Re: Broadcom NetXtreme II BCM5709 Gigabit Ethernet Dell 610 and Dell 710 servers

[Mel Flynn]

On Wednesday 27 May 2009 21:37:47 Wojciech Puchar wrote:
> >> Why would you need to do that?  There is a bce driver in the 7.x OS
> >> already.
> >
> > Cause it's unstable? Search these or many other archives. The short
> > answer: under IO load the card drops net link or panics kernel.
>
> is it any recipe to trigger that behaviour? i have that card (builtin) on
> loaded server with 2 months uptime and with quite lot of traffic going
> through that card.

Don't have local archives handy, but here ya go:

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

Re: Streaming server

[Mel Flynn]

On Monday 25 May 2009 15:41:04 Jos Chrispijn wrote:
> I have some short movies (a la YouTube) that I would like to show as
> video streams. Presenting them by download is messing up my bandwidth
> (...). Can someone tell me if there is a simple solution installing such a
> stream service/server into FreeBDS 7.2?

Err, this thread is very long...can we blame our snuggle Pole or did no one 
mention "convert to ogg-theora and install audio/ices2+audio/icecast2".
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Broadcom NetXtreme II BCM5709 Gigabit Ethernet Dell 610 and Dell 710 servers

[Mel Flynn]

On Tuesday 26 May 2009 02:08:38 Paul Schmehl wrote:
> --On May 25, 2009 4:02:57 PM -0700 Carlos Pardo 
>
> wrote:
> > We are back porting the bce driver from 8.0 to 7.0. We are still missing
> > some changes since cold boots work but rebooting (warm booting) fails!
> >
> > The error is:
> >
> > bce0: ../../../dev/bce/if_bce.c(1386); Unable to write CTX memory:
> > cid_addr = 0x, offset = 0x!
> >
> > files back ported:
> >
> > bce/if_bce.h
> > bce/if_bcefw.h
> > bce/if_bvereg.h
> > mii/brgphy.c
> > mii/brgphyreg.h
>
> Why would you need to do that?  There is a bce driver in the 7.x OS
> already.

Cause it's unstable? Search these or many other archives. The short answer: 
under IO load the card drops net link or panics kernel.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: How to detect when gmirror sync is complete?

[Mel Flynn]

On Tuesday 26 May 2009 22:33:54 Peter Steele wrote:
> I know I could have a script that continually checks "gmirror status" to
> detect when a gmirror synchronization has completed, but is there a more
> event-driven approach? Something that could be used to trigger and event
> like devd does for drive pulls/inserts would be nice. Is this possible?

Doesn't look like gmirror notifies devd. Look at devctl_notify examples[1] in 
the source tree to add it, where the hard part is figuring out when a gmirror 
sync is complete. Or you can file a PR requesting the feature, I'm sure you're 
not alone.

[1] Prototype: sys/bus.h, example: sys/dev/acpica/acpi.c.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Patching? Probably a trivial question, but...

[Mel Flynn]

On Wednesday 27 May 2009 17:21:42 Kurt Buff wrote:
> All,
>
> I've gotten a patch for a program in the ports tree from one of the
> authors of the program - not the port maintainer - to fix a small
> problem, but don't know how to install the updated port.
>
> I cd'ed into the
> /usr/ports/%CATEGORY%/%PROGRAM%/work/%PROGRAM-VERSION% directory, then
> performed 'patch 
> Then I did a make, but got no output.
>
> So - I'm obviously lacking clue here. Anyone have a spare set?

Don't feel like reading the entire thread atm, but for reference:
- Patches need to have relative paths, where the root of the path corresponds 
to the port's notion of $PATCH_WRKSRC
- You can find out this directory by running:
% make -C /usr/ports/category/portname -V PATCH_WRKSRC
  The default is $WRKSRC which is $WRKDIR/$DISTNAME by default.
  Example:
% make -C /usr/ports/sysutils/nagios-statd -V PATCH_WRKSRC
/stable/usr/obj/usr/ports/sysutils/nagios-statd/work/nagios-statd-3.12

- Patches are automatically applied if they reside in the port's notion of 
PATCHDIR and are named patch-*
- You can find out this directory by running:
%make -C /usr/ports/category/portname -V PATCHDIR
  The default is $.CURDIR/files.
  Example:
% make -C /usr/ports/sysutils/nagios-statd -V PATCHDIR
/usr/ports/sysutils/nagios-statd/files

- In order to apply a new patch after you have previously gone past the patch 
stage (configure, build, install), either run make clean or:
% rm $(make -C /usr/ports/category/portname -V PATCH_COOKIE)
  The above can cause problems, with the build. The normal course of action is 
to make clean.

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

Re: Formatted text conversion

[Mel Flynn]

On Wednesday 27 May 2009 17:41:56 Kelly Jones wrote:
> I have e-books in several formats (DOC, LIT, PDF, RTF, HTML, TXT,
> etc). Is there a Unix command-line tool that converts between these
> formats?
>
> If not, is there at least a tool that converts these formats to TXT?
>
> My goal is to read these books on my Kindle, even if it means losing
> some formatting/bells/whistles.

There isn't a one-for-all that I know of. Doc (if that's MS Word) and Lit are 
probably the hardest to find for command line, especially one that recognizes 
versions.
For the rest, check out print (for PDF mostly), converters and textproc 
category. There are a few available to you and best you check for yourself 
what features you need (I suppose batch processing).
grep -i rtf /usr/ports/$category/*/pkg-descr does wonders for a quick 
overview.

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

Re: How to recover disk space after "filesystem full"

[Mel Flynn]

On Friday 22 May 2009 18:19:25 Steve Bertrand wrote:

> # pkg_add -r lsof

Or use the native fstat(1).
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: synchronize time

[Mel Flynn]

On Friday 22 May 2009 06:57:59 Roy Stuivenberg wrote:
> Tnx guys,
>
> So now my /etc/ntp.conf is like :
> server ntp.xs4all.nl
>
> driftfile /var/db/ntp.drift
>
> Problem is, when I open time-admin from console (root),
> everything is greyed out, and I can't change from manual to
> time servers.
> I have added a screenshot.

That has probably to do with gnome/dbus/polkit, which I leave to other people 
to answer, as they're on stay far away from it list ;)
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Why so many ports have run-dependencies on non-system gcc versions?

[Mel Flynn]

On Friday 22 May 2009 02:00:29 Yuri wrote:
> When I tried to delete gcc-4.3.4_20090517 I got this message:
> pkg_delete: package 'gcc-4.3.4_20090517' is required by these other
> packages and may not be deinstalled:
> blas-1.0_3
> cgnslib-2.5.3_1
> fftw3-3.2
> fftw3-float-3.2_1
> fr-med-2.3.5
> getdp-1.2.1_7
> gmsh-2.3.1
> lapack-3.2.1
> libofa-0.9.3_3
> libsamplerate-0.1.7_1
> octave-3.0.5_1
> suitesparse-3.3.0

Probably only fftw3 requires gcc43 and the rest comes from needing fftw3. As 
others said, it's not just building, it will need runtime libraries and in the 
case of fftw3, fortran support.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Kernel Panic

[Mel Flynn]

On Friday 22 May 2009 05:30:42 Shawn Badger wrote:
> Hi, I installed FreeBSD 7.2 Release on a mini nettop (Intel
> Atom/945gc/ICH7 hardware), and everything seemed to go smoothly.
> However, when I boot the system and the filesystem checks have been
> going for awhile, it always ends in a panic.  Here's the dump:
>
> dev = ad4s1f, block = 1, fs = /usr
> panic: ffs_blkfree: freeing free block
> cpuid: 1
> uptime: 15m47s
> Physical memory: 2027 MB
> Dumping 180 MB:
>
> Fatal Trap 12: page fault while in kernel mode
> ...
>
> I am using the default filesystem.  Does anyone know what might cause
> this, and how I can fix it?

This will likely go away after booting into single user and running fsck -y. 
See the archives for various discussions about the problems with 
background_fsck.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: pkgdb -F problem

[Mel Flynn]

On Thursday 21 May 2009 18:18:16 Leslie Jensen wrote:
> I've just updated my 7.1-RELEASE to 7.2-RELEASE using freebsd-update.
>
> Everything went ok but I've got a problem when I do
>
> pkgdb -F
> /libexec/ld-elf.so.1: /usr/lib/libcrypt.so.4: unsupported file layout

What does file /usr/libl/libcrypt.so.4 tell us?
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: jail's adjkerntz

[Mel Flynn]

On Wednesday 20 May 2009 22:56:26 alexus wrote:
> On Wed, May 20, 2009 at 5:31 AM, Herbert J. Skuhra  
wrote:
> > 2009/5/20 alexus :
> >> inside of my jail i get following emails...
> >>
> >> adjkerntz[25058]: sysctl(set: "machdep.adjkerntz"): Operation not
> >> permitted
> >>
> >> i dont remember getting these before...
> >> i did changed time zone recently though...
> >
> > Hi!
> >
> > You can disable adjkerntz in /etc/crontab:
> >
> > #1,31   0-5 *   *   *   rootadjkerntz -a
> >
> > And then run '/etc/rc.d/cron restart'.
> >
> > - Herbert
> > ___
> > freebsd-questions@freebsd.org mailing list
> > http://lists.freebsd.org/mailman/listinfo/freebsd-questions
> > To unsubscribe, send any mail to
> > "freebsd-questions-unsubscr...@freebsd.org"
>
> would it brake something?
> i didn't find it in manual, if this is normal shouldn't it be cover there?

There are various things in cron and periodic that don't make sense to run in 
a jail, because a jail cannot modify kernel time and read various /dev 
devices. I have this line commented out in my jails and nothing breaks, just 
less annoying emails.

You might find this list useful as well:
# cat /data/jails/tpl/RELENG_7/etc/periodic.conf
daily_clean_rwho_enable="NO"
daily_accounting_enable="NO"
daily_status_disks_enable="NO"
daily_status_rwho_enable="NO"
daily_status_security_chksetuid_enable="NO"
daily_status_security_chkmounts_enable="NO"
daily_status_security_ipfwdenied_enable="NO"
daily_status_security_ipfdenied_enable="NO"
daily_status_security_pfdenied_enable="NO"
daily_status_security_ipf6denied_enable="NO"
daily_status_security_kernelmsg_enable="NO"
monthly_accounting_enable="NO"

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

Re: compiling FreeBSD date on Linux

[Mel Flynn]

#warning "Pedantic mode on"
#define TOPIC "BSD build system tricks"

On Wednesday 20 May 2009 21:57:02 Polytropon wrote:

> > DPADD=${LIBUTIL}
>
> Needs to compile what ${LIBUTIL} point to, usually the
> libutil directory in the src/ tree.

This is the actual build dependency and the var is defined in bsd.libnames.mk.

>
> > LDADD=-lutil
>
> Additional info for the linker: link against libutil.
> This indicates that libutil is a build dependency for
> the date program.

It's a linker statement, while it's logical that libutil has to be built 
before it can be linked against, this does in fact not have to be the same 
libutil, so what this line really cares about is that libutil.so or libutil.a 
is resolvable via the various linker rules.

One can in fact do:
LDFLAGS=-L/usr/local/lib
LDADD=${LIBUTIL} -lutil

This will link the static libutil from bsd.libnames.mk and then try to link 
with /usr/local/lib/libutil.so, before looking elsewhere.

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

Re: compiling FreeBSD date on Linux

[Mel Flynn]

On Wednesday 20 May 2009 19:45:59 francis keyes wrote:
> Hmm... the date program looks pretty simple but I don't understand the
> Makfile:
>
> #@(#)Makefile8.1 (Berkeley) 5/31/93
> # $FreeBSD: src/bin/date/Makefile,v 1.11.30.1 2009/04/15 03:14:26 kensmith
> Exp $
>
> PROG=date
> SRCS=date.c netdate.c vary.c
> DPADD=${LIBUTIL}
> LDADD=-lutil
>
> .include 
>
>
> Would it be possible to compile this without a makefile?

This is because the under appreciated FreeBSD make system does a *lot* of work 
for you.

- You need libutil: /usr/src/lib/libutil or (probably faster), see which 
functions from libutil date uses and re-implement them.
- Resolve header issues in the above source files
- And then it's most likely easiest to use cmake or autotools on linux to 
compile.

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

Re: Limiting resources in cron jobs

[Mel Flynn]

On Wednesday 20 May 2009 16:18:28 Kirk Strauser wrote:
> On May 20, 2009, at 7:00 AM, Mel Flynn wrote:
> > Check with top what the CPU time is, it's not the same as the wall
> > clock.
>
> Give me *some* credit. :-)

Sorry, haven't you heard? Financial crisis ;)
Are you sure cron respects login.conf? I don't see it mentioned in the man 
page. Have you tried modifying the offending crontab to run using limits(1) 
program?

