
I can build ports and so on without any problems.

Thank you.

Not enough data. Please send a full compressed version of the log to me.
Thanks!
-Garrett


That's really odd why the symbol was redefined. Here're the relevant
sections of the log:

c_tools/../../../../contrib/gcc/gencheck.c
In file included from ./tm.h:4,
from
/usr/src/gnu/usr.bin/cc/cc_tools/../../../../contrib/gcc/gencheck.c:25:
./options.h:901: error: redeclaration of enumerator 'OPT_w'
./options.h:899: error: previous definition of 'OPT_w' was here
*** Error code 1
1 error
*** Error code 2
1 error
*** Error code 2
1 error
*** Error code 2

Looking back, I'm not sure why OPT_w is printed into the file twice.
I wonder if you met a rare race condition where options.sh printed out
that line twice; then again that's unlikely, given the
reproducibility... I could be wrong however.
Is your source tree based off of RELENG_7 and what cvsup server is it
synced against?
-Garrett



I'm using cvsup.se.freebsd.org, cvsup.dk.freebsd.org, 
cvsup.no.freebsd.org I have tried to remove /usr/src/* right now and 
downloaded it again but it did not help after the build.


So I have no idea what to do from here.

But I guess It would be solved sometime in the future.

Thanks so far :)
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: Make world builderror on 7.1-BETA2 with latest cvsup

[Mattias Björk]

Hello again :)

Claus Guttesen wrote:

Hello Mattias.

On Fri, Jan 2, 2009 at 9:04 PM, Mattias Björk  wrote:

Hello Claus,

Claus Guttesen wrote:

I have a compile error/problem when building world.

My uname -a output is:

FreeBSD barabolaptop 7.1-BETA2 FreeBSD 7.1-BETA2 #0: Thu Dec  4 20:52:35
CET 2008 r...@barabolaptop:/usr/obj/usr/src/sys/BARABOLAPTOP  i386

Sync the time on your computer and try the buildworld again.



I have tried that several times, even removed /usr/src/* entirely and
started from scratch. And tried a couple of other cvsup.XX.freebsd.org
servers. Still no go Claus :(

Any more suggestions ?

Thanks so far.


Unfortunately not. :-/ The only times I've had problems recently
building world was because the time on my computer was (way)
incorrect.



Well I don't think anything is incorrect on the computer, can build
everything else fine and it goes on and on.



You can try removing /usr/obj as well before doing a new buildworld.
AFAIK you don't have to alter options.h to make a buildworld run
smoothly.



I have tried too make clean ; make clean ; make cleandir ; make cleandir
and then rm -rf /usr/obj. Still the same problem I'm afraid.



HTH.



___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: Make world builderror on 7.1-BETA2 with latest cvsup

[Mattias Björk]

Hello Claus,

Claus Guttesen wrote:

I have a compile error/problem when building world.

My uname -a output is:

FreeBSD barabolaptop 7.1-BETA2 FreeBSD 7.1-BETA2 #0: Thu Dec  4 20:52:35
CET 2008 r...@barabolaptop:/usr/obj/usr/src/sys/BARABOLAPTOP  i386


Sync the time on your computer and try the buildworld again.




I have tried that several times, even removed /usr/src/* entirely and 
started from scratch. And tried a couple of other cvsup.XX.freebsd.org 
servers. Still no go Claus :(


Any more suggestions ?

Thanks so far.
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: Make world builderror on 7.1-BETA2 with latest cvsup

[Mattias Björk]

Hello Garret,


Garrett Cooper wrote:

On Jan 1, 2009, at 19:23, Mattias Björk  wrote:


Hello everybody,

First of all, my bad if this get sent two times.

I have a compile error/problem when building world.

My uname -a output is:

FreeBSD barabolaptop 7.1-BETA2 FreeBSD 7.1-BETA2 #0: Thu Dec  4 20:52:35
CET 2008 r...@barabolaptop:/usr/obj/usr/src/sys/BARABOLAPTOP  i386

My make.conf looks like:

# added by use.perl 2008-12-03 00:58:09
PERL_VER=5.8.8
PERL_VERSION=5.8.8
WRKDIRPREFIX=/var/tmp


The options.h file in /usr/obj/usr/src/gnu/usr.bin/cc/cc_tools/options.h

Here are the output of options.h from pastebin:

http://pastebin.com/m254dfedf


When I run "make -j1 buildworld" i get the error is as following:

http://pastebin.com/m57738677

I don't know much of programming (if any), but I have tried to change
things in (remove OPT_w from both places) options.h and runned
"make -j1 -DNOCLEAN buildworld"

But that have not solve anything of this, perhaps some can shine some
light on this or have I missed something trivial?
I can build ports and so on without any problems.

Thank you.


Not enough data. Please send a full compressed version of the log to me.
Thanks!



No thank you, :) I will send it to you directly.



-Garrett___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: Make world builderror on 7.1-BETA2 with latest cvsup

[Claus Guttesen]

> I have a compile error/problem when building world.
>
> My uname -a output is:
>
> FreeBSD barabolaptop 7.1-BETA2 FreeBSD 7.1-BETA2 #0: Thu Dec  4 20:52:35
> CET 2008 r...@barabolaptop:/usr/obj/usr/src/sys/BARABOLAPTOP  i386

Sync the time on your computer and try the buildworld again.

-- 
regards
Claus

When lenity and cruelty play for a kingdom,
the gentler gamester is the soonest winner.

Shakespeare
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: Make world builderror on 7.1-BETA2 with latest cvsup

[Garrett Cooper]

On Jan 1, 2009, at 19:23, Mattias Björk   
wrote:



Hello everybody,

First of all, my bad if this get sent two times.

I have a compile error/problem when building world.

My uname -a output is:

FreeBSD barabolaptop 7.1-BETA2 FreeBSD 7.1-BETA2 #0: Thu Dec  4  
20:52:35

CET 2008 r...@barabolaptop:/usr/obj/usr/src/sys/BARABOLAPTOP  i386

My make.conf looks like:

# added by use.perl 2008-12-03 00:58:09
PERL_VER=5.8.8
PERL_VERSION=5.8.8
WRKDIRPREFIX=/var/tmp


The options.h file in /usr/obj/usr/src/gnu/usr.bin/cc/cc_tools/ 
options.h


Here are the output of options.h from pastebin:

http://pastebin.com/m254dfedf


When I run "make -j1 buildworld" i get the error is as following:

http://pastebin.com/m57738677

I don't know much of programming (if any), but I have tried to change
things in (remove OPT_w from both places) options.h and runned
"make -j1 -DNOCLEAN buildworld"

But that have not solve anything of this, perhaps some can shine some
light on this or have I missed something trivial?
I can build ports and so on without any problems.

Thank you.


Not enough data. Please send a full compressed version of the log to me.
Thanks!
-Garrett___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: make world broken for RELENG_6

[Rick C. Petty]

On Sat, Mar 31, 2007 at 09:22:58AM -0700, Julian Elischer wrote:
> David Wolfskill wrote:
> >
> >Right; I encountered the same thing.
> >
> >Locally reverting src/sys/netinet/ip_fw.h rev. 1.100.2.6 appears to have
> >fixed it for me:  after doing that, I was able to successfully build,
> >install, and boot.  And yes, I use IPFW.  :-}
> >
> >The issue appears to be that src/sbin/ipfw/ipfw2.c references
> >ipfw_insn_pipe, which 1.100.2.6 dyked out out ip_fw.h.
> >
> >I don't know that reverting 1.100.2.6 was the "correct" thing to do; it
> >may be better to change ipfw2.c to not try to refer to it.
> >
> >I've Cc:ed Julian, since he committed the changes.
> >
> >Peace,
> >david
> 
> 
> try just deleting the offending lines in ipfw2.c



It's just so rare that -stable breaks on buildworld (even -current isn't
broken often, in terms of build breakage)..  something I can't say about
other operating systems.  A csup this morning caused the problem to go
away.  Thanks, guys!

-- Rick C. Petty
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: make world broken for RELENG_6

[Julian Elischer]

David Wolfskill wrote:

On Sat, Mar 31, 2007 at 02:27:56AM -0600, Rick C. Petty wrote:

Is anyone else experiencing build problems with the latest RELENG_6 (sup'd
as of 2007-Mar-31 0800 UTC)?

A buildworld failed at:

===> sbin/ipfw (all)
...


Right; I encountered the same thing.

Locally reverting src/sys/netinet/ip_fw.h rev. 1.100.2.6 appears to have
fixed it for me:  after doing that, I was able to successfully build,
install, and boot.  And yes, I use IPFW.  :-}

The issue appears to be that src/sbin/ipfw/ipfw2.c references
ipfw_insn_pipe, which 1.100.2.6 dyked out out ip_fw.h.

I don't know that reverting 1.100.2.6 was the "correct" thing to do; it
may be better to change ipfw2.c to not try to refer to it.

I've Cc:ed Julian, since he committed the changes.

Peace,
david



try just deleting the offending lines in ipfw2.c

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: make world broken for RELENG_6

[Scott Robbins]

On Sat, Mar 31, 2007 at 03:47:19PM +0200, Stefan 'Steve' Tell wrote:
> * "Rick C. Petty" <[EMAIL PROTECTED]> wrote:
> 
> > Is anyone else experiencing build problems with the latest RELENG_6 (sup'd
> > as of 2007-Mar-31 0800 UTC)?
> 
> Jip, same here.

Tinderbox is showing it in the place--ipfw.  

In a way that's a good thing, usually, when I see a problem I'm having
on Tinderbox, it gets fixed in a few hours.  

Amazing how much we take for granted--thank you developers, for all your
hard work. (That's not saying anyone posting here is taking things for
granted--I just mean that we, in general,  see something broken 
on Tinderbox and expect that it will be fixed quickly.)


-- 

Scott Robbins

PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6

Xander: How could you let her go? 
Giles: As the soon-to-be-purple area on my jaw will attest, 
I did not 'let' her go. 

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: make world broken for RELENG_6

[David Wolfskill]

On Sat, Mar 31, 2007 at 02:27:56AM -0600, Rick C. Petty wrote:
> Is anyone else experiencing build problems with the latest RELENG_6 (sup'd
> as of 2007-Mar-31 0800 UTC)?
> 
> A buildworld failed at:
> 
> ===> sbin/ipfw (all)
> ...

Right; I encountered the same thing.

