multiple slices and invalid partition table problem

[Viren Patel]

Hello. I am trying to install 6.1 on a Dell 2950 with a Dell PERC 5/i (LSI
Logic) SAS RAID controller. The controller is supported by the mfi driver. The
system has six 300GB drives configured in a RAID 5 array for a total of 
1.5TB.

I would like to setup jails on this system with each jail in its own
partition. To maximize the number of jails I am trying to slice the RAID
volume into 4 slices and 7 partitions per slice. This is when problems start
cropping up. In the first instance I keep getting warning messages about CHS
geometry being incorrect and sysinstall defaults to what it thinks are the
correct values. Unfortunately Dell BIOS or the controller BIOS does not report
any CHS values. Is there a way to get the correct values from the controller?

But going with the sysinstall values, if I allocate the entire volume (1
slice) I can install the OS and it boots fine. If however I create two or more
slices, the OS does install but upon reboot produces an "Invalid partition
table" error.

Here is an odd thing - I created two RAID volumes, one 300GB RAID 1 mirror and
another 900GB RAID 5 array. When installing FreeBSD it sees the two volumes as
mfid0 and mfid1, however it reports the 300GB size for both volumes. In this
configuration the CHS errors are not reported for mfid0 but still come up for
mfid1. Also any partitions I create in mfid0 are also shown for mfid1 and I
still get the "Invalid partition table" error upon reboot.

I have tried this on two identical systems with the same results. Any ideas?
TIA for any assistance in resolving this puzzling issue.

Viren Patel
Chemistry & Biochemistry
University of Texas at Austin


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

Re: FreeBSD 6.1 - mount_nullfs broken?

[Viren Patel]

> On Fri, Mar 24, 2006 at 01:24:35PM -0600, Viren Patel
> wrote:
>> >
>> > Compile it into your kernel or load the module.  Read
>> the
>> > handbook for
>> > more help.
>> >
>> > Kris
>> >
>>
>> That did it! The nullfs module was not being loaded into
>> the kernel. I added it to loader.conf.local and it works
>> now. This is definitely a change from 6.0 to
>> 6.1-prerelease.
>
> I don't think so.  Modules are all installed by default.
>
> Kris
>

Then something is broken. The module file is present in
/boot/kernel. However it does *not* get loaded unless I
put it in loader.conf.local. As mentioned previously, I am
using the generic kernel. I followed the standard upgrade
procedure when going from 6.0 to 6.1.

Viren

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

Re: FreeBSD 6.1 - mount_nullfs broken?

[Viren Patel]

>
> Or you could come at it the other way; figure out how you
> did it when
> you were running 6.0.  The procedure has not changed.
>

I didn't change a thing, even in 6.0. The nullfs just
worked. I did the usual upgrade process to get from 6.0 to
6.1:

   1. change cvs tag from RELENG_6_0 to RELENG_6
   2. update sources
   3. make buildworld
   4. make buildkernel KERNCONF=GENERIC
   5. make installkernel KERNCONF=GENERIC
   6. reboot into single-user mode
   7. mergemaster -p
   8. make installworld
   9. mergemaster
   10. reboot

Viren


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

Re: FreeBSD 6.1 - mount_nullfs broken?

[Viren Patel]

>
> Compile it into your kernel or load the module.  Read the
> handbook for
> more help.
>
> Kris
>

That did it! The nullfs module was not being loaded into
the kernel. I added it to loader.conf.local and it works
now. This is definitely a change from 6.0 to
6.1-prerelease.

Viren

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

Re: FreeBSD 6.1 - mount_nullfs broken?

[Viren Patel]

> On Fri, Mar 24, 2006 at 08:05:31AM -0600, Viren Patel
> wrote:
>
>> # umount /export/jails/testjail/dev
>> # mount_nullfs -o ro /export/jails/basejail
>> /export/jails/testjail/basejail
>> mount_nullfs: Operation not supported by device
>>
>> After looking at the dmesg output more closely one
>> discrepancy I found was the second last line:
>>
>> ad6: 238475MB  at
>> ata3-master SATA300
>>
>> It was definitely SATA150 under FreeBSD 6.0, even after
>> updates. So maybe a problem in the SATA300 driver?
>
> Unlikely.  It looks like you don't have kernel nullfs
> support enabled.
>
> Kris
>

I am using the generic kernel. Looking through /etc there
is nothing regarding enabling nullfs. I also looked at
GENERIC kernel conf file without luck. How do I enable
nullfs?

Viren

Viren


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

Re: FreeBSD 6.1 - mount_nullfs broken?

[Viren Patel]

> On Thu, Mar 23, 2006 at 04:31:08PM -0600, Viren Patel
> wrote:
>> Hello. I just upgraded a FreeBSD 6.0 system to
>> 6.1-prerelease and now mount_nullfs seems to be broken.
>> I
>> get the error "mount_nullfs: Operation not supported by
>> device". Nothing has changed in the hardware or the
>> config
>> files. Here is the output dmesg:
>
> It's definitely working.
>
> Show us the exact command you're running.
>
> Kris
>

Thanks for your reply. Basically I run the ezjail.sh
script and I've shown the disk layout before and after. I
also tried running mount_nullfs directly and got the same
results.

