Re: How to force a static /etc/resolv.conf?
On 2013-06-13 05:02, Loic Capdeville wrote: You can configure it in your dhclient.conf file. Use the supersede keyword. For example, in your case add: supersede domain-search example.com example.net supersede domain-name-servers 2001:db8::53 That only addresses the DHCPv4 client. The DHCPv6 client doesn't have those options and neither do the VPN clients. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: How to force a static /etc/resolv.conf?
On 2013-06-12 17:46, Darren Pilgrim wrote: How do I tell resolvconf to always use a static configuration or, better yet, to not muck with /etc/resolv.conf at all? According to the project developer, the answer is to have resolvconf not touch /etc/resolv.conf by put the following in /etc/resolvconf.conf resolv_conf=/dev/null Then you just edit /etc/resolv.conf directly. ___ 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
How to force a static /etc/resolv.conf?
I'm running 9.1. I run a local recursive resolver, so my /etc/resolv.conf needs to remain static. I have DHCPv4, DHCPv6 and VPN clients running which all want to modify /etc/resolv.conf. I have set in /etc/resolvconf.conf: search_domains=example.com. example.net. name_servers=2001:db8::53 But that only prepends that information. Search domains and nameservers from other sources still get included. I can set /etc/resolv.conf as immutable, but's a hack and it generates errors from resolveconf. How do I tell resolvconf to always use a static configuration or, better yet, to not muck with /etc/resolv.conf at all? ___ 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
How to make local periodics run before base periodics?
I have some periodic scripts from ports that I need to run before periodic scripts in /etc/periodic; but I can't see how to make it so. Periodic always processes /etc/periodic before $local_periodic. If I move /etc/periodic/*/999.local file to 000.local and set: local_periodic= daily_local=/usr/local/etc/periodic/daily/* weekly_local=/usr/local/etc/periodic/weekly/* monthly_local=/usr/local/etc/periodic/monthly/* Then /usr/local/etc/periodic runs first, but it's not scalable since I moved a file that mergemaster cares about. It's also noisy because 999.local generates output for all scripts, even those that are not enabled. What is the proper way to make local periodics run before base periodics? ___ 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
Removing sendmail from an installed system
I'm removing sendmail entirely from an installed system. I had WITHOUT_SENDMAIL in /etc/src.conf when I updated to RELENG_8_3, but that left an old version of sendmail rotting away on disk. This is the list I have so far: /etc/mail/* (excluding mailer.conf) /etc/rc.d/sendmail /usr/bin/vacation /usr/libexec/mail.local /usr/libexec/sendmail /usr/libexec/smrsh /usr/sbin/editmap /usr/sbin/mailstats /usr/sbin/makemap /usr/sbin/praliases /usr/share/sendmail /var/spool/clientmqueue /var/spool/mqueue Is this list complete? I'm intentionally leaving the stuff for mailwrapper. I'm ok with leaving /etc/rc.d/sendmail behind as well, but it looks like it's not needed by anything (i.e., nothing requires mail). Even though I have WITHOUT_SENDMAIL specified and the world was built with that, mergemaster still installs /etc/mail/aliases and /etc/rc.d/sendmail. Is there a way to prevent this other than adding them to IGNORE_FILES in mergemasterrc? ___ 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: puc(4) not attaching to NM9845-based serial card in 6.3-p5
Darren Pilgrim wrote:
I have the following PCI, 4-port serial card:
[EMAIL PROTECTED]:2:0: class=0x070002 card=0x00041000 chip=0x98459710 rev=0x01
hdr=0x00
vendor = 'MosChip Semiconductors (Was: Netmos Technology)'
device = 'Nm9845 Parallel/Serial Port Adapter'
class = simple comms
subclass = UART
This is supposedly supported by puc(4); however, I'm unable to get the
driver to attach to the card. I tried both uart+puc and sio+puc. In
all both case, the uart/sio drivers attach to only the serial port on
the motherboard. There are a number of hits on the mailing list
archives that discuss puc(4) not attaching to this device, but none
provide any conclusive answers.