AFAIK, cron doesn't use login(1) or underlying infrastructure, yet it uses 
pam.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: proftpd TLS

[Mel Flynn]

On Wednesday 20 May 2009 16:13:15 alexus wrote:
> On Wed, May 20, 2009 at 7:46 AM, Mel Flynn
>
>  wrote:
> > On Tuesday 19 May 2009 21:18:48 alexus wrote:
> >> On Tue, May 19, 2009 at 2:26 PM, Mehul Ved  wrote:
> >> > On Tue, May 19, 2009 at 11:14 PM, alexus  wrote:
> >> >> i start it as a root, but it switchs to non-root
> >> >>
> >> >> nobody 52346  0.0  0.1 11820  4208  ??  SsJ  Sun06PM   0:00.66
> >> >> proftpd: (accepting connections) (proftpd)
> >> >
> >> > Check the value for 'user' in proftpd.conf. It will be nobody. Change
> >> > it to root.
> >> >
> >> > --
> >> >
> >> > Dyslexics have more fnu.  -
> >> > http://kingsly.net/tmp/fortune.php/1242364116
> >>
> >> wouldn't it sort of make it more risky in terms of security to run
> >> ftpd as root vs nobody?
> >> in general daemon do not run as root and thats for a reason..
> >
> > Yes, don't do it. Is proftpd started as root? Then this shouldn't occur,
> > although a forum post[1] suggests that mod_cap can fiddle with this.
> >
> > [1] http://forums.proftpd.org/smf/index.php?topic=1315.0
> > --
> > Mel
>
> if i set User in proftpd.conf to root, then it runs as a root

I said *start* as root. Theoretically, the pass phrase part for your 
certificate comes before dropping privileges. But maybe there's a bug in the 
code. Is proftpd running jailed or not?

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

Re: synchronize time

[Mel Flynn]

On Tuesday 19 May 2009 15:08:35 Greg Larkin wrote:
> Mel Flynn wrote:
> > On Saturday 16 May 2009 18:02:13 Roy Stuivenberg wrote:
> >> Hi Mel,
> >>
> >> /etc/ntp.conf is empty.
> >
> > You'd need a server...Just one line is enough, f.e.:
> > echo 'server ntp.xs4all.nl' >/etc/ntp.conf
>
> Hi all,
>
> I have been using the ntp.org server pool successfully.  There are
> instructions for configuring your NTP server to connect to it here:
>
> http://www.pool.ntp.org/en/use.html

I know about these, but they suffer from the same bug as Slashdot comments and 
Wikipedia: everybody's an expert syndrome.
Also, Xs4all is a dutch provider, running FreeBSD since Windriver dumped BSDi.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Limiting resources in cron jobs

[Mel Flynn]

On Saturday 16 May 2009 19:27:22 Kirk Strauser wrote:

> www:\
>
>  :cputime=300:\
>  :tc=default:
>
> I've run "cap_mkdb /etc/login.conf" to make that live.  Then, I used
> vipw to change www's class:
>
> www:*:80:80:www:0:0:World Wide Web Owner:/nonexistent:/usr/sbin/nologin
>
> However, I can trigger the error condition and watch the child
> Ghostscript process run for 6-7 minutes before I kill it.

Check with top what the CPU time is, it's not the same as the wall clock.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: proftpd TLS

[Mel Flynn]

On Tuesday 19 May 2009 21:18:48 alexus wrote:
> On Tue, May 19, 2009 at 2:26 PM, Mehul Ved  wrote:
> > On Tue, May 19, 2009 at 11:14 PM, alexus  wrote:
> >> i start it as a root, but it switchs to non-root
> >>
> >> nobody 52346  0.0  0.1 11820  4208  ??  SsJ  Sun06PM   0:00.66
> >> proftpd: (accepting connections) (proftpd)
> >
> > Check the value for 'user' in proftpd.conf. It will be nobody. Change
> > it to root.
> >
> > --
> >
> > Dyslexics have more fnu.  - http://kingsly.net/tmp/fortune.php/1242364116
>
> wouldn't it sort of make it more risky in terms of security to run
> ftpd as root vs nobody?
> in general daemon do not run as root and thats for a reason..

Yes, don't do it. Is proftpd started as root? Then this shouldn't occur, 
although a forum post[1] suggests that mod_cap can fiddle with this.

[1] http://forums.proftpd.org/smf/index.php?topic=1315.0
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: apache not starting on reboot

[Mel Flynn]

On Monday 18 May 2009 10:55:00 Odhiambo  ワシントン wrote:
> On Mon, May 18, 2009 at 11:03 AM, Brent Clark 
wrote:
> > Hiya
> >
> > I have the following in my /etc/rc.conf
> >
> > mitm# cat /etc/rc.conf | grep apache
> > apache22_enable="YES"
> > mitm#
> >
> > The problem I seem to be experiencing is that if I reboot the machine,
> > then apache does not come up. Its only on when I run
> >
> > /usr/local/etc/rc.d/apache22 restart
> >
> > that apache is running and available.
> >
> > If anyone could assist me on where I went wrong or on what route and / or
> > path to look, I would be most grateful.
>
> Start by looking at /var/log/messages

Better yet, /var/log/httpd-error.log. On restart, does it give a "apache not 
running?" message or is it running, but not responding to requests?
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Reading warnings when installing multiple ports

[Mel Flynn]

On Saturday 16 May 2009 21:21:54 Jerry McAllister wrote:
> On Sat, May 16, 2009 at 11:24:42AM +0200, Mel Flynn wrote:
> > On Friday 15 May 2009 19:26:00 mfv wrote:
> > > On Tuesday 12 May 2009 13:53:35 Mel Flynn wrote:
> > > > On Monday 11 May 2009 15:21:24 Jerry McAllister wrote:
> > > > > But, I very often wish there was a convenient way to see some of
> > > > > those messages and especially messages about things one has to do
> > > > > during the install, such as manually installing something or
> > > > > getting some license thing handled, before I start the port
> > > > > install.
> > > > >
> > > > > Sometimes I get a question in the middle of an install that I do
> > > > > not know how to answer and it is an awfully inconvenient time to
> > > > > have to start scrounging for information.Having a commannd that
> > > > > would display all those things and maybe some related information
> > > > > or pointers to information for making an intelligent response -
> > > > > before starting the make -  would be very helpful.
> > > >
> > > > Do you have specific examples? Cause I can't think of anything that
> > > > falls under your description.
> > >
> > > Hello All,
> > >
> > > I had a recent experience with editors/openoffice.org-2.  Based on past
> > > experience I know that the compile would take a few hours.  I started
> > > make and left to do some chores.  When I returned I discovered that the
> > > program had aborted as I did not have java installed and had to
> > > download a patch from the Sun website.
> >
> > This can and will not ever be fixed, because it's a legal and not a
> > technical issue.
> > Once you know this, you know to install diablio-jdk first, which takes
> > 5-10 minutes pending download speed.
>
> Yes, I know it is a legal/license issue at Sun and I sort of know to do
> it now, having installed OO a couple of times.   But, the point was and is,
> it would help people if the information about having to do it would
> come up right at the first, so a person could have it taken care of
> instead of starting an install which one knows will take hours so
> leaves to do something else and comes back and finds the install
> stopped hours ago because of something that could have and should
> have been taken care of before actually starting the install.

JDK's should really set IS_INTERACTIVE if distfiles that are not available 
cannot be downloaded without user intervention (MAINTAINER cc'd because of 
this). This would signal portmaster to present you with a message before 
starting the build and letting you go about your business.

If your ports management software does not recurs through all configuration 
dialogs before starting the build, you're not using the right tool for the 
job.

> Some variation of this, often involving entering a 'y' or 'n' at
> some point in the middle of an install that could have been done
> causing an environmental variable or some such to be set ahead of
> time exists in a number of ports I have installed.   It is annoying
> to come back from a bunch of tiring meetings only to see that an
> install that could be finished has several more hours to run because
> it was waiting all that time for a y or n.

This is where -DBATCH comes in. It silences all those. The ones that aren't 
silenced and aren't legal issues, should be considered bugs. Various ports 
management tools also support "automatic answer" features. Without ports 
management software you can always run yes|make -DBATCH.

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

Re: synchronize time

[Mel Flynn]

On Saturday 16 May 2009 18:02:13 Roy Stuivenberg wrote:
> Hi Mel,
>
> /etc/ntp.conf is empty.

You'd need a server...Just one line is enough, f.e.:
echo 'server ntp.xs4all.nl' >/etc/ntp.conf
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: synchronize time

[Mel Flynn]

Hoi Roy,

On Saturday 16 May 2009 17:01:25 Roy Stuivenberg wrote:

> I seem to be unable to change my systemtime to synchronize with time
> servers.
> I'm using 7.2 stable and gnome2.
> policykit.conf =
> 
> 
>   
> 
> 
> 
>   
> 
> 
> 
> And ntpd_enable="YES" in rc.conf.

Policykit should be irrelevant (at least I hope it doesn't put it's claws in 
there). WHat's your /etc/ntp.conf and anything ntpd-ish in /var/log/messages?
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: php5 pcre

[Mel Flynn]

On Friday 15 May 2009 11:13:41 Mark wrote:
> -Original Message-
> From: owner-freebsd-questi...@freebsd.org
> [mailto:owner-freebsd-questi...@freebsd.org] On Behalf Of Gorbatovsky
> Dmitry Sent: vrijdag 15 mei 2009 11:04
> To: freebsd-questions@freebsd.org
> Subject: Re: php5 pcre
>
> > When I run make config I see selected PCRE and SPL extensions and
> > other default the extensions. But I rebuilding all modules,
> > restarting apache and see same error:
> >
> > [15-May-2009 12:38:00] PHP Warning:  PHP Startup: Unable to load dynamic
> > library '/usr/local/lib/php/20060613/pcre.so' -
> > /usr/local/lib/php/20060613/pcre.so:
> > Undefined symbol "php_pcre_free" in Unknown on line 0
>
> I thought I was the only one this had happened to. :) In my case I
> resolved it by going into the pcre directory itself and build it from
> there.

That don't make sense. Any chance you had BUNDLED_PCRE selected and decided to 
untick it in the config dialog?
If this really is all you did, then maybe extension ordering is the culprit. 
Move pcre.so in extension.ini upwards and see if the bug resurfaces.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: csup source tree as of a certian date

[Mel Flynn]

On Thursday 14 May 2009 20:56:37 Sam Fourman Jr. wrote:
> > Have you tried:
> > *default date=2009.04.28.23.59.59
> >
> > Back when I had to get the old XOrg, *default date' in my ports-supfile
> > returned me to sanity ...
> >
> > --
> > Don Readdon_r...@att.net
>
> Worked Great, Thanks Don, although 4-29-2009 still caused the panic
> I could see from the csup output that I was getting the date specified.
> I will just keep going back further.

Before you end up in 1969, try to disable all 3rd party modules, like 
nvidia_load and fusefs_enable etc etc. A panic can be triggered when running 
sysctl -a, if modules aren't rebuilt, but maybe there's other code paths that 
trigger the same bug.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Reading warnings when installing multiple ports

[Mel Flynn]

On Friday 15 May 2009 19:26:00 mfv wrote:
> On Tuesday 12 May 2009 13:53:35 Mel Flynn wrote:
> > On Monday 11 May 2009 15:21:24 Jerry McAllister wrote:
> > > But, I very often wish there was a convenient way to see some of those
> > > messages and especially messages about things one has to do during
> > > the install, such as manually installing something or getting some
> > > license thing handled, before I start the port install.
> > >
> > > Sometimes I get a question in the middle of an install that I do not
> > > know how to answer and it is an awfully inconvenient time to have
> > > to start scrounging for information.Having a commannd that
> > > would display all those things and maybe some related information
> > > or pointers to information for making an intelligent response - before
> > > starting the make -  would be very helpful.
> >
> > Do you have specific examples? Cause I can't think of anything that falls
> > under your description.
>
> Hello All,
>
> I had a recent experience with editors/openoffice.org-2.  Based on past
> experience I know that the compile would take a few hours.  I started make
> and left to do some chores.  When I returned I discovered that the program
> had aborted as I did not have java installed and had to download a patch
> from the Sun website.

This can and will not ever be fixed, because it's a legal and not a technical 
issue.
Once you know this, you know to install diablio-jdk first, which takes 5-10 
minutes pending download speed.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: ghostscript8 - errors when compiling from ports

[Mel Flynn]

On Wednesday 13 May 2009 21:08:11 Jonathan Chen wrote:
> On Wed, May 13, 2009 at 05:54:09PM +0200, Zbigniew Komarnicki wrote:
> > On Wednesday 13 of May 2009 12:08:55 you wrote:
> > > just a guess, looks like a problems with cups-base, try to update it
> > > first. I recommend ports-mgmt/portmaster.
> >
> > Thank you for help :-)
> >
> > I upgraded all my installed ports, by
> > # portmaster -a
> >
> > but this error still remain. I also reinstall the cups-base port by
> > # portmaster -rf print/cups-base
> >
> > but still with no success. Here is the output:
>
> Here's the thing: Ghostscript has a hidden (and apparently broken)
> dependency on cups (ie, it's not in the Makefile).