# df -h
Filesystem SizeUsed   Avail Capacity  Mounted on
/dev/ad6s1a1.9G 80M1.7G 4%/
devfs  1.0K1.0K  0B   100%/dev
/dev/ad6s1g202G245M186G 0%/export
/dev/ad6s1f3.9G 18K3.6G 0%/home
/dev/ad6s1e7.7G1.9G5.2G27%/usr
/dev/ad6s1d7.7G 42M7.1G 1%/var

# /usr/local/etc/rc.d/ezjail.sh start
 ezjailConfiguring jails: set_hostname_allow=YES.
Starting jails:mount_nullfs: Operation not supported by
device
.

# df -h
Filesystem SizeUsed   Avail Capacity  Mounted on
/dev/ad6s1a1.9G 80M1.7G 4%/
devfs  1.0K1.0K  0B   100%/dev
/dev/ad6s1g202G245M186G 0%/export
/dev/ad6s1f3.9G 18K3.6G 0%/home
/dev/ad6s1e7.7G1.9G5.2G27%/usr
/dev/ad6s1d7.7G 42M7.1G 1%/var
devfs  1.0K1.0K  0B   100%   
/export/jails/testjail/dev

# umount /export/jails/testjail/dev
# mount_nullfs -o ro /export/jails/basejail
/export/jails/testjail/basejail
mount_nullfs: Operation not supported by device

After looking at the dmesg output more closely one
discrepancy I found was the second last line:

ad6: 238475MB  at
ata3-master SATA300

It was definitely SATA150 under FreeBSD 6.0, even after
updates. So maybe a problem in the SATA300 driver?

Viren


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

FreeBSD 6.1 - mount_nullfs broken?

[Viren Patel]

Hello. I just upgraded a FreeBSD 6.0 system to
6.1-prerelease and now mount_nullfs seems to be broken. I
get the error "mount_nullfs: Operation not supported by
device". Nothing has changed in the hardware or the config
files. Here is the output dmesg:

Copyright (c) 1992-2006 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991,
1992, 1993, 1994
The Regents of the University of California. All
rights reserved.
FreeBSD 6.1-PRERELEASE #0: Thu Mar 23 14:53:13 CST 2006
[EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC
Timecounter "i8254" frequency 1193182 Hz quality 0
CPU: AMD Athlon(tm) 64 X2 Processor 3800+ (2010.31-MHz
K8-class CPU)
  Origin = "AuthenticAMD"  Id = 0x20fb1  Stepping = 1
  
