Getting a new system to see an old disk

[Paul Hoffman]

Greetings again. I have a well-running 4.8 system. I have created a 
new 4.9 system on a new hard drive, and I want to be able to mount 
the old drive so I can copy all the relevant stuff from it. However, 
the simple-minded stuff fails. For example, trying to mount the old / 
partition on the new /mnt, I get:

mount: /dev/ad1s1a on /mnt: incorrect super block

But the drive is definitely seeable. Here's the output of "disklabel s1a":

# /dev/ad1c:
type: ESDI
disk: ad0s2
label:
flags:
bytes/sector: 512
sectors/track: 63
tracks/cylinder: 255
sectors/cylinder: 16065
cylinders: 4856
sectors/unit: 78011640
rpm: 3600
interleave: 1
trackskew: 0
cylinderskew: 0
headswitch: 0   # milliseconds
track-to-track seek: 0  # milliseconds
drivedata: 0
8 partitions:
#size   offsetfstype   [fsize bsize bps/cpg]
  a:   26214404.2BSD 2048 1638494   # (Cyl.0 - 16*)
  b:  1016912   262144  swap# (Cyl.   16*- 79*)
  c: 780116400unused0 0 # (Cyl.0 - 4855)
  e:   524288  12790564.2BSD 2048 1638494   # (Cyl.   79*- 112*)
  f:   524288  18033444.2BSD 2048 1638494   # (Cyl.  112*- 144*)
  g: 75684008  23276324.2BSD 2048 1638489   # (Cyl.  144*- 4855*)
What do I need to do to make the partitions of this new drive mountable?

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

portupgrade for binaries fails miserably, but not completely

[Paul Hoffman]

Greetings again. I used freebsd-update to upgrade a system from 7.1 to 7.2. The 
main upgrade went fine, but upgrading the ports had huge problems. It could not 
find most (but not all) of the packages to upgrade. A typical part of the 
failure looks like:

--->  Checking for the latest package of 'lang/python25'
--->  Fetching the package(s) for 'python25-2.5.4_3' (lang/python25)
--->  Fetching python25-2.5.4_3
fetch: 
ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/packages-7.2-release/All/python25-2.5.4_3.tbz:
 File unavailable (e.g., file not found, no access)
** The command returned a non-zero exit status: 1
** Failed to fetch 
ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/packages-7.2-release/All/python25-2.5.4_3.tbz
fetch: 
ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/packages-7.2-release/All/python25-2.5.4_3.tgz:
 File unavailable (e.g., file not found, no access)
** The command returned a non-zero exit status: 1
** Failed to fetch 
ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/packages-7.2-release/All/python25-2.5.4_3.tgz
** Failed to fetch python25-2.5.4_3
** Listing the failed packages (-:ignored / *:skipped / !:failed)
! python25-2.5.4_3  (fetch error)
--->  Fetching the latest package(s) for 'python25' (lang/python25)
--->  Fetching python25
fetch: 
ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/packages-7.2-release/Latest/python25.tbz:
 Protocol error
** The command returned a non-zero exit status: 1
** Failed to fetch 
ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/packages-7.2-release/Latest/python25.tbz
fetch: 
ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/packages-7.2-release/Latest/python25.tgz:
 Protocol error
** The command returned a non-zero exit status: 1
** Failed to fetch 
ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/packages-7.2-release/Latest/python25.tgz
** Failed to fetch python25
** Listing the failed packages (-:ignored / *:skipped / !:failed)
! python25@ (fetch error)
** Could not find the latest version (2.5.4_3)
** No package available: lang/python25

It then goes on to the next package. It only got about 10% of the ports.

Am I missing something obvious here?

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

Re: portupgrade for binaries fails miserably, but not completely

[Paul Hoffman]

At 12:50 AM + 11/29/09, Frank Shute wrote:
>You don't mention what command you are using to upgrade your
>ports/packages.

Sorry: portupgrade -aPPR

>You do realise that you don't have to upgrade your ports if you go
>from 7.1 to 7.2. You can do but don't have to.

No, I didn't realize that. The FreeBSD Handbook indicates differently.

>You should upgrade python25 to python26. See /usr/ports/UPDATING dated
>20090608 for instructions on how to do so.

Thanks, but that didn't help. So, I have now done a long, painful 'portrupgrade 
-a', having it stop regularly to prompt me for configuration settings. We'll 
see what happens when I go to 8.

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

Why do I need a bunch of mappings for ld-elf.so in FreeBSD 8?

[Paul Hoffman]

Greetings. I upgraded a 7.2 system to 8.0 using 'freebsd-update install'. At 
some time during the process, I could no longer log in remotely because bash 
could not start due to /libexec/ld-elf.so not finding the right libraries. I 
added a bunch of lines to /etc/libmap.conf so that I could continue, and did a 
full 'portupgrade -af'. 'freebsd-update fetch' reports nothing to fetch, but I 
can't remove the lines from /etc/libmap.conf.

How do I get a /libexec/ld-elf.so that has up-to-date mappings internal to it?

FWIW:
-r-xr-xr-x  1 root  wheel  220788 Dec  7 08:41 /libexec/ld-elf.so.1

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

Re: Why do I need a bunch of mappings for ld-elf.so in FreeBSD 8?

[Paul Hoffman]

At 11:05 AM +0100 12/8/09, Ruben de Groot wrote:
>On Mon, Dec 07, 2009 at 05:35:47PM -0800, Paul Hoffman typed:
> > Greetings. I upgraded a 7.2 system to 8.0 using 'freebsd-update install'. 
> > At some time during the process, I could no longer log ...
>...snip...
>>... t remove the lines from /etc/libmap.conf.
>>
>> How do I get a /libexec/ld-elf.so that has up-to-date mappings internal to 
>> it?
>
>These mappings are not internal to ld-elf.so.

Ah, I interpreted that from the error message.

>Could you post the contents of your libmap.conf

These are the four I needed to get the normal packages working on my system. My 
concern is that there may be more:
libcrypt.so.4   libcrypt.so.5
libncurses.so.7 libncurses.so.8
libutil.so.7libutil.so.8
libcrypto.so.5  libcrypto.so.6

>and the output of
>ldd /usr/local/bin/bash ?

/usr/local/bin/bash:
libncurses.so.7 => not found (0x0)
libintl.so.8 => /usr/local/lib/libintl.so.8 (0x2811f000)
libiconv.so.3 => /usr/local/lib/libiconv.so.3 (0x28131000)
libc.so.7 => /lib/libc.so.7 (0x28227000)

>Also the errors bash gives when started without the mappings in libmap.conf.

Just commenting out the libncurses line, then trying to log in as a user whose 
shell is bash:
/libexec/ld-elf.so.1: Shared object "libncurses.so.7" not found, required by 
"bash"
Removing the commmenting immediately changes this back to working.

In looking more closely, I can see that bash was *not* remade by 'portupgrade 
-af'. I suspect it is because I installed it from binary during the initial 
installation. If that's true, it completely sucks and is a serious bug in 
portupgrade.