After much searching, I finally found a message from Marcel Moolenaar to
freebsd-stable date 2008-05-06 regarding PCI serial cards working in 6.2
but not 6.3. The issue (and mine) were due to a problem with v1.51.2.3
of sys/dev/puc/pucdata.c and caused by the commit to add support for the
six-port version of my card. The following change:
--- pucdata.c.orig 2006-12-15 14:31:37.0 -0800
+++ pucdata.c 2008-11-16 15:34:50.0 -0800
@@ -946,7 +946,7 @@
/* NetMos 4S0P PCI: 4S, 0P */
{ NetMos NM9845 Quad UART,
- { 0x9710, 0x9845, 0, 0x0014 },
+ { 0x9710, 0x9845, 0, 0x0004 },
{ 0x, 0x, 0, 0x },
{
{ PUC_PORT_TYPE_COM, 0x10, 0x00, COM_FREQ },
Fixes the issue and I now have:
# dmesg | egrep '(uart|puc)'
Preloaded elf module /boot/kernel/uart.ko at 0xc08312c4.
Preloaded elf module /boot/kernel/puc.ko at 0xc0831370.
puc0: NetMos NM9845 Quad UART port
0x1060-0x1067,0x1058-0x105f,0x1050-0x1057,0
x1048-0x104f,0x1040-0x1047,0x1020-0x102f irq 18 at device 2.0 on pci6
puc: name: NetMos NM9845 Quad UART
puc0: Reserved 0x8 bytes for rid 0x10 type 4 at 0x1060
puc0: Reserved 0x8 bytes for rid 0x14 type 4 at 0x1058
puc0: Reserved 0x8 bytes for rid 0x18 type 4 at 0x1050
puc0: Reserved 0x8 bytes for rid 0x1c type 4 at 0x1048
puc: Using uart0
puc: type 1, bar 10, offset 0
puc_alloc_resource: pdev 0xc4c37d60, looking for t 4, r 0
puc_alloc_resource: pdev 0xc4c37d60, looking for t 4, r 0
uart0: Non-standard ns8250 class UART with FIFOs on puc0
puc_alloc_resource: pdev 0xc4c37d60, looking for t 4, r 0
puc_alloc_resource: pdev 0xc4c37d60, looking for t 1, r 0
puc: Using uart1
puc: type 1, bar 14, offset 0
puc_alloc_resource: pdev 0xc4c37ca0, looking for t 4, r 0
puc_alloc_resource: pdev 0xc4c37ca0, looking for t 4, r 0
uart1: 16950 or compatible on puc0
puc_alloc_resource: pdev 0xc4c37ca0, looking for t 4, r 0
puc_alloc_resource: pdev 0xc4c37ca0, looking for t 1, r 0
puc: Using uart2
puc: type 1, bar 18, offset 0
puc_alloc_resource: pdev 0xc4c37be0, looking for t 4, r 0
puc_alloc_resource: pdev 0xc4c37be0, looking for t 4, r 0
uart2: 16550 or compatible on puc0
puc_alloc_resource: pdev 0xc4c37be0, looking for t 4, r 0
puc_alloc_resource: pdev 0xc4c37be0, looking for t 1, r 0
puc: Using uart3
puc: type 1, bar 1c, offset 0
puc_alloc_resource: pdev 0xc4c37b20, looking for t 4, r 0
puc_alloc_resource: pdev 0xc4c37b20, looking for t 4, r 0
uart3: 16550 or compatible on puc0
puc_alloc_resource: pdev 0xc4c37b20, looking for t 4, r 0
puc_alloc_resource: pdev 0xc4c37b20, looking for t 1, r 0
uart4: 16550 or compatible port 0x3f8-0x3ff irq 4 on acpi0
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
puc(4) not attaching to NM9845-based serial card in 6.3-p5
I have the following PCI, 4-port serial card: [EMAIL PROTECTED]:2:0: class=0x070002 card=0x00041000 chip=0x98459710 rev=0x01 hdr=0x00 vendor = 'MosChip Semiconductors (Was: Netmos Technology)' device = 'Nm9845 Parallel/Serial Port Adapter' class = simple comms subclass = UART This is supposedly supported by puc(4); however, I'm unable to get the driver to attach to the card. I tried both uart+puc and sio+puc. In all both case, the uart/sio drivers attach to only the serial port on the motherboard. There are a number of hits on the mailing list archives that discuss puc(4) not attaching to this device, but none provide any conclusive answers. -- Darren Pilgrim ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Unable to unmount idle filesystem on 6.2
Kris Kennaway wrote: Darren Pilgrim wrote: softdep_waitidle: Failed to flush worklist for 0xc66e5298 A quick check and that message gets spit out whenever I issue any of the following commands: # mount -uo ro /usr/ports # umount /usr/ports # umount -f /usr/ports As luck would have it I ran into this in my own testing yesterday, and Kostik Belousov has a proposed fix (apply it with patch -p2). This is against 8.0 but should also appy to 7.0. Don't know about 6.x. It didn't apply cleanly, with hunks failing in: sys/kern/vfs_subr.c (all) sys/sys/vnode.h sys/ufs/ffs/ffs_softdep.c (hunk 4) sys/ufs/ffs/ffs_vfsops.c The chunks that succeeded all had offsets, so I'm guessing the patch isn't going to work for 6.x. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Unable to unmount idle filesystem on 6.2
