Re: bind round robin

2006-09-19 Thread Oliver Brandmueller 
Hi.

On Tue, Sep 19, 2006 at 02:00:23PM +0800, pinoyskull wrote:
> One of my client's domain has multiple IPs for redundancy, i configured 
> his www as such
> 
> wwwIN A   10.10.10.10
> wwwIN A   192.168.0.10
> 
> Is there a way to prioritize 10.10.10.10 over 192.168.0.10? How do i 
> configure it?

DNS round robin is not about redundancy, the only thing you could have
that way is a kind of load balancing (not the most sophisticated way,
though). Whenever one of the servers fails, around half of the requests
still goes there and then times out/gets conn refused or whatever the
problem is. Prioritizing is not easily possible. Probably it helps if
you add one of the IPs more often to the set, but I never tried that and
did not read the docs on this topic, so before breaking your zone first
read the specs, if this works!

For serious redundancy with failover and/or load balancing with a good 
leveling you should consider getting a load balancer (be it hardware or 
software), better two so you don't have the single point of failure 
there :-)

- Oliver

-- 
| Oliver Brandmueller | Offenbacher Str. 1  | Germany   D-14197 Berlin |
| Fon +49-172-3130856 | Fax +49-172-3145027 | WWW:   http://the.addict.de/ |
|   Ich bin das Internet. Sowahr ich Gott helfe.   |
| Eine gewerbliche Nutzung aller enthaltenen Adressen ist nicht gestattet! |


pgpIzG86sTPn3.pgp
Description: PGP signature

Re: ufs - fsck_ffs problem

2006-09-19 Thread David Malone 
On Tue, Sep 19, 2006 at 01:47:34AM -0400, [EMAIL PROTECTED] wrote:
> Alternate super block location: 160
> ** /dev/da0c (NO WRITE)
> ** Last Mounted on
> ** Phase 1 - Check Blocks and Sizes
> UNKNOWN FILE TYPE I=2
> CLEAR? no

If inode 2 is messed up, then you may have lost the root directory
of your filesystem. Repairing it probably won't be a lot of fun.

I'm guessing that the RAID somehow did the wrong thing - this seems
to actually be quite common.

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

Re: bind round robin

2006-09-19 Thread Dominic Marks 

Oliver Brandmueller wrote:

Hi.

On Tue, Sep 19, 2006 at 02:00:23PM +0800, pinoyskull wrote:
One of my client's domain has multiple IPs for redundancy, i configured 
his www as such


wwwIN A   10.10.10.10
wwwIN A   192.168.0.10

Is there a way to prioritize 10.10.10.10 over 192.168.0.10? How do i 
configure it?


DNS round robin is not about redundancy, the only thing you could have
that way is a kind of load balancing (not the most sophisticated way,
though). Whenever one of the servers fails, around half of the requests
still goes there and then times out/gets conn refused or whatever the
problem is. Prioritizing is not easily possible. Probably it helps if
you add one of the IPs more often to the set, but I never tried that and
did not read the docs on this topic, so before breaking your zone first
read the specs, if this works!

For serious redundancy with failover and/or load balancing with a good 
leveling you should consider getting a load balancer (be it hardware or 
software), better two so you don't have the single point of failure 
there :-)


A good software load balancer which supports weighting is pen. In ports.

http://siag.nu/pen

/usr/ports/net/pen

Dominic


- Oliver



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

Re: bind round robin

2006-09-19 Thread Dominik Zalewski 
On Tuesday 19 September 2006 11:50, Dominic Marks wrote:
> Oliver Brandmueller wrote:
> > Hi.
> >
> > On Tue, Sep 19, 2006 at 02:00:23PM +0800, pinoyskull wrote:
> >> One of my client's domain has multiple IPs for redundancy, i configured
> >> his www as such
> >>
> >> wwwIN A   10.10.10.10
> >> wwwIN A   192.168.0.10
> >>
> >> Is there a way to prioritize 10.10.10.10 over 192.168.0.10? How do i
> >> configure it?
> >
> > DNS round robin is not about redundancy, the only thing you could have
> > that way is a kind of load balancing (not the most sophisticated way,
> > though). Whenever one of the servers fails, around half of the requests
> > still goes there and then times out/gets conn refused or whatever the
> > problem is. Prioritizing is not easily possible. Probably it helps if
> > you add one of the IPs more often to the set, but I never tried that and
> > did not read the docs on this topic, so before breaking your zone first
> > read the specs, if this works!
> >
> > For serious redundancy with failover and/or load balancing with a good
> > leveling you should consider getting a load balancer (be it hardware or
> > software), better two so you don't have the single point of failure
> > there :-)
>
> A good software load balancer which supports weighting is pen. In ports.
>
> http://siag.nu/pen
>
> /usr/ports/net/pen
>
> Dominic
>
> > - Oliver
>
> ___
> freebsd-stable@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "[EMAIL PROTECTED]"

From pen homepage:

"This is pen, a load balancer for "simple" tcp based protocols such as http or 
smtp."

As I know DNS uses both tcp and udp protocols.

For failover you can try OpenBSD Packet Filter with CARP protocol. PF can do 
load-balacing using different algorithms also. CARP is ported to FreeBSD. 
More info on: http://pf4freebsd.love2party.net/carp.html

Enjoy

-- 
Dominik Zalewski | System Administrator
OpenCraft
t- +2 02 336 0003
w- http://www.open-craft.com
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: bind round robin

2006-09-19 Thread Dominic Marks 

From the original message:

[1]


wwwIN A   10.10.10.10
wwwIN A   192.168.0.10





From pen homepage:

"This is pen, a load balancer for "simple" tcp based protocols such as http or 
smtp."


As I know DNS uses both tcp and udp protocols.


From the posters excerpt it looks like they are looking to load balance 
HTTP. [1]


For failover you can try OpenBSD Packet Filter with CARP protocol. PF can do 
load-balacing using different algorithms also. CARP is ported to FreeBSD. 
More info on: http://pf4freebsd.love2party.net/carp.html


Also true. Last time I looked at this however there was a big disclaimer 
saying that
CARP's load balancing was likely to give a distorted distribution of 
load and I don't
believe it does weighting. I believe this would also be a problem 
considering the

example in [1]:

From carp(4):

 Note: ARP balancing only works on the local network segment.  It 
cannot

 balance traffic that crosses a router, because the router itself will
 always be balanced to the same virtual host.

Cheers,
Dominic

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

Re: bind round robin

2006-09-19 Thread Stefan Lambrev 



Dominic Marks wrote:

From the original message:

[1]


wwwIN A   10.10.10.10
wwwIN A   192.168.0.10





From pen homepage:

"This is pen, a load balancer for "simple" tcp based protocols such 
as http or smtp."


As I know DNS uses both tcp and udp protocols.


From the posters excerpt it looks like they are looking to load 
balance HTTP. [1]


For failover you can try OpenBSD Packet Filter with CARP protocol. PF 
can do load-balacing using different algorithms also. CARP is ported 
to FreeBSD. More info on: http://pf4freebsd.love2party.net/carp.html


Also true. Last time I looked at this however there was a big 
disclaimer saying that
CARP's load balancing was likely to give a distorted distribution of 
load and I don't
believe it does weighting. I believe this would also be a problem 
considering the

example in [1]:

From carp(4):

 Note: ARP balancing only works on the local network segment.  It 
cannot
 balance traffic that crosses a router, because the router itself 
will

 always be balanced to the same virtual host.

Cheers,
Dominic
Yes but the idea here is to use http balancer that runs on CARP 
interface(s) for fail-over.

Balancing will be done by balancer ;)


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


--
Best Wishes,
Stefan Lambrev
ICQ# 24134177

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

Re: bind round robin

2006-09-19 Thread Fred Clift 

On Tue, Sep 19, 2006 at 02:00:23PM +0800, pinoyskull wrote:

>> One of my client's domain has multiple IPs for redundancy, i configured
>> his www as such
>>
>> wwwIN A   10.10.10.10
>> wwwIN A   192.168.0.10
>>
>> Is there a way to prioritize 10.10.10.10 over 192.168.0.10? How do i
>> configure it?