It's called autotools magic. If cups.h is found, --disable-cups isn't set then 
the assumption is made that cups/raster.h also exists, but in FreeBSD ports 
cups/raster.h is not installed when only cups-client is installed.

If you also want gs binary to be able to print to cups directly, then after 
you did this upgrade, reinstall ghostscript 8 again, so it will pick up the 
cups support automatically. Most people don't care about this last bit, since 
gs is used under the hood to transform document formats and printing is done 
via a higher level application.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: subversion / dav_svn_module : Fatal error 'Recurse on a private mutex.'

[Mel Flynn]

On Friday 15 May 2009 02:27:43 Olivier Mueller wrote:
> Hi Mel,
>
> On Wed, 2009-05-13 at 22:21 +0200, Mel Flynn wrote:
> > I'm still thinking there's two different (threading|bdb) libraries linked
> > into httpd, but not sure to ask for which ldd...httpd or mod_dav. The db
> > version could be a red herring or that only one of the formats requires
> > this mutex .
>
> This is how it currently looks:(I'll try recompiling some packages
> later next week):
>
> $ ldd /usr/local/sbin/httpd
> /usr/local/sbin/httpd:
>   libz.so.3 => /lib/libz.so.3 (0x800681000)
>   libaprutil-0.so.9 => /usr/local/lib/apache2/libaprutil-0.so.9
> (0x800795000) libdb-4.2.so.2 => /usr/local/lib/libdb-4.2.so.2 (0x8008ab000)
>   libexpat.so.6 => /usr/local/lib/libexpat.so.6 (0x800a87000)
>   libiconv.so.3 => /usr/local/lib/libiconv.so.3 (0x800ba9000)
>   libapr-0.so.9 => /usr/local/lib/apache2/libapr-0.so.9 (0x800da2000)
>   libm.so.4 => /lib/libm.so.4 (0x800ec2000)
>   libcrypt.so.3 => /lib/libcrypt.so.3 (0x800fde000)
>   libc.so.6 => /lib/libc.so.6 (0x8010f7000)

No pthread in this. Could it be that apr is built without threads? It's the 
only thing out of the ordinary that I can see.
I use APR_FROM_PORTS (even though it has a be careful warning, that I don't 
understand) and as such can add threading support.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: How to move vi to /bin

[Mel Flynn]

On Friday 15 May 2009 08:46:46 Manish Jain wrote:
> Mel Flynn wrote:
> > On Wednesday 13 May 2009 09:21:46 manish jain wrote:
> >> I want to move vi to /bin so that I have an editor available in
> >> single-user mode.
> >
> > The only reason to need an editor and not have /usr and /var available is
> > to edit /etc/fstab. It is trivial to spot errors with /rescue/cat and fix
> > with /rescue/sed, without having to worry about a terminal.
> >
> > In all other cases:
> > fsck -p
> > /etc/rc.d/mountcritlocal start
> > /etc/rc.d/ldconfig start
> >
> > And one can use any editor one would want. Don't forget to export or
> > setenv TERM to cons25 from 'dumb'.
>
>  From all the discussion I have walked through on the issue of where to
> place vi, it does appear FreeBSD has a skewed policy on the issue. There
> are plenty of reasons you might need access an editor in single-user
> mode - editing  fstab is just one.

You didn't read what I wrote.
The above works in single user mode and does not magically transform you to 
multi-user. Just reboot in single user mode, type the above and then 
/usr/bin/vi /etc/rc.conf.

You will see that it works. In fact, if you have vim installed, 
/usr/local/bin/vim /etc/rc.conf will work.

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

Re: USB WLAN Atheros and USB Ethernet FBSD 7.2

[Mel Flynn]

On Friday 15 May 2009 13:04:50 Saša Stupar wrote:

> I suggest you to buy a good AP (Lynksys, Asus, etc.) and it will work much
> better than building it from FreeBSD.

And this is based on which assumption with what criteria for "working better"?
On my FreeBSD AP I can:
- view my logs in realtime
- shape traffic
- deny/grant access at will without requiring rule reloads (pf tables ftw)
- send custom DHCP info, like:
option wpad code 252 = text;
option wpad "http://10.0.0.1/proxy.pac";;
- configure over ssh
- add memory
- control internal and external DNS

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

Re: mergemaster -U overwriting modified files

[Mel Flynn]

On Sunday 26 April 2009 11:12:12 Frederique Rijsdijk wrote:
> Bruce Cran wrote:
> > On Sat, 25 Apr 2009 12:10:42 +0200
> >
> > Peter Schuller  wrote:
> >> I recently began testing mergemaster -U since the perpetual "review
> >> diff of file I never touched" grows annoying real quick.
> >>
> >> Unfortunately I recently discovered that it does not seem to do what
> >> you might expect. For example it nuked my mailer.conf on one machine,
> >> and my /etc/namedb/named.conf (!!!) on another machine.
> >>
> >> Is this a bug or intended? What is the intended functionality of -U?
> >
> > I noticed this recently too: after using mergemaster -U without
> > problems for a long time it suddenly went and overwrote named.conf on
> > a recently upgrade of 7-STABLE.
>
> I've seen this happen as well with named.conf.

I think I know the cause, not entirely the problem yet, as I just got hit by 
this too and right at the point where I upgraded source tree from cvs to svn 
so *all* files had different idents.

Before running mergemaster -iU I checked /var/db/mergemaster.mtree and it was 
zero-sized. Why, is not entirely clear, (hence, I don't know the real problem) 
but I thought I noticed mergemaster saving mtree database on the pre-world 
run. Looking at the code though, this should be impossible, so the more I 
think about it, the more I start to doubt. At the time I was thinking why is 
mergemaster saving the mtree and that's when I checked it's size.

Whatever the cause, this is where mergemaster fails:
CHANGED=
if [ -n "${AUTO_UPGRADE}" -a -f "${DESTDIR}${MTREEFILE}" ]; then
for file in `mtree -eq -f ${DESTDIR}${MTREEFILE} -p ${DESTDIR}/ \
2>/dev/null | awk '($2 == "changed") {print $1}'`; do
if [ -f "${DESTDIR}/$file" ]; then
CHANGED="${CHANGED} ${DESTDIR}/$file"
fi
done
fi

Because ${MTREEFILE} is empty, the mtree command will not produce output and 
CHANGED will not be populated. For things like this, it would be nice if mtree 
supported a 'lint' mode to check syntax, but at the very least could 
mergemaster test for -s rather then or in addition to -f?

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

Re: /tmp/security on 7.1-Release

[Mel Flynn]

On Thursday 14 May 2009 08:35:40 Zbigniew Szalbot wrote:
> Hello,
>
> >> Ever since upgrading from 7.0-Release to 7.1-Release I get the
> >> following in the daily security reports:
> >>
> >> master.lc-words.com kernel log messages:
> >> +++ /tmp/security.yMzCQbS8 2009-05-13 03:01:25.0 +0200
> >> +re0: watchdog timeout
> >> +re0: link state changed to DOWN
> >> +re0: link state changed to UP
> >>
> >> It just happens around 3 a.m. each day and so I wonder if I should be
> >> concerned about it...?
> >>
> >> Thank you in advance for your suggestions!
> >
> > I can bet it's NIC related problem, probably your re0 is sharing IRQ
> > with something and it's going up/down. The time you see (~3.oo AM) is
> > the moment periodic works, so it doesn't mean the problem happend then.
> >
> > I've seen similar problems on cheap PC hardware with integrated NIC
> > and/or external PCI NIC (re/rl/em). Search for 're0: watchdog timeout'
> > on google.
> >
> > The solution is to replace re with something else - em for example, but
> > there is no guarantee it's going to work - I had same problems on
> > crappy hardware with two identical interfaces plugged in (some model of
> > em - IRRC Intel PRO/1000 GT)
>
> Yes, you are most likely right but I am left wondering why this has never
> been a problem on 7.0-Release...?

Run a 7.0-RELEASE kernel to be sure of this. It can just be that the hardware 
went bad around the same time you upgraded.
Bad cable or negotiation with a switch are also causes of this. See the 
DIAGNOSTICS section of the re driver.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: How to move vi to /bin

[Mel Flynn]

On Wednesday 13 May 2009 09:21:46 manish jain wrote:

> I want to move vi to /bin so that I have an editor available in
> single-user mode.

The only reason to need an editor and not have /usr and /var available is to 
edit /etc/fstab. It is trivial to spot errors with /rescue/cat and fix with 
/rescue/sed, without having to worry about a terminal.

In all other cases:
fsck -p
/etc/rc.d/mountcritlocal start
/etc/rc.d/ldconfig start

And one can use any editor one would want. Don't forget to export or setenv 
TERM to cons25 from 'dumb'.

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

Re: How to move vi to /bin

[Mel Flynn]

On Thursday 14 May 2009 12:38:30 Chris Rees wrote:
> 2009/5/13 Mel Flynn :
> > On Wednesday 13 May 2009 11:34:43 Michael Powell wrote:
> >> Kind of like how those coming over from a
> >> Linux environment all seem to want to change root's shell to bash, it
> >> serves no purpose except foot-shooting.
> >
> > - csh cannot redirect stderr seperately from stdout
> > - on pipes the exit status from the first command is the exit status of
> > the total command
> > - will not expand matches without a user provided part, for *every*
> > component of a path
> >
> > There's plenty of reasons not to use csh and if you know what you're
> > doing, BSD lets you. And no, I don't want to type exec zsh when I'm
> > finally logged into the box that has a load of 100+.
>
> I think the problem with that is he meant changing the root shell to
> /usr/local/bin/bash. You're better off using /bin/sh if you want a
> Bourne-type shell, or using toor with /usr/local/bin/bash.

sh is worse then csh. And I said if you know what you're doing. My root shell 
is less prone to break then the standard csh shell, because I compile it 
statically (and also on the / partition).

On Thursday 14 May 2009 18:03:25 Chad Perrin wrote:

> I've never understood the resistance to just use toor instead of root if
> one wants a nonstandard administrative shell.

Habit, mostly. toor is one way of doing things, just changing the shell is the 
other. Maybe it's my paranoia that I might be running software that does 
string matches for root logins, rather then uid to disallow access ;)
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: make.conf options based on DESTDIR?

[Mel Flynn]

On Wednesday 13 May 2009 20:16:47 David Wassman wrote:
> I have setup several jails that use a master template similar to the
> advanced jail configuration in the handbook. Currently, I build the master
> jail with several build options turned off using a separate make.conf file
> and setting __MAKE_CONF to it.
>
> I was wondering if there is a way to set /etc/make.conf up so that if
> DESTDIR is set to the master jail location, make would use those options
> but not use them during a standard build process with DESTDIR not defined
> or pointed somewhere else.
>
> I have read through the man pages on make and make.conf, and have tried:
>
> .if DESTDIR=/usr/Jail/master
> WITHOUT_ACPI=YES
> WITHOUT_BOOT=YES
> ...
> .endif

Almost:
# head -3 /etc/make.conf
.if defined(DESTDIR) && (${DESTDIR:M/usr/jails/tpl} != "")
WITH_FOO=yes
.endif

# cat Makefile
all:
@echo "WITH_FOO=${WITH_FOO}"

# make
WITH_FOO=

# make DESTDIR=/usr/jails/tpl
WITH_FOO=yes

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

Re: scrotwm can't find

[Mel Flynn]

On Thursday 14 May 2009 10:54:43 Saifi Khan wrote:
> Hi:
>
> Trying to compile scrotwm (from OpenBSD ports)
> http://www.peereboom.us/scrotwm/html/scrotwm.html
> on my FreeBSD 8.x 200905 i386  system.
>
> and make can't find
>  .include 
>  .include 
>
> The directory /usr/ports/Mk/ has bsd.xorg.mk file but not
> bsd.prog.mk

The OpenBSD ports system is not the same as the FreeBSD ports one. FreeBSD 
ports all use bsd.port.mk, which pulls in prog.mk from /usr/share/mk 
automagically. xorg.mk is also automagically pulled in when USE_XORG is set.