There was no work directory in /usr/ports/shells/bash. Doing a 'make; make 
deinstall; make reinstall' makes it so I don't need the map any more.


At 1:45 PM +0300 12/8/09, Odhiambo Washington wrote:
>I usually think that including COMPAT_FREEBS7 in your new kernel during the 
>upgrade process would save one from such agony. I have never used 
>freebsd-update ever, and might never, because I prefer to build a new system 
>from scratch, but perhaps you could try it and see if it does resolve your 
>problem.

I'm using the GENERIC kernel, never re-built.

>There is always this instruction that you need to recompile all installed 
>ports, which I think you did not do. That instruction makes me sick, given the 
>time it would take on a critical server.

I *did* do it, as I said in my earlier message. It takes forever, but it 
doesn't cause any interruption on the server (which, in my case, is only 
semi-critical).

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

Suppressing "Limiting icmp unreach response" log messages

[Paul Hoffman]

How can I eliminate the "Limiting icmp unreach response" messages 
from getting to /var/log/messages or to the console? I have a spate 
of them that is causing log rollovers. I think I know the source of 
the problem, but need to get rid of the messages first.

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

Finding exactly which commands, and in which order, rc is running at startup

[Paul Hoffman]

Greetings again. After doing a freebsd-update, my system is starting up 
differently than it was before. I want to figure out why before I come here and 
say "it's broken".

Is there a way to say "show me all of the commands you are running during 
startup"? It would be grand if I could say "tell me what you would do next time 
(dry run)", but "what did you do last time" is OK too.

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

Re: Finding exactly which commands, and in which order, rc is running at startup

[Paul Hoffman]

Thanks for all the suggestions. Of them, this was the one that helped me with 
my issue:

On Aug 23, 2013, at 1:41 AM, Doug Hardie  wrote:

> You can add:
> 
> rc_debug="YES"
> 
> to /etc/rc.conf and that might give you what you need.  According to the man 
> page it will "produces copious output to the terminal and syslog(3)"

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

Getting USB2 on FreeBSD 5.3?

[Paul Hoffman]

Greetings again. I have a no-name USB 2 controller card in my server 
running FreeBSD 5.3. I rebuilt the kernel with the ehci driver 
(simply by adding a line to copy of GENERIC and rebuilding). The box 
already has two USB 1 ports. The dmesg says:

. . .
ohci0:  mem 0xfe123000-0xfe123fff irq 18 
at device 3.0 on pci0
ohci0: [GIANT-LOCKED]
usb0: OHCI version 1.0
usb0:  on ohci0
usb0: USB revision 1.0
uhub0: NEC OHCI root hub, class 9/0, rev 1.00/1.00, addr 1
uhub0: 3 ports with 3 removable, self powered
ohci1:  mem 0xfe122000-0xfe122fff irq 19 
at device 3.1 on pci0
ohci1: [GIANT-LOCKED]
usb1: OHCI version 1.0
usb1:  on ohci1
usb1: USB revision 1.0
uhub1: NEC OHCI root hub, class 9/0, rev 1.00/1.00, addr 1
uhub1: 2 ports with 2 removable, self powered
ehci0:  mem 0xfe124000-0xfe1240ff 
irq 18 at device 3.2 on pci0
ehci0: [GIANT-LOCKED]
ehci_pci_attach: companion usb0
ehci_pci_attach: companion usb1
usb2: EHCI version 0.95
usb2: companion controllers, 3 ports each: usb0 usb1
usb2:  on ehci0
usb2: USB revision 2.0
uhub2: NEC EHCI root hub, class 9/0, rev 2.00/1.00, addr 1
uhub2: 5 ports with 5 removable, self powered
. . .
ohci2:  mem 0xfe12-0xfe120fff irq 
3 at device 15.2 on pci0
ohci2: [GIANT-LOCKED]
usb3: OHCI version 1.0, legacy support
usb3: SMM does not respond, resetting
usb3:  on ohci2
usb3: USB revision 1.0
uhub3: (0x1166) OHCI root hub, class 9/0, rev 1.00/1.00, addr 1
uhub3: 2 ports with 2 removable, self powered

When I plug in a USB 2 IDE adapter to the USB 2 card, I see:
umass0: Myson Century, Inc. USB Mass Storage Device, rev 2.00/a3.01, addr 2
da0 at umass-sim0 bus 0 target 0 lun 0
da0:  Fixed Direct Access SCSI-0 device
da0: 1.000MB/s transfers
da0: 114473MB (234441648 512 byte sectors: 255H 63S/T 14593C)
The transfers happen at USB 1, not USB 2 speeds, which is of course 
not what I want.

I can't tell why da0 is getting the slow (USB 1) mode. I also can't 
figure out what the heck ohci2 is; there is only the on-board USB 1 
ports (which I assume are ohci0 and ohci1) and the card (which is 
clearly ehci0). Is ohci2 possibly also running on the add-in card and 
stomping on ehci0?

I guess a related question is how can I tell what device umass0 
thinks it is plugged into?

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

Re: Getting USB2 on FreeBSD 5.3?

[Paul Hoffman]

At 9:00 PM -0500 1/7/05, John Wilson wrote:
On Fri, 07 Jan 2005 16:32:31 -0800
Paul Hoffman <[EMAIL PROTECTED]> wrote:
[...]
 da0:  Fixed Direct Access SCSI-0 device
 da0: 1.000MB/s transfers
[...]
For what it's worth, I too am using ehci with a USB2 HD based MP3 player, and
it is being reported similarly as well in regard to the 1.000MB/s transfer
rate.  However, according to `systat -vmstat`, I am seeing transfers in the
area of 8MB/s.  In regard to stability, I've had no issues at all with it.
I'm sure this doesn't help at all. ;p
Actually, it helps some. I am seeing similar results (well, 7.1 
MB/s), so this is a bit heartening. But it doesn't explain why we're 
not seeing the 50 MB/s we're supposed to be seeing.

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

How do I add a third ATA controller to 4.8?

[Paul Hoffman]

Greetings again. I have a Dell 600SC that has three ATA controllers 
on the motherboard, and I'm running 4.8 quite happily on it. The 
GENERIC kernel has the following:

device  ata0at isa? port IO_WD1 irq 14
device  ata1at isa? port IO_WD2 irq 15
Since I didn't know the irq used by the third controller (it's not 
listed in their documentation, of course!), I blithely tried adding:

device  ata2at isa? port IO_WD3

However, when rebuilding this kernel, I get the following error:

cc -c -O -pipe  -Wall -Wredundant-decls -Wnested-externs 
-Wstrict-prototypes  -Wmissing-prototypes -Wpointer-arith -Winline 
-Wcast-qual  -fformat-extensions -ansi  -nostdinc -I- -I. -I../.. 
-I../../../include -I../../contrib/ipfilter  -D_KERNEL -include 
opt_global.h  -mpreferred-stack-boundary=2  ioconf.c
ioconf.c:48: `IO_WD3' undeclared here (not in a function)
ioconf.c:48: initializer element is not constant
ioconf.c:48: (near initialization for `ata2_resources[1].u.longval')
*** Error code 1

So, what am I supposed to add to the kernel to add this third controller?