Kris Kennaway wrote: Darren Pilgrim wrote: I'm unable to unmount an idle filesystem (or even drop it to read-only): # mount /dev/da0s1a on / (ufs, local, noatime) devfs on /dev (devfs, local) /dev/da0s1d on /var (ufs, local, noatime, soft-updates) /dev/da0s1e on /usr (ufs, local, noatime, soft-updates) /dev/da0s1fp1 on /usr/obj (ufs, asynchronous, local, noatime) /dev/da0s1fp2 on /usr/ports (ufs, local, soft-updates) /dev/da0s1fp3 on /usr/src (ufs, local, soft-updates) /dev/da0s2d on /data (ufs, local, noatime, soft-updates) # fstat -f /usr/ports USER CMD PID FD MOUNT INUM MODE SZ|DV R/W # umount /usr/ports umount: unmount of /usr/ports failed: Device busy # umount -f /usr/ports umount: unmount of /usr/ports failed: Device busy # mount -o ro /usr/ports mount: /dev/da0s1fp2: Operation not permitted # uname -r 6.2-RELEASE-p8 Strange, can you break to DDB and do 'show lockedvnods'? I don't have the necessary options compiled into the kernel. I'll build a kernel with the KDB and DDB options and hope the problem recurs. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Unable to unmount idle filesystem on 6.2
Kris Kennaway wrote: Darren Pilgrim wrote: I'm unable to unmount an idle filesystem (or even drop it to read-only): # mount /dev/da0s1a on / (ufs, local, noatime) devfs on /dev (devfs, local) /dev/da0s1d on /var (ufs, local, noatime, soft-updates) /dev/da0s1e on /usr (ufs, local, noatime, soft-updates) /dev/da0s1fp1 on /usr/obj (ufs, asynchronous, local, noatime) /dev/da0s1fp2 on /usr/ports (ufs, local, soft-updates) /dev/da0s1fp3 on /usr/src (ufs, local, soft-updates) /dev/da0s2d on /data (ufs, local, noatime, soft-updates) # fstat -f /usr/ports USER CMD PID FD MOUNT INUM MODE SZ|DV R/W # umount /usr/ports umount: unmount of /usr/ports failed: Device busy # umount -f /usr/ports umount: unmount of /usr/ports failed: Device busy # mount -o ro /usr/ports mount: /dev/da0s1fp2: Operation not permitted # uname -r 6.2-RELEASE-p8 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Strange, can you break to DDB and do 'show lockedvnods'? I haven't done that yet; however, I did find 12 instances of the following in the log: softdep_waitidle: Failed to flush worklist for 0xc66e5298 A quick check and that message gets spit out whenever I issue any of the following commands: # mount -uo ro /usr/ports # umount /usr/ports # umount -f /usr/ports A bit of searching on that error message tells me I've hit some kind of a corner case with soft-updates. The filesystem was mounted read-only, then upgraded to rw so I could update the ports tree. After cvsup was done, I tried to take the filesystem back down to read-only. The common case seems to be that the mount change is followed too quickly after the large number of writes and it somehow wedges soft-updates. Unfortunately, I haven't been able to find a fix other than rebooting the machine. The problem is that the search results[1] also tell me the filesystem may well be hosed and the reboot won't be clean. Luckily for me, I can just drop the FS from /etc/fstab and newfs the partition after the box comes back up. [1]: http://lists.freebsd.org/pipermail/freebsd-current/2007-February/069178.html ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Unable to unmount idle filesystem on 6.2
I'm unable to unmount an idle filesystem (or even drop it to read-only): # mount /dev/da0s1a on / (ufs, local, noatime) devfs on /dev (devfs, local) /dev/da0s1d on /var (ufs, local, noatime, soft-updates) /dev/da0s1e on /usr (ufs, local, noatime, soft-updates) /dev/da0s1fp1 on /usr/obj (ufs, asynchronous, local, noatime) /dev/da0s1fp2 on /usr/ports (ufs, local, soft-updates) /dev/da0s1fp3 on /usr/src (ufs, local, soft-updates) /dev/da0s2d on /data (ufs, local, noatime, soft-updates) # fstat -f /usr/ports USER CMD PID FD MOUNT INUM MODE SZ|DV R/W # umount /usr/ports umount: unmount of /usr/ports failed: Device busy # umount -f /usr/ports umount: unmount of /usr/ports failed: Device busy # mount -o ro /usr/ports mount: /dev/da0s1fp2: Operation not permitted # uname -r 6.2-RELEASE-p8 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: ipprecedence ?