Locally reverting src/sys/netinet/ip_fw.h rev. 1.100.2.6 appears to have
fixed it for me:  after doing that, I was able to successfully build,
install, and boot.  And yes, I use IPFW.  :-}

The issue appears to be that src/sbin/ipfw/ipfw2.c references
ipfw_insn_pipe, which 1.100.2.6 dyked out out ip_fw.h.

I don't know that reverting 1.100.2.6 was the "correct" thing to do; it
may be better to change ipfw2.c to not try to refer to it.

I've Cc:ed Julian, since he committed the changes.

Peace,
david
-- 
David H. Wolfskill  [EMAIL PROTECTED]
Believe SORBS at your own risk: 63.193.123.122 has been static since Aug 1999.

See http://www.catwhisker.org/~david/publickey.gpg for my public key.


pgpKb7Z9N9BlM.pgp
Description: PGP signature

Re: make world broken for RELENG_6

[Stefan 'Steve' Tell]

* "Rick C. Petty" <[EMAIL PROTECTED]> wrote:

> Is anyone else experiencing build problems with the latest RELENG_6 (sup'd
> as of 2007-Mar-31 0800 UTC)?

Jip, same here.

-- 
By(t)e,
Steve   /\  http://blog.crashmail.de

GnuPG/PGP: 0x9B6C7E15, encrypted mail prefered, see header

___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Make World Fails on Man Pages

[Jason C. Wells]

Mark Andrews wrote:

 From unlink(2).

 [ENOTDIR]  A component of the path prefix is not a directory.


I suppose that is why rm is not working.  The question was "Why don't 
some man pages install?"


I think I have found the problem.  Somewhere along the way a DOCSUPFILE 
make variable and separate doc-supfile came into existence.  I never 
learned this so my doc sources were out of date.  That gave me errors 
that caused me to set NO_SHARE which resulted in groff macros not being 
installed which resulted man page weirdness.


This answer to myself for posterity.

Later,
Jason C. Wells


___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Make World Fails on Man Pages

[Mark Andrews]

> Here is a snippet of my failed installworld logs as I move to 6.0R:
> 
> install -o root -g wheel -m 444 crypt.3.gz  /usr/share/man/man3
> /usr/share/man/man3/crypt_get_format.3.gz -> /usr/share/man/man3/crypt.3.gz
> rm: /usr/share/man/man3/crypt_get_format.3: Not a directory
> rm: /usr/share/man/man3/crypt_get_format.3.gz: Not a directory
> 
> I get the same error from a couple other man pages (e.g. bwrite) if I 
> run make from the directory where the pages source exists.  Not all man 
> pages cause an error, just some.  I tried removing /usr/obj for a fresh 
> build since this was a former 5.4R box.  Cvsup isn't fetching any new 
> sources so I must be up to date.
> 
> Any ideas why some man pages won't install?

 From unlink(2).

 [ENOTDIR]  A component of the path prefix is not a directory.

> 
> Thanks,
> Jason
> ___
> freebsd-stable@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "[EMAIL PROTECTED]"
--
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: [EMAIL PROTECTED]
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Make world fails

[Ruslan Ermilov]

On Thu, Nov 06, 2003 at 03:21:31PM -0500, condory wrote:
> There a bit of this I just posted part of it any ideas anyone ?
> 
> 
> magic, 34789: type @Initial revision invalid
> mkmagic: magic, 34790: offset @ invalid
> mkmagic: magic, 34790: type @ invalid
> mkmagic: magic, 34791: offset text invalid
> mkmagic: magic, 34791: type text invalid
> mkmagic: magic, 34792: offset @#
> -- invalid
> mkmagic: magic, 34792: type @#--
>  invalid
> mkmagic: magic, 34810: offset @ invalid
> mkmagic: magic, 34810: type @ invalid
> mkmagic: magic, 34813: type .1.1.1 invalid
> mkmagic: magic, 34814: offset log invalid
> mkmagic: magic, 34814: type log invalid
> mkmagic: magic, 34815: offset @Virgin import of Christos Zoulas's FILE 3.39. inv
> alid
> mkmagic: magic, 34815: type @Virgin import of Christos Zoulas's FILE 3.39. inval
> id
> mkmagic: magic, 34816: offset @ invalid
> mkmagic: magic, 34816: type @ invalid
> mkmagic: magic, 34817: offset text invalid
> mkmagic: magic, 34817: type text invalid
> mkmagic: magic, 34818: offset @@ invalid
> error.txt: unmodified: line 1
> 
No context -- no cookie.  :-(


Cheers,
-- 
Ruslan Ermilov  Sysadmin and DBA,
[EMAIL PROTECTED]   Sunbay Software Ltd,
[EMAIL PROTECTED]   FreeBSD committer


pgp0.pgp
Description: PGP signature

Re: make world

[Doug White]

On Sun, 28 Sep 2003, Peter J. Blok wrote:

> Hi, I was under the impression a successful make world was updating include
> files in /usr/include/netinet too.
>
> When I have a good make world, the files in /usr/include have new time stamps,
> but the ones in netinet have not!

Are you sure the files are different?

You can try doing 'make includes' from /usr/src/ to see if it fixes your
compile issues. This forcibly reinstalls the header files.

Also check that your checkout is complete.

>
> Am I missing something here? My current stable doesn't compille properly it
> fails in kdump on a missing ioctlcmd_t typedef.
>
> Peter
>
> ___
> [EMAIL PROTECTED] mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "[EMAIL PROTECTED]"
>

-- 
Doug White|  FreeBSD: The Power to Serve
[EMAIL PROTECTED]  |  www.FreeBSD.org
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: make world considered harmful

[Brian T. Schellenberger]

On Thursday 25 July 2002 12:18 am, Darren Pilgrim wrote:
| Matthew Whelan wrote:
| > 24/07/2002 14:59:42, Jamie Bowden <[EMAIL PROTECTED]> wrote:
| > [re-insert from Brian's post]
| >
| > >:Personally, I think it would be better to remove it; for those who
| > >: dislike typing and don't mind endangering their system, it would be
| > >: better to have instead a
| > >:
| > >:make universe
| >
| > Maybe 'make bravenewworld'?
|
| Is there a reason against changing the world target to just be an alias
| for "make buildworld && make kernel && make installworld && mergemaster"?
| Forgive me if I'm asking something already answered I haven't been able
| to follow this entire thread.

Far better IMHO to create a new target for this and delete the old "bad" 
target than to change the meaning of an existing name.  If we keep the 
exiting name we must keep the existing behavior or it will cause far more 
confusion than it prevents.


-- 
Brian, the man from Babble-On . . . .   [EMAIL PROTECTED] (personal)
http://www.babbleon.org

http://www.eff.org  http://www.programming-freedom.org 

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world considered harmful

[Darren Pilgrim]

Matthew Whelan wrote:
> 24/07/2002 14:59:42, Jamie Bowden <[EMAIL PROTECTED]> wrote:
> [re-insert from Brian's post]
> >:Personally, I think it would be better to remove it; for those who dislike
> >:typing and don't mind endangering their system, it would be better to have
> >:instead a
> >:
> >:make universe
> 
> Maybe 'make bravenewworld'?

Is there a reason against changing the world target to just be an alias
for "make buildworld && make kernel && make installworld && mergemaster"?
Forgive me if I'm asking something already answered I haven't been able
to follow this entire thread.

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world considered harmful

[Jonathan Chen]

On Tue, Jul 23, 2002 at 03:49:04AM -0700, Jamie Bowden wrote:
> On Mon, 22 Jul 2002, Barney Wolff wrote:
> 
> :But there's nothing right with it, either.  Is the demonstrated risk
> :of people using it inappropriately really worth not having to type
> :  make buildworld && make installworld
> :in the few cases where it's safe?  I claim not.
> :
> :I run cvsup nohup'd and look at the output, but I'd would never trust
> :that I'd notice a kernel interface change, and know when I could get
> :away with not building the kernel.
> 
> #
> # $FreeBSD: src/Makefile,v 1.234.2.14 2002/07/16 18:36:19 ru Exp $
> #
> # The user-driven targets are:
> #
> # buildworld  - Rebuild *everything*, including glue to help do
> #   upgrades.
> # installworld- Install everything built by "buildworld".
> # world   - buildworld + installworld.
> 
> Perhaps I'm missing something, but 'make world' appears to do nothing more
> than you've done above in fewer keystrokes.

Except for the fact that if the new kernel doesn't like your system,
you're SOL with out-of-sync userland.
-- 
Jonathan Chen <[EMAIL PROTECTED]>
--
 "Nyuck, nyuck, nyuck" - Curly

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world considered harmful

[Jamie Bowden]

On Mon, 22 Jul 2002, Barney Wolff wrote:

:But there's nothing right with it, either.  Is the demonstrated risk
:of people using it inappropriately really worth not having to type
:  make buildworld && make installworld
:in the few cases where it's safe?  I claim not.
:
:I run cvsup nohup'd and look at the output, but I'd would never trust
:that I'd notice a kernel interface change, and know when I could get
:away with not building the kernel.

#
# $FreeBSD: src/Makefile,v 1.234.2.14 2002/07/16 18:36:19 ru Exp $
#
# The user-driven targets are:
#
# buildworld  - Rebuild *everything*, including glue to help do
#   upgrades.
# installworld- Install everything built by "buildworld".
# world   - buildworld + installworld.

Perhaps I'm missing something, but 'make world' appears to do nothing more
than you've done above in fewer keystrokes.

Jamie Bowden

-- 
"It was half way to Rivendell when the drugs began to take hold"
Hunter S Tolkien "Fear and Loathing in Barad Dur"
Iain Bowen <[EMAIL PROTECTED]>



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world considered harmful

[Brian T. Schellenberger]

On Monday 22 July 2002 09:05 am, Ruben de Groot wrote:
| On Mon, Jul 22, 2002 at 08:38:50AM -0400, Brian T. Schellenberger typed:
| > On Monday 22 July 2002 07:41 am, Jamie Bowden wrote:
| > | On Sun, 21 Jul 2002, Barney Wolff wrote:
| > | :When is "make world" the right thing to do?  If the answer is "never"
| > | :then why don't we remove it from /usr/src/Makefile?  Its availability
| > | :just leads people into trouble.
| > |
| > | Since when?  I use it regularly after CVSupdating /usr/src, works fine.
| >
| > It's definately not recommended even if it will sometimes work.
| >
| > It causes you to try to start running with the new world while still
| > running the older kernel.  (Or, I guess, you could follow the sequence
| > make kernel, reboot, make world, mergemaster.  That sequence would seem
| > relatively safe, I guess.)
|
| Save, but not supported. You must buildworld before you can buildkernel.

