Re: Binary upgrade from legacy version + ports

2008-07-27 Thread Svein Halvor Halvorsen
Svein Halvor Halvorsen wrote:
 I want to upgrade two freebsd machines I have from 6.1-SECURITY
 and 5.3-RELEASE respectively, to the latest 7.0 release of
 FreeBSD. I don't want to cvsup and build, but prefer to use
 prebuilt binaries. Also I'd like to avoid wiping the systems,
 and starting afresh.

Matthew Seaman wrote:
 If your 6.1 system is using a system installed from one of the
 official iso images and hasn't been locally rebuilt (upgrading
 via freebsd-updates is OK though) then there is a quicker way.
 See
 
 http://www.daemonology.net/blog/2007-11.html

Now this is what I'm talking about!

I can't relly remember anymore, since it's so long ago, but I'm
pretty sure I upgraded the box that is now 6.1 from some 5.x
version, taking a binary route. I can't find the instructions
anymore. Maybe I'm mistaken :-/

Upgrading FreeBSD across major versions is such a drag, that I
allways postpone it. And now it's way overdue. I guess I just have
to bite the bullet and wipe the system clean, and start afresh.

The 6.1 system seem to be running 7.0-p2 now, though. Thanks!
It is still upgrading ports. Fortunately this box does nothing
important, and is just my personal playground. If something gets
messed up, noe problem. I'm trying to upgrade the ports binary as
well, using portupgrade -a -PP, and it seems to get the packages
from RELEASE, not the latest versions that are in the ports tree.
But as long as all ports use the same consistent set, I'm pretty
sure it will work out nicely. My local ports tree will be out of
sync, though, which might cause problems later?


Is there a problem using the prebuilt packages from STABLE on a
RELEASE box? If I want to run RELEASE, and still use the latest
packages? The ABI is consistent between STABLE and RELEASE, right?

Also, re: the freebsd-update, if I've built a system from source,
say 7.0-RELEASE. And I want to upgrade it binary to -p2. If my home
built system is basically like the prebuilt release, I should be
fine, but freebsd-update won't let me do this, right? Anyway to
overcome this?


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


Re: Root boot/mount Password?

2008-07-27 Thread DSA - JCR
On Sat, Jul 26, 2008 at 05:31:23PM -, DSA - JCR wrote:
 Hi all

 FreeBSD 6.2

 I would like to put a password when booting/mounting mi Freebsd box. is
it possible? How?

Yes. Use geli(8) encryption.

 is for protecting the system from unauthorized users

Disk encryption also protects your data if the PC or harddrive is stolen.
Roland


Yes, I had thinking of Geli, but my system is up and running and I don't
know if I can use geli for this without breaking all
I have used geli for unused disks and for swap but not for root, because i
dont know if I will break all

can I use it for root, when it is a live system?


Of course i don't think in methods like BIOS because simply taking the
battery out can reset all.



Juan Coruña
Desarrollo de Software Atlantico






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


Re: Binary upgrade from legacy version + ports

2008-07-27 Thread Matthew Seaman

Svein Halvor Halvorsen wrote:


The 6.1 system seem to be running 7.0-p2 now, though. Thanks!
It is still upgrading ports. Fortunately this box does nothing
important, and is just my personal playground. If something gets
messed up, noe problem. I'm trying to upgrade the ports binary as
well, using portupgrade -a -PP, and it seems to get the packages
from RELEASE, not the latest versions that are in the ports tree.
But as long as all ports use the same consistent set, I'm pretty
sure it will work out nicely. My local ports tree will be out of
sync, though, which might cause problems later?


Yes, this will work in the sense that the software will all be
functional.  If you're installing the RELEASE packages, you won't
get any security related fixes that have gone into the ports since 
7.0-RELEASE came out.  Try running 'portaudit -Fda' and see what it

shows up.

You should still be able to update using packages from the FTP servers
though -- the process of updating packages runs pretty frequently, but
it does take a while to chew through all of the updates so what's on the
FTP servers is usually some days adrift of the state of the ports tree
in CVS.  But really no more than that.


Is there a problem using the prebuilt packages from STABLE on a
RELEASE box? If I want to run RELEASE, and still use the latest
packages? The ABI is consistent between STABLE and RELEASE, right?