[Redirected to -questions from -net.] From: S.I How Can I set ipprecedence flag on FreeBSD? Precendence bits are part of the ip_tos bits in FreeBSD inet sockets. The ip(4) man page gives an example of using setsockopt(2) to set the ToS bits. See src/sys/netinet/ip.h (v1.29) lines 76 to 99 for a set of named precedence and ToS options (you don't have to use them, but it's recommended). ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: How to bind ntpd to a single address?
From: Matthew Seaman [mailto:[EMAIL PROTECTED] Darren Pilgrim wrote: I don't like (let alone want) ntpd binding to every IP address on the host. The man pages don't say anything about specifying a binding address for ntpd. A search of the sources and Google also failed to reveal anything useful. So how to I tell ntpd to bind to a specific IP address? ntpd doesn't have that functionality I'm afraid. The next best you can do is review your /etc/ntpd.conf 'restrict' rules carefully and implement a firewall to control access to port 123/UDP. The ntp.conf(5) man page isn't what I would consider well-written, so it's a bit difficult understand how rules are applied. For example, if I put: restrict default noquery nopeer limited restrict local_network/mask nomodify restrict peerhost nomodify restrict 127.0.0.1 Does that mean: - Provide only rate-limited, non-peering time service by default. - Provide unlimited time service to the local network and also let the local network make read-only mode 6/7 queries. - Peers are given the same treatment as the local network. - Let localhost do anything. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
How to bind ntpd to a single address?
I don't like (let alone want) ntpd binding to every IP address on the host. The man pages don't say anything about specifying a binding address for ntpd. A search of the sources and Google also failed to reveal anything useful. So how to I tell ntpd to bind to a specific IP address? ___ 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 mount partitions with soft-updates enabled with asyncoption
From: Matthias Buelow Lefteris Tsintjelis [EMAIL PROTECTED] writes: I am not sure if I do something wrong here or it is suppose to work that way but the async option doesn't seem to work for partitions that have soft-updates turned on. Can someone please clarify the difference and if the speed difference (if any) is significant when using the async option instead of the soft-updates for cases such as the /usr/obj or as a squid data storage? Is async preferred over soft-updates when data loss is not a big issue? With softupdates, everything is asynchronous so the option doesn't make sense. For improving squid filesystem performance, have you mounted the partition with noatime? That might make some difference. No. With softupdates, file writes are asynchronous, but writes to filesystem structures (metadata) are synchronous to prevent filesystem corruption if the machine crashes. The async mount option writes both asynchronously. You can't use the async mount option on a volume with softupdates turned on because the two options are mutually exclusive. [ Note: -stable trimmed for relevance. ] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Will an Intel PRO/Wireless 2915 mini-PCI NIC and WPA work on FreeBSD 5.3-R?
I have a laptop with an Intel PRO/Wireless 2915 (a/b/g) mini-pci card. The card is seen during boot, but no driver is attached. I haven't had much luck finding reference to it in any manual pages, src/sys/conf/NOTES, src/sys/i386/conf/NOTES, the mailing list archive or Google. Blindly stabbing away, I tried the wi, arl and ath, none of which attached. I did, however, notice the ndis driver. I'm not thrilled with the idea of using a Windows driver, but this ain't no server and I'm even less thrilled with not having wireless. Is the ndis driver the solution? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Finding the Right Sound Driver ...