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

Re: How do I add a third ATA controller to 4.8?

[Paul Hoffman]

At 9:54 PM -0800 1/11/04, Kent Stewart wrote:
I have used several mobos that had 4 controllers on them. I did't have to do
anything but add an HD and turn them on in the bios.
I would have hoped that that was the case here too, but it doesn't 
seem to be. The new drive is definitely listed in the Dell's BIOS (I 
set it to "automatic" and on reboot the BIOS recognized its size 
correctly).

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

Re: How do I add a third ATA controller to 4.8?

[Paul Hoffman]

At 9:20 AM + 1/12/04, Martyn Hill wrote:
Paul

I notice that the kernel config file you present is using the "old" device
ata lines - is the ATA controller a particularly old model?
If not, then the LINT kernel config file shows an alternative, viz

deviceata# just one entry for all ata controllers, no need
to reference IRQs etc...
deviceatadisk# for your IDE disks
deviceatapicd# for your CD-ROM like devices
You should comment out the ata0/ata1 lines before adding the above (and
additionally, remove any "device wd*" lines in that section.)
Then rebuild your kernel.

Hope that helps.
Thanks to all for the effort, but it didn't. The dmesg output still 
shows only ata0 and ata1. I am starting to believe that the problem 
lies in the Dell, not in FreeBSD.

I'll work around this by attaching the drive to the second IDE 
controller and later experiment with adding a third controller on the 
PCI bus.

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

Total amount of memory in my system?

[Paul Hoffman]

Greetings again. I'm running 4.8 on a remote server. I want to know 
how much RAM is in the server. 'dmesg -a' doesn't tell me because the 
boot information has scrolled off the top of the stack. Short of 
rebooting the system, how can I find out how much RAM is built in?

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

Re: Total amount of memory in my system?

[Paul Hoffman]

At 10:17 PM -0800 1/26/04, James Long wrote:
less /var/run/dmesg.boot
Bingo. Thanks! I knew that they would keep that around somewhere...
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Disk no longer valid

[Paul Hoffman]

Greetings. After rebooting a system that had been up for quite some 
time, I could not get the second drive (da1) to mount. The drive 
worked fine before I rebooted. dmesg reported:

. . .
da1s4: rejecting partition in BSD label: it isn't entirely within the slice
da1s4: start 63, end 71132959, size 71132897
da1s4d: start 0, end 71132959, size 71132960
. . .
When I tried to access the disk, dmesg said:

. . .
(da1:ahc0:0:1:0): SCB 0x36 - timed out
Dump Card State Begins <<<<<<<<<<<<<<<<<
ahc0: Dumping Card State while idle, at SEQADDR 0x9
Card was paused
ACCUM = 0x0, SINDEX = 0x1e, DINDEX = 0xe4, ARG_2 = 0x0
HCNT = 0x0 SCBPTR = 0x17
SCSIPHASE[0x0] SCSISIGI[0x0] ERROR[0x0] SCSIBUSL[0x0]
[ lots more debugging info]

sg[0] - Addr 0x706d000 : Length 4096
sg[1] - Addr 0xdb6e000 : Length 4096
(da1:ahc0:0:1:0): Queuing a BDR SCB
(da1:ahc0:0:1:0): no longer in timeout, status = 34a
Feb  5 00:01:03 above /kernel: <<<<<<<<<<<<<<<< Dump Card State 
Ends >>>>>>>>>>>

(da1:ahc0:0:1:0): Invalidating pack
. . .
The drive was inaccessible. I rebooted again. dmesg still says:
. . .
da1s4: rejecting partition in BSD label: it isn't entirely within the slice
da1s4: start 63, end 71132959, size 71132897
da1s4d: start 0, end 71132959, size 71132960
. . .
but now I can access the disk fine.

Can I fix the drive without losing the information on it? If so, how?

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

Migrating users and passwords from one system to another

[Paul Hoffman]

Greetings again. I'm kinda surprised that this isn't an FAQ, but...

How do I move users and passwords from a current system to a new one? 
Is it sufficient to move /etc/master.passwd and /etc/passwd, or are 
there other things?

I'm moving from an oldish 4.8 system to a brand new 4.9 system, and 
want to have all the users set up on the new machine before I start 
rsyncing everything over so that the users and groups come out right.

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

RE: Setting up a NAT without a firewall

[Paul Hoffman]

Off-list, someone pointed out to me that ipnat is *much* easier to 
deal with than IPFIREWALL and all its baggage. No kernel rebuilding, 
no juggling with the firewall. Nice. For those of you in the same 
situation as me, definitely look into ipnat.

My system gets its external address from my ISP's DHCP server on 
interface em0. The machines in my house are connected to a switch 
that is attached to itnerface rl0.

Relevant stuff in /etc/rc.conf:

ifconfig_em0="DHCP"
ifconfig_rl0="inet 10.20.30.1 netmask 255.255.255.0"
gateway_enable="YES"
ipfilter_enable="YES"
ipnat_enable="YES"
ipnat_rules="/etc/ipnat.conf"
Contents of /etc/ipnat.conf:

map em0 10.20.30.0/24 -> 0/32

Two notes not covered in the ipnat man pages:

- The man page doesn't say which interface name you use in the map 
statement; it's the external interface.

- If you get your external IP address from DHCP, you can use "0/32" 
as the target. This is very handy.

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

Not needing the console for a system reload

[Paul Hoffman]

Greetings again. Sometimes I want to do a "shutdown now; ; exit" to reload the upper layers of the system without 
doing a full reboot. If I'm not at the console, I can't do this, so I 
have to do the reboot, which takes much longer because of all the 
kernel loading and hardware probing.

Is there a way to give a "shutdown now; ; exit" command 
from the command line if I'm logged in remotely? Or do I really need 
to use "reboot" and go through the whole hardware reinitialization?

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

Re: Not needing the console for a system reload

[Paul Hoffman]

At 7:07 PM +0200 5/8/04, Ph. Schulz wrote:
Is there a way to give a "shutdown now; ; exit" 
command from the command line if I'm logged in remotely? Or do I 
really need to use "reboot" and go through the whole hardware 
reinitialization?

 I don't think this is possible. The reason is (if I understand 
things correctly) that if you're in single user mode, the network 
isn't started, so there's no way of accessing the machine through 
ssh, rlogin or something similar.
Exactly right. That's why I want a script that starts the process 
while I'm logged in over the network, but finishes the process even 
after I'm kicked off.

 I think you're best off if you hook up a serial console to your 
machine and remotely access that console. There are commercial 
solutions for this but any low-end PC will do.
That is massive overkill for something that should be much simpler 
and hopefully not involve new hard ware.

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

FTP not using the users database?

[Paul Hoffman]

Hi again. I want to set up an FTP server that does *not* pay 
attention to the FreeBSD login user database. That is, I want the 
server to look in some database (probably text file) that I create 
that has usernames and passwords. I'm not worried about file 
permissions, assuming that the FTP server takes care of that.