You need to port the port, compile by hand or ask someone to port the port.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: 7.0 -> 7.1 crash every two day

[Mel Flynn]

On Wednesday 13 May 2009 19:46:26 Gabri Mate wrote:
> On 19:57 Tue 12 May , Mel Flynn wrote:
> > On Sunday 10 May 2009 20:51:48 mailingl...@modernbiztonsag.org wrote:
> > > Dear List,
> > >
> > > I've upgraded from source my 7.0-RELEASE to 7.1-RELEASE. It's a general
> > > purpose server for a small company and it's clients, running web, sql,
> > > mail and recursive dns. The machine is a Compaq ML350 G2. Version 7.0
> > > was running on it for a year now without any problem. After 30 of April
> > > I've upgraded to 7.1 (I've followed the handbook) and since then nearly
> > > every two day the machine becomes unreachable except for echorequest.
> > > If i try to log in in the terminal it just hangs after i give my
> > > password.
> > >
> > > I'm out of any ideas. Please feel free to ask any more info that I've
> > > forgotten to provide. Thank You for your help in advance.
> >
> > If this is a not a GENERIC kernel, did you change scheduler or stuck with
> > SCHED_4BSD in your kernel?
>
> I've just added PF related stuff, so the scheduler is remained the
> default.

Just FYI: unless you use include GENERIC and added pf, rather then copy 
GENERIC to MYKERNEL, you're running with SCHED_4BSD, the default has changed 
in 7.1.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Reading warnings when installing multiple ports

[Mel Flynn]

On Tuesday 12 May 2009 21:04:57 Glen Barber wrote:
> On Tue, May 12, 2009 at 1:53 PM, Mel Flynn
>
>  wrote:
> > On Monday 11 May 2009 15:21:24 Jerry McAllister wrote:
> >> But, I very often wish there was a convenient way to see some of those
> >> messages and especially messages about things one has to do during
> >> the install, such as manually installing something or getting some
> >> license thing handled, before I start the port install.
> >>
> >> Sometimes I get a question in the middle of an install that I do not
> >> know how to answer and it is an awfully inconvenient time to have
> >> to start scrounging for information.Having a commannd that
> >> would display all those things and maybe some related information
> >> or pointers to information for making an intelligent response - before
> >> starting the make -  would be very helpful.
> >
> > Do you have specific examples? Cause I can't think of anything that falls
> > under your description.
>
> Perhaps he is talking about the 'make options' output, such as what is
> displayed right before x11-wm/fluxbox begins to build.  That is the
> only type of output I can ever remember seeing before a build begins.

Well, there's "would you like to activate postfix in mailer.conf [y/n]", but 
that never triggered me to scrounge for info and can be handled by -DBATCH and 
setting POSTFIX_DEFAULT_MTA in /etc/make.conf (or not setting it if you don't 
wanna).
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: subversion / dav_svn_module : Fatal error 'Recurse on a private mutex.'

[Mel Flynn]

Hi Olivier,

On Wednesday 13 May 2009 19:32:36 Olivier Mueller wrote:

> > > ==> /var/log/httpd/httpd-error.log <==
> > > Fatal error 'Recurse on a private mutex.' at line 986 in file
> > > /usr/src/lib/libpthread/thread/thr_mutex.c (errno = 86) [Tue May 12
> > > 11:56:02 2009] [notice] child pid 64353 exit signal Abort trap (6)
> > >
> > >
> > > The only difference between both repositories is the db/format file:
> > >
> > > diff -r repos/websites/testing/db/format
> > > repos/websites/testing2/db/format 1c1
> > > < 4
> > > ---
> > >
> > > > 3
> > >
> > > If format is "3" (subversion "pre-1.6"), everything works fine, but if
> > > the format is "4" the DAV/SVN part crashes (while everything remains
> > > fine via svn client / svnserve).
> >
> > this is with the same binary? Meaning there's no Berkeley db library
> > differences?
>
> Yes, the only recent change was a "portupgrade -rvbp subversion".

I'm still thinking there's two different (threading|bdb) libraries linked into 
httpd, but not sure to ask for which ldd...httpd or mod_dav. The db version 
could be a red herring or that only one of the formats requires this mutex .
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: How to move vi to /bin

[Mel Flynn]

On Wednesday 13 May 2009 11:34:43 Michael Powell wrote:

> Kind of like how those coming over from a
> Linux environment all seem to want to change root's shell to bash, it
> serves no purpose except foot-shooting.

- csh cannot redirect stderr seperately from stdout
- on pipes the exit status from the first command is the exit status of the 
total command
- will not expand matches without a user provided part, for *every* component 
of a path

There's plenty of reasons not to use csh and if you know what you're doing, 
BSD lets you. And no, I don't want to type exec zsh when I'm finally logged 
into the box that has a load of 100+.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: 7.0 -> 7.1 crash every two day

[Mel Flynn]

On Sunday 10 May 2009 20:51:48 mailingl...@modernbiztonsag.org wrote:
> Dear List,
>
> I've upgraded from source my 7.0-RELEASE to 7.1-RELEASE. It's a general
> purpose server for a small company and it's clients, running web, sql, mail
> and recursive dns. The machine is a Compaq ML350 G2. Version 7.0 was
> running on it for a year now without any problem. After 30 of April I've
> upgraded to 7.1 (I've followed the handbook) and since then nearly every
> two day the machine becomes unreachable except for echorequest. If i try to
> log in in the terminal it just hangs after i give my password.
>
> I'm out of any ideas. Please feel free to ask any more info that I've
> forgotten to provide. Thank You for your help in advance.

If this is a not a GENERIC kernel, did you change scheduler or stuck with 
SCHED_4BSD in your kernel?
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Reading warnings when installing multiple ports

[Mel Flynn]

On Monday 11 May 2009 15:21:24 Jerry McAllister wrote:

> But, I very often wish there was a convenient way to see some of those
> messages and especially messages about things one has to do during
> the install, such as manually installing something or getting some
> license thing handled, before I start the port install.
>
> Sometimes I get a question in the middle of an install that I do not
> know how to answer and it is an awfully inconvenient time to have
> to start scrounging for information.Having a commannd that
> would display all those things and maybe some related information
> or pointers to information for making an intelligent response - before
> starting the make -  would be very helpful.

Do you have specific examples? Cause I can't think of anything that falls 
under your description.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Booting question

[Mel Flynn]

On Monday 11 May 2009 03:57:21 Michel Di Croci wrote:

> Yes it is in. And it's not the sendmail that is slow, it's the detection /
> kernel step... not the service steps.

Please choose verbose boot from the menu and hand-copy the lines right 
before,right after and during which the hang occurs. You can use scroll-lock 
and page up/down to find them again after it scrolls up.
Disable any X/G/Kdm display manager you might have auto started or press ctrl-
alt-f1 to get back to the original console.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: subversion / dav_svn_module : Fatal error 'Recurse on a private mutex.'

[Mel Flynn]

On Tuesday 12 May 2009 12:10:58 Olivier Mueller wrote:
> Hello,
>
> I have a strange situation on our internal svn server. Since a few days
> and some upgrades, if I try to access a new created repository via
> apache, I get a blank page and this error in the apache error log:
>
> ==> /var/log/httpd/httpd-error.log <==
> Fatal error 'Recurse on a private mutex.' at line 986 in file
> /usr/src/lib/libpthread/thread/thr_mutex.c (errno = 86) [Tue May 12
> 11:56:02 2009] [notice] child pid 64353 exit signal Abort trap (6)
>
>
> The only difference between both repositories is the db/format file:
>
> diff -r repos/websites/testing/db/format repos/websites/testing2/db/format
> 1c1
> < 4
> ---
>
> > 3
>
> If format is "3" (subversion "pre-1.6"), everything works fine, but if
> the format is "4" the DAV/SVN part crashes (while everything remains
> fine via svn client / svnserve).

this is with the same binary? Meaning there's no Berkeley db library 
differences?
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Reformatting external harddrive

[Mel Flynn]

On Tuesday 12 May 2009 03:18:13 Daniel Underwood wrote:
> After unsuccessfully trying to reformat my external harddrive on my
> linux machine, I'm trying to reformat the disk in FreeBSD.  Frankly, I
> just don't know how to do that. Please help me get the disk back to
> working order; I don't need to keep any data that is currently on the
> disk.

> sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD)
> start 63, size 625137282 (305242 Meg), flag 80 (active)
>   beg: cyl 0/ head 1/ sector 1;
>   end: cyl 0/ head 254/ sector 63

And what's not working? this shows a 30G FreeBSD partition?
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: mount_nfs and fstab options

[Mel Flynn]

On Tuesday 12 May 2009 14:33:00 Martin Badie wrote:
> Hi,
>
> I have mounted an NFS share like:
>
> mount_nfs -LisT 10.10.10.199:/vol/share   /mnt
>
> but I can't use -LisT on fstab because man mount_nfs states:

Yes you can. Just comma seperate the arguments without whitespace in the 
appropreate column. The only thing you cannot do in fstab(5) is use arguments 
containing whitespace, like mounting your music collection on My Music folder.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: isc-dhcpd server, HOSTNAME

[Mel Flynn]

On Saturday 09 May 2009 15:09:45 Pieter Donche wrote:

> case DHCP server DHCP client   HOSTNAME env. var.
>
>   1   isc-dhcp30-server   FreeBSD7-i386 not set
>on FreeBSD-amd64
>   2   isc-dhcp30-server   SuSE Linux 10.3   set
>on FreeBSD-amd64
>
>   3   some DHCP serverFreeBSD7-i386 set
>on unkown serverOS
>   4   some DHCP serverSuSE Linux 10.3   set
> on unkown serverOS

Judging from this, you have a hostname set in /etc/rc.conf on freebsd 7 client 
and/or dhcpd isn't configured to send one as it receives one from the client 
and perhaps you have dynamic DNS configured?

If that's not the case, then you should add some debugging to /sbin/dhclient-
script in the check_hostname function.

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

Re: [warn] kevent: Bad file descriptor

[Mel Flynn]

On Friday 08 May 2009 23:23:32 Gary Gatten wrote:
> I just compiled and installed nTop 3.3.10 and now I'm getting this
> error.  Had an older version running before this with no problem.  I'm
> on 6.0 RELEASE.  I'm still googling, any quick fixes would be GREATLY
> appreciated!

Shot in the dark: mount /proc. 
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: sshfs

[Mel Flynn]

On Friday 08 May 2009 10:18:40 Antonio Tommasi wrote:

> "fuse: failed to open fuse device: No such file or directory"
> Where is the problem

There's no fuse device. Read the pkg-message again (fuse_enable in rc.conf and 
start fuse service - too rusty on the exact fuse* variable name).
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Frozen on Boot - Kernel Hanging?

[Mel Flynn]

On Thursday 07 May 2009 23:46:54 APseudoUtopia wrote:
> Hey,
>
> My server was fine when I went to work. When I got back, it was dead.
> I had the datacenter reboot it, and it refused to boot. It just hangs
> with no error message when booting. After the "Welcome to FreeBSD"
> menu, it just freezes up.
>
> I have no idea where to start to fix this. Any ideas?

Like Wojchiech said, most likely hardware. Try to boot a livecd and if that 
won't work either, it's time to yank out hardware. If you have a replacement  
or test machine, put the HDD in there and if the HDD isn't the problem you 
might be able to read it's log files to get hints about what hardware part is 
the problem.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Run script on boot, as ordinary user

[Mel Flynn]

On Thursday 07 May 2009 19:57:03 Nerius Landys wrote:
> So there's cron.  Is there anything that lets an ordinary user start
> his/her programs at bootup of the system?  And then run a script when
> the system is shutting down?  I'm familiar with /etc/rc.d/, but that's
> not really what I'm looking for.

You sure? You can simply write an rc.d script that iterates through 
/home/*/rc.d/* and invokes each enabled script in there as the user, using su 
or sudo.
This will cleanly shutdown stuff for them.

Whether they *should* be running their own instances is an entirely different 
question. VirtualHost can do a lot and with mod_vhost_alias you simplify the 
maintenance, while maintaining several instances complicates it.
--
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: isc-dhcp logging and status query

[Mel Flynn]

On Thursday 07 May 2009 12:00:10 Pieter Donche wrote:

> 2. Is there any tool to see what Statically assigned IP address are handed
> out at a given time?
> (I also see nothing in /var/db/dhcpd/dhcpd.leases file execpt comments)

Add omapi-port 7911; to dhcpd.conf.

Then, as follows:
$ omshell
> connect
obj: 
> new lease
obj: lease
> set ip-address = 192.168.2.253
obj: lease
ip-address = c0:a8:02:fd
> open
obj: lease
ip-address = c0:a8:02:fd
state = 00:00:00:02
client-hostname = "impy"


See omshell(1) for more info. Install isc-dhcp30-relay to get the omapi(3) and 
dhcpctl(3) programming interfaces to roll your own tools.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: basic

[Mel Flynn]

On Wednesday 06 May 2009 21:09:07 Jerry McAllister wrote:
> On Wed, May 06, 2009 at 06:00:32PM +0200, Polytropon wrote:
> > 10 GOTO 10
> >
> > On Wed, 6 May 2009 14:32:47 +0200, "giorgio novello"  
wrote:
> > > Do you want obtain new market share?
> > >
> > > Develop e visual-basic like language, or asp vb and  your OS will be a
> > > best seller
> >
> > FreeBSD isn't for beginners, it's for professionals.
>
> Everyone is a beginner sometime.   So, FreeBSD is for beginners.
> Otherwise there would be no FreeBSD --- or you.