I missed the original question about this but if you're really 
interested in doing somethign in software  rather than buying some kind 
of load-balancing hardware, then you could write your own (simple) 
backend for powerdns.


powerdns is in ports and it has a simple "pipe" interface to the daemon 
that would let you write your own "proporitional-share" dns responder 
for some set of RRs.


Check out the geographic load balancing that the blitzed.org guys use...

http://wiki.blitzed.org/DNS_balancing

Basically, you would delegate a subdomain (bar.example.org) to the 
server running your custom powerdns config and all lookups of 
foo.bar.example.org would end up at your server, where your custome (20 
or 60 lines of perl) powerdns module could return which ever of the two 
IPs you prefer, either statistically, or based on some kind of remotely 
fetched load average, ping time, other-availibility-metric, etc.


So, say you wanted 80% of your traffic to go to the primary box, then 
you turn off caching in your powerdns config file and have your pipe'd 
child return the primary IP 8 times for every 2 times you return the ip 
of the other box etc.  Or always return the primary server unless it is 
having problems, in which case you return the secondary.  Of course you 
want the TTL on these records, or perhaps the delegated subdomain to be 
low so the client doesn't cache it much.


I'm currently got a geo-balanced test setup I've been playing with - it 
returns CNAMES to XX.clift.org for any lookup of test.geo.clift.org, 
where XX are theoreticaly country codes based on what IP addresses you 
make the requests from.  The quality of the free geo-ip info isn't 
great, but at least it gets you on the right continent.  The geo-ip data 
is 'free' via rsync from countries.nerd.dk - see 
http://countries.nerd.dk/more.html for more information


Anyway, it'd take a bit of work, but would be doable.

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

Re: sed and comma-delimited file

2006-09-19 Thread Brooks Davis 
On Tue, Sep 19, 2006 at 12:26:54PM -0400, SigmaX asdf wrote:
> Yo;
> 
> I have a series of comma-delimited text files with fourteen columns of
> data and several hundred rows.  I want to use a short shell script to
> strip them of the last 9 columns, leaving the same file but with just
> five of its columns.  I can do it in C++, but that seems like
> overkill.  How would I go about doing it with sed or a similar
> utility?

See cut(1):

cut -d, -f1,2,3,4,5

-- Brooks


pgp9GMxOtSzTD.pgp
Description: PGP signature

Re: sed and comma-delimited file

2006-09-19 Thread Joan Picanyol i Puig 
* SigmaX asdf <[EMAIL PROTECTED]> [20060919 18:31]:
> I have a series of comma-delimited text files with fourteen columns of
> data and several hundred rows.  I want to use a short shell script to
> strip them of the last 9 columns, leaving the same file but with just
> five of its columns.  I can do it in C++, but that seems like
> overkill.  How would I go about doing it with sed or a similar
> utility?

cut -d ',' -f 1-5

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

[panic] iwi - sleeping thread owns a non-sleepable lock

2006-09-19 Thread Kipp Holger 

Hello,

I encounter a very annyoing panic with iwi0 on
'my' hp laptop 'Compaq nx8220'.

See kern/103368. Maybe the other open iwi-messages
are related (kern/89926 etc.)?

Is there a way to tell an Intel PRO/2200BG-card to
only use 802.11b and not 11g without crashing the
system?

Additional feedback can be provided if necessary.

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

Re: sed and comma-delimited file

2006-09-19 Thread N.J. Mann 
On Tue 19 Sep 12:26, SigmaX asdf wrote:

> I have a series of comma-delimited text files with fourteen columns of
> data and several hundred rows.  I want to use a short shell script to
> strip them of the last 9 columns, leaving the same file but with just
> five of its columns.  I can do it in C++, but that seems like
> overkill.  How would I go about doing it with sed or a similar
> utility?

cut -d, -f 1-5


Cheers,
   Nick.
-- 
"You call _that_ a knife?  _This_ is what _I_ call a knife!"
"Really?  No worries.  _This_ is what I call a crossbow!"
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

df -kP != df -Pk

2006-09-19 Thread Andriy Gapon 

It seems that -P flag to df resets previously specified -k for no good
reason. POSIX expressly talks about -P and -k being used together.
http://www.opengroup.org/onlinepubs/009695399/utilities/df.html

This is FreeBSD 6.1-RELEASE-p2 i386.

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

Re: df -kP != df -Pk

2006-09-19 Thread Andriy Gapon 
on 19/09/2006 19:17 Christian S.J. Peron said the following:
> Andriy Gapon wrote:
>> It seems that -P flag to df resets previously specified -k for no good
>> reason. POSIX expressly talks about -P and -k being used together.
>> http://www.opengroup.org/onlinepubs/009695399/utilities/df.html
>>
>> This is FreeBSD 6.1-RELEASE-p2 i386.
>>
>>   
> Please test the attached patch and let me know if it's good for you.
> 
> Thanks for bringing this to our attention.

Yes, the patch works very well. Thank you for the lightning-fast response.


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

sed and comma-delimited file

2006-09-19 Thread SigmaX asdf 

Yo;

I have a series of comma-delimited text files with fourteen columns of
data and several hundred rows.  I want to use a short shell script to
strip them of the last 9 columns, leaving the same file but with just
five of its columns.  I can do it in C++, but that seems like
overkill.  How would I go about doing it with sed or a similar
utility?

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

Re: df -kP != df -Pk

2006-09-19 Thread Christian S.J. Peron 

Andriy Gapon wrote:

It seems that -P flag to df resets previously specified -k for no good
reason. POSIX expressly talks about -P and -k being used together.
http://www.opengroup.org/onlinepubs/009695399/utilities/df.html

This is FreeBSD 6.1-RELEASE-p2 i386.

  

Please test the attached patch and let me know if it's good for you.

Thanks for bringing this to our attention.

--
Christian S.J. Peron
[EMAIL PROTECTED]
FreeBSD Committer
FreeBSD Security Team

? df
? df.1.gz
? df.c.diff
Index: df.c
===
RCS file: /home/ncvs/src/bin/df/df.c,v
retrieving revision 1.64
diff -u -r1.64 df.c
--- df.c10 Jan 2005 08:39:21 -  1.64
+++ df.c19 Sep 2006 16:14:37 -
@@ -93,7 +93,7 @@
return (a > b ? a : b);
 }
 
-static int aflag = 0, cflag, hflag, iflag, nflag;
+static int aflag = 0, cflag, hflag, kflag, iflag, nflag, Pflag;
 static struct  ufs_args mdev;
 
 int
@@ -123,6 +123,7 @@
case 'b':
/* FALLTHROUGH */
case 'P':
+   Pflag++;
putenv("BLOCKSIZE=512");
hflag = 0;
break;
@@ -143,6 +144,7 @@
iflag = 1;
break;
case 'k':
+   kflag++;
putenv("BLOCKSIZE=1k");
hflag = 0;
break;
@@ -171,6 +173,12 @@
argc -= optind;
argv += optind;
 
+   /*
+* POSIX specifies that if both -P and -k options are used together a
+* 1k blocksize should be used.
+*/
+   if (Pflag != 0 && kflag != 0)
+   putenv("BLOCKSIZE=1k");
mntsize = getmntinfo(&mntbuf, MNT_NOWAIT);
bzero(&maxwidths, sizeof(maxwidths));
for (i = 0; i < mntsize; i++)
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Seg fault error this morning

2006-09-19 Thread Brian 
I just cvsupped this morning after seeing the gzip announcement.  During 
a buildworld, I get the below.  I have replicated this a few times.


/usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/linker.c: 
In function `_bfd_default_link_order':
/usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/linker.c:2605: 
internal compiler error: Segmentation fault: 11

Please submit a full bug report,
with preprocessed source if appropriate.
See http://gcc.gnu.org/bugs.html> for instructions.
*** Error code 1

Stop in /usr/src/gnu/usr.bin/binutils/libbfd.
*** Error code 1

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

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

www# gcc -v
Using built-in specs.
Configured with: FreeBSD/i386 system compiler
Thread model: posix
gcc version 3.4.4 [FreeBSD] 20050518

System is a 1.2ghz p3
I was going to wait to file the gcc bug till I saw what folks had seen here.
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: bind round robin

2006-09-19 Thread Doug Barton 
Oliver Brandmueller wrote:
> 
> DNS round robin is not about redundancy, the only thing you could have
> that way is a kind of load balancing (not the most sophisticated way,
> though). Whenever one of the servers fails, around half of the requests
> still goes there and then times out/gets conn refused or whatever the
> problem is. Prioritizing is not easily possible. Probably it helps if
> you add one of the IPs more often to the set, but I never tried that and
> did not read the docs on this topic, so before breaking your zone first
> read the specs, if this works!

Just replying to this bit first, in BIND it does not work to specify
the same IP address multiple times for the same hostname. The server
will collapse the duplicates into one unique entry when it reads the
zone. I am not aware of any other authoritative name server for which
this would work either.

FYI,

Doug

-- 

This .signature sanitized for your protection

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

Re: bind round robin

2006-09-19 Thread Doug Barton 
pinoyskull wrote:
> Hi,
> 
> One of my client's domain has multiple IPs for redundancy, 

This really isn't on topic for any of the FreeBSD lists, FYI. If the
responses you have received so far haven't helped you, I would suggest
that you write up a little more detail about what you're trying to
achieve, and post a message to the [EMAIL PROTECTED] mailing list.

Briefly, if what you're trying to do is actually failover (if the
"primary" website is down, users should be directed to the "secondary"
site), then the answer is you can't do that in DNS alone. But the
bind-users folks can help you find some answers.

good luck,

Doug

-- 

This .signature sanitized for your protection

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

Re: sed and comma-delimited file

2006-09-19 Thread Julian Stacey 
Reference:
> From: "SigmaX asdf" <[EMAIL PROTECTED]> 
> Date: Tue, 19 Sep 2006 12:26:54 -0400 
> Message-id:   <[EMAIL PROTECTED]> 

"SigmaX asdf" wrote:
> Yo;
> 
> I have a series of comma-delimited text files with fourteen columns of
> data and several hundred rows.  I want to use a short shell script to
> strip them of the last 9 columns, leaving the same file but with just
> five of its columns.  I can do it in C++, but that seems like
> overkill.  How would I go about doing it with sed or a similar
> utility?

awk !
/usr/ports/lang/gawk  exists too.
PS
Here's odd notes from my syntax file, not exactly appropriate to you,
but near enough to give ideas with  `man awk' for ref, & `fun' to learn ;-)

 awk '{printf "rm -f %s ; ln -s ..%s %s\n",$1,$3,$1}' < /tmp/x
 awk -F = '{printf "%s\n",$1}' < /tmp/t
 awk --field-separator # '{printf "%s\n",$1}'
 awk -F # '{printf "%s\n",$1}'  # 5.0 has no --field-separator
 awk '{printf "%s\n",$1}' < distfiles.dump
 find /etc /var /usr -type l | sort | xargs ls -l | \
 awk '{printf "%s -> %s\n",$9,$11}' # List all links

-- 
Julian Stacey.  BSD Unix C Net Consultancy, Munich/Muenchen  http://berklix.com
Mail Ascii, not HTML.   Ihr Rauch = mein allergischer Kopfschmerz.
Don't buy it ! Get it free !  http://berklix.org/free-software
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: sed and comma-delimited file

2006-09-19 Thread Oleg D. 

SigmaX asdf wrote:

Yo;

I have a series of comma-delimited text files with fourteen columns of
data and several hundred rows.  I want to use a short shell script to
strip them of the last 9 columns, leaving the same file but with just
five of its columns.  I can do it in C++, but that seems like
overkill.  How would I go about doing it with sed or a similar
utility?

SigmaX


All of advices were cool
but!
there's a good idea to work around something like that, but I don't 
exactly know now how to solve your problem, the idea is:

`perl -pi~ -e 's/foo/bar/g' somefile`
works fine for `somefile'... I think TIMTOWTDI.


--
Oleg D.

--
don't believe every word people use to say, they might be wrong.

an undefined problem has infinitive number of solutions.


smime.p7s
Description: S/MIME Cryptographic Signature

Re: sed and comma-delimited file

2006-09-19 Thread Thomas David Rivers 
"SigmaX asdf" <[EMAIL PROTECTED]> wrote:
> Yo;
> 
> I have a series of comma-delimited text files with fourteen columns of
> data and several hundred rows.  I want to use a short shell script to
> strip them of the last 9 columns, leaving the same file but with just
> five of its columns.  I can do it in C++, but that seems like
> overkill.  How would I go about doing it with sed or a similar
> utility?

 Uh.. is there no reason the cut(1) program doesn't do this?

#!/bin/sh
cat file | cut -d',' -f1-5 > /tmp/t.$$
rm -f file
mv /tmp/t.$$ file

 This doesn't preserve permissions, etc... but - it's pretty
 straight-forward.

- Dave Rivers -

--
[EMAIL PROTECTED]Work: (919) 676-0847
Get your mainframe programming tools at http://www.dignus.com
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Seg fault error this morning

2006-09-19 Thread Don Wilde 

On 9/19/06, Brian <[EMAIL PROTECTED]> wrote:


I just cvsupped this morning after seeing the gzip announcement.  During
a buildworld, I get the below.  I have replicated this a few times.

buildworld from cvsup this morning (x86) worked fine for me, Brian, same

time frame. Suggest you update source again, 'make clean' and try again.

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

Kernel panic on boot (how to debug?)

2006-09-19 Thread Benjamin Lutz 
Hello,

Since I added a new P-ATA disk to my computer, FreeBSD 6.1/amd64 panics while 
booting:

[...]
acd0: DVDROM  at ata0-master UDMA33
acd1: DVDR  at ata0-slave UDMA33
ad2: 78533MB  at ata1-master UDMA100


Fatal trap 18: integer divide fault while in kernel mode
instruction pointer = 0x8:0x801bd9a5
stack pointer   = 0x10:0x80845a90
frame pointer   = 0x10:0x80845ae0
code segment= base 0x0, limit 0xf, type 0x1b
= DPL 0, pres 1, long 1, def32 0, gran 1
processor eflags= interrupt enabled, resume, IOPL = 0
current process = 0 (swapper)
trap number = 18
panic: integer divide fault
Uptime: 1s
Cannot dump. No dump device defined.
Automatic reboot in 15 seconds - press a key on the console to abort

If I just disable the second IDE channel, the system boots fine. FreeBSD 
5.5/i386, which is also installed on this computer, boots fine with or 
without that disk.

I've been trying to get into the kernel debugger to get a backtrace, but 
without luck. I've hooked up a serial console, but I just can't get into DDB. 
I've tried various option (adding debug.debugger_on_panic="1" 
to /boot/loader.conf, the BREAK_TO_DEBUGGER and KDB_UNATTENDED kernel options 
as well as specifying "boot -d" at the boot loader - without any effect. 
Since the panic happens before init is started, my dump device configuration 
in /etc/rc.conf isn't read either.

How can I get meaningful debugging data out of this system?

Oh, btw, notice the weird device name string for acd0. It's supposed to say 
, and in the BIOS as well as 
under FreeBSD 5.5/i386 it does. The erroneous strings shows up on FreeBSD 6.1 
with or without the second IDE channel being enabled.

Any idea what might cause this?

The full boot log follows bellow.