Features=0x178bfbff
  Features2=0x1
  AMD
Features=0xe2500800,LM,3DNow+,3DNow>
real memory  = 1073676288 (1023 MB)
avail memory = 1025048576 (977 MB)
ACPI APIC Table: 
ioapic0: Changing APIC ID to 4
ioapic0  irqs 0-23 on motherboard
acpi0:  on motherboard
acpi_bus_number: can't get _ADR
acpi_bus_number: can't get _ADR
acpi0: Power Button (fixed)
acpi_bus_number: can't get _ADR
acpi_bus_number: can't get _ADR
acpi_bus_number: can't get _ADR
acpi_bus_number: can't get _ADR
Timecounter "ACPI-fast" frequency 3579545 Hz quality 1000
acpi_timer0: <24-bit timer at 3.579545MHz> port
0x4008-0x400b on acpi0
cpu0:  on acpi0
acpi_button0:  on acpi0
pcib0:  port 0xcf8-0xcff on acpi0
pci0:  on pcib0
pci0:  at device 0.0 (no driver attached)
isab0:  at device 1.0 on pci0
isa0:  on isab0
pci0:  at device 1.1 (no driver attached)
ohci0:  mem
0xfebff000-0xfebf irq 21 at device 2.0 on pci0
ohci0: [GIANT-LOCKED]
usb0: OHCI version 1.0, legacy support
usb0:  on ohci0
usb0: USB revision 1.0
uhub0: nVidia OHCI root hub, class 9/0, rev 1.00/1.00, addr 1
uhub0: 10 ports with 10 removable, self powered
ehci0:  mem
0xfebfe000-0xfebfe0ff irq 22 at device 2.1 on pci0
ehci0: [GIANT-LOCKED]
usb1: EHCI version 1.0 usb1: companion controller, 4 ports
each: usb0 usb1:  on
ehci0
usb1: USB revision 2.0
uhub1: nVidia EHCI root hub, class 9/0, rev 2.00/1.00, addr 1
uhub1: 10 ports with 10 removable, self powered
pci0:  at device 4.0 (no driver attached)
atapci0:  port
0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xe000-0xe00f at
device 6.0 on pci0
ata0:  on atapci0
ata1:  on atapci0
atapci1:  port
0x9f0-0x9f7,0xbf0-0xbf3,0x970-0x977,0xb70-0xb73,0xcc00-0xcc0f
mem 0xfebfb000-0xfebfbfff irq 21 at device 7.0 on pci0
ata2:  on atapci1
ata3:  on atapci1
atapci2:  port
0x9e0-0x9e7,0xbe0-0xbe3,0x960-0x967,0xb60-0xb63,0xb800-0xb80f
mem 0xfebfa000-0xfebfafff irq 22 at device 8.0 on pci0
ata4:  on atapci2
ata5:  on atapci2
pcib1:  at device 9.0 on pci0
pci1:  on pcib1
fwohci0:  port 0xac00-0xac7f mem
0xfbfff000-0xfbfff7ff irq 16 at device 6.0 on pci1
fwohci0: OHCI version 1.10 (ROM=1)
fwohci0: No. of Isochronous channels is 4.
fwohci0: EUI64 00:01:29:20:00:03:4d:81
fwohci0: Phy 1394a available S400, 2 ports.
fwohci0: Link S400, max_rec 2048 bytes.
firewire0:  on fwohci0
fwe0:  on firewire0
if_fwe0: Fake Ethernet address: 02:01:29:03:4d:81
fwe0: Ethernet address: 02:01:29:03:4d:81
fwe0: if_start running deferred for Giant
sbp0:  on firewire0
fwohci0: Initiate bus reset
fwohci0: node_id=0xc800ffc0, gen=1, CYCLEMASTER mode
firewire0: 1 nodes, maxhop <= 0, cable IRM = 0 (me)
firewire0: bus manager 0 (me)
pci1:  at device 9.0 (no driver attached)
nve0:  port
0xb400-0xb407 mem 0xfebf9000-0xfebf9fff irq 23 at device
10.0 on pci0
nve0: Ethernet address 00:01:29:d2:25:69
miibus0:  on nve0
ukphy0:  on miibus0
ukphy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX,
1000baseT, 1000baseT-FDX, auto
nve0: Ethernet address: 00:01:29:d2:25:69
pcib2:  at device 11.0 on pci0
pci2:  on pcib2
pcib3:  at device 12.0 on pci0
pci3:  on pcib3
pcib4:  at device 13.0 on pci0
pci4:  on pcib4
pcib5:  at device 14.0 on pci0
pci5:  on pcib5
acpi_tz0:  on acpi0
fdc0:  port 0x3f0-0x3f5,0x3f7 irq
6 drq 2 on acpi0
fdc0: [FAST]
fd0: <1440-KB 3.5" drive> on fdc0 drive 0
sio0: <16550A-compatible COM port> port 0x3f8-0x3ff irq 4
flags 0x10 on acpi0
sio0: type 16550A
sio1:  port 0x2f8-0x2ff
irq 3 on acpi0
sio1: type 16550A
ppc0:  port 0x378-0x37f
irq 7 on acpi0
ppc0: Generic chipset (NIBBLE-only) in COMPATIBLE mode
ppbus0:  on ppc0
plip0:  on ppbus0
lpt0:  on ppbus0
lpt0: Interrupt-driven port
ppi0:  on ppbus0
atkbdc0:  port 0x60,0x64 irq
1 on acpi0
atkbd0:  flags 0x1 irq 1 on atkbdc0
kbd0 at atkbd0
atkbd0: [GIANT-LOCKED]
psm0:  irq 12 on atkbdc0
psm0: [GIANT-LOCKED]
psm0: model IntelliMouse Explorer, device ID 4
orm0:  at iomem
0xc-0xc7fff,0xc8000-0xc97ff on isa0
sc0:  at flags 0x100 on isa0
sc0: VGA <16 virtual consoles, flags=0x300>
vga0:  at port 0x3c0-0x3df iomem
0xa-0xb on isa0
Timecounter "TSC" frequency 2010312401 Hz quality 800
Timecounters tick every 1.000 msec
module_register_init: MOD_LOAD (amr_linux,
0x805f1c70, 0) error 6
acd0: DMA limited to UDMA33, device found non-ATA66 cable
acd0: DVDR  at ata0-master
UDMA33
ad6: 238475MB  at
ata3-master SATA30

apache13-modssl-1.3.33+2.8.22_1 port broken??

[Viren Patel]

I am trying to install apache+mod_ssl-1.3.33+2.8.22_1 and
get the following:

#make
===>  apache+mod_ssl-1.3.33+2.8.22_1 has known
vulnerabilities:
=> apache -- http request smuggling.
   Reference:

=> Please update your ports tree and try again.
*** Error code 1

Stop in /usr/ports/www/apache13-modssl.

I have updated the ports tree and the files directory
contains the patch for this vulnerability
(patch-secfix-CAN-2005-2088).

I need to install this port urgently. What am I doing
wrong?? If the vulnerability has not been fixed, how can I
force install? TIA.


--
Viren

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

Re: NFS broken after upgrading to 5.4

[Viren Patel]

> On Thu, May 26, 2005 at 07:49:53AM -0500, Viren Patel
> wrote:
>> >
>> > What does ^T show for the status of the 'hung'
>> process?
>> > Are you
>> > certain that DNS resolution is working correctly on
>> both
>> > machines?
>> >
>> > Kris
>> >
>> >
>>
>> Thanks for your help. Interestingly when I tried the
>> mounts this morning they all worked, so I can't provide
>> ^T
>> output. Go figure.
>>
>> Since my clients and the server are communicating over a
>> private LAN, they don't use DNS. However
>> /etc/nsswitch.conf contains:
>>
>>hosts: files dns
>>
>> and in /etc/hosts I have:
>>
>>127.0.0.1   localhost
>>192.168.0.10backuphost
>>
>> The problem occurred whether I used IP or hostname. I've
>> also noticed that mount_nfs tended to succeed if
>> preceded
>> by a ping to the backuphost. I rebooted a client and
>> tried
>> mounting without ping and it worked just fine. It's all
>> behaving really flakily. Basically it amounts to
>> sometimes
>> it works and sometimes it doesn't and there doesn't seem
>> to be a pattern.
>
> Are you sure they're not using DNS?  i.e. will DNS queries
> to those
> hostnames resolve?  Have you observed the data traffic
> using tcpdump
> to confirm no DNS lookups?
>
> Kris
>
>

I am not sure why the client would try to use DNS to
resolve backuphost if I've specified it in /etc/hosts and
/etc/nsswitch.conf specifically says to look at the files
first. I haven't done a tcpdump but I will try that and
report back.