Nope.  All 7.x releases should be ABI compatible -- including STABLE
and RELEASE.  So long as you get the inter-package dependencies right
everything should work fine.  I know the ABI promise guarantees that
anything compiled on an earlier 7.x version will continue to run on a
later one, and I believe it also now requires programs compiled on a
later 7.x system to run on an earlier one.

In any case, the move to versioned symbols makes the whole problem
pretty much go away.

Cheers,

Matthew

--
Dr Matthew J Seaman MA, D.Phil.   7 Priory Courtyard
 Flat 3
PGP: http://www.infracaninophile.co.uk/pgpkey Ramsgate
 Kent, CT11 9PW



signature.asc
Description: OpenPGP digital signature


Re: malloc options

2008-07-27 Thread Kris Kennaway

Doug Hardie wrote:


On Jul 26, 2008, at 19:03, Giorgos Keramidas wrote:


While that's understandable, the current malloc() has undergone quite
extensive testing by Jason Evans and a lot of people who use it in
FreeBSD 7.X or later.  Its ability to expose bugs in this way was deemed
important enough that it is now used by other projects too.


while in general I like the new approach, this problem has been a 
killer.  I did find a number of errors in my own code where I was not 
allocating enough space for some things.  Those showed up instantly with 
7.0 and were easy to fix.


I am not sure you are getting the point.  The bugs are exposed by 
accident, not by design, because the programs were only working by 
accident, not by design.  There will probably be a new subset of buggy 
programs that sometimes fail to crash under the new allocator.



What Kris wrote in:

   Finally, there is no way to revert to the old approach
   because the new allocator is completely new; it allocates
   memory based on its own strategy.  None of the malloc options
   affect the behaviour of correct programs (but some of them
   can help to improve performance, or to debug incorrect
   programs).

is a bit important.  Even if you tweak enough options the new malloc()
may *not* work similarly enough for the program to keep working.  If you
are lsing money _right_ _now_ because of problems in the program, it may
be worth going back to 6-STABLE and the old malloc() until the bugs of
the program have been fixed by the developers.


Unfortunately that is not possible.  We upgraded the hardware and some 
of the components were not supported very well under 6.x.  Despite 
several weeks of testing of the new hardware and 7.0, the problem did 
not arise till several weeks after going into production.  It takes 
about a week of real time before the problem tends to become visible.  
By compressing the workload I have been able to setup a test machine 
such that it takes 2-4 days before it occurs.


Your choices are:

1) Debug the application

2) Build a 6.x binary and use that forever under compatibility

3) Compile your own private copy of phkmalloc and link your application 
to it, and hope the bugs never return.


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


Re: Root boot/mount Password?

2008-07-27 Thread Roland Smith
On Sun, Jul 27, 2008 at 09:47:44AM -, DSA - JCR wrote:
 On Sat, Jul 26, 2008 at 05:31:23PM -, DSA - JCR wrote:
  Hi all
 
  FreeBSD 6.2
 
  I would like to put a password when booting/mounting mi Freebsd box. is
  it possible? How?
 
 Yes. Use geli(8) encryption.
 
  is for protecting the system from unauthorized users
 
 Disk encryption also protects your data if the PC or harddrive is stolen.
 Roland
 
 
 Yes, I had thinking of Geli, but my system is up and running and I don't
 know if I can use geli for this without breaking all
 I have used geli for unused disks and for swap but not for root, because i
 dont know if I will break all
 
 can I use it for root, when it is a live system?

You can encrypt the root filesystem, but in that case /boot must be on a
separate unencrypted partition, otherwise the OS cannot boot. So unless
you have a spare partition for /boot, you'll have to make backups and
re-partition your disk.

Note that encrypting the partitions where the OS lives is not
particularly usefull; there is nothing secret there. On the contrary, it
would potentially make the encrypted partition vulnerable to a known
plaintext attack.

So what I would recommend it to put all _your_ data (which you want to
protect from unauthorized access) on one partition (in case of a
desktop, I'd use /home), and encrypt that. To do this you should back up
all your data. Then you fill the partition with random noise using 'dd
if=/dev/random'. This can take some time depending on the size of the
partition. As soon as that is done you can use 'geli init' to initialize
a geli-encrypted device, and 'geli attach' to make a device node. Then
you can use newfs on the new device, mount it and restore your
backup. Now edit /etc/fstab to refer to the geli device. On the next
boot, the rc scripts will ask for the password and take care of the
mounting of the device.