[Redirected from -newbies.] On Tue, October 26, 2004 12:50 pm, Siavash EDRISI said: Hi! I have been reading the text Setting Up the Sound Card at http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/sound-setup.html in order to find and install the right driver for the sound card in my i386. The hardware is an ESS 1869 PCI. Before I had WinXP installed on this hardware and it reported that the location of the card is: PCI slot 3 (PCI bus 0, device 14, function 0) I did whatever is described in the section 7.2.1 of the manual: 1- I checked the entries in /boot/defaults/loader.conf. The right line was snd_sbc_load=NO I imagine this means that the driver is already somewhere on the harddisk! It means the kernel will not attempt to load the snd_sbc module at boot. 2- Then I inserted the following line into /boot/loader.conf snd_sbc_load=YES 3- I booted the system. Right after calling startx and entering KDE I got (as usual) the error message that the sound server could not find the file /dev/dsp. This may be due to a number of reasons. Among them are the snd_sbc driver not being the right driver for your hardware and KDE/X not being configured with the correct device. The first step is to provide copies of the outputs of the commands `uname -a`, `dmesg` and `kldstat`. Did you first try loading snd.ko (4.x) or snd_driver.ko (5.x) and see which driver finds your hardware? As I read in the manual configuring a custom kernel with sound support is just a second method and can be used alternatively. So I am not sure if I really have to do something in the kernel or not, since the first efforts did not help! Typically you don't need to recompile the kernel just to add something like a sound driver. Loading the module works just fine. Just to be sure, please provide a copy of your kernel configuration file so it can be checked. Could someone please tell me what I still have to do? Send a reply with the requested, necessary information. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Finding the Right Sound Driver ...
From: Matt Navarre [mailto:[EMAIL PROTECTED] On Tuesday 26 October 2004 01:55, Darren Pilgrim wrote: The first step is to provide copies of the outputs of the commands `uname -a`, `dmesg` and `kldstat`. Also the output of cat /dev/sndstat would help. Did you first try loading snd.ko (4.x) or snd_driver.ko (5.x) and see which driver finds your hardware? If you load the snd* kernel module you can cat /dev/sndstat and it should tell you which driver attached to you card. Or just read the attach messages. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: NTLDR missing after 5-RELEASE install
[ Note to Roman: Please adjust your mail client to wrap lines at more acceptable range of 72-76 characters. Thanks. ] From: Roman Kurakin I've seen this before 5.0 release and made some investigation of this proble. I didn't look this thread carefully so excuse me if information I give to all you is useless. My investigation show that FreeBSD reads full partition table, and after modification puts it back. It fix all entries from its own point of view. Windows dies from change of end of partition entry. As I understand with large disk it shouldn't mean anything at all. But windows checks it. You may save this entry and after installation of FreeBSD put it back. Just like in UFS, there are structures in NTFS that have to be changed if the size of the volume changes. Also like UFS, NTFS doesn't place data sequentially on the disk. A large, mostly-empty, NTFS volume can have data at or near the end of the volume. The slicing issue is well-known with NTFS. A reliable way to add a non-Windows slice to a computer with Windows installed is to use a volume management tool like PM. If you don't need to resize a slice, use the Disk Management administrative tool to create an unformatted partition, then change the media descriptor when you go to install FreeBSD. NT is so picky it's even recommended that you use the NT boot loader. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: What mouse? (was: Samsung Cordless Mouse)
From: Greg 'groggy' Lehey Can anybody recommend a good mouse? My criteria are: - Middle button easy to use. The current crop of mice has the middle button integrated with the roller, and that makes the middle button either heavy or easy to confuse with the roller. - Preferably cordless. Cord mice tend to wander a little when you let go of them, and that's a real nuisance on a high-resolution display. I have a Logitech MX700. Very solid mouse, excellent performance and rechargable battery life. It can also run on standard alkalines (though you can't charge them). The mouse is heavier than most, but this seems to help with making smooth movements. The weight makes some of the more fervid in-game mouse maneuvers a bit tiresome on the wrist, though. It does integrate the middle button with the wheel. But there is hope! The force needed to press the wheel-button isn't much more than that of the right and left buttons. The return spring on the wheel housing can be easily removed. Doing so makes the return tension the same as the left and right buttons without affecting the wheel's functionality. It also has five additional buttons which are presented as separate buttons (6 through 10, in xf86config). They could be mapped to the middle button if you don't want to do surgery on your mouse. I've used the MX700 in 5.1 with XF86 4.3.x with great success. The only thing I couldn't get to work was the AppSwitch button, but I ended up never needing to use it anyway. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Which version of FreeBSD to support a 3ware Escalade 7006 and 8006 controllers?