Viren


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

Re: NFS broken after upgrading to 5.4

[Viren Patel]

>
> What does ^T show for the status of the 'hung' process?
> Are you
> certain that DNS resolution is working correctly on both
> machines?
>
> Kris
>
>

Thanks for your help. Interestingly when I tried the
mounts this morning they all worked, so I can't provide ^T
output. Go figure.

Since my clients and the server are communicating over a
private LAN, they don't use DNS. However
/etc/nsswitch.conf contains:

   hosts: files dns

and in /etc/hosts I have:

   127.0.0.1   localhost
   192.168.0.10backuphost

The problem occurred whether I used IP or hostname. I've
also noticed that mount_nfs tended to succeed if preceded
by a ping to the backuphost. I rebooted a client and tried
mounting without ping and it worked just fine. It's all
behaving really flakily. Basically it amounts to sometimes
it works and sometimes it doesn't and there doesn't seem
to be a pattern.

Viren



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

Re: NFS broken after upgrading to 5.4

[Viren Patel]

> On Wed, May 25, 2005 at 07:31:20PM -0500, Viren Patel
> wrote:
>> Hello all. I have several FreeBSD clients connecting to
>> a
>> FreeBSD NFS server via a private gigabit LAN for backup
>> purposes. Both clients and server has IPFW with all
>> traffic between clients and server allowed on the
>> private
>> LAN. All clients and the server were running
>> 5.3-RELEASE-p10 with IPFW, and NFS was working just
>> fine.
>> Then I upgraded all clients to 5.4-RELEASE-p1 and the
>> NFS
>> was working fine. Finally I upgraded the NFS server to
>> 5.4-RELEASE-p1 and now NFS has stopped working. The
>> clients experience RPC timeouts:
>>
>>  RPCPROG_NFS: RPC: Port mapper failure - RPC: Timed out
>>
>> My NFS server rc.conf has the following as per manual:
>>
>>   nfs_server_enable="YES"
>>   rpcbind_enable="YES"
>>   mountd_flags="-r"
>>
>> Clients have:
>>
>>   nfs_client_enable="YES"
>>
>> IPFW rules amount to:
>>
>> allow all from NFS_SERVER to NFS_CLIENT in via
>> INTERNAL_IF
>> allow all from NFS_CLIENT to NFS_SERVER out via
>> INTERNAL_IF
>>
>> and appear at the top of the ruleset.
>
> Double check this, then triple-check it, because a local
> configuration
> difference like this is likely to be your cause.
>
>> Obviously something has changed in the NFS system under
>> 5.4
>
> Not that I can think of.
>
> Kris
>

I have checked my configuration multiple times. Nothing
there has changed between 5.3 and 5.4. I have even
disabled the firewalls on both clients and server but the
same problem is present.

Another interesting behavior: upon reboot, mount_nfs
occasionally succeeds. However if I then go to another
client and try to access the NFS server it doesn't work.
If I then go back to the first client and try to unmount
the NFS partition it also hangs.

Viren


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

NFS broken after upgrading to 5.4

[Viren Patel]

Hello all. I have several FreeBSD clients connecting to a
FreeBSD NFS server via a private gigabit LAN for backup
purposes. Both clients and server has IPFW with all
traffic between clients and server allowed on the private
LAN. All clients and the server were running
5.3-RELEASE-p10 with IPFW, and NFS was working just fine.
Then I upgraded all clients to 5.4-RELEASE-p1 and the NFS
was working fine. Finally I upgraded the NFS server to
5.4-RELEASE-p1 and now NFS has stopped working. The
clients experience RPC timeouts:

 RPCPROG_NFS: RPC: Port mapper failure - RPC: Timed out

My NFS server rc.conf has the following as per manual:

  nfs_server_enable="YES"
  rpcbind_enable="YES"
  mountd_flags="-r"

Clients have:

  nfs_client_enable="YES"

IPFW rules amount to:

allow all from NFS_SERVER to NFS_CLIENT in via INTERNAL_IF
allow all from NFS_CLIENT to NFS_SERVER out via INTERNAL_IF

and appear at the top of the ruleset.

I even added

   nfs_reserved_port_only="YES"
   nfs_server_flags="-u -t -n 4 -h "

to the server but it makes no difference.

/etc/hosts and /etc/hosts.allow are also configured
correctly (re: NFS was working with 5.3-RELEASE-p10).

Obviously something has changed in the NFS system under
5.4, but I have no clue where to look. Any help is greatly
appreciated. Thanks.