Roland
-- 
R.F.Smith   http://www.xs4all.nl/~rsmith/
[plain text _non-HTML_ PGP/GnuPG encrypted/signed email much appreciated]
pgp: 1A2B 477F 9970 BA3C 2914  B7CE 1277 EFB0 C321 A725 (KeyID: C321A725)


pgpcuWzENqk5E.pgp
Description: PGP signature


Deinstalling X and all dependencies

2008-07-27 Thread bsd

Hello,

I have just received a new system that's planned to be a large scale  
DNS server.

I have asked the guy who has setup the hardware not to install X…

This has been useless!!

I am now ending up with 250 apps in the port tree!!

Is there a good way to get rid of all these useless apps without  
breaking the system…

What would you suggest?

Like removing X and It's dependencies…


I can also remove all apps in the port tree and recompile only the one  
needed…




What's best what do you suggest.




I'd rather do painful jobs now than in 6 months when everything will  
be up and running !!



Thanks.


Gregober --- PGP ID -- 0x1BA3C2FD
bsd @at@ todoo.biz


P Please consider your environmental responsibility before printing  
this e-mail



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


7.0-make.conf

2008-07-27 Thread JD Bronson

Has this been removed or is it still supported?
It does not appear in the man page or examples...


NO_BIND=true



-JD

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


Re: Deinstalling X and all dependencies

2008-07-27 Thread N. Raghavendra
At 2008-07-27T12:52:56+02:00, bsd wrote:

 Is there a good way to get rid of all these useless apps without
 breaking the system… What would you suggest?

One way is to use the `ports-mgmt/pkg_cutleaves' port to iteratively
remove the superfluous leaves of your package tree.  It can also be
done with `ports-mgmt/portmanager' (with the -slid option) or
`ports-mgmt/portmaster' (with the -s option).

Raghavendra.

-- 
N. Raghavendra [EMAIL PROTECTED] | http://www.retrotexts.net/
Harish-Chandra Research Institute   | http://www.mri.ernet.in/
See message headers for contact and OpenPGP information.

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


Re: Deinstalling X and all dependencies

2008-07-27 Thread Bernt Hansson

bsd :

Hello,

I have just received a new system that's planned to be a large scale DNS 
server.

I have asked the guy who has setup the hardware not to install X…

This has been useless!!


Better doing it your self.


I am now ending up with 250 apps in the port tree!!

Is there a good way to get rid of all these useless apps without 
breaking the system…

What would you suggest?


pkg_delete *

This will remove ALL packages installed.

It has worked for me in the past but be careful!
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: Deinstalling X and all dependencies

2008-07-27 Thread RW
On Sun, 27 Jul 2008 12:52:56 +0200
bsd [EMAIL PROTECTED] wrote:

 Hello,
 
 I have just received a new system that's planned to be a large scale  
 DNS server.
 I have asked the guy who has setup the hardware not to install X___
 
 This has been useless!!
 
 I am now ending up with 250 apps in the port tree!!
 
 Is there a good way to get rid of all these useless apps without  
 breaking the system___

If you want to remove X you can use  a leaf-cutting tool like
ports-mgmt/pkg_cutleaves.

But I would have thought that a dns server would require only very few
ports (possibly even zero if you use the default BIND), so it might
be simpler to start over.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: 7.0-make.conf

2008-07-27 Thread JD Bronson

At 03:03 PM 7/27/2008 +0300, Reko Turja wrote:
make.conf has been split into two, the actual make.conf which has 
variables for the make process and generic make environment and 
src.conf which controls the building of add-on software. Check 
src.conf for details.


-Reko


so something like this it seems:

WITHOUT_BIND=true
WITHOUT_GAMES=true
WITHOUT_MAILWRAPPER=true
WITHOUT_OPENSSH=true
WITHOUT_SENDMAIL=true

I am not sure of the need for the 'true' or not. it seems it is not
required but should work either way?

-JD 


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


Re: 7.0-make.conf

2008-07-27 Thread Reko Turja

Has this been removed or is it still supported?
It does not appear in the man page or examples...


NO_BIND=true


make.conf has been split into two, the actual make.conf which has 
variables for the make process and generic make environment and 
src.conf which controls the building of add-on software. Check 
src.conf for details.


-Reko 


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


Re: 7.0-make.conf

2008-07-27 Thread Reko Turja

WITHOUT_SENDMAIL=true

I am not sure of the need for the 'true' or not. it seems it is not
required but should work either way?