I'm looking at getting a 3ware Escalade 7006 or 8006 RAID controller for one of my servers. The machine presently runs RELENG_4_8. The twe man page for that version doesn't list the 7000 or 8000 series controllers. However, 3ware lists 4.8 as the supported version of FreeBSD for both. Which is correct? More to the point: What would be the recommended version of FreeBSD I should use for these controllers? Does it really matter? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Booting from a 3ware ATA RAID
I'm looking at getting a 3ware Escalade 7006. Can FreeBSD 4.x and 5.x boot from a RAID on this controller? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
How to move a disk that's part of a vinum mirror?
On a machine running RELENG_4_8, I have two parititions, ad0s1d and ad4s1e, configured as a mirror using vinum. I need to move one of the drives to another controller, resulting in ad4 changing to ad2. I read through the vinum man page, saw the move command, then read elsewhere that vinum's internal naming is independent of the underlying disk device name. Can I just move the disk and have it work or will I need to do adjust my vinum configuration? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Different sysinstall labelling behaviour when run post-install vs. during install?
From: Matthew Seaman [mailto:[EMAIL PROTECTED] How do I manually specify or change the partition letter to use? Don't use sysinstall(8). sysinstall is like the training wheels children use when learning to ride a bike: essential in the beginning, but once they've achieved some proficiency, the training wheels just get in the way. Reread the man page The last time I used disklabel was on a 4.2-R system. That nasty little creature required you to calculate sector counts and offsets by hand and hope it didn't barf with errors about c not covering the entire disk or partition overlaps, despite your calculations indicating otherwise. It's nice to see it has evolved a bit since then. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Different sysinstall labelling behaviour when run post-install vs. during install?
Yesterday I was upgrading the disk in one of my workstations. Since I was planning on growing the filesystems for my FreeBSD install at the same time, I needed to manually create a new disklabel. I used sysinstall for this, but ran into a problem: sysinstall doesn't allocate the partition letters, with 'a' first, if you aren't doing an initial install. I ended up using the auto defaults option to create the 'a' partition, then remove everything and create them as I wanted. I'm glad to say that it worked out just fine, but this problem raises a couple of questions: Why does sysinstall behave differently when run post-install? How do I manually specify or change the partition letter to use? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: How to install a custom built world+kernel to a machine with no OS?
From: Chris Vance [mailto:[EMAIL PROTECTED] Darren Pilgrim wrote: From: Sergey Zaharchenko [mailto:[EMAIL PROTECTED] On Mon, Jul 19, 2004 at 04:21:09AM -0700, Darren Pilgrim probably wrote: What do I need to us to accomplish the above? man release. Is there a way to skip the build portions of `make release` and just create the distributions and ISOs with what's already built? I typically use: setenv CVSROOT foo time make -DNOPORTS -DNODOC -DMAKE_ISOS release CHROOTDIR=/usr/release EXTSRCDIR=/usr/src /tmp/release.out This presumes you've already done a buildworld in /usr/src; it will also use that version of the source code, rather than pulling everything down from the CVS server. Because of the way the release process uses a chroot environment, it's still going to have to re-build everything. So, it's quite a time-intensive operation no matter how many shortcuts you take. I ended up creating FreeSBIE auto-install CDs with some scripts to do the disk layout, installworld/kernel and copy over the configuration. It worked very well. I just sat there, jockeyed discs and watched the scroll. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
How to install a custom built world+kernel to a machine with no OS?
I have a number of machines onto which I want to install a custom-built
world, kernel and set of packages, all of which have been built on another
machine. The target machines presently have no OS on them and can only be
booted by CD.
Installing a minimal FreeBSD and then mounting /usr/src and /usr/obj via NFS
isn't attractive because of the extra work and uncertainty involved with
removing and changing files left over from the initial install. I tried
using discs 1 and 2 from a FreeBSDMall CD set to run install{world,kernel}
directly, but they don't seem to have the necessary bits. The ports system
has the package make target. I couldn't, however, find a make target to
turn a built world and kernel into the distribution chunks sysinstall can
use to do a binary install.
The solutions I've come up with so far are:
- Create a custom CD that contains the necessary bits to partition, label,
newfs, and run installworld/installkernel.
- Do my own release engineering.
- Some third option I haven't thought of yet.
What do I need to us to accomplish the above?
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: How to install a custom built world+kernel to a machine with no OS?