My intuition is that it would be almost certain to fail to compile if it ewere 
to fail rather than leaving you completely dead in the water.  But I always 
do it per UPDATING:

make buildworld
make kernel   # dunno why UPDATING says to do *this* in two steps
reboot (-s)
make installworld
mergemaster

-- 
Brian, the man from Babble-On . . . .   [EMAIL PROTECTED] (personal)
http://www.babbleon.org

http://www.eff.org  http://www.programming-freedom.org 

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world considered harmful

[Ruben de Groot]

On Mon, Jul 22, 2002 at 08:38:50AM -0400, Brian T. Schellenberger typed:
> On Monday 22 July 2002 07:41 am, Jamie Bowden wrote:
> | On Sun, 21 Jul 2002, Barney Wolff wrote:
> | :When is "make world" the right thing to do?  If the answer is "never"
> | :then why don't we remove it from /usr/src/Makefile?  Its availability
> | :just leads people into trouble.
> |
> | Since when?  I use it regularly after CVSupdating /usr/src, works fine.
> 
> It's definately not recommended even if it will sometimes work.
> 
> It causes you to try to start running with the new world while still running 
> the older kernel.  (Or, I guess, you could follow the sequence make kernel, 
> reboot, make world, mergemaster.  That sequence would seem relatively safe, I 
> guess.)

Save, but not supported. You must buildworld before you can buildkernel.

> 
> |
> | Jamie Bowden
> 
> -- 
> Brian, the man from Babble-On . . . .   [EMAIL PROTECTED] (personal)
> http://www.babbleon.org
> 
> http://www.eff.org  http://www.programming-freedom.org 
> 
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-stable" in the body of the message

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world considered harmful

[Brian T. Schellenberger]

On Monday 22 July 2002 07:41 am, Jamie Bowden wrote:
| On Sun, 21 Jul 2002, Barney Wolff wrote:
| :When is "make world" the right thing to do?  If the answer is "never"
| :then why don't we remove it from /usr/src/Makefile?  Its availability
| :just leads people into trouble.
|
| Since when?  I use it regularly after CVSupdating /usr/src, works fine.

It's definately not recommended even if it will sometimes work.

It causes you to try to start running with the new world while still running 
the older kernel.  (Or, I guess, you could follow the sequence make kernel, 
reboot, make world, mergemaster.  That sequence would seem relatively safe, I 
guess.)

|
| Jamie Bowden

-- 
Brian, the man from Babble-On . . . .   [EMAIL PROTECTED] (personal)
http://www.babbleon.org

http://www.eff.org  http://www.programming-freedom.org 

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world every week?

[John Merryweather Cooper]

On 2001.07.24 19:41 jett wrote:
> im running freebsd 4.3-stable and every friday night i'm updating my
> sources via cvsup. my question is do i have to always build the world
> after every cvsup so that i may stay stable?
> 
> jett
> 

Please turn off the double-pump of HTML.  It makes replying a mess . . .

No.  I cvsup frequently, but I only make when something of interest
has changed.  Then I make world && make kernel KERNCONF=your_kernel_config.

jmc

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

RE: make world every week?

[Todd Punderson]

I 
don't. I just run a buildworld/makeworld when there is a security advisory. That 
gives me an excuse to go through the process.
Your 
milage my vary..
Todd

  -Original Message-From: 
  [EMAIL PROTECTED] 
  [mailto:[EMAIL PROTECTED]]On Behalf Of 
  jettSent: Tuesday, July 24, 2001 10:41 PMTo: 
  freebsd-stableSubject: make world every week?
  im running freebsd 4.3-stable and every friday 
  night i'm updating my sources via cvsup. my question is do i have to always 
  build the world after every cvsup so that i may stay stable?
   
  jett

Re: make world fails in RELENG_4_3

[A. L. Meyers]

On Sat, 7 Jul 2001, Chad R. Larson wrote:

> On Sat, Jul 07, 2001 at 03:05:06PM +0200, A. L. Meyers wrote:
> > Had exactly the same problem using RELENG_4 (choking on pod2man)
> > and received a solution from Kent.
>
> My solution was to cd to the perl source directory, do a "make
> install" and then restart the install world.
(snip)

Kent's solution was to dig down in perl source directory all the
way to pod2man and make it. Then restart installworld. It worked.

>
> Anyway, I posted the original not particularly because I needed help
> getting through the build, but because I thought someone with commit
> privileges would want to try to fix it.

Well, let's hope a friendly committer is reading these postings
and fixes the problem :-) .

Lucien


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world still broken around krb4_version bug--when will the patch reach stable?

[John Merryweather Cooper]

Nat Lanza wrote:
> 
> John Merryweather Cooper <[EMAIL PROTECTED]> writes:
> 
> > I'm open to other suggestions, and I'll do the rm -rfd /usr/src if I
> > absolutely have to, but I'd rather avoid it.
> 
> I just tried deleting /usr/src, re-cvsupping, and rebuilding, and it
> still dies at the same spot as others have reported:
> 
>   cc -O -pipe -march=pentiumpro 
>-I/usr/src/kerberos5/lib/libasn1/../../../crypto/heimdal/include 
>-I/usr/src/kerberos5/lib/libasn1/../../../crypto/heimdal/lib/asn1 
>-I/usr/src/kerberos5/lib/libasn1/../../../crypto/heimdal/lib/roken 
>-I/usr/src/kerberos5/lib/libasn1/../../include 
>-I/usr/obj/usr/src/kerberos5/lib/libasn1 -Wall 
>-I/usr/src/kerberos5/lib/libasn1/../../include 
>-I/usr/src/kerberos5/lib/libasn1/../../include -DHAVE_CONFIG_H -DKRB5_KRB4_COMPAT 
>-DKRB4 -DINET6 -static -o make-print-version 
>/usr/src/kerberos5/lib/libasn1/../../../crypto/heimdal/lib/roken/make-print-version.c
>   In file included from 
>/usr/src/kerberos5/lib/libasn1/../../../crypto/heimdal/lib/roken/make-print-version.c:49:
>   /usr/src/kerberos5/lib/libasn1/../../include/version.h:3: conflicting types for 
>`krb4_version'
>   
>/usr/src/kerberos5/lib/libasn1/../../../crypto/heimdal/lib/roken/make-print-version.c:47:
> previous declaration of `krb4_version'
>   *** Error code 1
> 
>   Stop in /usr/src/kerberos5/lib/libasn1.
>   *** Error code 1
> 
>   Stop in /usr/src.
>   *** Error code 1
> 
>   Stop in /usr/src.
>   *** Error code 1
> 
> Clearly, this isn't an artifact of a corrupted local copy of the tree;
> I think it's genuinely broken, and would love to see a fix.
> 
> --nat
> 
> --
> nat lanza --- there are no whole truths;
> [EMAIL PROTECTED]  all truths are half-truths
> http://www.cs.cmu.edu/~magus/ ---  -- alfred north whitehead
> 
Then I won't bother zapping /usr/src.  Your make "tail" is identical to
mine.

jmc

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world?

[David W. Chapman Jr.]

That doesn't sound like a freebsd issue, why don't you use
/usr/ports/deve/automake ?

- Original Message -
From: "Harlan Stenn" <[EMAIL PROTECTED]>
To: "Eric M Logan" <[EMAIL PROTECTED]>
Cc: "Erik Trulsson" <[EMAIL PROTECTED]>; "FreeBSD stable"
<[EMAIL PROTECTED]>; "FreeBSD Questions"
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>
Sent: Sunday, May 13, 2001 1:41 PM
Subject: Re: make world?


> Thanks!
>
> I have been using
>
>   :pserver:[EMAIL PROTECTED]:/cvs/automake
>
> to get automake.
>
> What, if any, is the difference between this repository and the one at
>
>  <http://sources.redhat.com/automake/>
>
> ???
>
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-stable" in the body of the message
>


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world?

[Erik Trulsson]

On Sun, May 13, 2001 at 11:23:53AM -0700, Eric M Logan wrote:
> Thanks for responding but no.  I did a "du" on each of /usr/src, /usr/ports, and
> /usr/obj.  There's nothing unexpected with src and ports and obj is always
> deleted after the make world (mfs).  The system just ballons up.  I know that
> when one "makes world", everything gets installed.  So my question is, is there
> anyway to make world and *only* update/install stuff that was part of the
> original installation.  Again, I've already played with the options in
> /etc/make.conf but with no luck.  Any ideas or help would be appreciated,
> thanks.

That sounds weird. 
To answer your question: No, there is no such way. The system does not keep
track of what was originally installed.

I would suggest using "du" to find out where in the file system the bloat
occurs.

> 
> Erik Trulsson wrote:
> 
> > On Sun, May 13, 2001 at 10:55:49AM -0700, Eric M Logan wrote:
> > > I recently installed a bare bones installation of FreeBSD that came out
> > > to under 200MB.  After a "make world" the system went over 500MB?!  How
> > > can I cvsup my sources and "make world" w/o adding/updating additional
> > > stuff other than the original barebones installation?  I've already
> > > tried uncommenting the options in /etc/make.conf such as no_cvs, etc but
> > > with no success.  Any ideas or help would be appreciated, thanks.
> > >
> >
> > Those 500 MB would not by any chance include the contents of /usr/src or
> > /usr/obj, each of which is about 300 MB ?
> > If that is the case 'rm -fr /usr/src/* /usr/obj/*' should free a lot of
> > diskspace.
> > Otherwise I have no idea. My barebones installation is still at about 150 MB
> > after several upgrades but then neither /usr/src nor /usr/obj reside on that
> > machine.
> >

-- 

Erik Trulsson
[EMAIL PROTECTED]


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world?

[Harlan Stenn]

Thanks!

I have been using

  :pserver:[EMAIL PROTECTED]:/cvs/automake

to get automake.

What, if any, is the difference between this repository and the one at

 

???

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world fails on -STABLE - solved?

[Bjarne Wichmann Petersen]

On Tue, 17 Apr 2001 19:08:31 +0200
Bjarne Wichmann Petersen <[EMAIL PROTECTED]> wrote:

Seems to be the way I had set my CXX-flags in make.conf that screwed
things up.

Bjarne

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: Make world Broken

[Kris Kennaway]

On Wed, Mar 07, 2001 at 10:27:15AM +0200, Jacques Marneweck wrote:
> Hi all,
> 
> This what I'm getting where it breaks:
> 
> ===> libdisk
> make: don't know how to make dkcksum.c. Stop
> 
> The file dkcksum.c does not exist in the directory where libdisk.