Cheers
Benjamin


KDB: debugger backends: ddb
KDB: current backend: ddb
Copyright (c) 1992-2006 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
The Regents of the University of California. All rights reserved.
FreeBSD 6.1-RELEASE-p7 #4: Tue Sep 19 21:56:53 CEST 2006
[EMAIL PROTECTED]:/usr/obj/usr/src/sys/MERLIN64
Timecounter "i8254" frequency 1193182 Hz quality 0
CPU: AMD Athlon(tm) 64 Processor 3500+ (2210.09-MHz K8-class CPU)
  Origin = "AuthenticAMD"  Id = 0xff0  Stepping = 0
  
Features=0x78bfbff
  AMD Features=0xe0500800
real memory  = 1073676288 (1023 MB)
avail memory = 1026371584 (978 MB)
ACPI APIC Table: 
ioapic0  irqs 0-23 on motherboard
kbd1 at kbdmux0
acpi0:  on motherboard
acpi0: Power Button (fixed)
Timecounter "ACPI-fast" frequency 3579545 Hz quality 1000
acpi_timer0: <24-bit timer at 3.579545MHz> port 0x4008-0x400b on acpi0
cpu0:  on acpi0
powernow0:  on cpu0
acpi_button0:  on acpi0
pcib0:  port 0xcf8-0xcff,0xcf0-0xcf3 on acpi0
pci0:  on pcib0
agp0:  mem 0xf000-0xf7ff at device 
0.0 on pci0
isab0:  at device 1.0 on pci0
isa0:  on isab0
pci0:  at device 1.1 (no driver attached)
ohci0:  mem 0xfe02f000-0xfe02 irq 21 at 
device 2.0 on pci0
ohci0: [GIANT-LOCKED]
usb0: OHCI version 1.0, legacy support
usb0: SMM does not respond, resetting
usb0:  on ohci0
usb0: USB revision 1.0
uhub0: nVidia OHCI root hub, class 9/0, rev 1.00/1.00, addr 1
uhub0: 4 ports with 4 removable, self powered
ohci1:  mem 0xfe02e000-0xfe02efff irq 22 at 
device 2.1 on pci0
ohci1: [GIANT-LOCKED]
usb1: OHCI version 1.0, legacy support
usb1: SMM does not respond, resetting
usb1:  on ohci1
usb1: USB revision 1.0
uhub1: nVidia OHCI root hub, class 9/0, rev 1.00/1.00, addr 1
uhub1: 4 ports with 4 removable, self powered
ehci0:  mem 0xfe02d000-0xfe02d0ff irq 
23 at device 2.2 on pci0
ehci0: [GIANT-LOCKED]
usb2: EHCI version 1.0
usb2: companion controllers, 4 ports each: usb0 usb1
usb2:  on ehci0
usb2: USB revision 2.0
uhub2: nVidia EHCI root hub, class 9/0, rev 2.00/1.00, addr 1
uhub2: 8 ports with 8 removable, self powered
uhub3: vendor 0x0409 product 0x013e, class 9/0, rev 2.00/0.07, addr 2
uhub3: multiple transaction translators
uhub3: 4 ports with 4 removable, self powered
pci0:  at device 6.0 (no driver attached)
atapci0:  port 
0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xe000-0xe00f at device 8.0 on pci0
ata0:  on atapci0
ata1:  on atapci0
atapci1:  port 
0x9e0-0x9e7,0xbe0-0xbe3,0x960-0x967,0xb60-0xb63,0xcc00-0xcc0f,0xc800-0xc87f 
irq 22 at device 9.0 on pci0
ata2:  on atapci1
ata3:  on atapci1
atapci2:  port 
0x9f0-0x9f7,0xbf0-0xbf3,0x970-0x977,0xb70-0xb73,0xb400-0xb40f,0xb000-0xb07f 
irq 23 at device 10.0 on pci0
ata4:  on atapci2
ata5:  on atapci2
pcib1:  at device 11.0 on pci0
pci1:  on pcib1
pci1:  at device 0.0 (no driver attached)
pcib2:  at device 14.0 on pci0
pci2:  on pcib2
pcm0:  port 0x9c00-0x9c1f irq 18 at device 8.0 on pci2
pcm0: 
pci2:  at device 9.0 (no driver attached)
pci2:  at device 9.1 (no driver attached)
fwohci0:  port 0x9400-0x947f mem 0xfdfff000-0xfdfff7ff 
irq 19 at device 12.0 on pci2
fwohci0: OHCI version 1.0 (ROM=1)
fwohci0: No. of I

Re: bind round robin

2006-09-19 Thread Chris H. 

Greetings all,
...
Quoting Doug Barton <[EMAIL PROTECTED]>:


Oliver Brandmueller wrote:


DNS round robin is not about redundancy, the only thing you could have
that way is a kind of load balancing (not the most sophisticated way,
though). Whenever one of the servers fails, around half of the requests
still goes there and then times out/gets conn refused or whatever the
problem is. Prioritizing is not easily possible. Probably it helps if
you add one of the IPs more often to the set, but I never tried that and
did not read the docs on this topic, so before breaking your zone first
read the specs, if this works!


Just replying to this bit first, in BIND it does not work to specify
the same IP address multiple times for the same hostname. The server
will collapse the duplicates into one unique entry when it reads the
zone. I am not aware of any other authoritative name server for which
this would work either.


While this /might/ hold true in some/certain situations. I /can/ say
after 3.5 yrs. of doing exactly this, that it does not collapse the
namespace into a single IP<-->name. Here is the excerpt from the zone
file(s) running a recent BIND version:


#hostA.domain.tld.zone
hostA  IN  A  XXX.XXX.XXX.XA
  IN  HINFO  IBM-PC/AT  UNICS/UNIX
  IN  MX 10  mx
  IN  MX 60  mx2
graphics   IN  A  XXX.XXX.XXX.XA
nameA  IN  A  XXX.XXX.XXX.XA
nameB  IN  A  XXX.XXX.XXX.XA
...
wwwIN  A  XXX.XXX.XXX.XA
nameC  IN  CNAME  nameB
etc...

#hostB.domain.tld.zone
hostB  IN  A  XXX.XXX.XXX.XB
  IN  HINFO  IBM-PC/AT  UNICS/UNIX
  IN  MX 10  mx
  IN  MX 60  mx2
nameD  IN  A  XXX.XXX.XXX.XB
graphics   IN  A  XXX.XXX.XXX.XB
hostE  IN  A  XXX.XXX.XXX.XB
etc...
#
Please note the RR (PTR) zone only lists RR's for
hostA and hostB. It is the responsibility of the hosts
own zones to delegate the hostnames for their own zones.
Both of these hosts are running Apache for the HTTPd
service, and both of them serve pages for graphics.domain.tld.

Now, on to your initial question...
Speaking of Apache; Apache has provided a solution for
the /exact/ situation you are enquiring about since v.1.2.
You will find it in the documentation that comes with the
installation. I will endevour to find it's whereabouts in
the doc's and provide a link. As I host those doc's. It
involves DNS and either the use of Perl, or http(s)d.conf
trickery. This, of course, all assumes that you are working
with Apache. :)

Best wishes,
Chris H.



FYI,

Doug

--

   This .signature sanitized for your protection

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





--
panic: kernel trap (ignored)



-
FreeBSD 5.4-RELEASE-p12 (SMP - 900x2) Tue Mar 7 19:37:23 PST 2006
/

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

Re: Seg fault error this morning

2006-09-19 Thread Brian 

Don Wilde wrote:


On 9/19/06, *Brian* <[EMAIL PROTECTED] > wrote:

I just cvsupped this morning after seeing the gzip
announcement.  During
a buildworld, I get the below.  I have replicated this a few times.

buildworld from cvsup this morning (x86) worked fine for me, Brian, 
same time frame. Suggest you update source again, 'make clean' and try 
again.


:D


Tried a couple more times.