Which FTP servers do this? Reviews?
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Installing over NetBSD

[Paul Hoffman]

Greetings. I have a box with NetBSD 1.6 on it that I want to turn 
into a FreeBSD box, while retaining the information in the user 
directories. Is this possible with the FreeBSD 4.7 install CD-ROM? 
That is, can I say during setup "don't reformat or re-partition, but 
just use the / and /usr that is already on this disk"?

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

A vi for /bin?

[Paul Hoffman]

I'm kinda surprised this isn't in the FAQ (or at least not in a place 
that I could find it). It is really impossible to build a vi with no 
external dependencies that can be installed in /bin? All I want is 
something that knows how to full-screen edit on the console, nothing 
else. I dread the day that I can only mount / and not /usr and need 
to edit fstab or rc.conf, and have to use ed.

(Just in case the answer is "no, you really can't do that", I have 
put a plain-text copy of the ed man page in /bin, but still...)

--Paul Hoffman

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

RE: A vi for /bin?

[Paul Hoffman]

At 1:36 AM + 1/26/03, Petersen wrote:

Paul Hoffman wrote:


 I'm kinda surprised this isn't in the FAQ (or at least not in a place
 that I could find it). It is really impossible to build a vi with no
 external dependencies that can be installed in /bin?


What made you think it was impossible?


The book "FreeBSD" by Anderson (which was highly recommended by some) 
says so on page 371. I'm glad to hear that's wrong.

Assuming you have the object files from a buildworld hanging around, then
cd /usr/obj/usr/src/usr.bin/vi
cc -O -pipe -o vi *.o -lncurses -static && strip vi && mv vi /bin/
should probably supply you with what you want.


Thanks, I'll try that. (I haven't done a buildworld yet...)

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

Re: A vi for /bin?

[Paul Hoffman]

At 1:36 AM + 1/26/03, Petersen wrote:

Assuming you have the object files from a buildworld hanging around, then
cd /usr/obj/usr/src/usr.bin/vi
cc -O -pipe -o vi *.o -lncurses -static && strip vi && mv vi /bin/
should probably supply you with what you want.


Two modifications made this work fine:
- I hadn't done a buildworld, but doing the following got the same result:
   cd /usr/src/usr.bin/vi
   make && cc -O -pipe -o vi *.o -lncurses -static && strip vi && mv vi /bin/
- As pointed out off-line, you also need to get it the termcap library. Doing
   cp /usr/share/misc/termcap.db /root/.termcap.db
fixes that.

At 8:30 PM -0600 1/25/03, Dan Nelson wrote:

You
might want to install e3, which is only 13k statically linked and uses
generic ANSI escape codes so it doesn't need a termcap file.  If run as
e3vi, it acceptes vi keystrokes.


Nice! e3 built from the ports collection linked statically 
automatically. e3 didn't work correctly on my console (it didn't 
recognize the Alt key), but e3vi worked fine and felt just like vi.

Thanks! I now feel better about emergencies.

--Paul Hoffman

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

Re: Which files and directories to backup?

[Paul Hoffman]

At 8:16 PM -0500 1/26/03, Francisco Reyes wrote:

On Sun, 26 Jan 2003, Mike Meyer wrote:


 It's a bad idea to exclude fstab.


Why? At one point I had it included and it actually clobered a working one
and just caused much more headaches.


You should still back it up; you just need to be more careful with 
your restores. The same will be true for many of the files in /etc.

Other files won't create as much headaches as fstab if copied over by
mistake.


You say that now, but I would bet that for many people, restoring an 
old rc.conf would be more prone to making them yell "Doh!" than 
restoring an old fstab.

In other words, you are better off doing more complete backups and 
just being that much more careful when you do restores. You can also 
do backups very often to minimize the problems of a backup being too 
old. I back up something similar to what you have listed here (with 
fewer exceptions of directories because most of the things you 
carefully excluded are really small) every week, and always just 
before I do a major upgrade. I then shove the backup offsite via ftp.

Much of what you are backing up is quite compressable, so you should 
most likely be using 'tar -czf' instead of just 'tar -cf'

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

Re: Which files and directories to backup?

[Paul Hoffman]

At 4:18 PM -0500 1/27/03, Francisco Reyes wrote:

On Mon, 27 Jan 2003, Paul Hoffman wrote:


 before I do a major upgrade. I then shove the backup offsite via ftp.


I have not been sending the files out, but working on that.
First will encryp the files with gpg (GNU privacy) and then will use scp
to send the files out. I don't have FTP enabled on any of the, very few,
machines I admin.


Of course, that works too. I use FTP because I can use 'mirror' (from 
/usr/ports/net/mirror) to pull down the automatic backups reliably, 
and because I simply don't back up any file that has passwords (none 
of the password-containing files in /etc, nor any directory called 
'.ssh', nor any private key file used for writing certificates). I 
could probably hack up mirror or some other system to use scp, but 
mirror works and I can quickly get the files to where I want them 
(namely, away from the server they are backing up).

 > Much of what you are backing up is quite compressable, so you should

 most likely be using 'tar -czf' instead of just 'tar -cf'


I don't recall posting the line I use to run tar, but I use bzip2 instead
of gzip so it's something like 'tar -cyf '


gzip is more universally known that bzip2, so using it means that you 
are more likely to be able to recover your data in an emergency, such 
as on a non-BSD system.

--Paul Hoffman

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

Safe to change the position of "local" in the path?

[Paul Hoffman]

Greetings again. The default path in FreeBSD 4.7 is:

PATH=/sbin:/bin:/usr/sbin:/usr/bin:/usr/games:/usr/local/sbin:/usr/local/bin:/usr/X11R6/bin:/home/you/bin

OpenSSL comes as part of FreeBSD in /usr/bin. If you add a newer 
version of OpenSSL, it goes into /usr/local/bin, which is later in 
the path and therefore not found without explicitly calling it each 
time.

Is it safe to change the default path to:

PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin:/usr/games:/usr/X11R6/bin:/home/you/bin

Is there a generic way to make things in /usr/local/bin take 
precedence over things in /usr/bin? For example, I remember that the 
Perl 5.8 port comes with a command that you can run that makes the 
new Perl take precedence (although I don't remember what it is). 
Should I be doing that (whatever it is) to make just OpenSSL take 
precedence, instead of changing the search path?

--Paul Hoffman

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

dhcpd subnets?

[Paul Hoffman]

Greetings again. I have a DSL connection with 32 addresses. My 
ifconfig line in rc.conf looks like:
ifconfig_tx0="inet a.b.c.130  netmask 255.255.255.224"
defaultrouter="a.b.c.158"

This works fine. I am trying to configure dhcpd (from the ports 
collection). My config file says:

default-lease-time 86400;
option subnet-mask 255.255.255.224;
option routers a.b.c.158;
option domain-name-servers a.b.c.130;
ddns-update-style none;

subnet a.b.c.158 netmask 255.255.255.224 {
range a.b.c.150 a.b.c.157;
}

This gets the fatal startup error:

/etc/dhcpd.conf line 7: subnet a.b.c.158: bad subnet number/mask combination.
subnet a.b.c.158 netmask 255.255.255.224
 ^
Configuration file errors encountered -- exiting

This is a valid subnet, and it works just fine for everything else. 
What is dhcpd wanting?