This was fixed about 3 days ago.  Please update your sources when you
run into a build problem before reporting it to the list, because most
of them are fixed within hours of being introduces.

Kris

 PGP signature

Re: Make world failure on ssl23.h

[Jim Durham]

On Sat, 17 Feb 2001, Alex M wrote:

> If you have followed the postings here u'd probably notice that i had the
> same prob, make world always failed on openssl stuff.
> I got rid of this prob by having src-all in my cvsupfile.
> 
> > My system is currently 4.2-RELEASE. I have been trying to bring
> > it up to 4.2-STABLE.
> >
> > make world fails on
> >
> > /usr/obj/usr/src/secure/lib/libssl/openssl/ssl23.h complaining of
> > an unterminated character constant.
> >
> > Upon examination of ssl23.h, it is not a text file, or at least
> > it is extremely corrupted.
> >

Hmmm... I just joined, but I did check the archives and saw nothing.
I just must have missed it.

However, I did have "src-all" in my cvsupfile when I originally
cvsup'd the sources. I figured there might have been a fix, so
I did a cvsup with just "secure" and "crypto" and that didn't
fix it.

I'll try "src-all" again.

Thanks,

Jim




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: Make world failure on ssl23.h

[Alex M]

If you have followed the postings here u'd probably notice that i had the
same prob, make world always failed on openssl stuff.
I got rid of this prob by having src-all in my cvsupfile.


--
Alex M aka TZapper.

- Original Message -
From: "Jim Durham" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Saturday, February 17, 2001 5:22 AM
Subject: Make world failure on ssl23.h


> My system is currently 4.2-RELEASE. I have been trying to bring
> it up to 4.2-STABLE.
>
> make world fails on
>
> /usr/obj/usr/src/secure/lib/libssl/openssl/ssl23.h complaining of
> an unterminated character constant.
>
> Upon examination of ssl23.h, it is not a text file, or at least
> it is extremely corrupted.
>
> I have tried cvsup'ing secure several times over the last 3 days
> and no change.
>
> I have also removed /usr/obj and started "clean". No help.
>
> What am I doing wrong?
>
> Thanks,
>
> Jim Durham
>
>
>
>
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-stable" in the body of the message
>




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failure on telnetd

[Kris Kennaway]

On Tue, Feb 13, 2001 at 06:07:32PM -0800, Isaac Mushinsky wrote:
> make buildworld fails in /usr/src/libexec/telnetd.
> 
> This is the latest cvs stable source. Would anybody know why is
> pathnames.h missing?

Nope, look for a local problem. Perhaps your source tree is corrupt.

Kris

 PGP signature

Re: make world

[William Wong]

Ah. Thanks for the automate command.

I just find it strange since if 'make world' was designed to update the
entire system, having an older kernel.GENERIC lying around by default is a
bit weird.

I don't think the "make world" page in the handbook states that
kernel.GENERIC isn't updated by default.  Maybe just a note that you could
update that too. (Like how it makes a note about updating sysinstall for
completeness).

- Will

> > Just a suggestion for the make world process.  How about by default
> > compiling a kernel.GENERIC (just in case)?
>
> How about just doing
>
> make buildkernel KERNEL=kernel.GENERIC
>
> after a successful make world?  For example, to automate this, you might
> do
>
> make buildworld installworld buildkernel installkernel
> KERNEL=kernel.GENERIC



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world

[Donn Miller]

William Wong wrote:

> Just a suggestion for the make world process.  How about by default
> compiling a kernel.GENERIC (just in case)?

How about just doing

make buildkernel KERNEL=kernel.GENERIC

after a successful make world?  For example, to automate this, you might
do

make buildworld installworld buildkernel installkernel
KERNEL=kernel.GENERIC


-Donn


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world on 4.2-R breaking

[Nevermind]

Hello, David Kelly!

On Wed, Nov 22, 2000 at 06:41:17PM -0600, you wrote:

> "Antonio Carlos Pina" writes:
> > David, I have a little program (I don't recall who wrote it, sorry) 
> > that tests memory, stressing it. I can tell that it hangs a bad 
> > machine (cpu or bus or memory) in 2 or 3 minutes. It fits in a floppy 
> > and it is a DOS program. If you want a copy, I can send it to you.
> 
> Yeah. But. Remember a memory test program can not prove memory is good.
> It can only prove memory is bad. And then only if it happens to be very
> bad. Memory can and does fail when certian sequences of events happen.
> The trick is to reproduce that sequence.
There is a program called testmem.com (for DOS, sure) which tests memory very
good. I know, here is not much DOS-lovers, but it is the only program I found
which gives true results. It is so small that it can be loaded into L1 cache
so, no RAM access occurs during memory test except by this prog.

-- 
Alexandr P. Kovalenko   http://nevermind.kiev.ua/
NEVE-RIPE


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world on 4.2-R breaking

[Cyrille Lefevre]

"Antonio Carlos Pina" <[EMAIL PROTECTED]> writes:

> David, I have a little program (I don't recall who wrote it, sorry) 
> that tests memory, stressing it. I can tell that it hangs a bad 
> machine (cpu or bus or memory) in 2 or 3 minutes. It fits in a floppy 
> and it is a DOS program. If you want a copy, I can send it to you.

also, there is :

http://reality.sgi.com/~cbrady_denver/memtest86/
   ^ not sure about the ~.

Cyrille.
--
home: mailto:[EMAIL PROTECTED] work: mailto:[EMAIL PROTECTED]


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world on 4.2-R breaking

[Jordan Hubbard]

> I'm convinced that it is NOT a memory or CPU problem, because
>  A) the problem is 100% repeatable

Well, it may be 100% repeatable on your machine but not on any of the
machines here.  Since we're using the same software, ostensibly,
how would YOU explain it then? :-)

- Jordan


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world dance / ports

[Evan S]

Well, I mean, ld didn't recogniz any of the libraries I had installed. In
order to get X to work I had to type 'make install' in
/usr/ports/x11/XFree86-4/work/xc

Thanks,

Evan Sarmiento ([EMAIL PROTECTED])
http://sekt7.org/es

On Mon, 13 Nov 2000, Chris Faulhaber wrote:

> On Mon, Nov 13, 2000 at 06:52:24AM -0500, Evan S wrote:
> > Hey,
> > 
> > I'm such an utter newbie, I've read the handbook, etc. I just moved from
> > Linux. a few days ago, I cvsup FreeBSD 4.2-BETA, did the make
> > buildworld/installworld/buildkernel/installkernel/mergemaster dance. I
> > rebooted. It worked, I could login.
> > 
> > But, all my ports were dead. I had to recompile every  one of them, it got
> > so tedious that I decided to reinstall 4.1.1-RELEASE from a cd I burned.
> > 
> 
> Define 'dead'.  Were there errors?
> 
> -- 
> Chris D. Faulhaber - [EMAIL PROTECTED] - [EMAIL PROTECTED]
> 
> FreeBSD: The Power To Serve   -   http://www.FreeBSD.org
> 
> 
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-stable" in the body of the message
> 



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world

[Jeffrey J. Mountin]

At 10:36 PM 10/10/00 -0400, Nader Turki wrote:
>OK guys, I understand all that, but the thing is. I already did "make 
>world" and
>got disconnected and i did telnet to the box again and did "make world" again!
>So, Is that going to be a problem?

Possibly.

Strongly suggest you start over and don't use 'make world' again.  It is 
not guaranteed to work properly and should you encounter any problems will 
be most likely be told to start over doing things the supported way.

Looking at various files in the source tree I see way too much mention of 
'make world' and there is one little mention in UPDATING:

 To rebuild everything
 -
 make world

 Except when it doesn't work :-)


Not much of an implied warning.

That starts at line 139.  Around line 192 is what you want to follow.  Not 
sure why the order is such, but think imp only listens to diffs.  Think 
I'll look over things a bit and get back to that.


Jeff Mountin - [EMAIL PROTECTED]
Systems/Network Administrator
FreeBSD - the power to serve



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world

[Chris BeHanna]

On Wed, 11 Oct 2000 [EMAIL PROTECTED] wrote:

> > "Nader Turki" <[EMAIL PROTECTED]> wrote:
> > > Sorry I didn't explain the problem.  I'm doing make world
> > > remotely using telnet.  I got disconnected from the Network, had
> > > to wait for a while till my cable modem work and i telnet again
> > > to my server and did make world again 'cause it was stoped.  Is
> > > that going to be a problem? I mean should I do somethin' else
> > > them make world? 'cause the first time it was done almost half
> > > way.

IIRC, "make world" first cleans /usr/obj unless -DNOCLEAN is set,
right?  In that case, you should be alright.

>   Not forget to capture and examine the output from make.  e.g.
>   make buildworld >& world.out &

For sh, bash, and ksh users:

nohup make buildworld > buildworld.out 2>&1 &

--
Chris BeHanna
Software Engineer (at yourfit.com)
[EMAIL PROTECTED]




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world

[Chad R. Larson]

As I recall, Robert Banniza wrote:
> Rather than doing a 'make buildworld &', wouldn't you want to do a
> 'nohup make buildworld' to keep the process running even after getting
> disconnected?  I thought the & only put the process in the background
> and that the bg process would stop after being disconnected.  Maybe
> I'm misled here.

Depends on your shell.  The original Bourne shell needs the "nohup".
The C-shell inherently nohups background jobs.

-crl
--
Chad R. Larson (CRL15)   602-953-1392   Brother, can you paradigm?
[EMAIL PROTECTED] [EMAIL PROTECTED]  [EMAIL PROTECTED]   
DCF, Inc. - 14623 North 49th Place, Scottsdale, Arizona 85254-2207


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world

[Scot W. Hetzel]

From: "Robert Banniza" <[EMAIL PROTECTED]>
> Rather than doing a 'make buildworld &', wouldn't you want to do a 'nohup
> make buildworld' to keep the process running even after getting
> disconnected? I thought the & only put the process in the background and
> that the bg process would stop after being disconnected. Maybe I'm misled
> here.
>
The process will continue running until it has completed.