First attempt;
cvsup4 used
Updating collection src-all/cvs
TreeList failed: Error in "/var/db/sup/src-all/checkouts.cvs:RELENG_6": 
43352: File is not sorted properly.  Delete it and try again.


rm the file in question, then repeat.
cvsup4 again
Updating collection src-all/cvs
Checkout src/lib/libcompat/4.1/ftime.3
Updater failed: Cannot install 
"/usr/src/lib/libcompat/4.1/#cvs.cvsup-12210.1" to 
"/usr/src/lib/libcompat/4.1/ftime.3": Input/output error


Out of frustration I mv /usr/src to /usr/src-old, then re cvsup, getting 
cvsup4 again.

Make it thru cvsup, get into buildworld.
cc -O2 -fno-strict-aliasing -pipe  -DTERMIOS -DANSI_SOURCE 
-I/usr/src/secure/lib/libcrypto/../../../crypto/openssl 
-I/usr/src/secure/lib/libcrypto/../../../crypto/openssl/crypto 
-I/usr/obj/usr/src/secure/lib/libcrypto -DOPENSSL_THREADS -DDSO_DLFCN 
-DHAVE_DLFCN_H -DOPENSSL_NO_IDEA -DL_ENDIAN -DNO_IDEA -std=gnu89  -c 
/usr/src/secure/lib/libcrypto/../../../crypto/openssl/crypto/pkcs7/example.c

make: don't know how to make /usr/obj/usr/src/tmp/usr/knclude/stdarg.h. Stop
*** Error code 2

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

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

I've seen some 6.2 chatter recently, I wonder if I am getting a 
prerelease build as a result of tag=RELENG_6.






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

PERC trouble?

2006-09-19 Thread Ivan Voras 
I had a chance today to play a little with a server that was later
passed on for deployment, and one of the thing I tried to do was create
something unusual - three disk groups/virtual disks on the PERC5/i RAID
controller, with a single drive in each group (entered as RAID0).

All went fine until I booted FreeBSD 6.1-release (amd64) and tried to do
something with the drives. It turned out that, while there WERE three
devices mfid[0,1,2], they all "pointed" to the same hardware - the first
drive. I.e. accessing either of these would access the first virtual
drive, and this is confirmed by watching drive LEDs blinking.

The server went away later so I couldn't dig deeper, but I'm wondering
if this is a bug in PERC or the driver?

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

Re: Seg fault error this morning

2006-09-19 Thread David Wolfskill 
On Tue, Sep 19, 2006 at 02:41:29PM -0700, Brian wrote:
> Don Wilde wrote:
> >
> >On 9/19/06, *Brian* <[EMAIL PROTECTED] > wrote:
> >
> >I just cvsupped this morning after seeing the gzip
> >announcement.  During
> >a buildworld, I get the below.  I have replicated this a few times.
> >
> >buildworld from cvsup this morning (x86) worked fine for me, Brian, 
> >same time frame. Suggest you update source again, 'make clean' and try 
> >again.
> >
> >:D
> >
> Tried a couple more times.
> ...

I've been tracking RELENG_6 on slice 1 of my laptop on a daily basis; no
problems in the last several weeks (at least).

Recent updates to my private mirror of the CVS repository:

localhost(6.2-P)[1] tail -20 /var/log/cvsup-history.log
CVSup begin from cvsup12.freebsd.org at Sun Sep 17 01:17:37 PDT 2006
CVSup begin from cvsup14.freebsd.org at Sun Sep 17 03:47:03 PDT 2006
CVSup begin from cvsup12.freebsd.org at Sun Sep 17 03:48:31 PDT 2006
CVSup ended from cvsup12.freebsd.org at Sun Sep 17 03:54:18 PDT 2006
CVSup begin from cvsup14.freebsd.org at Mon Sep 18 01:17:02 PDT 2006
CVSup begin from cvsup12.freebsd.org at Mon Sep 18 01:17:09 PDT 2006
CVSup begin from cvsup13.freebsd.org at Mon Sep 18 01:17:22 PDT 2006
CVSup begin from cvsup7.freebsd.org at Mon Sep 18 01:17:38 PDT 2006
CVSup begin from cvsup12.freebsd.org at Mon Sep 18 01:17:40 PDT 2006
CVSup begin from cvsup14.freebsd.org at Mon Sep 18 03:47:03 PDT 2006
CVSup begin from cvsup12.freebsd.org at Mon Sep 18 03:48:42 PDT 2006
CVSup ended from cvsup12.freebsd.org at Mon Sep 18 03:54:48 PDT 2006
CVSup begin from cvsup14.freebsd.org at Tue Sep 19 01:17:05 PDT 2006
CVSup begin from cvsup12.freebsd.org at Tue Sep 19 01:17:10 PDT 2006
CVSup begin from cvsup13.freebsd.org at Tue Sep 19 01:17:13 PDT 2006
CVSup begin from cvsup10.freebsd.org at Tue Sep 19 01:17:15 PDT 2006
CVSup ended from cvsup10.freebsd.org at Tue Sep 19 01:22:17 PDT 2006
CVSup begin from cvsup14.freebsd.org at Tue Sep 19 03:47:03 PDT 2006
CVSup begin from cvsup12.freebsd.org at Tue Sep 19 03:47:07 PDT 2006
CVSup ended from cvsup12.freebsd.org at Tue Sep 19 03:52:38 PDT 2006
localhost(6.2-P)[2] 


Presently running:

localhost(6.2-P)[2] uname -a
FreeBSD localhost 6.2-PRERELEASE FreeBSD 6.2-PRERELEASE #180: Tue Sep 19 
06:30:23 PDT 2006 [EMAIL PROTECTED]:/common/S1/obj/usr/src/sys/LAPTOP_30W  
i386
localhost(6.2-P)[3] 

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

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


pgpE3TESAj8Cc.pgp
Description: PGP signature

Re: PERC trouble?

2006-09-19 Thread Paul Saab 

this was fixed after 6.1-RELEASE.  You need to grab the driver from -stable.

Ivan Voras wrote:

I had a chance today to play a little with a server that was later
passed on for deployment, and one of the thing I tried to do was create
something unusual - three disk groups/virtual disks on the PERC5/i RAID
controller, with a single drive in each group (entered as RAID0).

All went fine until I booted FreeBSD 6.1-release (amd64) and tried to do
something with the drives. It turned out that, while there WERE three
devices mfid[0,1,2], they all "pointed" to the same hardware - the first
drive. I.e. accessing either of these would access the first virtual
drive, and this is confirmed by watching drive LEDs blinking.

The server went away later so I couldn't dig deeper, but I'm wondering
if this is a bug in PERC or the driver?

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

  

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

Re: bind round robin

2006-09-19 Thread Chris H. 

Greetings,
...
Quoting "Chris H." <[EMAIL PROTECTED]>:


Greetings all,
...
Quoting Doug Barton <[EMAIL PROTECTED]>:


Oliver Brandmueller wrote:


DNS round robin is not about redundancy, the only thing you could have
that way is a kind of load balancing (not the most sophisticated way,
though). Whenever one of the servers fails, around half of the requests
still goes there and then times out/gets conn refused or whatever the
problem is. Prioritizing is not easily possible. Probably it helps if
you add one of the IPs more often to the set, but I never tried that and
did not read the docs on this topic, so before breaking your zone first
read the specs, if this works!


Just replying to this bit first, in BIND it does not work to specify
the same IP address multiple times for the same hostname. The server
will collapse the duplicates into one unique entry when it reads the
zone. I am not aware of any other authoritative name server for which
this would work either.


While this /might/ hold true in some/certain situations. I /can/ say
after 3.5 yrs. of doing exactly this, that it does not collapse the
namespace into a single IP<-->name. Here is the excerpt from the zone
file(s) running a recent BIND version:


#hostA.domain.tld.zone
hostA  IN  A  XXX.XXX.XXX.XA
  IN  HINFO  IBM-PC/AT  UNICS/UNIX
  IN  MX 10  mx
  IN  MX 60  mx2
graphics   IN  A  XXX.XXX.XXX.XA
nameA  IN  A  XXX.XXX.XXX.XA
nameB  IN  A  XXX.XXX.XXX.XA
...
wwwIN  A  XXX.XXX.XXX.XA
nameC  IN  CNAME  nameB
etc...