--Paul Hoffman

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

Re: dhcpd subnets?

[Paul Hoffman]

At 5:50 PM -0600 2/8/03, Daniel Schrock wrote:

a.b.c.158 is the last usable address, not the network address.  try 
a.b.c.128 instead, which is the network address for you /27.

Er, right. I figured that out by playing with the subnet calculator 
at .

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

How important is CPU speed for a bridging NAT box?

[Paul Hoffman]

Greetings again. I'm about to set up a box that is dedicated as a 
bridging NAT and firewall. I was going to use an old P133 box I had 
laying around. Will this be fast enough for typical Internet access 
at 384Kbps if the box isn't doing anything else, or do I need a 
faster machine?

--Paul Hoffman

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

USB hard drives?

[Paul Hoffman]

Greetings. I have a USB hard drive attached to a FreeBSD 4.7 box. The 
first part of the dmesg looks good:

. . .
ohci0:  mem 0xfe12-0xfe120fff irq 
5 at device 15.2 on pci0
usb0: OHCI version 1.0, legacy support
usb0: SMM does not respond, resetting
usb0:  on ohci0
usb0: USB revision 1.0
uhub0: (0x1166) OHCI root hub, class 9/0, rev 1.00/1.00, addr 1
uhub0: 2 ports with 2 removable, self powered
umass0: ScanLogic USBIDE ScanLogic USBIDE, rev 1.10/2.60, addr 2
. . .

But at the end:
umass0: Invalid CSW: tag 2359384 should be 1
umass0: Invalid CSW: tag 2359384 should be 2
umass0: Invalid CSW: tag 2359384 should be 3
umass0: Invalid CSW: tag 2359384 should be 4
umass0: Invalid CSW: tag 2359384 should be 5

That may be because there is already data on the drive (which I don't 
care about).

And there is no mention of da0 in the dmesg.

All of the devices listed in the man page for umass are in the 
kernel. However, I cannot write a disk label on the drive:

# disklabel -w da0 auto
disklabel: /dev/da0c: Device not configured

How do I move forwards?

--Paul Hoffman

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

Compiling OpenSCEP under FreeBSD 4.7?

[Paul Hoffman]

Does anyone have experience with compiling OpenSCEP under 4.7? It 
compiles under NetBSD, but hangs under FreeBSD. I'd rather run it on 
a FreeBSD box.

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

Using different auth method for telnet?

[Paul Hoffman]

Hi again. On a test machine, I have changed the line in inetd.conf to 
something like:
   telnet  stream  tcp nowait  root/usr/libexec/telnetd 
telnetd -a none -p /path/to/my-auth-program
and I have hup'd inetd. However, when I try to telnet to this 
machine, I still go through the standard telnet prompting for 
username and password. I get the same result with "-a valid" and "-a 
user".

What else do I need to modify to get my program to be used instead of 
the default login?

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

Getting a specific value from netstat

[Paul Hoffman]

Greetings again. If I do a 'netstat -I em0 -b', I get:

NameMtu Network   Address  Ipkts Ierrs Ibytes 
Opkts Oerrs Obytes  Coll
em01500   00:0e:0c:67:c8:04 93555198 0 2179562966 
114493253 0  723565977 0
em01500 fe80:1::20e:c fe80:1::20e:cff:f0 -  0 
4 -288 -
em01500 192.245.12Balder-22735399016 - 1770283188 
114484197 - 3415268168 -
em01500 192.245.12.22 Balder-22827063120 - 1655024896 
0 -  0 -
em01500 192.245.12.22 Balder-22947427840 - 3954775975 
18975500 - 2445620452 -


What I care about is the number of input and output bytes (in this 
case, 2179562966 and 723565977). I can write a short Perl script to 
parse the netstat output, but I would rather just get the numbers 
directly from the OS. Are these values available without going 
through netstat?

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

Re: Getting a specific value from netstat

[Paul Hoffman]

At 9:42 AM -0500 9/18/06, Dan Nelson wrote:

In the last episode (Sep 17), Paul Hoffman said:
 > Greetings again. If I do a 'netstat -I em0 -b', I get:


 NameMtu Network   Address  Ipkts Ierrs 
Ibytes  Opkts Oerrs Obytes  Coll
 em01500   00:0e:0c:67:c8:04 93555198 0 
2179562966  114493253 0  723565977 0
 em01500 fe80:1::20e:c fe80:1::20e:cff:f0 - 
0  4 -288 -
 em01500 192.245.12Balder-22735399016 - 
1770283188  114484197 - 3415268168 -
 em01500 192.245.12.22 Balder-22827063120 - 
1655024896  0 -  0 -
 em01500 192.245.12.22 Balder-22947427840 - 
3954775975  18975500 - 2445620452 -


 What I care about is the number of input and output bytes (in this
 case, 2179562966 and 723565977). I can write a short Perl script to
 parse the netstat output, but I would rather just get the numbers
 directly from the OS. Are these values available without going
 through netstat?


If you use the same code netstat does, yes :)  It looks like
per-interface stats are still obtained by grovelling through /dev/kmem,
though, so it may be easier to just parse netstat's output.


Yes, probably so. The quick-and-dirty Perl script I wrote is:

$ThisNetStat = `/usr/bin/netstat -I em0 -b`;
@Lines = split(/^/, $ThisNetStat);
$TheLine = $Lines[1];
$TheLine =~ s/ ( )*/ /g;
@Fields = split(/ /, $TheLine);
$InBytes =  $Fields[6];
$OutBytes = $Fields[9];

*However*, I now see that the byte numbers from netstat seem to wrap 
around at about 4 gigabytes. I'll have to add some code to handle 
that over the long term, given that my system puts out that much in a 
day...



Another
alternative would be to install net-snmp and ask it for the stats.


I thought SNMP stood for "Simply Not My Problem"? :-)

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

Making a mounted ISO image of a CD-ROM writable

[Paul Hoffman]

Greetings again. I want to make an ISO image of the FreeBSD 
distribution with a boot.config file that contains "/boot/loader -h". 
I have the ISO image as a file on my hard drive, and have mounted it 
on /mnt:


mdconfig -a -t vnode -f /path/to/my/bootable.iso -u 0
mount -w -t cd9660 /dev/md0 /mnt

However, when I try to create files on /mnt, I get:

-su: boot.config: Read-only file system

How do I make it so that I can write into /mnt so that I can then 
later save those back to the ISO image?