Scot




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world

[Jonel Rienton]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

heh, to answer your question, i would suggest just removing whatever
is in your /usr/obj/* and start from the beginning.
and my 2 cents, use screen if that helps :)

Jonel Rienton
http://qmail.freebsduser.org
sent by qmail-1.03 on a FreeBSD 4.1-STABLE
- - Original Message - 
From: "Nader Turki" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Cc: "Scot W. Hetzel" <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>
Sent: Tuesday, October 10, 2000 9:36 PM
Subject: Re: make world


| OK guys, I understand all that, but the thing is. I already did
| "make world" and got disconnected and i did telnet to the box again
| and did "make world" again! So, Is that going to be a problem?
| 
| Thanks,
| 
| - Nader
| 
| Robert Banniza wrote:
| 
| > Rather than doing a 'make buildworld &', wouldn't you want to do
| > a 'nohup make buildworld' to keep the process running even after
| > getting disconnected? I thought the & only put the process in the
| > background and that the bg process would stop after being
| > disconnected. Maybe I'm misled here.
| >
| > Robert
| >
| > -Original Message-
| > From: [EMAIL PROTECTED]
| > [mailto:[EMAIL PROTECTED]]On Behalf Of Scot W.
| > Hetzel Sent: Tuesday, October 10, 2000 9:20 PM
| > To: Nader Turki
| > Cc: [EMAIL PROTECTED]
| > Subject: Re: make world
| >
| > From: "Nader Turki" <[EMAIL PROTECTED]>
| > > Sorry I didn't explain the problem.
| > > I'm doing make world remotely using telnet.
| > > I got disconnected from the Network, had to wait for a while
| > > till my cable modem work and i telnet again to my server and
| > > did make world again 'cause it was stoped.
| > > Is that going to be a problem? I mean should I do somethin'
| > > else them make world? 'cause the first time it was done almost
| > > half way.
| > >
| > In stead of doing a "make world", you should do a "make
| > buildworld &".  This will start building your sources in the
| > background,
| > then if you get disconnected the "make buildworld" will continue
| > in the background.  Then all you need to do is "make
| > build|installkernel" and "make installworld" after you
| > reconnected and verified that the build process has completed.
| >
| > Scot
| >
| > To Unsubscribe: send mail to [EMAIL PROTECTED]
| > with "unsubscribe freebsd-stable" in the body of the message
| 
| 
| 
| To Unsubscribe: send mail to [EMAIL PROTECTED]
| with "unsubscribe freebsd-stable" in the body of the message

-BEGIN PGP SIGNATURE-
Version: PGPfreeware 6.5.8 for non-commercial use <http://www.pgp.com>

iQA/AwUBOePVrqQ0pAI9Fl/WEQI8bACfb59b3buUUY6OnmAk6fFGzO1cVS0AoPWD
A9ba0Ddzr20oHeUs6pTFjgtN
=g/HJ
-END PGP SIGNATURE-




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: Make world is dying...

[Rodney W. Grimes]

[Charset iso-8859-1 unsupported, filtering to ASCII...]
> All-
> 
> I think I found the problem.  I have used the memory before, it was known
> good.  I swapped it out, and had the same problem.  I tried a second,
> identical, CPU with a new heat sink, and still had the same problem.  I
> received a thought-provoking email from one kind FreeBSD user that pointed
> me to the CPU I was using:  The Cyrix P-166 itself, was the problem.  The
> email is reproduced below:
> 
>   If it's occurring in the same place consistently, it's not a RAM failure;
> I used to have a Cx-6x86L-PR166 (ie. Pre-MMX Cx) and it died at roughly the
> same place in a makeworld consistently; at the time I figured that I'd
> cooked the CPU in it's lifetime (was prone to overheating), ditched it in
> favour of a Pentium-100 and it all worked well no changes made other
> than the CPU. That RAM is still doing sterling service in a dual PPro system
> now.

Our experience (which is somewhat more than an end user) as a computer
repair facility with the Cyrix PRxxx line of CPU chips is that they seem
to develope strange problems over time.  Under FreeBSD it is usually signal
11's, 10's and 6's (often in that order of frequency), under Windblows 9X
it is usually fatal exception 0xE's and or lockups.  We often find that
by slowing the CPU down one step (ie run a PR166 as a PR150 or PR133) the
problem goes away.  I suspect that the the chip has a minor internal
thermal problem that leads to electromigration related timing problems.

>   If that's still the case regardless, then it's either a Cyrix CPU bug or a
> code generation bug in GCC that affects them specifically.
> 
> This turned out to be exactly my case.  I installed a spare Pentium-100 I
> have, and buildworld is working flawlessly, so far.  Evidently, there's a
> CPU-specific bug in "cc1" that causes it to crash on this particular Cyrix
> processor.  I do not know enough about the differences between the
> older-generation Cyrix processors and the Pentium processors of the same
> vintage to draw any conclusions, but I would be learn more about this
> phenomenon.

Try putting your PR166 back in, set up to run as a PR133 and see if
the problem is still there if you wish to confirm my hypothsis of the
cyrix problem.

> Paul A. Howes
> [EMAIL PROTECTED]

-- 
Rod Grimes - KD7CAX @ CN85sl - (RWG25)   [EMAIL PROTECTED]


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world fails

[Chris BeHanna]

Doug Barton wrote:

> Signals 10 and 11 are almost always hardware.

Not in my experience.  Signal 11 is usually the result of attempting to
dereference a NULL pointer.  Signal 10 is usually the result of attempting to
dereference a pointer that contains a garbage address.

At least, that's been my experience.  YMMV.

Regards,
Chris BeHanna
[EMAIL PROTECTED]



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world fail (at) (fwd)

[Rasmus Skaarup]

On Wed, 26 Jul 2000, Marcel Moolenaar wrote:

> Rasmus Skaarup wrote:
> > 
> > I did a fresh cvsup in a empty /usr/src directory, and got the exact same
> > error:
> 
> I did not follow this thread, but I assume you also had an empty
> /usr/obj?

No. I removed the obj directory, and tried again and now it's working. I
figured that the exsting obj-files would be replaced, but apparently not.

> > ** snip snip **
> > ===> usr.bin/apply
> > rm -f apply apply.o apply.1.gz apply.1.cat.gz
> > rm -f .depend /usr/src/usr.bin/apply/GPATH /usr/src/usr.bin/apply/GRTAGS
> > /usr/src/usr.bin/apply/GSYMS /usr/src/usr.bin/apply/GTAGS
> > ===> usr.bin/at
> > ".depend", line 3: Need an operator
> > ".depend", line 4: Need an operator
> > ".depend", line 5: Need an operator
> > ".depend", line 6: Need an operator
> > ".depend", line 7: Need an operator
> > ".depend", line 10: Need an operator
> > make: fatal errors encountered -- cannot continue
> > *** Error code 1
> 
> Can you mail what's on these lines?

Somehow a lot of the .depend files (not only at, also banner.. etc) had
binary content (!?). 

Well, it's working now.

Sorry for the inconvience.


Best regards
Rasmus



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world fail (at) (fwd)

[Rasmus Skaarup]

Hi,

I did a fresh cvsup in a empty /usr/src directory, and got the exact same
error:

** snip snip **
===> usr.bin/apply
rm -f apply apply.o apply.1.gz apply.1.cat.gz
rm -f .depend /usr/src/usr.bin/apply/GPATH /usr/src/usr.bin/apply/GRTAGS
/usr/src/usr.bin/apply/GSYMS /usr/src/usr.bin/apply/GTAGS
===> usr.bin/at
".depend", line 3: Need an operator
".depend", line 4: Need an operator
".depend", line 5: Need an operator
".depend", line 6: Need an operator
".depend", line 7: Need an operator
".depend", line 10: Need an operator
make: fatal errors encountered -- cannot continue
*** Error code 1

Stop in /usr/src/usr.bin.
*** Error code 1

** snip snip **


Best regards
Rasmus

-- Forwarded message --
Date: Wed, 26 Jul 2000 08:54:14 +0200 (CEST)
From: Rasmus Skaarup <[EMAIL PROTECTED]>
To: Kris Kennaway <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Subject: Re: make world fail (at)

On Tue, 25 Jul 2000, Kris Kennaway wrote:

> On Tue, 25 Jul 2000, Rasmus Skaarup wrote:
> 
> > ** snip snip **
> > ===> usr.bin/at
> > ".depend", line 3: Need an operator
> 
> You have something wrong with your usr.bin/at/Makefile, it looks
> like. Check you haven't locally modified the file - if you're using CVS to
> update, you may have a CVS conflict in this file (look for <<< and >>>
> markers)

I haven't changed a bit. Is there anything wrong with the file below?

** usr.bin/at/Makefile **
# $FreeBSD: src/usr.bin/at/Makefile,v 1.10 1999/08/28 00:59:08 peter Exp $

.include "${.CURDIR}/Makefile.inc"

PROG=   at
CFLAGS+=-Wall
SRCS=   at.c panic.c parsetime.c perm.c
LINKS=  ${BINDIR}/at ${BINDIR}/atq \
${BINDIR}/at ${BINDIR}/atrm \
${BINDIR}/at ${BINDIR}/batch
MLINKS= at.1 batch.1 \
at.1 atq.1 \
at.1 atrm.1

BINMODE= 4555
MANSRC= .
CLEANFILES += ${MAN1}
MANDEPEND = ${MAN1}

.include 

${MAN1}: at.man
@${ECHO} Making ${.TARGET:T} from ${.ALLSRC:T}; \
sed -e \
"s@_ATSPOOL_DIR@$(ATSPOOL_DIR)@g; \
s@_ATJOB_DIR@$(ATJOB_DIR)@g; \
s@_DEFAULT_BATCH_QUEUE@$(DEFAULT_BATCH_QUEUE)@g; \
s@_DEFAULT_AT_QUEUE@$(DEFAULT_AT_QUEUE)@g; \
s@_LOADAVG_MX@$(LOADAVG_MX)@g; \
s@_PERM_PATH@$(PERM_PATH)@g; \
s@_LOCKFILE@$(LOCKFILE)@g" \
< ${.OODATE} > ${.TARGET}
** snip snip **


Best regards
Rasmus




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world of 4.1-RC

[Eric Ogren]

 
You're running Berkeley make, not GNU make, correct?