#hostB.domain.tld.zone
hostB  IN  A  XXX.XXX.XXX.XB
  IN  HINFO  IBM-PC/AT  UNICS/UNIX
  IN  MX 10  mx
  IN  MX 60  mx2
nameD  IN  A  XXX.XXX.XXX.XB
graphics   IN  A  XXX.XXX.XXX.XB
hostE  IN  A  XXX.XXX.XXX.XB
etc...
#
Please note the RR (PTR) zone only lists RR's for
hostA and hostB. It is the responsibility of the hosts
own zones to delegate the hostnames for their own zones.
Both of these hosts are running Apache for the HTTPd
service, and both of them serve pages for graphics.domain.tld.

Now, on to your initial question...
Speaking of Apache; Apache has provided a solution for
the /exact/ situation you are enquiring about since v.1.2.
You will find it in the documentation that comes with the
installation. I will endevour to find it's whereabouts in
the doc's and provide a link. As I host those doc's. It
involves DNS and either the use of Perl, or http(s)d.conf
trickery. This, of course, all assumes that you are working
with Apache. :)

Best wishes,
Chris H.


O.K. Here's the link(s) I promised:

You will/should find these two links /extremely/ valuable:

http://hosting.1command.com/manual/misc/rewriteguide.html

These are the things dreams are made of. ;)

the link I indicated you'd find as a good solution is here:

http://hosting.1command.com/manual/misc/rewriteguide.html

Just scroll down to the topic:
*Load Balancing*

Best wishes,
Chris H.

P.S. you will/should also spend some time on a DNS/BIND
newsgroup. As the knowledge gained there is invaluable.
I spend quite alot of time there answering questions,
and it is probably a better place to ask questions of
this nature. Because this list really isn't designed
for this kind of topic.





FYI,

Doug

--

   This .signature sanitized for your protection

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





--
panic: kernel trap (ignored)



-
FreeBSD 5.4-RELEASE-p12 (SMP - 900x2) Tue Mar 7 19:37:23 PST 2006
/

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





--
panic: kernel trap (ignored)



-
FreeBSD 5.4-RELEASE-p12 (SMP - 900x2) Tue Mar 7 19:37:23 PST 2006
/

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

Re: Seg fault error this morning

2006-09-19 Thread Brian 

David Wolfskill wrote:

On Tue, Sep 19, 2006 at 02:41:29PM -0700, Brian wrote:
  

Don Wilde wrote:


On 9/19/06, *Brian* <[EMAIL PROTECTED] > wrote:

   I just cvsupped this morning after seeing the gzip
   announcement.  During
   a buildworld, I get the below.  I have replicated this a few times.

buildworld from cvsup this morning (x86) worked fine for me, Brian, 
same time frame. Suggest you update source again, 'make clean' and try 
again.


:D

  

Tried a couple more times.
...



I've been tracking RELENG_6 on slice 1 of my laptop on a daily basis; no
problems in the last several weeks (at least).

Recent updates to my private mirror of the CVS repository:

localhost(6.2-P)[1] tail -20 /var/log/cvsup-history.log
CVSup begin from cvsup12.freebsd.org at Sun Sep 17 01:17:37 PDT 2006
CVSup begin from cvsup14.freebsd.org at Sun Sep 17 03:47:03 PDT 2006
CVSup begin from cvsup12.freebsd.org at Sun Sep 17 03:48:31 PDT 2006
CVSup ended from cvsup12.freebsd.org at Sun Sep 17 03:54:18 PDT 2006
CVSup begin from cvsup14.freebsd.org at Mon Sep 18 01:17:02 PDT 2006
CVSup begin from cvsup12.freebsd.org at Mon Sep 18 01:17:09 PDT 2006
CVSup begin from cvsup13.freebsd.org at Mon Sep 18 01:17:22 PDT 2006
CVSup begin from cvsup7.freebsd.org at Mon Sep 18 01:17:38 PDT 2006
CVSup begin from cvsup12.freebsd.org at Mon Sep 18 01:17:40 PDT 2006
CVSup begin from cvsup14.freebsd.org at Mon Sep 18 03:47:03 PDT 2006
CVSup begin from cvsup12.freebsd.org at Mon Sep 18 03:48:42 PDT 2006
CVSup ended from cvsup12.freebsd.org at Mon Sep 18 03:54:48 PDT 2006
CVSup begin from cvsup14.freebsd.org at Tue Sep 19 01:17:05 PDT 2006
CVSup begin from cvsup12.freebsd.org at Tue Sep 19 01:17:10 PDT 2006
CVSup begin from cvsup13.freebsd.org at Tue Sep 19 01:17:13 PDT 2006
CVSup begin from cvsup10.freebsd.org at Tue Sep 19 01:17:15 PDT 2006
CVSup ended from cvsup10.freebsd.org at Tue Sep 19 01:22:17 PDT 2006
CVSup begin from cvsup14.freebsd.org at Tue Sep 19 03:47:03 PDT 2006
CVSup begin from cvsup12.freebsd.org at Tue Sep 19 03:47:07 PDT 2006
CVSup ended from cvsup12.freebsd.org at Tue Sep 19 03:52:38 PDT 2006
localhost(6.2-P)[2] 



Presently running:

localhost(6.2-P)[2] uname -a
FreeBSD localhost 6.2-PRERELEASE FreeBSD 6.2-PRERELEASE #180: Tue Sep 19 
06:30:23 PDT 2006 [EMAIL PROTECTED]:/common/S1/obj/usr/src/sys/LAPTOP_30W  
i386
localhost(6.2-P)[3] 


Peace,
david
  
I believe I have the dreaded disk drive failure approaching, sorry for 
the wasted cycles.


Sep 19 11:47:48 www kernel: 
g_vfs_done():ad0s1f[READ(offset=150880354304, length=16384)]error = 5
Sep 19 12:00:59 www kernel: 
g_vfs_done():ad0s1f[READ(offset=150880354304, length=16384)]error = 5


Bri

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

Re: Seg fault error this morning

2006-09-19 Thread Kris Kennaway 
On Tue, Sep 19, 2006 at 02:41:29PM -0700, Brian wrote:
> Don Wilde wrote:
> >
> >On 9/19/06, *Brian* <[EMAIL PROTECTED] > wrote:
> >
> >I just cvsupped this morning after seeing the gzip
> >announcement.  During
> >a buildworld, I get the below.  I have replicated this a few times.
> >
> >buildworld from cvsup this morning (x86) worked fine for me, Brian, 
> >same time frame. Suggest you update source again, 'make clean' and try 
> >again.
> >
> >:D
> >
> Tried a couple more times.
> 
> First attempt;
> cvsup4 used
> Updating collection src-all/cvs
> TreeList failed: Error in "/var/db/sup/src-all/checkouts.cvs:RELENG_6": 
> 43352: File is not sorted properly.  Delete it and try again.
> 
> rm the file in question, then repeat.
> cvsup4 again
> Updating collection src-all/cvs
> Checkout src/lib/libcompat/4.1/ftime.3
> Updater failed: Cannot install 
> "/usr/src/lib/libcompat/4.1/#cvs.cvsup-12210.1" to 
> "/usr/src/lib/libcompat/4.1/ftime.3": Input/output error

You have failing hardware.

Kris

pgpoAsPVcFVP6.pgp
Description: PGP signature

Re: sed and comma-delimited file

2006-09-19 Thread Freddie Cash 
On Tue, September 19, 2006 9:26 am, SigmaX asdf wrote:
> I have a series of comma-delimited text files with fourteen columns
> of data and several hundred rows.  I want to use a short shell script
> to strip them of the last 9 columns, leaving the same file but with
> just five of its columns.  I can do it in C++, but that seems like
> overkill.  How would I go about doing it with sed or a similar
> utility?

cat file | awk -F"," '{ printf "%s,%s,%s,%s,%s\n",$1,$2,$3,$4,$5  }' >
newfile

You can probably even remove the cat and just use awk on the file
directly.