What he means is that FreeBSD does no hand holding or hide stuff "because you 
don't need access to it anyway". Also, there aren't many that started 
computing on FreeBSD.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Safe to 'make installkernel' in multi-user mode?

[Mel Flynn]

On Wednesday 06 May 2009 22:52:12 Modulok wrote:
> Just making sure I'm not brewing a disaster...
>
> Is it 'safe' to install a kernel (i.e. 'make installkernel') on a
> system while in multi-user mode?

It's the best and prefered way. Dropping to single user for installkernel has 
very little advantages, the running kernel doesn't change, only the on-disk 
version and if something goes wrong, you have full tools available to track 
the problem down.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Maintaining a FreeBSD system - Workcycle

[Mel Flynn]

On Thursday 07 May 2009 05:48:10 Tim Judd wrote:

> 2) Install portaudit and watch the periodic mailings that are sent to you.
> They list vulnerabilities in ports that really should be addressed. 

Not really. You can use the same common sense as with the base system and even 
more so (for the base system I just install them always, as it doesn't pay off 
in the long run to skip them). Portaudit for a (web)server has a lot of 
notifications that "are not critical", like several issues over the last year 
with php's safe mode that any sane webserver admin doesn't use.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: unable to boot with Nvidia AGP graphics card

[Mel Flynn]

On Thursday 07 May 2009 04:35:30 Scott Parrish wrote:
> Hi all,
> Recently my PCI graphics card failed on my Dell Dimension 4100.  I replaced
> it with a known good card I had lying around:  an Nvidia GeForce 3 TI200
> with an AGP interface.  My FreeBSD installation will not boot with this
> graphics card.  The boot loader hangs at the twirling baton as follows:
> /boot/kernel/acpi.ko text=0x43698 data=0x23c0+0x10f0
> syms=[0x4+0x7ba0+0x4+0xa828]
> \
>
> I'm running 7.1-RELEASE generic kernel.
> Anyone ever see anything like this before?  Any ideas on how I can debug? 

Most common cause is a faulty hints file, since you can't install from CD 
either, the GENERIC hints aren't working for this system. You could try 7.2-
RELEASE cd and file a PR otherwise. The dmesg from linux would be useful 
information in this PR.

> Is there anything I can do with the loader prompt to see what is happening
> when this occurs? The keyboard still seems responsive when this happens
> (the caps-lock, num-lock, etc. still work).  It is almost as if the boot
> loader is unsure how to send output to the AGP bus.

You can disable the AGP at loader prompt, similar to how one would do that in 
the hints file:
hint.agp.0.disabled="1"
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: make - reassign variable using if-then ?

[Mel Flynn]

On Wednesday 06 May 2009 11:31:17 Anton Shterenlikht wrote:

> I'm trying to build gcc43 on alpha 6.4.
> In /usr/ports/lang/gcc43/Makefile I have:
>
> # grep NOT_FOR_ARCHS /usr/ports/lang/gcc43/Makefile
> NOT_FOR_ARCHS=  alpha ia64 powerpc
> #
>
> In /etc/make.conf I have:
>
> .if ${.CURDIR:M*/lang/gcc43*}
> NOT_FOR_ARCHS= ia64
> USE_GCC=4.3+
> .endif
>
> This used to work fine until some update. Not anymore.
> The second setting is being used, i.e. the port is being built
> with gcc43. But the NOT_FOR_ARCHS is not changed, so I have
> to do it manually each time.
>
> So I tried to experiment with changing variable values withing if-then.

Your only option is overriding in /usr/portslang/gcc43/Makefile.local. This is 
because make.conf is read *before* the Makefile and the Makefile simply 
overrides your values. Makefile.local is read *after* the Makefile.
csup will leave it alone, however portsnap will delete the entire directory 
before upgrading the port, so your Makefile.local will be shot.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: make - reassign variable using if-then ?

[Mel Flynn]

On Wednesday 06 May 2009 10:31:53 Anton Shterenlikht wrote:
> I've this simple makefile:
>
> VAR=one
>
> all   :   main
>
> main  :
>   @echo ${.CURDIR}
> .if ${.CURDIR}
>   @echo ${VAR}
>   VAR=two
>   @echo ${VAR}
> .endif
>
> When I output VAR second time, the value is still "one", and not the
> new value "two". Why?

Because it is expanded before being passed to the shell. Sh sees:
echo one
VAR=two
echo one

What are you really trying to accomplish?
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: local security scanner for vulnerable common opensource www projects

[Mel Flynn]

On Wednesday 06 May 2009 00:01:12 Jeroen Hofstee wrote:
> Mel Flynn schreef:
> > You can do that, the issue is plugins:
> > 0) SuperCMS v 1.0 installed
> > 1) CoolStuff via webinterface, by SuperCMSNr1Fan, version 0.1.0.1beta
> > 2) SuperCMS v 1.0.1 security release, changes some issues with plugin
> > handling 3) CoolStuff's maintainer is now known as CompetitorCMSNr1Fan
> > 4) CoolStuff still works, because of backwards compatibility, but now is
> > insecure.
> >
> > Stuff like this goes back to the phpNukeYourSite days.
>
> I understand that there are allot of caveats and that is quite some work
> to create a full blown checker, especially with
> plugins. But as far as I am corcerned, finding the easy to locate
> vultnerable script is already better then doing nothing.

Agreed, as long as the client does not assume you are responsible. Portaudit 
will go a long way then. Which version of a plugin is installed is not always 
available in the file system, some store that in the database.
To ease your work, you may want to replace custom installed software with the 
corresponding port if available. This will go for a lot of stuff, including 
joomla and the various nuke forks. 
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: local security scanner for vulnerable common opensource www projects

[Mel Flynn]

On Tuesday 05 May 2009 22:04:27 Jeroen Hofstee wrote:
> Mel Flynn schreef:
> > On Saturday 02 May 2009 14:50:14 Jeroen Hofstee wrote:
> >> I tried to find a program which could scan the local filesystem and
> >> extract a lists of well known web projects (joomla, wordpress etc)
> >
> > Not that I'm aware of and it's hell to write and keep current.
>
> k, pitty. Although user can be jailed, it is still a bit unconfortable
> experience for users if their website looks
> somewhat different then they are used to; or their message board
> suddenly contains 2 additional post,
> albeit due to their own lack of maintaining the scripts behind it. A
> reminder that their script has known
> vulnerabities would therefore be nice, even if it doesn't pose a direct
> risk to the system as a whole.

I understand the problem.

> Most of these open source projects are in the ports, so the portaudit db
> will contain vulnerability information
> for them. If I find time, I will have a look if it is possible to match
> against that db.

You can do that, the issue is plugins:
0) SuperCMS v 1.0 installed
1) CoolStuff via webinterface, by SuperCMSNr1Fan, version 0.1.0.1beta
2) SuperCMS v 1.0.1 security release, changes some issues with plugin handling
3) CoolStuff's maintainer is now known as CompetitorCMSNr1Fan
4) CoolStuff still works, because of backwards compatibility, but now is 
insecure.

Stuff like this goes back to the phpNukeYourSite days.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Where am I wasting resources? How to fix this problem?

[Mel Flynn]

On Thursday 30 April 2009 21:49:13 Chuck Swiger wrote:
> Hi, VeeJay--
>
> On Apr 30, 2009, at 12:02 PM, VeeJay wrote:
> > Guys, I am not very good on freebsd, its you guys who help me to
> > keeping my
> > server up... I hope you can spare a few minutes to sort this
> > problem...
> >
> > last pid: 19656;  load averages:  1.00,  1.00,
> > 1.00
> > up 2+05:00:12  19:18:47
> > 3049 processes:2 running, 3047 sleeping
> > CPU: 12.5% user,  0.0% nice,  0.1% system,  0.0% interrupt, 87.4% idle
> > Mem: 6253M Active, 3810M Inact, 921M Wired, 128K Cache, 214M Buf,
> > 4683M Free
> > Swap: 32G Total, 32G Free
> >
> > PID USERNAMETHR PRI NICE   SIZERES STATE  C   TIME   WCPU
> > COMMAND
> > 830 mysql  1500  440  1670M   813M ucond  1   0:00 100.00%
> > mysqld
>
> It sure looks like you're running into a system limit with the maximum
> # of threads available to the mysql process.

There's no such limit, see pthread_create(3) and pthread.h:
#define PTHREAD_THREADS_MAX __ULONG_MAX

but the 1500 is suspicious. Suspicious enough to be a MySQL configuration 
value or compile time option. The only way to get to the bottom of it, is to 
watch the number of threads in the mysql process and attach ktrace to it the 
moment it approaches 1500, to see if pthread_create actually does return 
EAGAIN and get a hint as to where. My suspicion however is that the thread 
abstraction of MySQL sets EAGAIN.

A my.cnf certainly would help.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: PostgreSQL 8.3.7 builds incomplete

[Mel Flynn]

On Wednesday 29 April 2009 23:32:37 Mark wrote:

> I should have looked better, instead of just picking the highest-version
>
> server. Still, makes you wonder, if the postgresql84-server port is so
>
> incredibly broken, why even include it?

It's a "repo copy stub" for the forthcoming release, that the maintainer 
really should've marked as broken, until he and PostgreSQL is done with it.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: kernel errors - watchdog timeout

[Mel Flynn]

On Friday 01 May 2009 16:12:50 Seur Bors wrote:

> I'm constantly getting the following repeated in my /var/log/messages:
>
> kernel: re0: watchdog timeout
> kernel: re0: link state changed to DOWN
> kernel: re0: link state changed to UP
>
> This was happening right from the get-go on new hardware running
> 7.1-Release-p4, but only happened infrequently. Apparently as well,
> although everything seems to be working, the server seems to be responding
> very sluggish (over 10 minutes to work with a 1MB file through a Samba
> share, no exageration on the time).
>
> Can someone point me to required reading for these types of networking
> errors?

DIAGNOSTICS
...
 re(4):
 re%d: watchdog timeout  The device has stopped responding to the network,
 or there is a problem with the network connection (cable).

In short: cable or hardware. If you're sure this is not the case, you should 
file a PR and mention you switched out cables and (if possible) the card.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Poor ZFS performance

[Mel Flynn]

On Monday 04 May 2009 02:07:41 nf wrote:
> On Sat, May 2, 2009 at 7:22 PM, Frederique Rijsdijk
>
>  wrote:
> > nf wrote:
> >> 733843456 bytes transferred in 61.124812 secs (12005656 bytes/sec)
> >
> > That is very low. I get about 60MB/sec in this way. Adding bs=1m it'll go
> > up to 240MB/sec even (raidz1 with 4*1TB).
> >
> > Could you show top -S ?
>
> I will, the next time I experience the issue.
>
> I had already rebooted the box, which immediately alleviated the
> issue. I can only presume, at this point, that it seems to have been
> related to the 2gb of allocated/active memory shown by top. I had no
> memory intensive apps running, merely an idle lighttpd, mysqld, and
> rtorrent (which only occupied about 90mb).

There's a thread or 2 on -current list about ZFS arc cache growth, that you 
may or may not be seeing as well.


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

Re: local security scanner for vulnerable common opensource www projects

[Mel Flynn]

On Saturday 02 May 2009 14:50:14 Jeroen Hofstee wrote:
> I tried to find a program which could scan the local filesystem and
> extract a lists of well known
> web projects (yoomla, wordpress etc), extract the installed version
> number and match it against
> a database of known vulnerabilities. Similiar to portaudit, but then for
> the standard scripts users
> install themselves. I was unable to find such a program in the ports.
>
> Does such an utilities exists for FreeBSD ?

Not that I'm aware of and it's hell to write and keep current.
There's 2 good policies for this kind of thing:
- Don't allow any plugins of any kind to be installed via CMS/Gallery software 
etc. and deal with the complaints
- Put them in a seperate jail and make sure client understands he's 
responsible for getting hacked and loosing hours of work by installing unsafe 
plugins.

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

Re: Apache errors.

[Mel Flynn]

On Monday 04 May 2009 15:57:02 Martin Smith wrote:
> Armin Pirkovitsch wrote:
> > Hi!
> >
> > Have you tried to recompile the port from which that library came?
> > (pkg_info -W /usr/local/lib/libavahi-common.so.3 should help you finding
> > the correct port if you do not know which port that is)
> >
> > Armin
> >
> > On Sun 03 May 2009, Jeff Molofee wrote:
> >> Just started getting this.. can anyone tell me how to fix it?
> >>
> >> Performing sanity check on apache22 configuration:
> >> httpd: Syntax error on line 104 of /usr/local/etc/apache22/httpd.conf:
> >> Cannot load /usr/local/libexec/apache22/mod_dnssd.so into server:
> >> /usr/local/lib/libavahi-common.so.3: Undefined symbol
> >> "libintl_bindtextdomain"
>
> Now I have the same problem so I went to recompile avahi-app, did rmconfig
> and then make config, it said no user config options, or something very
> similar.
> So where to now?