From the manpage:



The values of variables are ignored regardless of their setting; 
even if
they would be set to ``FALSE'' or ``NO''.  Just the existence of 
an

option will cause it to be honoured by make(1).

so the plain option is enough. If the machines are in environment 
where they might be someday administered by someone else, I'd use pure 
option to avoid confusing someone who hasn't read the manpage and 
thinks setting variables to false will void them.


-Reko 


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


Re: Deinstalling X and all dependencies

2008-07-27 Thread andrew clarke
On Sun 2008-07-27 12:52:56 UTC+0200, bsd ([EMAIL PROTECTED]) wrote:

 I have just received a new system that's planned to be a large scale DNS 
 server.
 I have asked the guy who has setup the hardware not to install X?

 This has been useless!!

 I am now ending up with 250 apps in the port tree!!

He probably just went with the defaults.

 Is there a good way to get rid of all these useless apps without  
 breaking the system?
 What would you suggest?

 Like removing X and It's dependencies?


 I can also remove all apps in the port tree and recompile only the one  
 needed?

 What's best what do you suggest.

FreeBSD provides a base system with software such as a SSH daemon,
Sendmail, BIND, etc.  You can uninstall all the packages on your
system, but the FreeBSD base system will still remain.  This allows
FreeBSD to boot normally without any packages installed.

I recommend you uninstall all packages (with 'pkg_delete -a', or
'pkg_delete -av' if you want to watch all the files being deleted),
then install only what you need from the Ports tree.

Your DNS server should probably not require any packages to be
installed, as DNS server software (BIND) is provided with the FreeBSD
base system.  But that really depends what your requirements are.

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


Re: Deinstalling X and all dependencies

2008-07-27 Thread Polytropon
On Sun, 27 Jul 2008 23:17:44 +1000, andrew clarke [EMAIL PROTECTED] wrote:
 On Sun 2008-07-27 12:52:56 UTC+0200, bsd ([EMAIL PROTECTED]) wrote:
 
  I have just received a new system that's planned to be a large scale DNS 
  server.
  I have asked the guy who has setup the hardware not to install X?
 
  This has been useless!!
 
  I am now ending up with 250 apps in the port tree!!

In or from? If they are in the ports tree but not installed,
don't mind. If they are installed, deinstall them as recommended,
using pkg_delete for example.



  I can also remove all apps in the port tree and recompile only the one  
  needed?
 
  What's best what do you suggest.
 
 FreeBSD provides a base system with software such as a SSH daemon,
 Sendmail, BIND, etc.  You can uninstall all the packages on your
 system, but the FreeBSD base system will still remain.  This allows
 FreeBSD to boot normally without any packages installed.

Very well you mentioned this - this difference between base OS
and installable packages is one of the most important features
of FreeBSD to me.

Having said this, all the additional software (from ports or from
packages) reside within /usr/local; everything outside /usr/local
belongs to the OS.

This means you can

# rm -rf /usr/local
# mtree -f /etc/mtree/BSD.local.dist

and then start installing the software you want. The base system
won't be affected at all.

# cd /usr/ports/category/port
# make install package clean

or

# pkg_add -r what you want

So you end up only with the things you intendedly install (including
the needed dependencies).

This is the way I did setup a 5.X system many years ago which died
this month due to a problem killing various inodes... :-(



 I recommend you uninstall all packages (with 'pkg_delete -a', or
 'pkg_delete -av' if you want to watch all the files being deleted),
 then install only what you need from the Ports tree.

You can, of course, just deinstall the packages you know you won't
need, but as you said, if there are more than 250 of them installed
(related to X, maybe Gnome or KDE, too), it's easier to invest some
time and build from scratch, just as you need.



 Your DNS server should probably not require any packages to be
 installed, as DNS server software (BIND) is provided with the FreeBSD
 base system.

See? Everything there from the base install. :-)



 But that really depends what your requirements are.

Exactly.


-- 
Polytropon
From Magdeburg, Germany
Happy FreeBSD user since 4.0
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


pptp and routing

2008-07-27 Thread Andrew D

G'Day all,

got a freebsd Box
FreeBSD gw.ade.eltrak.com.au 7.0-STABLE FreeBSD 7.0-STABLE #0: Wed Jul 
9 03:46:03 CST 2008 
[EMAIL PROTECTED]:/usr/obj/usr/src/sys/ELKERN  i386


that has a poptop server on it. When a client logs in they get authed 
fine and get issued an IP. However when connecting or pinging no data 
comes back from the server.  arpproxy is set, as is forwarding.

net.inet.ip.forwarding: 1
net.link.ether.inet.proxyall: 1

The server for some reason puts a route for the client ip on the 
ethernet interface rather than the tun interface the client has come in on.