Freddie Cash
[EMAIL PROTECTED]

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

Re: isc-dhcpd and jails bound to an aliased ip

2006-09-19 Thread Russell Jackson 
On Mon, Sep 18, 2006 at 01:08:28PM -0700, Russell Jackson wrote:
> Attempting to run isc-dhcpd (using USE_SOCKETS) inside a jail
> bound to an aliased ip does not appear to work. The process never seems
> to recieve any broadcast traffic; however, it does see unicast traffic
> as would be expected. I'm not sure how to debug this since one cannot
> run tcpdump in the jail to see what traffic is getting there obviously.
> 
> It works fine if I change the jail to bind to the primary ip on the
> interface. Not surprisingly, it also works fine if I run it outside of a
> jail using BPF. Changing the broadcast addresses on the aliases does not
> seem to change anything.
> 
> It is just that the kernel will not deliver broadcasts to jails on ip
> aliases as I suspect? Yes, I now I have a "zombied" jail in the jls
> listing. There are no processes with a JID of 2 running, and I'm
> reluctant to reboot the machine because it's in production.
> 
> If I have to run the jail on the primary ip address, that's okay. I
> would just prefer to have it running in a seperate jail and still have
> ssh running on the standard port (less confusing to users).
> 
> Relevant configuration:
> 
> em0: flags=8843 mtu 1500
>   options=b
>   inet6 fe80::213:72ff:fe4b:70e7%em0 prefixlen 64 scopeid 0x1 
>   inet 136.168.1.5 netmask 0x broadcast 136.168.255.255
>   inet 136.168.1.8 netmask 0x broadcast 136.168.1.8
>   inet 136.168.1.91 netmask 0x broadcast 136.168.1.91
>   ether 00:13:72:4b:70:e7
>   media: Ethernet autoselect (1000baseTX )
>   status: active
> 
> # global jail knobs
> jail_enable="YES"
> jail_list="ns1 netstat"
> jail_set_hostname_allow="NO"
> 
> # ns1 jail
> jail_ns1_rootdir="/usr/jail/ns1"
> jail_ns1_hostname="ns1.csub.edu"
> jail_ns1_ip="136.168.1.91"
> jail_ns1_exec_start="/bin/sh /etc/rc"
> jail_ns1_devfs_enable="YES"
> jail_ns1_mount_enable="YES"
> 
> # netstat jail
> jail_netstat_rootdir="/usr/jail/netstat"
> jail_netstat_hostname="netstat.csub.edu"
> jail_netstat_ip="136.168.1.8"
> jail_netstat_exec_start="/bin/sh /etc/rc"
> jail_netstat_devfs_enable="YES"
> jail_netstat_mount_enable="YES"
> 
>JID  IP Address  Hostname  Path
>  8  136.168.1.91ns1.csub.edu  /usr/jail/ns1
>  4  136.168.1.8 netstat.csub.edu  /usr/jail/netstat
>  2  136.168.1.91ns1.csub.edu  /usr/jail/ns1
> 

I should have mentioned I'm running a 6.1-STABLE system built on the
21st of Aug. RELEASE had problems with interrupt storms if I recall
correctly.

Here's dmesg.boot if it helps any:

Copyright (c) 1992-2006 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
The Regents of the University of California. All rights reserved.
FreeBSD 6.1-STABLE #0: Mon Aug 21 00:59:05 PDT 2006
[EMAIL PROTECTED]:/usr/obj/usr/src/sys/NETSTAT
ACPI APIC Table: 
Timecounter "i8254" frequency 1193182 Hz quality 0
CPU: Intel(R) Xeon(TM) CPU 3.00GHz (2992.71-MHz 686-class CPU)
  Origin = "GenuineIntel"  Id = 0xf43  Stepping = 3
  
Features=0xbfebfbff
  Features2=0x641d>
  AMD Features=0x2010
real memory  = 2147221504 (2047 MB)
avail memory = 2096189440 (1999 MB)
FreeBSD/SMP: Multiprocessor System Detected: 2 CPUs
 cpu0 (BSP): APIC ID:  0
 cpu1 (AP): APIC ID:  6
ioapic0: Changing APIC ID to 7
ioapic1: Changing APIC ID to 8
ioapic1: WARNING: intbase 32 != expected base 24
ioapic2: Changing APIC ID to 9
ioapic2: WARNING: intbase 64 != expected base 56
ioapic3: Changing APIC ID to 10
ioapic3: WARNING: intbase 96 != expected base 88
ioapic0  irqs 0-23 on motherboard
ioapic1  irqs 32-55 on motherboard
ioapic2  irqs 64-87 on motherboard
ioapic3  irqs 96-119 on motherboard
acpi0:  on motherboard
acpi0: Power Button (fixed)
Timecounter "ACPI-fast" frequency 3579545 Hz quality 1000
acpi_timer0: <24-bit timer at 3.579545MHz> port 0x808-0x80b on acpi0
cpu0:  on acpi0
cpu1:  on acpi0
pcib0:  port 0xcf8-0xcff on acpi0
pci0:  on pcib0
pcib1:  at device 2.0 on pci0
pci1:  on pcib1
pcib2:  at device 0.0 on pci1
pci2:  on pcib2
amr0:  mem 0xf80f-0xf80f,0xfeac-0xfeaf 
irq 46 at device 14.0 on pci2
amr0: delete logical drives supported by controller
amr0:  Firmware 521X, BIOS H430, 256MB RAM
pcib3:  at device 0.2 on pci1
pci3:  on pcib3
pcib4:  at device 4.0 on pci0
pci4:  on pcib4
pcib5:  at device 5.0 on pci0
pci5:  on pcib5
pcib6:  at device 0.0 on pci5
pci6:  on pcib6
em0:  port 0xecc0-0xecff 
mem 0xfe7e-0xfe7f irq 64 at device 7.0 on pci6
em0: Ethernet address: 00:13:72:4b:70:e7
em0: [FAST]
pcib7:  at device 0.2 on pci5
pci7:  on pcib7
em1:  port 0xdcc0-0xdcff 
mem 0xfe5e-0xfe5f irq 65 at device 8.0 on pci7
em1: Ethernet address: 00:13:72:4b:70:e8
em1: [FAST]
pcib8:  at device 6.0 on pci0
pci8:  on pcib8
pcib9:  at device 0.0 on pci8
pci9:  on pcib9
pcib10:  at device 0.2 on pci8
pci10:  on pcib10
pcib11:  at device 30.0 on pci0
pci11:  on pcib11
pc

6.2/pxe blues, was Re: pxe boot size limit?

2006-09-19 Thread Danny Braniss 
> it seems that pxeboot has a limit with respect to the kernel size,
> which prevents the kernel to get loaded, the error printed is
> slightly misleading.
> 
> the solution is to make a kernel with loadable modules, instead of
> compiled in.

not entirely true :-(
but changing kernel size has different results.

kernel is 6.2-PRERELEASE.
hosts are 64bits:
1 - amd Athlon dual core - nic is em
2 - intel Xeon dual core - nic is bc
3 - intel Pentium 4 K8   - nic is bge
1 & 3 are now booting, but 2 is panics with:

/boot/kernel/kernel text=0x430118 data=0x79450+0x3ccf0 
syms=[0x8+0x73278+0x8+0x62861]
panic: free: guard1 fail @ 0x7c6fe from /r+d/6.2/src/lib/libstand/ufs.c:699

a slightly older kernel, and bigger, boots just fine.
booting from local disk is also ok.

so, what is causing pxeboot to fail?


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

Re: loader can't load kernel on FreeBSD 6.1-p6.

2006-09-19 Thread Nikolay Pavlov 
On Saturday, 16 September 2006 at  9:48:09 +0300, Nikolay Pavlov wrote:
> Hi, folks.
> While booting 6.1-RELEASE-p6 i have this error on /boot/loader stage:
> 
> can't load kernel
> 
> ls shows nothing...
> 
> lsdev shows me:
> disk devices:
>   disk0: BIOS drive C:
>  disk0s1: FFS bad disklabel

Sorry for all that noise, guys. The problem is fixed. BIOS on that motherboard
don't use LBA mode while accessing disks. This "Auto" settings realy
stupid.