Rebuild gettext, then avahi. The bindtextdomain is from gettext. If that don't 
help, disable mod_dnssd and ping avahi/apache maintainers. Most users don't 
need this module. It is used to advertise your http server to the local 
network, just in case you have it on another port.

gnome-user-share and as such x11/gnome2 finds it necessary to install this 
stuff. Only way to get rid of it, seems to be to install gnome2-lite and then 
pick what extras you do want.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Upgrading jails to 7.2-RELEASE with freebsd-update

[Mel Flynn]

On Tuesday 05 May 2009 17:37:42 Martin Turgeon wrote:
> Hi everyone,
>
> It's the first time I'm using freebsd-update to upgrade my system to a
> new release. I just upgraded the base system from 7.1-RELEASE to
> 7.2-RELEASE and everything went fine. I now wanted to upgrade my jails
> to the new release but freebsd-update is telling me that's already
> updated...
>
> freebsd-update -b /usr/jail/mysql/ upgrade -r 7.2-RELEASE
> freebsd-update: Cannot upgrade from 7.2-RELEASE to itself

Should be able to trick it using:
env UNAME_r=7.1-RELEASE-p5 freebsd-update -b /path/to/jail

(based on a quick source scan, untested).
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Is it necessary to generate a new SSL request each year?

[Mel Flynn]

On Thursday 30 April 2009 01:05:50 Robert Huff wrote:
> Dan Nelson writes:
> >  > When buying a new SSL cert, I've been generating a new
> >  > request each year...  I am just about to buy another and it
> >  > occurred to me that I'm entering the same info.  Do I really
> >  > need a new request file each year?  Or can I just reuse the
> >  > same one (presuming none of the info has changed.)
> >
> >  You can reuse the old one.
>
>   I'm not an expert on these, but it was my understanding that
> certificates carry in internal "expiration date" after which the
> application may respond as it pleases.

Yes, but the *request* does not.
Also, if using openssl, just set the defaults in /etc/ssl/openssl.cnf to your 
values, so you can enter through the questions.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Where am I wasting resources? How to fix this problem?

[Mel Flynn]

On Tuesday 28 April 2009 14:21:45 VeeJay wrote:
> Hello Peter
>
> Thanks... I have tried the values but even after rebooting, I am still
> getting the same old values as:
>
> server1# sysctl -a | grep maxdsiz
> compat.ia32.maxdsiz: 536870912
>
> server1# sysctl -a | grep maxssiz
> compat.ia32.maxssiz: 67108864

It's a kenv(1) variable. Either way I don't think it's the problem. mysqld 
uses 1500 threads and many apache processes waiting for mysql to reply. You 
should figure out why that is, cause that sounds like a query that's holding a 
table lock and needing to sort the intermediate result set, stalling all other 
queries.
If you really have ~1500 connections and consider that normal operation, then 
you may need more kernel memory.
amd64 doesn't have a process memory limit (feature or bug I'm undecided on), 
so you can delete those.
Instead set:
vm.kmem_size_max="1024M"
vm.kmem_size="1024M"

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

Re: Where am I wasting resources? How to fix this problem?

[Mel Flynn]

On Tuesday 28 April 2009 14:29:42 Wojciech Puchar wrote:
> [r...@wojtek ~]# sysctl -a |grep maxpr
> kern.maxproc: 5266
> kern.maxprocperuid: 4739
>
>
> i don't know if there is limit

Not relevant. See pthread_create(): EAGAIN is returned for lack of kernel 
memory or going over PHTREAD_THREADS_MAX which is ULONG_MAX. 1500 threads 
isn't even close to USHORT_MAX.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: bsnmpd vs net-snmp

[Mel Flynn]

On Monday 27 April 2009 20:19:33 Anton Yuzhaninov wrote:
> On Mon, 27 Apr 2009 12:28:01 -0400, Maxim Khitrov wrote:
> MK> I'm setting up a firewall and would like to monitor certain system
> MK> parameters like network, cpu, and memory usage. SNMP is an obvious
> MK> choice to do the monitoring and I'm planning to set up rrdtool to
> MK> generate graphs of captured data. The question is what SNMP agent to
> MK> use. I found net-snmp and bsnmpd (which is included in the base
> MK> system). Has anyone here used both implementations, and if so, what
> MK> are the basic differences?
>
> main difference is the set of supported MIBs.
>
> In general net-snmp supports more MIBs than bsnmpd.
>
> E. g. BEGEMOT-PF-MIB supported only by bsnmpd and useful for monitoring
> pf(4), UCD-SNMP-MIB supported only by net-snmp and useful for monitoring
> CPU load (ssCpuRaw* counters).

There is the GoC 2008 project:
% cat /usr/ports/net-mgmt/bsnmp-ucd/pkg-descr
bsnmp-ucd is a module for bsnmpd which allows you to get memory, load
average, cpu utilization and other system statistics. It implements parts
of UCD-SNMP-MIB for this.

WWW: http://bsnmp-ucd.googlecode.com/
Author: Mikolaj Golub 

Maybe if more people started testing/using this, it could some day be in base.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: portmaster -a on a live server

[Mel Flynn]

On Sunday 26 April 2009 01:12:48 Tom Worster wrote:

> thanks for the tip, mel. i got rid of the ports involved and reinstalled
> with WITHOUT_X11=yes and the install was faster and things are a lot
> tidier.
>
> i had no idea that i ought to be configuring port builds with env vars. is
> there documentation anywhere so i find out about these options in gneral?

/usr/ports/Mk/bsd.port.mk comments section and /usr/ports/Mk/bsd.*.mk for 
specific collections, like bsd.databases.mk lets you set a default version for 
various database engines, for which multiple versions exist in the port.

It does not have to be env vars, make.conf works as well, but you will forget 
in 3 months you have set it, so be careful what you choose and where you put 
it.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: portmaster -a on a live server

[Mel Flynn]

On Wednesday 22 April 2009 17:37:06 Tom Worster wrote:

> by the by, on my test machine i ended up with python installed. seems to be
> because i needed php5-gd which now depends on python. all for some simple
> freetype2 calls.

Wrong assumption. php-gd doesn't depend on python at all. devel/apr does, so 
you've built php with apache module and that pulls in python OR you didn't set 
WITHOUT_X11=yes when building, so libxcb is pulled in which uses python:

% sudo /stable/root/bin/finddep.php graphics/php5-gd lang/python26
/usr/ports/x11/xcb-proto: /usr/local/bin/python2.6 => /usr/ports/lang/python26
/usr/ports/x11/xcb-proto: /usr/local/bin/python2.6 => /usr/ports/lang/python26
/usr/ports/x11/libxcb: /usr/local/bin/python2.6 => /usr/ports/lang/python26
/usr/ports/x11/libxcb: /usr/local/bin/python2.6 => /usr/ports/lang/python26
/usr/ports/devel/apr: /usr/local/bin/python2.6 => /usr/ports/lang/python26

% sudo env WITHOUT_X11=yes /stable/root/bin/finddep.php graphics/php5-gd 
lang/python26
/usr/ports/devel/apr: /usr/local/bin/python2.6 => /usr/ports/lang/python26

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

Re: libc or OpenSSL patches break ssh?

[Mel Flynn]

On Wednesday 22 April 2009 20:11:09 Jake Evans wrote:
> Am running 6.4-PRERELEASE.
>
> Received FreeBSD-SA-09:07.libc / FreeBSD-SA-09:08.openssl notifications
> today, applied applicable patches correctly.
>
> However, now when anyone attempts to connect to my server via SSH, the
> connection is closed after they enter their login & password.
>
> /var/log/messages says the following when they try to login:
>
> Apr 22 12:53:12 x sshd[75505]: fatal: openpty returns device for which
> ttyname fails.
> Apr 22 12:53:12 x sshd[75505]: error: chown  0 0 failed: No such file or
> directory
> Apr 22 12:53:12 x sshd[75505]: error: chmod  0666 failed: No such file or
> directory

You did a full buildworld and restarted sshd, or just updated openssl?
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Preventing ntpd from adjusting time (backwards)

[Mel Flynn]

On Tuesday 21 April 2009 21:07:34 Chuck Swiger wrote:

>  Try contacting your ISP for nearby NTP
> sources,

Anchorage, AK,  is special that way. I'll check with ACS if they have one, but 
if they don't, even traffic to the local competitor (GCI) goes through 
Seattle.

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

Re: Preventing ntpd from adjusting time (backwards)

[Mel Flynn]

On Tuesday 21 April 2009 19:43:30 Chuck Swiger wrote:
> Hi, Mel--
>
> On Apr 21, 2009, at 2:06 AM, Mel Flynn wrote:
> > Some coarse reading of ntpd(8) and ntp.conf(5) doesn't lead me to
> > believe it's
> > possible to make ntpd *not* adjust the time. With adjust I don't
> > mean the skew
> > operation, but really change the time.
>
> Perhaps I've missed it elsewhere in this thread, but I don't believe
> anyone actually answered the original question, which would be to use:
>
> -x, --slew
>Slew up to 600 seconds.
>
>Normally, the time is slewed if the offset is less than the
> step
>threshold,  which is 128 ms by default, and stepped if
> above the
>threshold.  This option sets the threshold to 600  s,
> which  is
>well  within  the  accuracy  window  to  set the clock
> manually.

Hmm, that might work. Thanks!


> It should be surprising that your clock would jump by 6 seconds.  Do
> you have adequate upstream timesources (ie, at least 4) configured, is
> your local HW clock busted somehow, or are you doing something odd
> with power-savings mode or running in a VM or something...?

One timesource, shared on local network, this machine is a client of the 
gateway, which uses only one source (ntp.alaska.edu, which is geographically 
10 minutes by car but thanks to Alaska bad peering, we go through Seattle 
anyway). I checked the logs, that machine didn't step at all that day (or any 
other day, as far as my logs go). It always happens after reboot, as Matthew 
indicated. No VM, no power-savings. The only odd things are Hyperthreading and 
the reboot.

Well, I abuse the machine quite heavily from time to time, but then I'd expect 
the clock to be slow, not 6 seconds faster.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Preventing ntpd from adjusting time (backwards)

[Mel Flynn]

On Tuesday 21 April 2009 20:29:18 Chuck Swiger wrote:
> On Apr 21, 2009, at 11:23 AM, Mel Flynn wrote:
> > Now I'm also wondering how ntpd handles securelevel 2.
>
> "man init" suggests that stepping the clock by more than a second is
> disallowed:

yes, so does it bail or retry till skew wins over the failed steps?
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Sudden /libexec/ld-elf.so.1: Shared object "libintl.so.8" not found, required by errors

[Mel Flynn]

On Tuesday 21 April 2009 19:33:37 Agus wrote:
> 2009/4/14 Ruben de Groot :
> > On Fri, Apr 10, 2009 at 08:40:51PM -0300, Agus typed:
> >> Hi guys,
> >>
> >> Yesterday i suddenly start receiving this errors... first i noticed it
> >> cause i couldnt login and bash threw it.. then su...
> >>
> >> /libexec/ld-elf.so.1: Shared object "libintl.so.8" not found, required
> >> by "-su"
> >>
> >> Then i remove bash and change to tcsh and was "ok"... but then again,
> >> this one when using sendmail..
> >>
> >> /libexec/ld-elf.so.1: Shared object "libsasl2.so.2" not found,
> >> required by "send-mail"
> >>
> >>
> >> I didnt update anything... and it was all running fine for months...
> >> what can it be? the files i have are..
> >>
> >> -rwxr-xr-x  1 root  wheel92K Mar 11 15:26 libsasl2.so.2
> >> lrwxr-xr-x  1 root  wheel13B Mar 11 15:26 libsasl2.so ->
> >> libsasl2.so.2
> >>
> >> -r--r--r--  1 root  wheel39K Aug 23  2008 libintl.so.8
> >> lrwxr-xr-x  1 root  wheel12B Aug 23  2008 libintl.so -> libintl.so.8
> >
> > What is the output of "ldconfig -r" ?
>
> Sorry for the delay.. was too busyy...
>
> No output... just this
>
> ldconfig -r
> /var/run/ld-elf.so.hints:
>   search directories:

Possible causes:
/var/run/ld-elf.so.hints is reduced to 0 by program foo, operator bar or 
hacker baz.
You or operator bar ran ldconfig -s without arguments.

/rescue/ldconfig /lib
/etc/rc.d/ldconfig start

Should get you back up and running.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Preventing ntpd from adjusting time (backwards)

[Mel Flynn]

On Tuesday 21 April 2009 19:31:33 RW wrote:
> On Tue, 21 Apr 2009 16:43:32 +0200
>
> Mel Flynn  wrote:
> > On Tuesday 21 April 2009 16:20:52 RW wrote:
> > > The bottom line though, is that ntpdate_enable=yes solves the
> > > problem entirely, since the real problem is not the step, but the
> > > fact that it happens in the background, and after a delay.
> >
> > Care to expand on that? Dovecot won't stop if root issues a date
> > command that sets time to the past, for example?
>
> I was assuming that since you're running ntpd you wouldn't be doing
> that.