Eric

On Fri, Jul 21, 2000 at 08:46:57PM +0200, [EMAIL PROTECTED] wrote:
> Hi,
> 
> I get an error when I do a make world on my 4.1-RC.
> I CVSupped the sourcetree lots of times, redownloaded the sourcetree
> of 4.1-RELEASE, CVSupped it again, but every make world gives out
> the same error:
> 
> [...]
> ".depend", line 396: Need an operator
> ".depend", line 397: Missing dependency operator
> [...]
> *** Error 1 /usr/src/usr.sbin
> [...]
> 
> 
> How can I fix that?
> I would be very happy about suggestions how to fix that...
> 
> 
> Thanks,
>   Freddy
> 
> 
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-stable" in the body of the message
> 


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world fails : file exists

[Dan Langille]

On 5 Jul 2000, at 12:28, Kent Stewart wrote:

> 
> 
> Dan Langille wrote:
> > 
> > On 5 Jul 2000, at 12:17, Kent Stewart wrote:
> > 
> > > Dan Langille wrote:
> > > >
> > > > In my continuing 4-day saga to get 4.0-stable compiled, I encountered
> > > > the following error during make world:
> > > >
> > > > cc -O -pipe -I/usr/obj/usr/src/gnu/usr.bin/perl/libperl -
> > > > I/usr/src/gnu/usr.bin/perl/libperl/../../../../contrib/perl5 -
> > > > I/usr/obj/usr/src/i386/usr/include -c
> > > > /usr/src/gnu/usr.bin/perl/libperl/../../../../contrib/perl5/toke.c -o
> > > > toke.o mkdir: build: File exists *** Error code 1
> > > >
> > > > Stop in /usr/src/gnu/usr.bin/perl/libperl.
> > > > *** Error code 1
> > > >
> > > > Stop in /usr/src.
> > > > *** Error code 1
> > > >
> > > > Stop in /usr/src.
> > > > *** Error code 1
> > > >
> > > > Stop in /usr/src.
> > >
> > > Hi Dan,
> > >
> > > Did you remove the files from /usr/obj/*?
> > >
> > > I've done 3 cvsups and builds in the last two days on 3 different
> > > machines and haven't had any problems.
> > 
> > Yep.  I did this:
> > 
> > chflags -R noschg /usr/obj/;
> > rm -rf /usr/obj/*;
> > rm -rf /usr/tmp/*;
> 
> That takes care of original files with permission problems. I don't do make
> world's. I read where that was a good way to ruin your universe and started
> doing buildworlds, creating and installing the kernel, rebooting to single
> user mode to do the installworld. The way it was explained to me was that
> left you with your original world via /kernel.old in case something
> happened.
> 
> Since I'm not having any problems and don't have any ideas at this
> point, are your just updating your 4.0-Stable or are you moving from
> 4.0R to 4.0S?

My third attempt to go from 4.0R to 4.0S.
--
Dan Langille - DVL Software Limited
FreshPorts - http://freshports.org/ - the place for ports
well, it was the place for ports.  But because of a stuff
up, it's http://freshports.freebsddiary.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world fails : file exists

[Kent Stewart]

Dan Langille wrote:
> 
> On 5 Jul 2000, at 12:17, Kent Stewart wrote:
> 
> > Dan Langille wrote:
> > >
> > > In my continuing 4-day saga to get 4.0-stable compiled, I encountered the
> > > following error during make world:
> > >
> > > cc -O -pipe -I/usr/obj/usr/src/gnu/usr.bin/perl/libperl -
> > > I/usr/src/gnu/usr.bin/perl/libperl/../../../../contrib/perl5 -
> > > I/usr/obj/usr/src/i386/usr/include -c
> > > /usr/src/gnu/usr.bin/perl/libperl/../../../../contrib/perl5/toke.c -o
> > > toke.o mkdir: build: File exists *** Error code 1
> > >
> > > Stop in /usr/src/gnu/usr.bin/perl/libperl.
> > > *** Error code 1
> > >
> > > Stop in /usr/src.
> > > *** Error code 1
> > >
> > > Stop in /usr/src.
> > > *** Error code 1
> > >
> > > Stop in /usr/src.
> >
> > Hi Dan,
> >
> > Did you remove the files from /usr/obj/*?
> >
> > I've done 3 cvsups and builds in the last two days on 3 different
> > machines and haven't had any problems.
> 
> Yep.  I did this:
> 
> chflags -R noschg /usr/obj/;
> rm -rf /usr/obj/*;
> rm -rf /usr/tmp/*;

That takes care of original files with permission problems. I don't do
make world's. I read where that was a good way to ruin your universe
and started doing buildworlds, creating and installing the kernel,
rebooting to single user mode to do the installworld. The way it was
explained to me was that left you with your original world via
/kernel.old in case something happened.

Since I'm not having any problems and don't have any ideas at this
point, are your just updating your 4.0-Stable or are you moving from
4.0R to 4.0S?

Kent

-- 
Kent Stewart
Richland, WA

mailto:[EMAIL PROTECTED]
http://www.3-cities.com/~kstewart/index.html
FreeBSD News http://daily.daemonnews.org/

Bomber dropping fire retardant in front of Hanford Wild fire.
http://www.3-cities.com/~kstewart/bomber.jpg


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world fails : file exists

[Dan Langille]

On 5 Jul 2000, at 12:17, Kent Stewart wrote:

> Dan Langille wrote:
> > 
> > In my continuing 4-day saga to get 4.0-stable compiled, I encountered the
> > following error during make world:
> > 
> > cc -O -pipe -I/usr/obj/usr/src/gnu/usr.bin/perl/libperl -
> > I/usr/src/gnu/usr.bin/perl/libperl/../../../../contrib/perl5 -
> > I/usr/obj/usr/src/i386/usr/include -c
> > /usr/src/gnu/usr.bin/perl/libperl/../../../../contrib/perl5/toke.c -o
> > toke.o mkdir: build: File exists *** Error code 1
> > 
> > Stop in /usr/src/gnu/usr.bin/perl/libperl.
> > *** Error code 1
> > 
> > Stop in /usr/src.
> > *** Error code 1
> > 
> > Stop in /usr/src.
> > *** Error code 1
> > 
> > Stop in /usr/src.
> 
> Hi Dan,
> 
> Did you remove the files from /usr/obj/*?
> 
> I've done 3 cvsups and builds in the last two days on 3 different
> machines and haven't had any problems.

Yep.  I did this:

chflags -R noschg /usr/obj/;
rm -rf /usr/obj/*;
rm -rf /usr/tmp/*;
--
Dan Langille - DVL Software Limited
FreshPorts - http://freshports.org/ - the place for ports
well, it was the place for ports.  But because of a stuff
up, it's http://freshports.freebsddiary.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world fails : file exists

[Kent Stewart]

Dan Langille wrote:
> 
> In my continuing 4-day saga to get 4.0-stable compiled, I encountered
> the following error during make world:
> 
> cc -O -pipe -I/usr/obj/usr/src/gnu/usr.bin/perl/libperl -
> I/usr/src/gnu/usr.bin/perl/libperl/../../../../contrib/perl5 -
> I/usr/obj/usr/src/i386/usr/include -c
> /usr/src/gnu/usr.bin/perl/libperl/../../../../contrib/perl5/toke.c -o toke.o
> mkdir: build: File exists
> *** Error code 1
> 
> Stop in /usr/src/gnu/usr.bin/perl/libperl.
> *** Error code 1
> 
> Stop in /usr/src.
> *** Error code 1
> 
> Stop in /usr/src.
> *** Error code 1
> 
> Stop in /usr/src.

Hi Dan,

Did you remove the files from /usr/obj/*?

I've done 3 cvsups and builds in the last two days on 3 different
machines and haven't had any problems.

Kent

-- 
Kent Stewart
Richland, WA

mailto:[EMAIL PROTECTED]
http://www.3-cities.com/~kstewart/index.html
FreeBSD News http://daily.daemonnews.org/

Bomber dropping fire retardant in front of Hanford Wild fire.
http://www.3-cities.com/~kstewart/bomber.jpg


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: Make world fails on latest 2.2.8...

[Kris Kennaway]

On Sun, 21 May 2000, Warner Losh wrote:

> : I think this was an overenthusiastic commit by Joe Karthauser (strlcpy()
> : doesnt exist in 2.x)
> 
> Has this been reverted yet?

No, it's only been a few hours since it was discovered. Give Joe some time
to correct the mistake..

Kris


In God we Trust -- all others must submit an X.509 certificate.
-- Charles Forsythe <[EMAIL PROTECTED]>



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: Make world fails on latest 2.2.8...

[Warner Losh]

In message <[EMAIL PROTECTED]> Kris 
Kennaway writes:
: On Sun, 21 May 100, Andrew Wilson wrote:
: 
: > sorry if this seems like a trip back in time.  Today I cvsup'd
: > 2.2.8-stable and ran make world.  It croaked while compiling bin/ed:
: 
: I think this was an overenthusiastic commit by Joe Karthauser (strlcpy()
: doesnt exist in 2.x)

Has this been reverted yet?

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Jukka Simila]

On 05-Apr-00 David Murphy wrote:
> Quoting <[EMAIL PROTECTED]>
> by Jukka Simila <[EMAIL PROTECTED]>:
> 
>> that showed me a list of files, and there was BIG LETTERS: UPDATING
>> I thought, well, I'm now running 3.4, and I'm about to install 4.0,
>> so I'd guess that's called UPDATING. Some common sense, I think? I
>> read the UPDATING and followed those very simple instructions there
>> (well I have to admit I couldn't build a custom kernel before I had
>> updated the system with GENERIC, but that really isn't a big
>> problem)
> 
> Where in UPDATING did you find the update instructions? I did the same
> as you, except when I came to read UPDATING, it looked like a
> changelog, so i stopped reading after a 5 pages or so - little did I
> know at the time that the critical instructions were at the end of the
> file. Shortly after, those paragraphs were moved to the top of
> UPDATING, which is a vast improvement, and greatly improves the
> chances of basically clueful people finding the instructions. It's
> this level of improvement that the docs need most - moving/putting
> pointers to pre-existing text to places where the basically clueful
> will look. 

apparently i was lucky: 
info was at the top of the file, so the file had already been updated that way
before i got it.

@-`--,--`--- 
Jukka Simila


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Brad Knowles]

At 10:42 AM -0600 2000/4/4, Warner Losh wrote:

>  Yes.  Also, for -current's UPDATING I specifically keep things short
>  and to the point, lest I encourage people to follow -current who don't
>  have the skills, time or patience that is required of -current.

Unfortunately, this may come back to bite you when -CURRENT 
becomes -STABLE.