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

Re: Making a mounted ISO image of a CD-ROM writable

[Paul Hoffman]

At 10:53 PM +0200 7/12/07, Roland Smith wrote:

On Thu, Jul 12, 2007 at 12:38:10PM -0700, Paul Hoffman wrote:

 Greetings again. I want to make an ISO image of the FreeBSD distribution
 with a boot.config file that contains "/boot/loader -h". I have the ISO
 image as a file on my hard drive, and have mounted it on /mnt:

 mdconfig -a -t vnode -f /path/to/my/bootable.iso -u 0
 mount -w -t cd9660 /dev/md0 /mnt

 However, when I try to create files on /mnt, I get:

 -su: boot.config: Read-only file system

 How do I make it so that I can write into /mnt so that I can then later
 save those back to the ISO image?


Try the sysutils/isomaster port.


Er, thanks, but I am running on a text-only system. Looks nice, however.

Other thoughts? I would hope this would be as easy as "run this 
program to change the image to read-write".


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

Can't do an "make installworld"

[Paul Hoffman]

Any help would be appreciated here. I'm on a clean 6.1-RELEASE sysem. 
I created /home/pxe. I cd'd to /usr/src. I gave 'make installworld 
DESTDIR=/home/pxe'. It ends with:

. . .
--

 Installing everything

--
cd /usr/src; make -f Makefile.inc1 install
===> share/info (install)
install -o root -g wheel -m 444  dir-tmpl /home/pxe/usr/share/info/dir
install:No such file or directory
*** Error code 1

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

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

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

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

Stop in /usr/src.


OK, that seems bad, so I tried just doing a 'mkdir 
/home/pxe/usr/share/info/dir' and running again. This time, it gets 
past that step, but fails on the next:

. . .
--

 Installing everything

--
cd /usr/src; make -f Makefile.inc1 install
===> share/info (install)
===> include (install)
install -C -o root -g wheel -m 444  a.out.h ar.h assert.h bitstring.h 
complex.h cpio.h _ctype.h ctype.h db.h dirent.h dlfcn.h elf.h 
elf-hints.h err.h fmtmsg.h fnmatch.h fstab.h fts.h ftw.h getopt.h 
glob.h grp.h hesiod.h histedit.h ieeefp.h ifaddrs.h inttypes.h 
iso646.h kenv.h langinfo.h libgen.h limits.h link.h locale.h malloc.h 
memory.h monetary.h mpool.h ndbm.h netconfig.h netdb.h nl_types.h 
nlist.h nss.h nsswitch.h objformat.h paths.h proc_service.h pthread.h 
pthread_np.h pwd.h ranlib.h readpassphrase.h regex.h regexp.h 
resolv.h runetype.h search.h setjmp.h sgtty.h signal.h stab.h 
stdbool.h stddef.h stdio.h stdlib.h string.h stringlist.h strings.h 
sysexits.h tar.h tgmath.h time.h timeconv.h timers.h ttyent.h 
ulimit.h unistd.h utime.h utmp.h uuid.h varargs.h vis.h wchar.h 
wctype.h wordexp.h osreldate.h /home/pxe/usr/include

install:No such file or directory
*** Error code 1

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

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

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

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

Stop in /usr/src.


However, there *is* already a /home/pxe/usr/include, and it is 
populated with a whole bunch of directories, just none of the regular 
files.


Is there a better way for me to do an 'make installworld'? What am I 
missing here?


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

Re: Can't do an "make installworld"

[Paul Hoffman]

At 5:38 PM -0500 7/17/07, Shaun Meyer wrote:

On Tue, July 17, 2007 4:14 pm, Paul Hoffman wrote:

 Any help would be appreciated here. I'm on a clean 6.1-RELEASE sysem.
 I created /home/pxe. I cd'd to /usr/src. I gave 'make installworld
 DESTDIR=/home/pxe'. It ends with:
 . . .



Worked fine on my 6.2 just now.


Interestingly, after I sent this, I tried it on a 6.0 system, and had 
the same problem (with fewer files):


--

 Installing everything

--
cd /usr/src; make -f Makefile.inc1 install
===> share/info (install)
===> include (install)
creating osreldate.h from newvers.sh
touch: not found
*** Error code 127



This isn't a permissions problem, is it?


Shouldn't be, given that I am running as root.

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

Re: Can't do an "make installworld"

[Paul Hoffman]

At 4:12 PM +1200 7/18/07, Jonathan Chen wrote:

On Tue, Jul 17, 2007 at 03:49:02PM -0700, Paul Hoffman wrote:

 At 5:38 PM -0500 7/17/07, Shaun Meyer wrote:
 >On Tue, July 17, 2007 4:14 pm, Paul Hoffman wrote:
 >> Any help would be appreciated here. I'm on a clean 6.1-RELEASE sysem.
 >> I created /home/pxe. I cd'd to /usr/src. I gave 'make installworld
 >> DESTDIR=/home/pxe'. It ends with:
 >> . . .
 >
 >
 >Worked fine on my 6.2 just now.

 Interestingly, after I sent this, I tried it on a 6.0 system, and had
 the same problem (with fewer files):

 --
 >>> Installing everything
 --
 cd /usr/src; make -f Makefile.inc1 install
 ===> share/info (install)
 ===> include (install)
 creating osreldate.h from newvers.sh
 touch: not found
 *** Error code 127


This is commonly caused by a bad date. Check the system time.


That doesn't seem to be it. The system time is fine, and none of the 
files in /usr/src/include have "funny" times (either in the future or 
near 1970).


Any other clues? Does 'make installworld DESTDIR=/home/something' 
work for other people running 6.1-RELEASE?

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

Creating an ext2 file system on FreeBSD?

[Paul Hoffman]

Hi again. Is there a FreeBSD equivalent of Linux's 'mke2fs'? I want 
to create a disk image that is in ext2 format. On Linux, I would do:

  dd if=/dev/zero of=some.img bs=1M count=1 seek=1024
  /sbin/mke2fs -F -j some.img
Can I do something similar on FreeBSD?

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

Monitoring CPU usage on multi-core system

[Paul Hoffman]

Hi again. On a dual-core system, how do I tell how much of each of 
the CPU cores are in use? Is the CPU usage in 'top' for the two CPUs 
at once? Is there something in ports (that works without X...) that 
will give good info?

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

What does an * in master.passwd (not passwd) mean?

[Paul Hoffman]

In 5.4 (and probably lots of other versions), the master.passwd file 
is pre-seeded with lots of accounts such as daemon, operator, and so 
on. The master.passwd file looks like:


daemon:*:1:1::0:0:Owner of many system processes:/root:/usr/sbin/nologin
operator:*:2:5::0:0:System &:/:/usr/sbin/nologin
. . .

The man page for master.passwd and passwd say what an "*" in the 
second field means in passwd, but not in master.passwd. Any clues 
would be appreciated (and I will put in a documentation pr when I 
have an answer).


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

RE: What does an * in master.passwd (not passwd) mean?

[Paul Hoffman]

At 3:27 PM -0700 9/16/05, Chris St Denis wrote:

It means an account that can not be logged in to.

The in the hash algorithm used in master.password nothing encrypts to * so
no possible password will ever match the encrypted value * thus locking out
the account from login.