Right, then this works because ntpdate is started before dovecot in rcorder, 
like Tim Judd said else in thread.

> > > ntpdate may be deprecated, but it's been deprecated for years, and I
> > > doubt it will go away until ntpd fully replaces it's functionality.
> > > ntpd -gq can replace ntpdate in a crontab, but ntpd -gqn doesn't
> > > really replace ntpdate -b in the boot-sequence.
> >
> > I'm actually counting on it to be gone in 8.0.
>
> Is that official?

Nope, but 3 major releases of mourning should be enough ya think? 
Maybe not, given the problems above.

Now I'm also wondering how ntpd handles securelevel 2.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Compiling FreeBSD with GCC 4.3+

[Mel Flynn]

On Tuesday 21 April 2009 17:37:50 David Naylor wrote:
> On Tuesday 21 April 2009 10:32:04 Mel Flynn wrote:
> > Hi David,
> >
> > On Monday 20 April 2009 21:48:39 David Naylor wrote:
> > > There has been an article recently published by phoronix
> > > (http://www.phoronix.com/scan.php?page=article&item=pcbsd_vs_kubuntu&nu
> > >m= 1) that compares PC-BSD to Kubuntu.  Kubuntu uses GCC 4.3.3 compared
> > > to FreeBSD's GCC 4.2.2.  There is a considerable performance difference
> > > between the two OS's, the article contributes this difference to the
> > > compiler.
> >
> > Nice shot in the dark, since except the calculations a lot of these are
> > influenced by "journaled FS vs stock UFS".
>
> I know, benchmarking anything but the simplest things are influenced by too
> many factors.  Pity it doesn't provide an unbiased comparison of FreeBSD
> and Linux.

That and comparing apples and pears as default configured fruit, don't usually 
work well. Of course it appeals to the end user "which fruit is healthier".

> > > In order to check if this is so (and to get the speed improvements of
> > > GCC 4.3+) one needs to compile the ports (and preferable world/kernel
> > > as well) with GCC 4.3+.
> >
> > It's license is incompatible with world/kernel.
>
> What type of incompatibility.  I know FreeBSD has reservations about GPLv3
> (I personally don't understand why everyone cannot be friends and use BSD
> Licenses).  So is this a policy incompatibility or a legal one (i.e. would
> it be 'illegal' for me to use GCC 4.3+ to compile world/kernel, as an
> end-user/consumer of FreeBSD).  I assume the same discussion applies to
> binutils.

Policy. Only legal issue in FreeBSD for the end user is WITH_IDEA.

> > That said, install
> > lang/gcc43 and set CC/CXX for ports. World/kernel would be a lot harder.
> > Maybe setting WITHOUT_GCC in /etc/src.conf and setting CC/CXX would work,
> > but there's quite a few modifications to gcc that aren't in ports
> > lang/gcc, so I have my doubts.
>
> I suppose it would be nice if there was an easy way to use an out-of-source
> compiler in FreeBSD.  Like set PORTS_COMPILER=gcc43 and the port will
> installed and used... One may have dreams.

cat <<'EOF' >> /etc/make.conf
.if !empty(.CURDIR:M/usr/ports/*)
CC=/usr/local/bin/gcc43
CXX=/usr/local/bin/g++43
.endif
EOF

Pretty close, huh?

> > > Is there an easy way to set this up and does anyone know the
> > > compatibility of world/kernel/ports with GCC 4.3+?
> > >
> > > Also has anyone tried this and benchmarked the result?
> >
> > Not me, but be sure to stick around for the new non-gcc compiler coming
> > to a FreeBSD near you. And with the work done by Marcel Molenaar on
> > gpart, hopefully we can have ZFS and gjournal as choices in the
> > installer.
>
> You mean llvm, waiting patiently.  I suppose my suggestion above will
> become even more important (at least for compiling ports) since it will be
> a while till llvm has decent c++ support.

Yeah, I don't know how that's gonna work if llvm is ready for base, but no 
c++. I guess we'll have to sit out g++ 4.2 for a while. If you're in the 
position to do so, I'd do their benchmarks with ZFS and see how much 
difference that already makes.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Sorting out owner and group permissions...

[Mel Flynn]

On Tuesday 21 April 2009 15:13:47 Mister Olli wrote:

> no does not work, since using SSH / SFTP does not involve starting a
> shell. so umask settings don't work.

Then you're using the wrong system for the task. The OS can't make assumptions 
about "what the ownership/modes of a file should really be, if an application 
is telling it they should be different".
This is why more mature FTP daemons allow modes/ownerships to be set on 
upload.

The OS already:
- gives a new file group of the containing directory so it is easy to create 
"shared files" in a "shared directory"
- has a default umask that is world readable
- allows changing a users umask

The application (sftp) overrides all this and now you're expecting the OS to 
override that again. Don't think so ;)
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Preventing ntpd from adjusting time (backwards)

[Mel Flynn]

On Tuesday 21 April 2009 16:11:52 Tim Judd wrote:
> On Tue, Apr 21, 2009 at 3:39 AM, Matthew Seaman <
>
> m.sea...@infracaninophile.co.uk> wrote:
> > Mel Flynn wrote:
> > > Hi,
> > >
> > > Some coarse reading of ntpd(8) and ntp.conf(5) doesn't lead me to
> > > believe
> >
> > it's
> >
> > > possible to make ntpd *not* adjust the time. With adjust I don't mean
> > > the
> >
> > skew
> >
> > > operation, but really change the time. Backwards is my primary concern
> >
> > but if
> >
> > > it can be turned off completely it's fine with me.
> > >
> > > Reason being dovecot bailing out when this happens:
> > > Apr  1 16:18:26 squish ntpd[1353]: time reset -6.711955 s
> > >
> > > Apr  1 16:18:26 mx1 dovecot: Fatal: Time just moved backwards by 6
> >
> > seconds.
> >
> > > This might cause a lot of problems, so I'll just kill myself now.
> > > http://wiki.dovecot.org/TimeMovedBackwards
> >
> > This seems to be a bete-noir for the dovecot developer.  Whatever, it is
> > a royal pain in the arse, as my mailserver always steps the time
> > backwards on each reboot, and then dovecot does it's dying swan thing.
> >
> > Three choices:
> >
> >  * Don't run 'ntpd -g' as the documentation tells you is the modern and
> >accepted method.  Instead, run 'ntpdate' as a separate process and
> >run 'ntpd' without the '-g' flag.
> >
> >  * Don't run dovecot.  Other IMAP servers do not suffer in the same
> >way.
> >
> >  * Put up with it.  Avoid reboots, and swear at all concerned any time
> >you really do have to reboot.
> >
> >Cheers,
> >
> >Matthew
>
> How about adding ntpdate's provided string to dovecot's required string in
> their respective startup rc.d scripts?  This forces dovecot to wait until
> ntpdate has been called, assuming time has actually been set/changed, then
> dovecot may start?

That could work, if ntpd_sync_on_start would actually sync on start. Trying 
not to enable ntpdate unless I really have to, since I expect it to be gone in 
8.0.
Still, there's a chance ntp steps backwards during the runtime, but then my 
CMOS battery probably needs replacing anyway.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Preventing ntpd from adjusting time (backwards)

[Mel Flynn]

On Tuesday 21 April 2009 16:20:52 RW wrote:

> The bottom line though, is that ntpdate_enable=yes solves the problem
> entirely, since the real problem is not the step, but the fact that it
> happens in the background, and after a delay.

Care to expand on that? Dovecot won't stop if root issues a date command that 
sets time to the past, for example?

> ntpdate may be deprecated, but it's been deprecated for years, and I
> doubt it will go away until ntpd fully replaces it's functionality.
> ntpd -gq can replace ntpdate in a crontab, but ntpd -gqn doesn't really
> replace ntpdate -b in the boot-sequence.

I'm actually counting on it to be gone in 8.0.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Sorting out owner and group permissions...

[Mel Flynn]

On Tuesday 21 April 2009 11:17:40 Mister Olli wrote:
> hi,
>
> I have the same problem on some fileservers I do the administration for.
> But in my case the users send the files via SSH to the server.
>
> A solution for this, based on some OS mechanism would be really
> great :-)

umask(1).
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: named fails to start on boot on FreeBSD 6.1, complains about libxml2.so.5

[Mel Flynn]

On Tuesday 21 April 2009 14:21:12 Manolis Kiagias wrote:
> Mark Stosberg wrote:
> >>> We had problem with "named" starting on boot on a FreeBSD 6.1 server,
> >>> managed
> >>> by /etc/rc.conf.
> >>>
> >>> The startup script failed with errors about shared library "libm.so.2"
> >>> failing
> >>> to load because of something related to libxml2.so.5.
> >>>
> >>> Later, when I then tried starting it via /etc/rc.d/named it worked
> >>> fine.
> >>>
> >>> I include the following "ldd" output in case it's helpful. What could
> >>> possibly
> >>> be the issue here?
> >>>
> >>>Mark
> >>>
> >>> ###
> >>>
> >>> # ldd /usr/sbin/named
> >>> /usr/sbin/named:
> >>>libcrypto.so.4 => /lib/libcrypto.so.4 (0x281ff000)
> >>>libxml2.so.5 => /usr/local/lib/libxml2.so.5 (0x282f1000)
> >>>libiconv.so.3 => /usr/local/lib/libiconv.so.3 (0x283ef000)
> >>>libm.so.4 => /lib/libm.so.4 (0x284e3000)
> >>>libc.so.6 => /lib/libc.so.6 (0x284f9000)
> >>>libm.so.2 => /usr/local/lib/compat/libm.so.2 (0x285d1000)
> >>
> >> I also see to LOCAL libraries in it.  named is part of base, unless you
> >> compiled and installed the port version and maybe told it to overwrite
> >> the base.
> >>
> >> None of this adds up.
> >>
> >> %ldd /usr/sbin/named
> >> /usr/sbin/named:
> >> libcrypto.so.5 => /lib/libcrypto.so.5 (0x281fe000)
> >> libthr.so.3 => /lib/libthr.so.3 (0x28357000)
> >> libc.so.7 => /lib/libc.so.7 (0x2836a000)
> >>
> >> There's a named on 7.1p4
> >
> > Thanks for the response. I've now compared this named 'ldd' outfit to
> > another 6.1 install we have that also runs named. It has the exact same
> > file size and version, but slightly different ldd output:
> >
> > -- from the second machine with FreeBSD 6.1
> > # ldd /usr/sbin/named
> > /usr/sbin/named:
> > libcrypto.so.4 => /lib/libcrypto.so.4 (0x281ff000)
> > libxml2.so.5 => /usr/local/lib/libxml2.so.5 (0x282f1000)
> > libz.so.3 => /lib/libz.so.3 (0x283ff000)
> > libiconv.so.3 => /usr/local/lib/libiconv.so.3 (0x2840f000)
> > libm.so.4 => /lib/libm.so.4 (0x284fc000)
> > libc.so.6 => /lib/libc.so.6 (0x28512000)
> >
> > 
> >
> > XML is still there, but the mention of libm.so no longer points into
> > /usr/local/lib/compat This other FreeBSD user also found the libxml link:
> > https://lists.isc.org/pipermail/bind-users/2008-November/073929.html
> >
> > I also checked "/etc/make.conf" on both machines. They mentioned X11_BASE
> > and Perl... nothing about XML. However, these machines have evolved some
> > over time. Perhaps something with there in the past.
> >
> > It sounds like advisable paths forward include re-compiling or
> > re-installing named.
> >
> > Mark
>
> As it is already pointed out, you probably have a bind version installed
> from ports.
>
> Try:
>
> pkg_info -Ix bind
>
> and check if it produces anything. On a 6.4 box, the base system bind
> shows: ldd /usr/sbin/named
> /usr/sbin/named:
> libcrypto.so.4 => /lib/libcrypto.so.4 (0x80077c000)
> libc.so.6 => /lib/libc.so.6 (0x8009c3000)
>
> However, the port dns/bind96 for example:
>
> # pwd
> /usr/ports/dns/bind96
> # make run-depends-list
> /usr/ports/textproc/libxml2
>
> which looks suspiciously similar to your dependency there.