--
Viren Patel
Chemistry & Biochemistry
University of Texas at Austin





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

Aide port broken in 5.4?

[Viren Patel]

Hello. I just upgraded my servers to 5.4 and find that the
Aide port is broken. The message I get is:

   aide-0.10_1 is marked as broken. Incomplete pkg-plist

Is this port truly broken or did I break something in the
upgrade? Thanks.

Viren

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

Re: Transferring dump file to tape?

[Viren Patel]

>>
>> > On Mon, May 16, 2005 at 11:36:18AM -0500, Viren Patel
>> > wrote:
>> >> Hello. I am using 'dump' to backup client filesystems
>> to
>> >> a
>> >> disk file on an NFS mounted partition, e.g.
>> >>
>> >>dump -0aLu -f /bk/var-20050516 /var
>> >>
>> >> Now I'd like to take the backup file (var-200500516)
>> and
>> >> move it to tape (on the NFS server). However I'd like
>> >> the
>> >> tape to have the data just as if it had been created
>> >> directly by 'dump', e.g.
>> >>
>> >>dump -0aLu -f /dev/nsa0 /var
>> >>
>> >> Any way to do this?
>> >
>> > You can use dd to write a file to a device (e.g.
>> tape).
>> >
>> > Kris
>> >
>>
>> I thought so too, but it doesn't work. When I issue
>>
>>dd if=var-20050516 of=/dev/nsa0
>>
>> I get the following error:
>>
>> dd: /dev/nsa0: Invalid argument
>> 1+0 records in
>> 0+0 records out
>> 0 bytes transferred in 0.000786 secs (0 bytes/sec)
>>
>> Since 'dump' works with /dev/nsa0 I am not sure what the
>> problem is.
>
> I would expect the dd to work also.   You may need to do
> something
> with blocksize.
>
> But, we see that error a lot on different machines while
> attempting
> to write a file to a tape.   We have code that opens the
> drive and
> writes directly to it, other places that attempt to use dd
> and also
> some places that try to cat to the drive.   All of them
> work most of
> the time, but fail consistently on some machines that
> appear to be
> otherwise identical.   We have resorted to numerous tricks
> to write
> to a tape on a system that is failing, usually to no
> avail.   The error
> you report is the one we see with dd.   I generally don't
> see any
> other console message or anything that shows in a dmesg.
>
> I would very much like to see someone who really knows
> about writing
> tapes respond to this.   I have begun to be suspicious
> that there is
> something weird about the the sa driver or something along
> the line,
> but don't know enough to really dig it out.   I posted
> some questions
> around 3 years ago, but got no response.
>
> By the way, this has occurred in various versions of
> FreeBSD from
> around 3.2 through 4.9.   I haven't experimented with it
> on 5.xx yet.
>
> jerry
>
>>
>> Viren
>

Thanks for your input. As per Kris' suggestion adding the
blocksize worked for me. 'mt' reports my tapedrive having
a blocksize of 1024 bytes (1k), so my command was

  dd if=var-200501516 of=/dev/nsa0 bs=1k

I transfered the dump file to tape and then was able to
restore directly from tape. My test dump file was small
(140K) but today I plan to test with a full-size
multi-gigabyte dump file. I haven't played long with dd to
judge its reliability. While I was looking around, I also
came across sdd (/usr/ports/sysutils/sdd), which claims to
be a better dd replacement.

Viren

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

Re: Transferring dump file to tape?

[Viren Patel]

> On Mon, May 16, 2005 at 01:52:18PM -0500, Viren Patel
> wrote:
>> > On Mon, May 16, 2005 at 11:36:18AM -0500, Viren Patel
>> > wrote:
>> >> Hello. I am using 'dump' to backup client filesystems
>> to
>> >> a
>> >> disk file on an NFS mounted partition, e.g.
>> >>
>> >>dump -0aLu -f /bk/var-20050516 /var
>> >>
>> >> Now I'd like to take the backup file (var-200500516)
>> and
>> >> move it to tape (on the NFS server). However I'd like
>> >> the
>> >> tape to have the data just as if it had been created
>> >> directly by 'dump', e.g.
>> >>
>> >>dump -0aLu -f /dev/nsa0 /var
>> >>
>> >> Any way to do this? I could use rdump/rmt to backup
>> >> directly to the remote tape, but I am doing it this
>> way
>> >> to
>> >> do fast disk-based backups during night and move to
>> >> slower
>> >> tapes during day. For normal restores I can use
>> >> NFS-based
>> >> files but have tapes for single-user-mode type
>> restores.
>> >> Thanks.
>> >
>> > You can use dd to write a file to a device (e.g.
>> tape).
>> >
>> > Kris
>> >
>>
>> I thought so too, but it doesn't work. When I issue
>>
>>dd if=var-20050516 of=/dev/nsa0
>>
>> I get the following error:
>>
>> dd: /dev/nsa0: Invalid argument
>> 1+0 records in
>> 0+0 records out
>> 0 bytes transferred in 0.000786 secs (0 bytes/sec)
>>
>> Since 'dump' works with /dev/nsa0 I am not sure what the
>> problem is.
>
> You need to specify a block size compatible with your tape
> drive,
> e.g. 32k.
>
> Kris
>


doh! Thanks.

Viren

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

Re: Transferring dump file to tape?