/etc/ppp/ppp.conf


loop:
set timeout 0
set log phase chat connect lcp ipcp command
set device localhost:pptp
set dial
set login
set ifaddr 10.10.1.5 10.10.1.20-10.10.1.60 255.255.255.0
add default HISADDR
set server /tmp/loop  0177

loop-in:
set timeout 0
set log phase lcp ipcp command
allow mode direct

pptp:
load loop
disable pap
enable passwdauth
disable ipv6cp
enable proxy
accept dns
enable MSChapV2
enable mppe
disable deflate pred1
deny deflate pred1
set dns 10.10.1.5
set device !/etc/ppp/secure



/etc/ppp/secure

#!/bin/sh
exec /usr/sbin/ppp -direct loop-in



/usr/local/etc/pptpd.conf


localip 10.10.1.5
remoteip 10.10.1.20-60
pidfile /var/run/pptpd.pid
noipparam
debug



$ ifconfig
fxp0: flags=8943UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST metric 
0 mtu 1500

options=8VLAN_MTU
ether 00:04:ac:98:d2:c6
inet 10.10.1.5 netmask 0xff00 broadcast 10.10.1.255
media: Ethernet autoselect (100baseTX full-duplex)
status: active
re0: flags=8802BROADCAST,SIMPLEX,MULTICAST metric 0 mtu 1500

options=399bRXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,TSO4,WOL_UCAST,WOL_MCAST,WOL_MAGIC
ether 00:40:ca:23:ed:5f
media: Ethernet autoselect (100baseTX half-duplex)
status: no carrier
lo0: flags=8049UP,LOOPBACK,RUNNING,MULTICAST metric 0 mtu 16384
inet6 fe80::1%lo0 prefixlen 64 scopeid 0x3
inet6 ::1 prefixlen 128
inet 127.0.0.1 netmask 0xff00
tun0: flags=8051UP,POINTOPOINT,RUNNING,MULTICAST metric 0 mtu 1500
inet 10.8.0.1 -- 10.8.0.2 netmask 0x
Opened by PID 775
tun1: flags=8051UP,POINTOPOINT,RUNNING,MULTICAST metric 0 mtu 1398
inet 10.10.1.5 -- 10.10.1.34 netmask 0xff00
Opened by PID 14740

$ netstat -nr
Routing tables

Internet:
DestinationGatewayFlagsRefs  Use  Netif Expire
default10.10.1.254UGS 029107   fxp0
10.8.0.0/2410.8.0.2   UGS 0  215   tun0
10.8.0.2   10.8.0.1   UH  2   45   tun0
10.10.1.0/24   link#1 UC  00   fxp0
10.10.1.5  00:04:ac:98:d2:c6  UHLW2  322lo0
10.10.1.34 10.10.1.5  UGH 00   fxp0
10.10.1.25400:1e:be:97:95:23  UHLW20   fxp0 
10.10.2.0/24   10.8.0.2   UGS 0  918   tun0

127.0.0.1  127.0.0.1  UH  0  208lo0



As you can see the client (10.10.1.34) is routed on the fxp0 interface 
rather than the tun1.


Anyone got any Ideas?

Cheers
cya
Andrew

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


Re: graid3

2008-07-27 Thread Wojciech Puchar

why it can't be say 5 disks+parity?


The reason is in the definition on RAID 3, which says the updates to the 
RAID device must be atomic. In some ideal universe, RAID 3 is implemented in 
hardware and on individual bytes, but here we cannot write to the drives in 
units other than sectorsize and sectorsize is 512 bytes.


OK i understand - the RAID sectors must be  2^something, so amount of 
drives must be 2^something+1.