Agreed. Bind 9.5 and higher from ports has "XML statistics" support. That 
explains the xml and iconv. ldd -a /usr/sbin/named should show you which one 
wants libm.so.2 which is from the 4.x days.
If you don't need these statistics, I would suggest turning them off through 
make config.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Preventing ntpd from adjusting time (backwards)

[Mel Flynn]

On Tuesday 21 April 2009 11:39:32 Matthew Seaman wrote:
> Mel Flynn wrote:
> > Hi,
> >
> > Some coarse reading of ntpd(8) and ntp.conf(5) doesn't lead me to believe
> > it's possible to make ntpd *not* adjust the time. With adjust I don't
> > mean the skew operation, but really change the time. Backwards is my
> > primary concern but if it can be turned off completely it's fine with me.
> >
> > Reason being dovecot bailing out when this happens:
> > Apr  1 16:18:26 squish ntpd[1353]: time reset -6.711955 s
> >
> > Apr  1 16:18:26 mx1 dovecot: Fatal: Time just moved backwards by 6
> > seconds. This might cause a lot of problems, so I'll just kill myself
> > now. http://wiki.dovecot.org/TimeMovedBackwards
>
> This seems to be a bete-noir for the dovecot developer.  Whatever, it is
> a royal pain in the arse, as my mailserver always steps the time
> backwards on each reboot, and then dovecot does it's dying swan thing.
>
> Three choices:
>
>   * Don't run 'ntpd -g' as the documentation tells you is the modern and
> accepted method.  Instead, run 'ntpdate' as a separate process and
> run 'ntpd' without the '-g' flag.

Hmm, isc sure knows how to abstract something as simple as command line 
options into several levels. From the source, -q activates mode_ntpdate which 
is one path for time reset. Since not using that, it's not that path.

The other codepath, has 4 possibles, 2 of which relating to step-in and step-
out, which I could increase to values that are less likely to cause a step. 
Would be worthwhile if there aren't 2 other possibilities which most likely 
cause the "step back after reboot" syndrome:
 * In S_NSET state an initial frequency correction is
 * not available, usually because the frequency file has
 * not yet been written. Since the time is outside the
 * step threshold, the clock is stepped. The frequency
 * will be set directly following the stepout interval.
 *
 * In S_FSET state the initial frequency has been set
 * from the frequency file. Since the time is outside
 * the step threshold, the clock is stepped immediately,
 * rather than after the stepout interval. Guys get
 * nervous if it takes 17 minutes to set the clock for
 * the first time.


>   * Don't run dovecot.  Other IMAP servers do not suffer in the same
> way.

Since this is the only "issue" I have with dovecot, I don't think so. ;)

>   * Put up with it.  Avoid reboots, and swear at all concerned any time
> you really do have to reboot.

* Patch ntpd (most likely option now)
* Do something smart with init, restarting dovecot when this happens.

To be continued (on TODO somewhere on the bottom :/). Thanks for the input 
Matt.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Preventing ntpd from adjusting time (backwards)

[Mel Flynn]

Hi,

Some coarse reading of ntpd(8) and ntp.conf(5) doesn't lead me to believe it's 
possible to make ntpd *not* adjust the time. With adjust I don't mean the skew 
operation, but really change the time. Backwards is my primary concern but if 
it can be turned off completely it's fine with me.

Reason being dovecot bailing out when this happens:
Apr  1 16:18:26 squish ntpd[1353]: time reset -6.711955 s

Apr  1 16:18:26 mx1 dovecot: Fatal: Time just moved backwards by 6 seconds. 
This might cause a lot of problems, so I'll just kill myself now. 
http://wiki.dovecot.org/TimeMovedBackwards

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

Re: portmaster -a on a live server

[Mel Flynn]

On Tuesday 21 April 2009 01:38:26 Tom Worster wrote:

> portmaster -a -x mysql-server
> portmaster mysql-server
> reboot

No no no. /usr/local/etc/rc.d/mysql-server start. Reboot is for kernel 
upgrades. And never use reboot unless in single user mode, cause reboot is 
really fast reboot: it doesn't stop services nicely. Use shutdown -r now.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Compiling FreeBSD with GCC 4.3+

[Mel Flynn]

Hi David,

On Monday 20 April 2009 21:48:39 David Naylor wrote:

> There has been an article recently published by phoronix
> (http://www.phoronix.com/scan.php?page=article&item=pcbsd_vs_kubuntu&num=1)
> that compares PC-BSD to Kubuntu.  Kubuntu uses GCC 4.3.3 compared to
> FreeBSD's GCC 4.2.2.  There is a considerable performance difference
> between the two OS's, the article contributes this difference to the
> compiler.

Nice shot in the dark, since except the calculations a lot of these are 
influenced by "journaled FS vs stock UFS".

> In order to check if this is so (and to get the speed improvements of GCC
> 4.3+) one needs to compile the ports (and preferable world/kernel as well)
> with GCC 4.3+.

It's license is incompatible with world/kernel. That said, install lang/gcc43 
and set CC/CXX for ports. World/kernel would be a lot harder. Maybe setting 
WITHOUT_GCC in /etc/src.conf and setting CC/CXX would work, but there's quite 
a few modifications to gcc that aren't in ports lang/gcc, so I have my doubts.

> Is there an easy way to set this up and does anyone know the compatibility
> of world/kernel/ports with GCC 4.3+?
>
> Also has anyone tried this and benchmarked the result?

Not me, but be sure to stick around for the new non-gcc compiler coming to a 
FreeBSD near you. And with the work done by Marcel Molenaar on gpart, 
hopefully we can have ZFS and gjournal as choices in the installer.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: named fails to start on boot on FreeBSD 6.1, complains about libxml2.so.5

[Mel Flynn]

On Monday 20 April 2009 23:48:47 Tim Judd wrote:

> > I include the following "ldd" output in case it's helpful. What could
> > possibly
> > be the issue here?
> >
> >Mark
> >
> > ###
> >
> > # ldd /usr/sbin/named
> > /usr/sbin/named:
> >libcrypto.so.4 => /lib/libcrypto.so.4 (0x281ff000)
> >libxml2.so.5 => /usr/local/lib/libxml2.so.5 (0x282f1000)
> >libiconv.so.3 => /usr/local/lib/libiconv.so.3 (0x283ef000)
> >libm.so.4 => /lib/libm.so.4 (0x284e3000)
> >libc.so.6 => /lib/libc.so.6 (0x284f9000)
> >libm.so.2 => /usr/local/lib/compat/libm.so.2 (0x285d1000)

As far as I know, named never supported or needed xml. I would check if this 
really is named and for rootkits while you're at it. Either that, or you have 
LDFLAGS set in your /etc/make.conf that make everything link with these 
libraries.
Backup data and configs and reinstall from CD if you can't find a sane answer 
for this.

> libm.so.2, in /usr/local/lib/compat/libm.so.2 is for Linux (and might I
> add a possibly older version).

No, it's from compat4x. Linux would be in /compat/linux/lib.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: mount_cd9660 - /dev/md0: Invalid Arguement

[Mel Flynn]

On Monday 20 April 2009 18:32:58 Евгений Л wrote:
> Hello everyone, I am trying to mount an ISO image (which was converted with
> with help from ccd2iso tool)

Tried that too, I think that tool is broken, cause known working ISOs work 
with mdconfig as you described :/
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Dump | Restore

[Mel Flynn]

On Monday 20 April 2009 14:59:55 cpghost wrote:
> On Mon, Apr 20, 2009 at 12:46:05PM +0200, Wojciech Puchar wrote:
> > use rsh not ssh unless you really need encryption.
>
> Sure, you *could* do that, but be sure to encrypt *and* sign the
> backup stream beforehand, e.g. using openssl or gnupg... And even
> then, anyone sniffing that poorly encrypted (at layer 2) wireless LAN
> connection could still hijack the password, log into the backup host,
> and delete or corrupt the (encrypted) dump files.
>
> Perhaps it's better to use ssh anyway, even for encrypted and signed
> dump files. Creating and transfering a couple of key files to the
> clients and backup host and using ssh(1) is not hard. Really not. ;-)

But doesn't use full network capacity. Closed circuit LAN's (yes, they still 
do exist) don't need ssh, but a level 0 dump of several TB of data does need 
full lan speed.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: snd-hda no sound whatsoever

[Mel Flynn]

On Monday 20 April 2009 03:11:08 Eitan Adler wrote:
> > If it still doesn't work put a verbose boot dmesg of snd_hda and pcm
> > somewhere (see the man page of snd_hda) and ask on the
> > freebsd-multimedia@ mailing list.
>
> #sysctl hw.snd.default_unit=0
> Works - thanks.  Now - to make this change stick I add it to
> /boot/loader.conf ?

As said in sound(4) manpage, that I already mentioned in previous mail, it's a 
sysctl and as such belongs in /etc/sysctl.conf.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Upgrading from 6.3 to 7.1 -- how dangerous?

[Mel Flynn]

On Sunday 19 April 2009 19:42:39 Doug Hardie wrote:
> While most of the update process is
> waiting for things to complete, mergemaster requires a lot of
> responses to a ton of questions about updates to configuration files.
> The vast majority of those will be to install the new version.
> However, there are some where you really need to review the changes
> and make sure your unique configuration gets carried over into the new
> files.  Its really easy to get into the "i" mode and skip right
> through some of those.  The recovery from that will be painful.

-iU is your friend. Only painful if you have custom rc.d files that depend on 
functionality in other rc.d files.

-i => install files that do not exist yet
-U => upgrade files that you have not modified

It's a blessing and reduces much of mergemaster's operator attention.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Upgrading from 6.3 to 7.1 -- how dangerous?

[Mel Flynn]

On Sunday 19 April 2009 19:06:55 John Almberg wrote:

> I've thought about setting up a dummy server, just to practice on. Is
> this a good idea?

If this is to get a feel for the upgrade process, sure. But if the hardware is 
different, you won't be much wiser for the production box in question. I'm not 
upgrading a 6.4 for a client, because a previous upgrade to 7.0-STABLE hosed 
the data on the disk, due to an ata-regression. This is fixed most likely in 
7.1-STABLE, but I'm not gambling just yet. Regressions are rare in FreeBSD 
(technically it's not a regression, the original 6.3 ran in UDMA mode and the 
7.0 thought it could handle SATA, but it didn't), but they do happen.

Either way with 7.2 around the corner, I would wait till that's out or do the 
testing with 7.2-PRERELEASE, since it's getting an awful lot of testing now.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Fetchmail problem

[Mel Flynn]

On Monday 20 April 2009 09:12:32 Annelise Anderson wrote:
> Fbsd1 wrote:
> > Annelise Anderson wrote:
> >> I am trying to use fetchmail on FreeBSD but get the message:
> >>
> >> fetchmail: SMTP< 451 4.1.8 Domain of sender address
> >> owner-freebsd-questi...@freebsd.org does not resolve

The question isn't whether you can resolve it, but whether your mailserver 
can. Check your mail log for the smtp server. If the mailserver is chrooted, 
check for $chrootdir/etc/resolv.conf.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: How to create a livecd without compiling everything.

[Mel Flynn]

On Monday 20 April 2009 08:23:42 Peter Wang wrote:

> I want to create a customized freebsd livecd, and i have read quite a
> lot guides about how to do that. but the problem is: most of these
> need make buildworld, make buildkernel ... I think it's hard for my
> notebook do that. so is there a simpler/quicker way in which i can
> create a livecd, what i want to be included in the livecd are the base
> freebsd system plus some customized packages/config files.

If you already built them, skip buildworld/kernel in whichever guide you're 
using. Can't skip the install part tho.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: About fetchmail

[Mel Flynn]

On Monday 20 April 2009 12:20:15 张臻 wrote:
> Today when I used fetchmail to get my mails, it suddenly said that if
> failed to connect to localhost:25 and failed to send the mail to myself,
> does anyone know why?
> The information that fetch out put
> reading message x...@xxx.xx:1 of 3 (2223 octets)
> Trying to connect to 127.0.0.1/25...connection failed.
> fetchmail: connection to localhost:smtp [127.0.0.1/25] failed: Connection

sendmail/postfix/whatever_mta not running.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Fan Speed

[Mel Flynn]

On Sunday 19 April 2009 03:06:07 Christopher Chambers wrote:

> My laptop runs a little hot. Is it possible to increase the default fan
> speed? I'm running version 7.1 on an acer.

Active cooling is under control of ACPI, typically not exposed to non-windows 
operating systems:
hw.acpi.thermal.tz0.active: -1

Value other then -1 you might actually be able to change it.

But...your best 15USD spent will be a laptop cooling pad, best one powered by 
the USB port. Some lappies just can't be fixed with fanspeed, because of bad 
design and a cooling pad will take at least 5C off.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: lightweight webserver that can run php

[Mel Flynn]

On Saturday 18 April 2009 16:34:52 Rodrigo Gonzalez wrote:
> As a fastcgi not cgi, there is a differenceimportant one.
> for cgi request a new php process is started for a request.
> fastcgi is php listening for network requests and process can be reused.

And the more important difference with the php module for apache, is 
php_value/php_flag directives: per host/dir/location configuratioion of php. 
Which is much harder to do with fcgi.

Something to consider /before/ migrating if this affects you.
-- 
Mel
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"