[Viren Patel]

> On Mon, May 16, 2005 at 11:36:18AM -0500, Viren Patel
> wrote:
>> Hello. I am using 'dump' to backup client filesystems to
>> a
>> disk file on an NFS mounted partition, e.g.
>>
>>dump -0aLu -f /bk/var-20050516 /var
>>
>> Now I'd like to take the backup file (var-200500516) and
>> move it to tape (on the NFS server). However I'd like
>> the
>> tape to have the data just as if it had been created
>> directly by 'dump', e.g.
>>
>>dump -0aLu -f /dev/nsa0 /var
>>
>> Any way to do this? I could use rdump/rmt to backup
>> directly to the remote tape, but I am doing it this way
>> to
>> do fast disk-based backups during night and move to
>> slower
>> tapes during day. For normal restores I can use
>> NFS-based
>> files but have tapes for single-user-mode type restores.
>> Thanks.
>
> You can use dd to write a file to a device (e.g. tape).
>
> Kris
>

I thought so too, but it doesn't work. When I issue

   dd if=var-20050516 of=/dev/nsa0

I get the following error:

dd: /dev/nsa0: Invalid argument
1+0 records in
0+0 records out
0 bytes transferred in 0.000786 secs (0 bytes/sec)

Since 'dump' works with /dev/nsa0 I am not sure what the
problem is.

Viren

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

Transferring dump file to tape?

[Viren Patel]

Hello. I am using 'dump' to backup client filesystems to a
disk file on an NFS mounted partition, e.g.

   dump -0aLu -f /bk/var-20050516 /var

Now I'd like to take the backup file (var-200500516) and
move it to tape (on the NFS server). However I'd like the
tape to have the data just as if it had been created
directly by 'dump', e.g.

   dump -0aLu -f /dev/nsa0 /var

Any way to do this? I could use rdump/rmt to backup
directly to the remote tape, but I am doing it this way to
do fast disk-based backups during night and move to slower
tapes during day. For normal restores I can use NFS-based
files but have tapes for single-user-mode type restores.
Thanks.

--
Viren Patel
Chemistry & Biochemistry
University of Texas at Austin

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

5.3 sendmail delayed delivery from cron

[Viren Patel]

Hello. We have a FreeBSD 5.3 box with stock sendmail
running in delivery only mode (sendmail="NO"). When a cron
job generates output, the delivery of that message to
external address is delayed by one day. Nothing obvious is
in the maillog. The system can resolve hostnames and MX
records. The proper aliases are in /etc/aliases,
/usr/bin/newaliases has been run, sendmail has been
restarted. Anything else/obvious I may be overlooking?
Message from non-cron script or command line (e.g. mailx
-s 'test' external_address < file) gets delivered
immediately. Thanks.

--
Viren Patel
Chem. & Biochem.
Univ. of Texas @ Austin

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

Re: Sharing directories with jails

[Viren Patel]

> On Thursday 03 March 2005 05:23 pm, Ean Kingston wrote:
>> > On Thursday 03 March 2005 12:42 pm, Chris Hodgins
>> wrote:
>>
>> [cut original question and answer]
>>
>> >> Ok perhaps I should clarify what my intentions are a
>> little
>> >> more. I am planning on providing a FreeBSD jail for
>> any member
>> >> of a geek society I am a member of.  When I say they
>> are
>> >> untrusted, I mean that I won't be giving them full
>> root access
>> >> to my server but I trust them enough not to do
>> anything
>> >> malicious inside a jail.  It is just like a fun place
>> they can
>> >> play and not have to worry to much about breaking
>> things.
>> >>
>> >> How easy is it exactly to break out of a jail if you
>> have access
>> >> to development tools?
>> >
>> > http://www.securiteam.com/unixfocus/5WP031535U.html
>>
>> How current is this? The article appears to be dated
>> 2001. Are
>> there still buffer-overflow issues with /proc?
>>
>
> 5.3 and later no longer need proc and it's not mounted by
> default.
>
>> > If you use securelevels you can a sigificantly improve
>> security.
>
> --
> Anish Mistry
>

The jail manpage instructs to mount proc when starting a
jail and the /etc/rc.d/jail scripts mounts both devfs and
procfs. Are you saying this is not needed and if so why
and how to disable? Thanks.

--
Viren Patel


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

Compiling linux_base in jail fails

[Viren Patel]

Hello. I am trying to install linux_base inside a jail and
it fails with the following:

#make
===>  Extracting for linux_base-rh-7.3
=> Checksum OK for
rpm/rh-7.3/glibc-common-2.2.5-44.legacy.3.i386.rpm.
=> Checksum OK for
rpm/rh-7.3/glibc-2.2.5-44.legacy.3.i386.rpm.
=> Checksum OK for
rpm/rh-7.3/redhat-release-7.3-1.noarch.rpm.
=> Checksum OK for rpm/rh-7.3/setup-2.5.12-1.noarch.rpm.
=> Checksum OK for rpm/rh-7.3/filesystem-2.1.6-2.noarch.rpm.
=> Checksum OK for rpm/rh-7.3/basesystem-7.0-2.noarch.rpm.
=> Checksum OK for rpm/rh-7.3/termcap-11.0.1-10.noarch.rpm.
=> Checksum OK for rpm/rh-7.3/db1-1.85-8.i386.rpm.
=> Checksum OK for rpm/rh-7.3/db3-3.3.11-6.i386.rpm.
=> Checksum OK for rpm/rh-7.3/gdbm-1.8.0-14.i386.rpm.
=> Checksum OK for rpm/rh-7.3/glib-1.2.10-5.i386.rpm.
=> Checksum OK for rpm/rh-7.3/glib2-2.0.1-2.i386.rpm.
=> Checksum OK for rpm/rh-7.3/libtermcap-2.0.8-28.i386.rpm.
=> Checksum OK for rpm/rh-7.3/bash-2.05a-13.i386.rpm.
=> Checksum OK for rpm/rh-7.3/bzip2-libs-1.0.2-2.i386.rpm.
=> Checksum OK for rpm/rh-7.3/bzip2-1.0.2-2.i386.rpm.
=> Checksum OK for rpm/rh-7.3/libstdc++-2.96-110.i386.rpm.
=> Checksum OK for
rpm/rh-7.3/compat-libstdc++-6.2-2.9.0.16.i386.rpm.
=> Checksum OK for rpm/rh-7.3/ncurses-5.2-26.i386.rpm.
=> Checksum OK for rpm/rh-7.3/info-4.1-1.i386.rpm.
=> Checksum OK for rpm/rh-7.3/fileutils-4.1-10.i386.rpm.
=> Checksum OK for rpm/rh-7.3/grep-2.5.1-1.i386.rpm.
=> Checksum OK for rpm/rh-7.3/popt-1.6.4-7x.18.i386.rpm.
=> Checksum OK for rpm/rh-7.3/readline-4.2a-4.i386.rpm.
=> Checksum OK for rpm/rh-7.3/setserial-2.17-5.i386.rpm.
=> Checksum OK for rpm/rh-7.3/slang-1.4.5-2.i386.rpm.
=> Checksum OK for rpm/rh-7.3/sh-utils-2.0.11-14.i386.rpm.
=> Checksum OK for rpm/rh-7.3/rpm-4.0.4-7x.18.i386.rpm.
=> Checksum OK for rpm/rh-7.3/zlib-1.1.3-25.7.i386.rpm.
=> Checksum OK for
rpm/rh-7.3/XFree86-libs-4.3.0-78.EL.tj.i386.rpm.
===>   linux_base-rh-7.3 depends on executable: rpm - found
LC_ALL=C rpm --initdb --root
/usr/ports/emulators/linux_base/work/linux_base-rh-7.3
--dbpath /var/lib/rpm
kern.fallback_elf_brand: -1
sysctl: kern.fallback_elf_brand: Operation not permitted
ELF binary type "0" not known.
execution of glibc-2.2.5-44.legacy.3 script failed, exit
status 255
*** Error code 1

Stop in /usr/ports/emulators/linux_base.

Any ideas on how to get linux emulator running inside a
jail? Thanks.

--
Viren Patel
Chemistry & Biochemistry
University of Texas at Austin

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

Strange IP problem with 5.3 STABLE #2

[Viren Patel]

Hello. We have a Dell PE 1850 (dual Intel PRO/1000 nics,
em driver) running FreeBSD 5.3 STABLE #2. The generic
kernel has been configured to include IPFW and SMP as
follows:

   options   IPFIREWALL
   options   IPFIREWALL_VERBOSE
   options   IPDIVERT
   options   SMP

No other changes have been made. The firewall options in
rc.conf are as follows:

   firewall_enable="YES"
   firewall_type="OPEN"
   firewall_quiet="NO"

The problem is that some IP addresses on our class C
subnet do not work. I mean the NIC can be configured to
use the IP address and it shows up in ifconfig output, but
the system cannot be pinged or otherwise accessed from
outside the subnet. Within the subnet it is fine.

The only other reference I have found to a similar issue
is one involving VLANs

http://lists.freebsd.org/pipermail/freebsd-current/2004-November/042591.html

but I think my problem is different as I am not
configuring any VLANs. Also rebooting the system makes no
difference.

I had been using the RELENG_5_3 branch and run into this
problem. I switched to RELENG_5, rebuilt the system, and
the IPs that were giving me problems started working. Now,
however, I have found another one that does not work.

I am running 5_3 on other Dell and non-Dell systems but
with different network cards and have not encountered this
problem. The problem also occurs intermittently with the
GENERIC kernel. This leads me to believe the cause may be
related to Intel PRO/1000 and/or em driver.

Any help is much appreciated?