Arrrgh. Whomever decided to use the same character for "password not 
shown in passwd" and "cannot ever log in in master.passwd" should be, 
well, punished somehow.


Thanks, this makes perfect sense. I'll craft up a short explanation 
for the man page and send in a pr.

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

Short HOWTO on reading a core to determine why my server is rebooting?

[Paul Hoffman]

Hi again. I have a server running 6.0 that has been spontaneously 
rebooting every few weeks. Is there a short HOWTO that tells me how 
to read the files in /var/crash to at least find out what the kernel 
thinks the issue is? There is nothing in /var/log/messages of 
interest before the crash, nothing in 'dmesg -a' in coming up that 
says anything interesting, and there's plenty of room on the drives.


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

Fast partial reboot?

[Paul Hoffman]

Greetings again. Under 5.x, is there a way to quickly reboot FreeBSD 
if I'm not sitting at the console? I want the equivalent of, if I 
were sitting at the console, 'shutdown now' followed by specifying 
'/bin/sh' followed by 'exit'.

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

Re: Fast partial reboot?

[Paul Hoffman]

At 10:02 AM -0500 6/6/05, Tim Erlin wrote:

Sounds like you're looking for 'shutdown -r now'


Nope. 'shutdown -r now' does a full reboot of the PC, which means 
re-loading the kernel and all the devices. That's quite a bit slower 
than 'shutdown now' and exiting the single-user shell.

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

Laptop with PCcard ethernet: how to set up?

[Paul Hoffman]

I have a laptop with an Ethernet PCcard which comes up as "ep0". I 
want to use DHCP on it. My rc.conf has:

pccard_enable="YES"
pccard_ifconfig="YES"
ifconfig_ep0="DHCP"
The card comes up fine, but it doesn't get ifconfig'd. Do I need to 
add something else to rc.conf?
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Laptop with PCcard ethernet: how to set up?

[Paul Hoffman]

At 8:14 AM -0500 3/30/03, Dan Pelleg wrote:
 > I have a laptop with an Ethernet PCcard which comes up as "ep0". I want to
 use DHCP on it. My rc.conf has:

 pccard_enable="YES"
 pccard_ifconfig="YES"
 ifconfig_ep0="DHCP"
 The card comes up fine, but it doesn't get ifconfig'd. Do I need to add
 something else to rc.conf?
1. You don't need the ifconfig_ep0 line.

2. Change the pccard_ifconfig value to either "DHCP" or something like
"inet 192.168.1.1/24".
That worked fine, thanks!
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Best X configurator for laptops?

[Paul Hoffman]

Hi again. I have a Dell Inspiron 3500 laptop, now running 4.7. 
xf86cfg and xf86config both give (different) unusable results for my 
system. Which of the other X configurators in the ports collection 
seem to do a good job on laptops, if any?

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

Re: Best X configurator for laptops?

[Paul Hoffman]

At 8:29 AM -0800 3/30/03, Nathan Kinkade wrote:
On Sun, Mar 30, 2003 at 07:59:59AM -0800, Paul Hoffman wrote:
 Hi again. I have a Dell Inspiron 3500 laptop, now running 4.7.
 xf86cfg and xf86config both give (different) unusable results for my
 system. Which of the other X configurators in the ports collection
 seem to do a good job on laptops, if any?
 --Paul Hoffman
You need to be more specific about the problems you are encountering.
I was trying to avoid that because it doesn't seem like this is a 
good place to debug particular XWindows problems.

But, since you asked, the screen comes up blank. There are no 
XWindows errors, just a blank screen.

Thus, my quest for a better configuration...
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Smarter 'make buildkernel'?

[Paul Hoffman]

Hi again. Is there a way to get 'make buildkernel' in /usr/src to not 
rebuild things that it already compiled? I'm playing around on a 
not-very-fast laptop, and the rebuilds take forever.

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

Re: Best X configurator for laptops?

[Paul Hoffman]

At 9:43 AM +0930 3/31/03, Greg 'groggy' Lehey wrote:
 > Thus, my quest for a better configuration...

You're jumping to conclusions that it's the configurator.
Turns out I wasn't. None of the configuration programs got me 
anywhere close. They either got the monitor wrong, the card wrong, 
the screen wrong, or a combination.

I ended up cobbling it together from some advice for Linux, some 
other snippets and so on. In case anyone cares, the relevant hard 
parts of the XF86Config for (my/the) Inspiron 3500 are:

Section "Monitor"
Identifier   "Monitor0"
HorizSync31.5-48.5
VertRefresh  60
EndSection
Section "Device"
Identifier  "Card0"
Driver  "neomagic"
VendorName  "Neomagic"
BoardName   "NM2200"
EndSection
Section "Screen"
Identifier "Screen0"
Device "Card0"
Monitor"Monitor0"
DefaultDepth 16
SubSection "Display"
Depth 16
Modes"1024x768"
EndSubSection
EndSection
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Updating /usr/src after updating 4.7->4.8

[Paul Hoffman]

Greetings again. On a test machine, I upgraded from fairly vanilla 
4.7 to 4.8 using CD-ROM and /stand/sysintall. At the beginning of the 
upgrade, it told me that it would not upgrade /usr/src. After the 
upgrade, I see by the dates that it upgraded some of /usr/src, but 
not most of it.

What is the proper way to bring /usr/src up to date so that I can 
make kernel mods?

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

Re: Updating /usr/src after updating 4.7->4.8

[Paul Hoffman]

At 9:56 PM -0500 4/5/03, taxman wrote:
On Saturday 05 April 2003 09:26 pm, Paul Hoffman wrote:
 Greetings again. On a test machine, I upgraded from fairly vanilla
 4.7 to 4.8 using CD-ROM and /stand/sysintall. At the beginning of the
 upgrade, it told me that it would not upgrade /usr/src. After the
 upgrade, I see by the dates that it upgraded some of /usr/src, but
 not most of it.
 What is the proper way to bring /usr/src up to date so that I can
 make kernel mods?
cvsup is one of the most common ways.  You need to install it first.  See
http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/cvsup.html
I was assuming that there was a way to get it off the CD-ROM. Is that not true?
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

SSH login banner: IP address instead of DNS name

[Paul Hoffman]

Hi again. When I SSH into my 4.8 box, the first line of the banner is:

Last login: Mon Jun 30 19:31:44 2003 from 15-characters-of-a-host-name

That DNS name is truncated to 15 characters. I would much prefer an 
IP address. What do I need to change to get this?

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

Re: SSH login banner: IP address instead of DNS name

[Paul Hoffman]

At 9:04 PM +0200 7/1/03, David Siebörger wrote:
At 12:41 PM on Tuesday  1 July 2003, Supote leelasupphakorn wrote:
 > Hi again. When I SSH into my 4.8 box, the first line of the banner is:
 > Last login: Mon Jun 30 19:31:44 2003 from 15-characters-of-a-host-name
 > That DNS name is truncated to 15 characters. I would much prefer an
 > IP address. What do I need to change to get this?
 > --Paul Hoffman
Run sshd with the -u 15 option.  You can do so by adding this line to
/etc/rc.conf:
sshd_flags="-u 15"
Perfect, that's exactly what I needed. Thanks!
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Laptops as routers

[Paul Hoffman]

Greetings again. I'm looking to buy a couple of cheap old laptops to 
be used as temporary routers. They just need to be able to handle 
PCMCIA Ethernet cards, not much more (having an Ethernet connector on 
the motherboard is fine, of course.) I don't want to run XWindows, 
and I'm sure 64 MB and a 1gig hard drive would suffice.