Funny, it's always the edge conditions that tend to cause the 
most problems.  ;-)

--
   These are my opinions -- not to be taken as official Skynet policy
==
Brad Knowles, <[EMAIL PROTECTED]>|| Belgacom Skynet SA/NV
Systems Architect, Mail/News/FTP/Proxy Admin || Rue Colonel Bourg, 124
Phone/Fax: +32-2-706.13.11/12.49 || B-1140 Brussels
http://www.skynet.be || Belgium


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Warner Losh]

In message <[EMAIL PROTECTED]> Nik Clayton writes:
: If you (or anyone else reading this) wants to step forward and maintain
: either of these files, please yell now.  For example, /usr/src/UPDATING
: is currently empty in 3-stable because no one's volunteered to maintain
: it.

I have some entries in my mailbox for the 3.x version of UPDATING.
I'll see about getting them committed.

: However, at certain times, in *both* trees, you might need to jump through
: more hoops than that.  The hoops differ depending on where you're coming
: from, and where you're going to.  UPDATING aims to list all those hoops
: (and this is dynamic information, which makes it less suitable for the
: Handbook).

Yes.  Also, for -current's UPDATING I specifically keep things short
and to the point, lest I encourage people to follow -current who don't
have the skills, time or patience that is required of -current.

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Warner Losh]

: > It is time to put in the Propellerheads and chill with a some good tunes.

i've used the following image
http://www.village.org/~imp/propeller.gif
on my website for years.  The picture was taken in 92 using the then
newfangled frame grabber attached to a Solbourne S4000 via some funky
sbus card.

Now, where did I set down my Stout.  Or was I drinking Porter?

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[James Housley]

Brennan W Stehling wrote:
> 
> It looks like that would have helped me a great deal.  Thanks for making
> that change.  I am sure it will help someone else down the road.
> 
> http://www.freebsd.org/handbook/makeworld.html
> 
Another question.  It mentions that in -CURRENT you can use -j4.  Is
that -CURRENT 5.x or the now -STABLE 4.x and -CURRENT 5.x ?

Jim
-- 
The wise man built his network upon U*nx.
The foolish man built his network upon Windows.


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[James Housley]

Brennan W Stehling wrote:
> 
> It looks like that would have helped me a great deal.  Thanks for making
> that change.  I am sure it will help someone else down the road.
> 
> http://www.freebsd.org/handbook/makeworld.html
> 
Nice page.  But 

# cd /usr/obj
# rm -rf *
# chflags -R noschg *
# rm -rf *

Has been mentioned to be faster on most systems, yes it produces error
messages about not being able to remove some files.

Jim
-- 
The wise man built his network upon U*nx.
The foolish man built his network upon Windows.


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Brennan W Stehling]

It looks like that would have helped me a great deal.  Thanks for making
that change.  I am sure it will help someone else down the road.

http://www.freebsd.org/handbook/makeworld.html

Brennan Stehling - web developer and sys admin
projects: www.onmilwaukee.com | www.sncalumni.com

fortune:
Never underestimate the bandwidth of a station wagon full of tapes.
-- Dr. Warren Jackson, Director, UTCS

On Tue, 4 Apr 2000, Michel TALON wrote:

> I have upgraded from source yesterday to 4.0-Stable following the nice
> instructions at the beginning of UPDATING and everything turned out fine.
> Congratulations to the developpers and particularly to W. Losh for its
> addenda to UPGRADING. However, i am reading the stable mailing list and
> was aware that several people had screwed their machine doing this upgrade.
> Two weeks ago, the instructions in UPGRADING were everything except clear.
> So, as is often the case one needs to wait a little bit that several people
> run into trouble before plunging in the game. This is unfortunate, because
> it means that ordinary users (not following CURRENT) cannot test the
> candidate release as soon as possible, thereby detecting unusual bugs without
> risking very much. Had the correct and detailed explanations been written
> a month ago, it would have been very possible.
> 
> -- 
> 
> Michel TALON
> 
> 
> 
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-stable" in the body of the message
> 



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Eric Ogren]

Hi-

IMO this thread has gone on long enough. However, since I'm hypocritical,
I'm going to post one (and hopefully only one) message regarding it.

> > 1) The responsibility for locating available documentation
> > rests with the user.

 Who else should the responsibility fall on? Blindly doing things on
ANY operating system is a bad idea. Just because WinNT/Win2000
has a more user-friendly administration interface doesn't mean that
your average person should just start screwing around in the control panel.
  Any user should read the documentation, especially with anything
related to system maintenance. I know that people generally don't like
to do this, but it's still necessary: we try to make software
as user-friendly as possible, but computers, and specifically OSs are
still complicated.

 and, just a kind of general comment on the documentation:

 The RELNOTES, ERRATA, and README files AFAIK have always
been mostly associated with a binary release. If you were simply doing a
binary
upgrade (which I believe we reccommend for those who don't follow
the -stable/-current
lists), you wouldn't have to know that you can't simply do a "cd /usr/src &&
make world"
to upgrade between major version #s.
 Personally, I feel that, given as probably 90% of FreeBSD users buy/burn a
CD or
simply download FreeBSD from the net, cluttering up the
relnotes/errata/readme files
with instructions on how to upgrade from source would just needlessly a)
confuse the
majority of users and b) put unnecessary information into these files.

 As has been pointed out to you, the handbook does say that those who wish
to use -stable / -current (ie upgrade via source) MUST read the -stable
and -current
mailing lists. I don't know how else we can make it more clear that
upgrading from
source is not advised for newbies or people that are not willing to put in
the time
to read [or at least skim] the mailing list.

 The one big problem that I see is that the makeworld.html file does not yet
have
any caveats regarding the 3.x -> 4.0 upgrade. This should definitely be
fixed, and
I believe Nik Clayton (doc project manager) is/has prepared patches to fix
this.

 How else would you suggest that this is made more clear?

 I'm not trying to flame you; I'm trying to get your opinion. I've been
using FreeBSD
for 3 or 4 years now, and I've been thinking of trying to get more involved
in some way.
Maybe writing some documentation to clarify the upgrade process could be a
way
to get started. Since it seems you don't feel that users should be forced to
submit
their own docs, could you at least tell a [prospective] doc-guy where else
he should put pointers in?


thanks,
Eric



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Walter Brameld]

On Mon, 03 Apr 2000, in a never-ending search for enlightenment, David Murphy wrote:
> Quoting <[EMAIL PROTECTED]>
> by Doug Barton <[EMAIL PROTECTED]>:
> 
> > As I said, the documentation project can always use fresh
> > blood. I suggest you subscribe to [EMAIL PROTECTED] so you can get an
> > idea of what's happening currently. Or, you can just use this whole
> > pointless exercise to promote your web site. Either is fine with me,
> > as long as you spend more time making constructive contributions
> > instead of sending pointless e-mails to the list.
> 
> I've been on the receiving end of a previous iteration of this
> flamewar, and I don't personally see that your position is any more
> constructive.
> 
> At the end of the day, if someone wants to write documentation, they
> will. If they don't, they won't. In your model, the person who needs
> documentation is assigned the task of writing documentation.
> 
> There seem to be two widely held opinions on this list, and I assume,
> perhaps incorrectly, that they are held by the majority of FreeBSD
> developers:
> 
>   1) The responsibility for locating available documentation
> rests with the user.
> 
>   2) If the user finds the available documentation insufficient,
> the responsibility for creating sufficient documentation rests with
> the user.
> 
> It seems to me that the difference between users and developers is
> that developers generally find the above opinions reasonable, users
> generally do not.
> 
> There are, broadly speaking, two classes of users of any software
> system: those who use the system, find problems with the system, and
> fix those problems; and those who use the system, find problems with
> the system, and report those problems. I think of the former group of
> people as "developers", and I call the latter group of people "users".
> 
> The bottom line seems to be that, while developers are greatly
> desired, users are tolerated, to the extent that they don't get in the
> way of development.
> 
> This makes FreeBSD an excellent choice of system for developers. It
> makes it a poor choice of system for users.
> 
> This, in and of itself, is no problem if you are interested in
> creating and using a system by developers, for developers. Good for
> you, and much success.
> 
> Just don't promote it to users.

Are you saying that FreeBSD is just a toy for developers, and not meant
to be actually USED by any one?

-- 
Walter Brameld

Microsoft: Where do you want to go today?
Linux: Where do you want to go tomorrow?
BSD:   Are you guys coming, or what?
Walter:And what does THIS button do??



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Bryan Bursey]

> I'd be glad to help in this area, as I'm a very good writer.  Can someone
  

Gotta love the modesty...  ;)  I wouldn't mind getting more involved in
this project as well (as time permits).  Maybe we should continue this
type of discussion on -doc?

Cheers,
Bryan



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Nik Clayton]

On Mon, Apr 03, 2000 at 03:55:59PM -0700, R Joseph Wright wrote:
> Sorry to jump in in the middle of this thread, but I must agree with
> Brennan that FreeBSD's online documentation is woefully incomplete.  A
> person upgrading from 3.x to 4.0-release shouldn't have to subscribe to a
> mailing list.

Indeed, and they don't.  You could download 4.0 and install it over your
existing 3.0 installation, or use /stand/sysinstall's "Upgrade" option.

It only becomes tricky when you want to upgrade using "make world".

Trying to bootstrap one system from another is a *hard* problem.  You
should not expect to be able to do it by reading three lines and running
one command.

> I'd like to see the FreeBSD handbook achieve that level of quality and
> detail.  At that point, I think the flames people get for not reading the
> docs would make more sense.
>
> I'd be glad to help in this area, as I'm a very good writer.  Can someone
> point me in the right direction?   

[EMAIL PROTECTED]
http://www.freebsd.org/docproj/
http://www.freebsd.org/tutorials/docproj-primer/

N
-- 
Internet connection, $19.95 a month.  Computer, $799.95.  Modem, $149.95.
Telephone line, $24.95 a month.  Software, free.  USENET transmission,
hundreds if not thousands of dollars.  Thinking before posting, priceless.
Somethings in life you can't buy.  For everything else, there's MasterCard.
  -- Graham Reed, in the Scary Devil Monastery


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Nik Clayton]

On Mon, Apr 03, 2000 at 09:26:27PM +0200, Brad Knowles wrote:
> At 11:54 AM -0700 2000/4/3, Doug Barton wrote:
> > I realize that you don't want to accept responsibility for your
> >  actions, but please stop posting hear trying to convince us that there is
> >  some way we could have unloaded the gun before you pointed it at your
> >  foot.
> 
>   I know you don't want to hear this, but the reality of it is that 
> we really should be doing a better job of keeping the web pages 
> up-to-date with regards to things like this.

It's not so much that we don't want to hear it, more that we'd rather 
hear

This page isn't quite right.  Please review this patch, which adds
all the necessary information, and then commit it.