Viren

--
Viren Patel
Chemistry & Biochemistry
University of Texas at Austin


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

Re: problems getting a jail running on 5.3-RELEASE using man 8 jail

[Viren Patel]

> I am running a 5.3-RELEASE-p5 machine.
>
> $uname -a
> FreeBSD twinmp.tcbug.org 5.3-RELEASE-p5 FreeBSD
> 5.3-RELEASE-p5 #0: Wed
> Feb  9 16:54:40 CST 2005
> [EMAIL PROTECTED]:/usr/obj/usr/src/sys/TWINMP
> i386
>
> Hardware is basic PC sound/lan/video, 4 IDE channels, dual
> AMD MP
> processors. dmesg is attached.
>
> I am following the following procedure in the jail
> manpage:
>
> #D=/usr/local/jail
> #cd /usr/src
> #mkdir -p $D
> #make world DESTDIR=$D
> #cd etc
> #make distribution DESTDIR=$D
> #mount_devfs devfs $D/dev
> #cd $D
> #ln -sf dev/null kernel
>
> I get to:
> #make world DESTDIR=$D
> and the compile fails at the following point:
>
> ===> games/fortune/strfile
> /usr/obj/usr/src/i386/usr/src/games/fortune/strfile
> created
> for /usr/src/games/fortune/strfile
> rm -f .depend
> mkdep -f .depend -a
> -I/usr/obj/usr/src/i386/legacy/usr/include
> /usr/src/games/fortune/strfile/strfile.c
> echo
> strfile: /storage/jail/usr/lib/libc.a
> /usr/obj/usr/src/i386/legacy/usr/lib/libegacy.a
>>> .depend
> cc -O -pipe  -I/usr/obj/usr/src/i386/legacy/usr/include
> -c /usr/src/games/fortune/strfile/strfile.c
> make: don't know how to make /storage/jail/usr/lib/libc.a.
> Stop
> *** Error code 2
>
> Stop in /usr/src.
> *** Error code 1
>
> Stop in /usr/src.
> *** Error code 1
>
> Stop in /usr/src.
>
> This is reproducable on my hardware with RELEASE-5.3,
> RELEASE-5.3-p2
> and RELEASE-5.3-p5.  I found an email in the -questions
> mailing list
> in which the author has the same errors that I do
> attempting to build
> world.
>
> <http://docs.FreeBSD.org/cgi/mid.cgi?57d71501131800788ec662>
>
> If I replace:
> #make world DESTDIR=$D with
> #env DESTDIR=$D make world everything goes fine.
>
> (just for kicks I tested with /bin/sh /bin/csh and
> /usr/local/etc/zsh)
>
> I considered sending a patch for man 8 jail but thought it
> would be
> better to get input first.  If a solution isn't found I'll
> send-pr
> the docproj with a patch.
>
> --
> Thanks,
>
> Josh Paetzel
> ___
> freebsd-questions@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-questions
> To unsubscribe, send any mail to
> "[EMAIL PROTECTED]"
>

Hello. I have also had success with:

#/bin/tcsh
#setenv DESTDIR /path/to/jail
then follow the instructions for "Setting up a Jail
Directory Tree".

FYI: also look at "Managing Jails on FreeBSD 5" at
http://www.devco.net - invaluable and should be added, if
possible, to the jail man pages IMHO.

--
Viren Patel
Chemistry & Biochemistry
University of Texas at Austin

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

Re: dell poweredge servers

[Viren Patel]

> I was looking at the support hardware list for Fbsd 5.x
> and could find
> no mention of the PERC3-DI scsi controller.. so I was
> wondering if
> anyone has used a dell poweredge 2650, and what your
> experience was
> running Freebsd 4.X and 5.x on it.
>
> --
> David Bear
> phone:480-965-8257
> fax:  480-965-9189
> College of Public Programs/ASU
> Wilson Hall 232
> Tempe, AZ 85287-0803
>  "Beware the IP portfolio, everyone will be suspect of
> trespassing"

We've been running FreeBSD 5.3 on PE2650 with no issues.

--
Viren Patel
Chemistry & Biochemistry
University of Texas at Austin

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

make installworld fails

[Viren Patel]

Hello. I am trying to upgrade FreeBSD 5.3 and performed
the following steps on a clean install and after upgrading
sources:

Booted into single-user mode

   cd /usr/src
   make buildworld
   make buildkernel KERNCONF=GENERIC
   make installkernel KERNCONF=GENERIC

Then rebooted the system back into single-user mode

   cd /usr/src
   make installworld

However this fails and the last few lines on screen are:

install -o root -g wheel -m 444 zzz.8.gz /usr/share/man/man8
===> etc
===> etc/sendmail
cd /usr/src/etc/../share/man; make makedb
makewhatis /usr/share/man
pid 17053 (makewhatis), uid 0: exited on signal 11 (core
dumped)
*** Signal 11

Stop in /usr/src/share/man
*** Error code 1

Stop in /usr/src/etc
*** 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

I couldn't find a core file on the system.

Please help. Thank you.

--
Viren Patel
Chemistry & Biochemistry
University of Texas at Austin

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

Apache+mod-ssl dumps core on shutdown only

[Viren Patel]

Hello. I am running apache+mod_ssl-1.3.29+2.8.16 with
mod_php4-4.3.4_2,1 on FreeBSD 4.8-p13. Everything works fine.
However, upon shutting down apache (/usr/local/etc/rc.d/apache.sh
stop), apache dumps core. The following are the syslog entries to
the console:

pid 13918 (httpd), uid 80: exited on signal 11
pid 13916 (httpd), uid 80: exited on signal 11
pid 13915 (httpd), uid 80: exited on signal 11
pid 13912 (httpd), uid 80: exited on signal 11
pid 13914 (httpd), uid 80: exited on signal 11
pid 13913 (httpd), uid 80: exited on signal 11
pid 13910 (httpd), uid 0: exited on signal 11 (core dumped)

Does anyone have any idea of why this is happening? Thanks.



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