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


Re: SATA300

2008-07-27 Thread Wojciech Puchar

ad5: 476940MB MAXTOR STM3500630AS 3.AAE at ata2-slave SATA150
ad6: 305245MB Hitachi HDT725032VLA360 V54OA7EA at ata3-master SATA150

Does this mean I'm only getting half the throughput I could be getting?

still more that actually drive can get from media (about 100MB/s)
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: malloc options

2008-07-27 Thread Ivan Voras

Doug Hardie wrote:


On Jul 26, 2008, at 19:03, Giorgos Keramidas wrote:


While that's understandable, the current malloc() has undergone quite
extensive testing by Jason Evans and a lot of people who use it in
FreeBSD 7.X or later.  Its ability to expose bugs in this way was deemed
important enough that it is now used by other projects too.


while in general I like the new approach, this problem has been a 
killer.  I did find a number of errors in my own code where I was not 
allocating enough space for some things.  Those showed up instantly with 
7.0 and were easy to fix.


As Kris said, you can run the 6.x binary on 7.x with compatibility 
libraries. This is the worst long-term option but it *will* gain you the 
7.x kernel (with all its benefits like multi-CPU support, new drivers, 
etc.) with the old allocator from the old libraries. This will be hard 
to maintain but it will work.




signature.asc
Description: OpenPGP digital signature


Re: malloc options

2008-07-27 Thread Wojciech Puchar
/etc/malloc.conf.  The default for that file is to not exist.  The man page 
does not indicate which settings are used in that situation.  After reading 
through it I get the feeling that the default settings for D and M are 'dM'. 
Hence, to return to the older malloc aproach to see if the problems go away I 
would need to set Dm.  But some of the descriptions seem to indicate that 
might not be correct.  What are the default settings?

no idea.

anyway - if this makes any non-performance difference to your programs - 
it's something wrong with it by design.


it should not.

if you need to manage memory your way, just use sbrk only and then manage 
it your way - as i do in most of my programs.

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


Re: Sendmail local LAN delivery

2008-07-27 Thread Bruno Joho
Hi Derek

thanks for the reply.
My intention was to deliver the mails between the workstations on the
LAN directly. Every Workstation
on the LAN would have an appropriate cf file which forwards mails with
a destination on the WAN - to the
WAN-Smarthost, any mail going to a destination from inside the LAN
would be delivered
directly to the destination host without involving a (LAN) smarthost.
Is that possible somehow?

Thanks for your help

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


Re: {Spam?} Re: Sendmail local LAN delivery

2008-07-27 Thread Derek Ragona

At 03:49 PM 7/27/2008, Bruno Joho wrote:

Hi Derek

thanks for the reply.
My intention was to deliver the mails between the workstations on the
LAN directly. Every Workstation
on the LAN would have an appropriate cf file which forwards mails with
a destination on the WAN - to the
WAN-Smarthost, any mail going to a destination from inside the LAN
would be delivered
directly to the destination host without involving a (LAN) smarthost.
Is that possible somehow?

Thanks for your help

Bruno


That scenario is possible, but you may still need DNS MX records to point 
to the smart host as well.  You can have multiple MX records for a domain 
(or subdomain) where the value field is higher for a farther away mail 
server, or for a secondary mail server.  Sendmail uses DNS to figure 
routing along with the internal configuration file UNLESS you specify to 
sendmail to NOT use DNS.


So you have some choices in how you configure sendmail and/or DNS.

Also it may help you to test things on one system and bump up the sendmail 
logging so the /var/log/maillog file has more information.  Use the option:

-O LogLevel=80
Or some other value than 80.  You can add these options to your sendmail 
flags in /etc/rc.conf


-Derek

--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

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


Algis Kimbaras is out of the office.

2008-07-27 Thread a . kimbaras

I will be out of the office starting  2008.07.27 and will not return until
2008.07.28.

Busiu isvykes nuo 2008-07-28 iki 2008-08-10. I Jusu laiska atsakysiu
grizes. Skubiais atvejais prasau kreiptis i Lina Linkaite tel. 8 (5)
2786278 arba e-mail'u [EMAIL PROTECTED]

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


freebsd networking issues

2008-07-27 Thread gahn
hello:

i have two interfaces installed on my freebsd machine (desktop); one is 
wireless (ath0, facing the internet, 192.168.1.10/24) and another is internal 
(fxp0, 192.168.2.1/24). the internet facing interface of the freebsd works 
fine: from my xp laptop (192.168.1.2), i can ping the interface of freebsd 
(ath0, facing internet, 192.168.1.10). also from the freebsd machine, i can 
ping the default gateway (192.168.1.1) and get access to internet. 

but the second interface doesn't work. from the freebsd machine, i can't ping 
anywhere on the 192.168.2.0/24 except own ip address. with arp -a on freebsd, 
i can only see mac address of own interface (fxp0), but not mac addresses of 
other machinese on the net 192.168.2.0/24.

i connected internet side directly (without ath0) with fxp0 and the interface 
works fine.

the bottom line is: on the net 192.168.2.0/24, i can't see any mac addresses. 

in the file /etc/rc.conf, i have the line gateway_enbale=YES but i don't 
think it matters since i have no intention to use the freebsd machine as a 
router anyway.

any ideas?

thanks in advance.