If you do that two or three times, preferably as PRs, and your patches
apply cleanly each time, and are accurate, I'll be banging on -core's 
door to get you added as a committer very shortly afterwards.

Seriously folks, this is one of the simplest ways to get one of those 
coveted @FreeBSD.org e-mail addresses.  Just write good quality 
documentation, or submit updates to the existing documentation, and keep
submitting it.

13 of the 55 or so committers added over the past 14 months or so have
been Documentation Project committers.

>   In particular, when there is a major change coming and we know 
> that the necessary information is to be found in the appropriate 
> UPDATING file (or wherever), then we should change the web pages to 
> reflect the fact and point to the appropriate specific file.

Patch patch patch. . .

I'm about to commit a change to the Handbook which tells the user to 
read the UPDATING file first.

N
-- 
Internet connection, $19.95 a month.  Computer, $799.95.  Modem, $149.95.
Telephone line, $24.95 a month.  Software, free.  USENET transmission,
hundreds if not thousands of dollars.  Thinking before posting, priceless.
Somethings in life you can't buy.  For everything else, there's MasterCard.
  -- Graham Reed, in the Scary Devil Monastery


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Walter Brameld]

On Mon, 03 Apr 2000, you wrote:
> Thanks for getting that updated.  I am sure it will help many people.
> 
> As for improving documentation I may just make a bsd search site.  I think
> I can configure htdig to search multiple sites and hope to point them to
> several bsd sites which carry articles and documentation.  I know that
> would help me a great deal.
> 
> I purchased the domain the other day, greasydaemon.com.  The focus of the
> site is meant to be bsd hardware, but I can add the search page for bsd
> sites onto it.  Perhaps I will post instructions on how to get htdig
> configured for anyone who hopes to create their own search page.  I am not
> sure how much traffic my server can handle. (not millions a day)
> 
> Brennan Stehling - web developer and sys admin
> projects: www.onmilwaukee.com | www.sncalumni.com

I for one would appreciate that information. I was not even aware of
htdig, but am now in the process of installing it.

Sorry about all the flames you've been getting. These lists do seem to
be a bit more hostile lately, not sure what the reason might be. I've
been prompted to respond to a few of those idiots about their
behavior. I see no reason for it. One should either provide help, or
keep his/her mouth shut.

-- 
Walter Brameld

Microsoft: Where do you want to go today?
Linux: Where do you want to go tomorrow?
BSD:   Are you guys coming, or what?
Walter:And what does THIS button do??



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Nik Clayton]

On Mon, Apr 03, 2000 at 04:42:34PM -0500, Brennan W Stehling wrote:
> How about having an UPDATING and an UPGRADING file?

Lack of people to update them.  Then people complain when they get out
of sync.

If you (or anyone else reading this) wants to step forward and maintain
either of these files, please yell now.  For example, /usr/src/UPDATING
is currently empty in 3-stable because no one's volunteered to maintain
it.

> One could explain a simple upgrade of a similar branch while the upgrade
> file will explain how to go from 3.x to 4.0 and above.  That would make it
> clear that there is a different method for the two situations.

There isn't really a different method.

% cd /usr/src
% make world

However, at certain times, in *both* trees, you might need to jump through
more hoops than that.  The hoops differ depending on where you're coming
from, and where you're going to.  UPDATING aims to list all those hoops
(and this is dynamic information, which makes it less suitable for the
Handbook).

> If anyone has suggestions for documenation, perhaps it should copied to
> Nick Clayton so he can take some action.  And if anyone would like to
> assist in creating the documentation.. he's the man to contact as well.

Suggestions should always be sent to the [EMAIL PROTECTED] mailing list as
well.  If I can't field them, there are people there who can -- some of
them seem to do little except commit patches that other people send in,
and that's one of the most valuable, and thankless, roles there is.

N
-- 
Internet connection, $19.95 a month.  Computer, $799.95.  Modem, $149.95.
Telephone line, $24.95 a month.  Software, free.  USENET transmission,
hundreds if not thousands of dollars.  Thinking before posting, priceless.
Somethings in life you can't buy.  For everything else, there's MasterCard.
  -- Graham Reed, in the Scary Devil Monastery


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Nik Clayton]

On Mon, Apr 03, 2000 at 03:03:33PM -0500, Brennan W Stehling wrote:
> As for improving documentation I may just make a bsd search site.  

http://www.google.com/bsd

N
-- 
Internet connection, $19.95 a month.  Computer, $799.95.  Modem, $149.95.
Telephone line, $24.95 a month.  Software, free.  USENET transmission,
hundreds if not thousands of dollars.  Thinking before posting, priceless.
Somethings in life you can't buy.  For everything else, there's MasterCard.
  -- Graham Reed, in the Scary Devil Monastery


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Doug Barton]

On Mon, 3 Apr 2000, Brennan W Stehling wrote:

> 
> The problem was that the makeworld.html page did not warn that moving from
> 3.x to 4.0 needed to be done differently. 

The actual problem was that you stuck to that web page as your
only source of information. Right at the very top of that page, in bold
letters it says that you need to subscribe to and read the -stable and/or
-current mailing lists before trying to track -stable. If you had done
that, you would have seen ample discussion about the particular problems
related to that upgrade. 

I realize that you don't want to accept responsibility for your
actions, but please stop posting hear trying to convince us that there is
some way we could have unloaded the gun before you pointed it at your
foot. 

Thanks,

Doug
-- 
"So, the cows were part of a dream that dreamed itself into
existence? Is that possible?" asked the student incredulously.
The master simply replied, "Mu."




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failed

[Jim Weeks]

Which much venom, Brennan W Stehling wrote:
 
> I am not a moron that does not read any documentation.  So let me lash
> back a bit here, maybe you should read someones full email and the rest of
> the thread before flaming them.
> 
> You only come off as a prick and discourage people from using FreeBSD.  I
> love it and have been using it for over 2 years but the flames I got
> recently when asking for a few suggestions have been discouraging.

First, I apologize if that is the way it seemed.  I only meant to point
out what has bitten me in the back side so often, and I had read the rest
of the thread.  I know how hard it can be to keep up with the mailing
list, but this exact procedure has been worn out on the mailing list ever
since the release. 

> It's like I should just stick to reading the README's and online
> documentation and avoid all contact with any mailing list or chat room.

Not at all.  In fact you will find this on the Make World page.

If you try and track -STABLE or -CURRENT and do not read the
<[EMAIL PROTECTED]> or <[EMAIL PROTECTED]> mailing lists then you are
asking for trouble. 

Which brings us back to the fact that the correct procedure has
been discussed at length on this list for the past two or three weeks.

No flame intended,

Jim



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world fails

[Brandon D. Valentine]

On Tue, 21 Mar 2000, Carl Johan Madestrand wrote:

>rpcgen -C -h -DWANT_NFS3 /usr/src/include/rpcsvc/spray.x -o spray.h
>unsigned hnt usec;

Last time I checked there was no type unsigned hnt in the standard
library.  ;-)  It should read unsigned int.  Not sure who's fault that is.

-
| -Brandon D. Valentinebandix at looksharp.net
|  bandix on EFnet IRC BVRiker on AIM
-
| "...and as for hackers, we note that all of those known to
| The Register are so strapped financially that seizing their
| property would be tantamount to squeezing blood from a
| stone." -- The Register, 02/17/2000
-



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failing because of missing unroff

[Brad Knowles]

At 1:39 PM +0200 1999/10/13, Sheldon Hearn wrote:

> The doc tree isn't the src tree at all. Just because the source tree is
> supplied with source for all the programs used to build it doesn't mean
> it should be supplied with the source for the programs needed to build
> the doc tree.

So, what's required to build the programs to support the doc/ 
tree?  If all they need is unroff, I'd suspect that all you'd need in 
the source tree is gcc/egcs, which should already be there, right?

If nothing has to be added to the source tree to support the doc 
tree, and the doc tree includes all the tools it will need to support 
itself, then we're fine and we haven't added any unnecessary 
dependancies, right?

> What's next, adding a web server to the src tree to support the www
> tree?

No, the web server should be a part of the www tree, but the 
programs to build the web server should be part of the source tree, 
at least the way I understand things.

-- 
   These are my opinions -- not to be taken as official Skynet policy
  
|o| Brad Knowles, <[EMAIL PROTECTED]>Belgacom Skynet NV/SA |o|
|o| Systems Architect, News & FTP Admin  Rue Col. Bourg, 124   |o|
|o| Phone/Fax: +32-2-706.11.11/12.49 B-1140 Brussels   |o|
|o| http://www.skynet.be Belgium   |o|
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/
  Unix is like a wigwam -- no Gates, no Windows, and an Apache inside.
   Unix is very user-friendly.  It's just picky who its friends are.


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failing because of missing unroff

[Sheldon Hearn]

On Wed, 13 Oct 1999 13:35:03 +0200, Brad Knowles wrote:

>   That's fine, but you shouldn't need anything outside of the 
> normal source tree to support those tools, right?

Wrong.

The doc tree isn't the src tree at all. Just because the source tree is
supplied with source for all the programs used to build it doesn't mean
it should be supplied with the source for the programs needed to build
the doc tree.

What's next, adding a web server to the src tree to support the www
tree?

Leave this alone now.

Ciao,
Sheldon.


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message

Re: make world failing because of missing unroff

[Brad Knowles]

At 1:28 PM +0200 1999/10/13, Sheldon Hearn wrote:

> If you want tools distributed with the sources which they support, you
> should incorporate the tools into the doc/ tree. But then you'd still
> need non-doc tools to build _those_.

That's fine, but you shouldn't need anything outside of the 
normal source tree to support those tools, right?  In other words, 
you wouldn't be dependant on yet another program from the ports 
system, right?

I'm trying to understand just precisely what depends on what, and 
to help ensure that there are no circular dependancies.  I'm also 
trying to help suggest a path whereby we don't have more important 
things (like the doc/ tree) dependant on lesser things (like the 
ports system).

-- 
   These are my opinions -- not to be taken as official Skynet policy
  
|o| Brad Knowles, <[EMAIL PROTECTED]>Belgacom Skynet NV/SA |o|
|o| Systems Architect, News & FTP Admin  Rue Col. Bourg, 124   |o|
|o| Phone/Fax: +32-2-706.11.11/12.49 B-1140 Brussels   |o|
|o| http://www.skynet.be Belgium   |o|
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/
  Unix is like a wigwam -- no Gates, no Windows, and an Apache inside.
   Unix is very user-friendly.  It's just picky who its friends are.


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message