> 
> 
> But when i am booting from LiveCD i see all my labels, i can mount them,
> i can produce read/write operations.
> Other thing that confused me is that when i am using second stage  
> bootstrap file i see my root partition:
> 
> FreeBSD/i386 BOOT
> Default: 0:ad(0,a)/boot/loader
> boot: ?. .. .snap bin boot dev etc lib libexec mnt proc and so on...
> 
> But when i am trying to boot kernel (GENERIC) directly from bootstrap 
> i have "BTX halted" error, don't know if this actually supported.
> World was builded on other box and installed to this drive
> using tar.
> 
> Here is additional information that i can get from LiveCD:
> 
> [EMAIL PROTECTED]:~# fdisk /dev/ad0
> *** Working on device /dev/ad0 ***
> parameters extracted from in-core disklabel are:
> cylinders=19679 heads=16 sectors/track=255 (4080 blks/cyl)
> 
> Figures below won't work with BIOS for partitions not in cyl 1
> parameters to be used for BIOS calculations are:
> cylinders=19679 heads=16 sectors/track=255 (4080 blks/cyl)
> 
> Media sector size is 512
> Warning: BIOS sector numbering starts with sector 1
> Information from DOS bootblock is:
> The data for partition 1 is:
> sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD)
> start 63, size 80293185 (39205 Meg), flag 80 (active)
> beg: cyl 0/ head 1/ sector 1;
> end: cyl 807/ head 15/ sector 63
> The data for partition 2 is:
> 
> The data for partition 3 is:
> 
> The data for partition 4 is:
> 
> 
> [EMAIL PROTECTED]:~# disklabel /dev/ad0s1
> # /dev/ad0s1:
> 8 partitions:
> #size   offsetfstype   [fsize bsize bps/cpg]
>   a:   52428804.2BSD 2048 16384 32776
>   b:  4161280   524288  swap
>   c: 802931850unused0 0 # "raw" part,
> don't edit
>   d:  2097152  46855684.2BSD 2048 16384 28552
>   e: 20971520  67827204.2BSD 2048 16384 28552
>   f: 10485760 277542404.2BSD 2048 16384 28552
>   g: 42053185 38244.2BSD 2048 16384 28552
> 
> 
> [EMAIL PROTECTED]:~# atacontrol list
> ATA channel 0:
> Master:  ad0  ATA/ATAPI revision 7
> Slave:   no device present
> ATA channel 1:
> Master:  no device present
> Slave:   no device present
> 
> [EMAIL PROTECTED]:~# mount
> /dev/iso9660/FreeSBIE on / (cd9660, local, read-only)
> devfs on /dev (devfs, local)
> /dev/md0.uzip on /usr (ufs, local, read-only)
> /dev/md1 on /etc (ufs, local)
> /dev/md2 on /usr/local/etc (ufs, local)
> /dev/md3 on /root (ufs, local)
> /dev/md4 on /var (ufs, local)
> /dev/md5 on /tmp (ufs, local)
> /dev/ad0s1a on /mnt/ufs.1 (ufs, local, read-only)
> /dev/ad0s1d on /mnt/ufs.2 (ufs, local, read-only)
> /dev/ad0s1e on /mnt/ufs.3 (ufs, local, read-only)
> /dev/ad0s1f on /mnt/ufs.4 (ufs, local, read-only)
> /dev/ad0s1g on /mnt/ufs.5 (ufs, local, read-only)
> 
> [EMAIL PROTECTED]:~# ls -al /mnt/ufs.1/boot/
> total 588
> drwxr-xr-x   5 root  wheel 512 Sep 16 00:52 .
> drwxr-xr-x  20 root  wheel 512 Sep 15 10:05 ..
> -r--r--r--   1 root  wheel7638 Sep  6 18:43 beastie.4th
> -r--r--r--   1 root  wheel8192 Sep  6 18:43 boot
> -r--r--r--   1 root  wheel 512 Sep  6 18:43 boot0
> -r--r--r--   1 root  wheel 512 Sep  6 18:43 boot0sio
> -r--r--r--   1 root  wheel 512 Sep  6 18:43 boot1
> -r--r--r--   1 root  wheel7680 Sep  6 18:43 boot2
> -r--r--r--   1 root  wheel1200 Sep  6 18:43 cdboot
> drwxr-xr-x   2 root  wheel 512 Sep  6 18:43 defaults
> -r--r--r--   1 root  wheel1741 Sep  6 18:43 device.hints
> -r--r--r--   1 root  wheel2249 Sep  6 18:43 frames.4th
> drwxr-xr-x   2 root  wheel9216 Sep  6 18:44 kernel
> -r-xr-xr-x   1 root  wheel  212992 Sep  6 18:43 loader
> -r--r--r--   1 root  wheel7772 Sep  6 18:43 loader.4th
> -rw-r--r--   1 root  wheel 463 Sep  6 18:44 loader.conf
> -r--r--r--   1 root  wheel   15058 Sep  6 18:43 loader.help
> -r--r--r--   1 root  wheel 389 Sep  6 18:43 loader.rc
> -r--r--r--   1 root  wheel 512 Sep  6 18:43 mbr
> drwxr-xr-x   2 root  wheel 512 Sep  6 18:41 modules
> -r--r--r--   1 root  wheel  215040 Sep  6 18:43 pxeboot
> -r--r--r--   1 root  wheel 692 Sep  6 18:43 screen.4th
> -r--r--r--   1 root  wheel   36440 Sep  6 18:43 support.4th
> 
> [EMAIL PROTECTED]:~# ls -al /mnt/ufs.1/boot/kernel/kernel
> -r-xr-xr-x  1 root  wheel  6640118 Sep  6 18:21 /mnt/ufs.1/boot/kernel/kernel
> 
> [EMAIL PROTECTED]:~# file /mnt/ufs.1/boot/kernel/kernel
> /mnt/ufs.1/boot/kernel/kernel: ELF 32-bit LSB executable, Intel 80386,
> version 1 (FreeBSD), dyn

Re: bind round robin

2006-09-19 Thread Doug Barton 
Chris H. wrote:
> Greetings all,
> ...
> Quoting Doug Barton <[EMAIL PROTECTED]>:
> 
>> Oliver Brandmueller wrote:
>>>
>>> DNS round robin is not about redundancy, the only thing you could have
>>> that way is a kind of load balancing (not the most sophisticated way,
>>> though). Whenever one of the servers fails, around half of the requests
>>> still goes there and then times out/gets conn refused or whatever the
>>> problem is. Prioritizing is not easily possible. Probably it helps if
>>> you add one of the IPs more often to the set, but I never tried that and
>>> did not read the docs on this topic, so before breaking your zone first
>>> read the specs, if this works!
>>
>> Just replying to this bit first, in BIND it does not work to specify
>> the same IP address multiple times for the same hostname. The server
>> will collapse the duplicates into one unique entry when it reads the
>> zone. I am not aware of any other authoritative name server for which
>> this would work either.
> 
> While this /might/ hold true in some/certain situations.

Under the circumstances that Oliver suggested, what I said holds true
in every situation (assuming you are using BIND). The example you
pasted, while colorful, is not actually an example of what Oliver
suggested. If you would like me to write out an example I will, but:
A) This subject is already off topic, and
B) It would more usefully be left as an exercise for the reader.

> I /can/ say after 3.5 yrs. of doing exactly this,

Bzzzt. See above.

> that it does not collapse the namespace into a single IP<-->name.

It might also be useful to note here that nothing about DNS is
(automatically) bi-directional in the manner you imply here.

I do concur with your suggestion to move this thread to a list that is
focused on DNS, however 

Doug

-- 

This .signature sanitized for your protection

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