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


Re: malloc options

2008-07-27 Thread Karl Vogel
 On Sat, 26 Jul 2008 17:36:35 -0700,
 Doug Hardie [EMAIL PROTECTED] wrote:

D The program has worked under considerable load for many years with versions
D 3.7 to 6.2.  Problems only occur with 7.0.  The program is quite complex
D and big.  It uses probably hundreds of mallocs in a typical use.  The
D problems only occur reasonably randomly and only under quite heavy load.
D The developer is looking into it, but the problem only occurs on FreeBSD
D 7.0, not any other Unix systems.  In the meantime I am losing money because
D of it.

 On Sun, 27 Jul 2008 05:03:58 +0300,
 Giorgos Keramidas [EMAIL PROTECTED] said:

G While that's understandable, the current malloc() has undergone quite
G extensive testing by Jason Evans and a lot of people who use it in FreeBSD
G 7.X or later.  Its ability to expose bugs in this way was deemed important
G enough that it is now used by other projects too.

   I ran into a similar problem with the BSD allocator running under heavy
   load that didn't happen under any Solaris or Linux system I used.  I
   finally fixed it by using Doug Lea's malloc just for this one application:
 http://shell.siscom.net/~vogelke/Software/Languages/C/Libraries/malloc/

   This was under FreeBSD 6.*, but it might provide another data point if you
   want to give it a try.

-- 
Karl Vogel  I don't speak for the USAF or my company
vogelke at pobox dot com   http://www.pobox.com/~vogelke

And God said, Let there be vodka!  And saw that it was good.  Then God
said, Let there be light!  And then said, Whoa - too much light.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: malloc options

2008-07-27 Thread Giorgos Keramidas
On Sun, 27 Jul 2008 21:55:39 -0400 (EDT), [EMAIL PROTECTED] (Karl Vogel) wrote:
 On Sat, 26 Jul 2008 17:36:35 -0700,
 Doug Hardie [EMAIL PROTECTED] wrote:

 D The program has worked under considerable load for many years with
 D versions 3.7 to 6.2.  Problems only occur with 7.0.  The program is
 D quite complex and big.  It uses probably hundreds of mallocs in a
 D typical use.  The problems only occur reasonably randomly and only
 D under quite heavy load.  The developer is looking into it, but the
 D problem only occurs on FreeBSD 7.0, not any other Unix systems.  In
 D the meantime I am losing money because of it.

 On Sun, 27 Jul 2008 05:03:58 +0300,
 Giorgos Keramidas [EMAIL PROTECTED] said:

 G While that's understandable, the current malloc() has undergone
 G quite extensive testing by Jason Evans and a lot of people who use
 G it in FreeBSD 7.X or later.  Its ability to expose bugs in this way
 G was deemed important enough that it is now used by other projects
 G too.

 I ran into a similar problem with the BSD allocator running under
 heavy load that didn't happen under any Solaris or Linux system I
 used.  I finally fixed it by using Doug Lea's malloc just for this one
 application:

  http://shell.siscom.net/~vogelke/Software/Languages/C/Libraries/malloc/

 This was under FreeBSD 6.*, but it might provide another data point if
 you want to give it a try.

I'm not sure how similar the two problems are.  I quite frankly know
_very_ little of what the original problem was, other than I am
encountering issues where values just seem to arbitrarily change.

Memory exhaustion is a potential problem with almost any sort of
allocator that fragments memory in any way, but random corruption of
user data is probably a different issue :/

If you have some sort of description of the workload that triggered the
memory exhaustion with jemalloc (the current malloc implementation in
FreeBSD), it's probably a good idea to talk to Jason Evans about it (his
email is jasone at FreeBSD.org).  He may be able to help you tune
malloc or even make changes to the system version of malloc that make it
less vulnerable to this sort of problem.

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