Are there any brands/models I should lean towards? Ones I should avoid?
--Paul Hoffman
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Admin-visible differences between 4.10 and 5.3

[Paul Hoffman]

Greetings again. At one point, I think I heard that there was going 
to be a big change in the bootup (rc.foo) stuff in FreeBSD 5, but I 
don't see anything about that in the Early Adopter Guide. Assuming 
that I'm quite comfortable with day-to-day system administration 
under FreeBSD 4 but want to start using 5 on new systems when 5.3 
comes out, what do I need to know?

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

Re: Admin-visible differences between 4.10 and 5.3

[Paul Hoffman]

At 8:02 PM + 10/31/04, David Jenkins wrote:
On Sun, 31 Oct 2004 10:49:07 -0700, Paul Hoffman <[EMAIL PROTECTED]> wrote:
 Greetings again. At one point, I think I heard that there was going
 to be a big change in the bootup (rc.foo) stuff in FreeBSD 5, but I
 don't see anything about that in the Early Adopter Guide. Assuming
 that I'm quite comfortable with day-to-day system administration
 under FreeBSD 4 but want to start using 5 on new systems when 5.3
 comes out, what do I need to know?
Maybe this helps.
http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/configtuning-rcng.html
Bingo, that is exactly what I was looking for. Thanks!
Are there any other admin-visible changes in 5.3 I should be looking for?
--Paul Hoffman
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Knoppix-like FreeBSD-on-a-CD?

[Paul Hoffman]

Sorry if I'm missing something obvious, but is there a 
FreeBSD-on-a-CD project similar to Knoppix for Linux?

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

Want to use two PCCards on a laptop

[Paul Hoffman]

Greetings again. I want to use a laptop as a router under 4.10. I 
have two PCcards that it recognizes, but when starting up, after 
connecting to the first card, I get the message:
   pccard0: Can no attach more than one child.
So, is this a limitation in the kernel that I can fix with a 
re-build? Something in one of the config files that I haven't figured 
out? Something that is fixed in 5.3? Or... ?

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

Re: Want to use two PCCards on a laptop

[Paul Hoffman]

At 12:03 PM +1030 11/16/04, Greg 'groggy' Lehey wrote:
On Monday, 15 November 2004 at 17:02:40 -0800, Paul Hoffman wrote:
 Greetings again. I want to use a laptop as a router under 4.10. I
 have two PCcards that it recognizes, but when starting up, after
 connecting to the first card, I get the message:
pccard0: Can no attach more than one child.
 So, is this a limitation in the kernel that I can fix with a
 re-build? Something in one of the config files that I haven't figured
 out? Something that is fixed in 5.3? Or... ?
This isn't a general problem.  It must be related to your hardware,
but you don't say what it is.
Sorry; didn't realize it might be relevant. Dell Inspiron 3500. 
FreeBSD is definitely seeing both cards; it responds nicely when I 
eject either of the cards. It just won't do anything useful when I 
insert the second card. More clues appreciated!

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

Re: Want to use two PCCards on a laptop

[Paul Hoffman]

At 12:28 PM +1030 11/16/04, Greg 'groggy' Lehey wrote:
On Monday, 15 November 2004 at 17:45:42 -0800, Paul Hoffman wrote:
 At 12:03 PM +1030 11/16/04, Greg 'groggy' Lehey wrote:
 On Monday, 15 November 2004 at 17:02:40 -0800, Paul Hoffman wrote:
 Greetings again. I want to use a laptop as a router under 4.10. I
 have two PCcards that it recognizes, but when starting up, after
 connecting to the first card, I get the message:
 >>>pccard0: Can no attach more than one child.
 So, is this a limitation in the kernel that I can fix with a
 re-build? Something in one of the config files that I haven't figured
 out? Something that is fixed in 5.3? Or... ?
 This isn't a general problem.  It must be related to your hardware,
 but you don't say what it is.
 Sorry; didn't realize it might be relevant. Dell Inspiron 3500.
 FreeBSD is definitely seeing both cards; it responds nicely when I
 eject either of the cards. It just won't do anything useful when I
 insert the second card. More clues appreciated!
More information appreciated.  People shouldn't have to ask you twice
to know what cards you're using, or what the messages were.  You might
like to take a look at http://www.lemis.com/questions.html.
Doh! I saw "hardware" and I thought "computer". You are right, of course.
3Com EtherLink III, 3C569D
3Com Megahertz, 3CCE589ET
The actual message seen in dmsg when putting in the second card is as above:
   pccard0: Can no attach more than one child.
I can't get a file off the system yet, unfortunately, so I can't give 
the full dmesg output.
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

newsyslog not reading /ect/rc.conf arguments?

[Paul Hoffman]

Greetings again. Running FreeBSD 8.0, I have added the following to 
/etc/rc.conf:
  newsyslog_flags="-a /usr/old-log/"
I have stopped and started newsyslog. However, the rotated logs are still being 
written into /var/log. No errors appear in /var/log/messages or in dmesg.

Any clues?

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

Re: newsyslog not reading /ect/rc.conf arguments?

[Paul Hoffman]

At 6:14 PM +0300 5/5/10, Giorgos Keramidas wrote:
>How did you start newsyslog?  There's an rc.d script that should *read*
>the flags from rc.conf:
>
>/etc/rc.d/newsyslog start

Yes, exactly. I did '/etc/rc.d/newsyslog stop', then '/etc/rc.d/newsyslog 
start'.


At 11:14 AM -0400 5/5/10, Greg Larkin wrote:
>newsyslog is invoked at boot time by the /etc/rc.d/newsyslog script to
>create missing log files, but after that, it's invoked regularly by cron
>to do the actual rotations.  Check the /etc/crontab file and add your
>flags there, and you should be all set.

Thanks, I see that now.

This seems like a broken model: intial boot and later restarts uses arguments 
from /etc/rc.conf, but the periodic call does not. I don't think we want people 
modifying /etc/crontab, do we? Shouldn't /etc/crontab be calling 
'/etc/rc.d/newsyslog restart' instead?

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

Python 3 from ports: alternate install?

[Paul Hoffman]

Greetings. I want to have both python 2.x and 3.0 on a FreeBSD 7.0 box. 2. I 
already installed 2.5.1 from ports; 'which python' reports 
/usr/local/bin/python, and one would hope.

I would like to install python 3 as /usr/local/bin/python3 or somesuch; is that 
possible from ports? If not, I can probably install from source code, but I 
trust FreeBSD's ports installer more...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"