[Private email redirected back to list] From: 3BSD [mailto:[EMAIL PROTECTED] On Mon, 19 Jul 2004 04:21:09 -0700, Darren Pilgrim [EMAIL PROTECTED] wrote: I have a number of machines onto which I want to install a custom-built world, kernel and set of packages, all of which have been built on another machine. The target machines presently have no OS on them and can only be booted by CD. How similar are those machines hardware wise? Because if they are really similar, you could use an imaging utility such as norton ghost to replicate one install on any number of machines, provided their hardware is similar, but if you use a Generic kernel, I'd imagine you don't even need the machines to be that similar at all. That doesn't solve the problem, though. I would still need to install the OS onto at least one machine. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: How to install a custom built world+kernel to a machine with no OS?
From: Andrew L. Gould [mailto:[EMAIL PROTECTED] On Monday 19 July 2004 03:32 pm, Darren Pilgrim wrote: [Private email redirected back to list] From: 3BSD [mailto:[EMAIL PROTECTED] On Mon, 19 Jul 2004 04:21:09 -0700, Darren Pilgrim [EMAIL PROTECTED] wrote: I have a number of machines onto which I want to install a custom-built world, kernel and set of packages, all of which have been built on another machine. The target machines presently have no OS on them and can only be booted by CD. How similar are those machines hardware wise? Because if they are really similar, you could use an imaging utility such as norton ghost to replicate one install on any number of machines, provided their hardware is similar, but if you use a Generic kernel, I'd imagine you don't even need the machines to be that similar at all. That doesn't solve the problem, though. I would still need to install the OS onto at least one machine. So are you saying that the custom world and kernel are not the system in use on the computer on which they exist? Is the build machine part of the build set? No. Assuming you've already edited the configuration files (/etc/rc.conf, /etc/fstab, etc): 1. Could you take the harddrives out of the destination computers, mount them on the source computer, and copy the system onto the hard drives? The target machines will be using UFS2. The build machine runs 4.9. 2. Couldn't you use a live cdrom distribution (Freesbie, Knoppix, BBC Linux, etc) to boot the machine, setup networking, mount the hard drive and transfer (ftp, nfs, rsync, etc) the custom system to the new computer? No version of Linux is going to have the programs or environment to run installworld. I spoke with one of the FreeSBIE people (Drizzt) and s/he said a FreeSBIE CD would have the necessary tools for installing via makeworld so that looks like a good option. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: How to install a custom built world+kernel to a machine with no OS?
From: Sergey Zaharchenko [mailto:[EMAIL PROTECTED] On Mon, Jul 19, 2004 at 04:21:09AM -0700, Darren Pilgrim probably wrote: What do I need to us to accomplish the above? man release. Is there a way to skip the build portions of `make release` and just create the distributions and ISOs with what's already built? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: FreeBSD 5.1 - WinXP Networking Problem UPDATE
From: freebsder [mailto:[EMAIL PROTECTED] Hi Everyone ... thanks for your help thus far. I've made some changes below. [I have Not made all the changes that you've kindly suggested but enough that I am able to ping back and forth ... if I have ignored your suggestion and you still see a gapping error, please feel free to reinterate, I won't hold it again you!] ... I think that the natd_flags=redirect_port tcp 192.168.0.3:80 80 should be: natd_flags=redirect_port tcp 192.168.0.1:80 80 natd_flags=redirect_port tcp 192.168.1.1:80 80 ifconfig vr0= media 10baseT/UTP up ifconfig_ed0=inet 192.168.0.1 netmask 255.255.0.0 ifconfig_vr0=inet 192.168.0.1/24 media 10baseT/UTP up ifconfig_ed0=inet 192.168.1.1/24 You will then need to change the IP addresses of the two WinXP machines to use addresses starting with 192.168.1 (excluding .0, .1 and .255), a netmask of 255.255.255.0 and a gateway of 192.168.1.1. Thanks to Matthew Seaman for bringing to my attention that ifconfig now supports CIDR notation. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Freebsd 5.1 - Win XP Networking problems
[Note: cross-post removed. -questions is the appropriate place for this.]
As your rc.conf contents show, you have the same block of addresses assigned
to both interfaces. This is a broken configuration. You need to renumber
one of these networks into non-overlapping space. I would change the
DSL-modem side to use 192.168.0.0/24, then use 192.168.1.0/24 on the LAN.
From: freebsder [EMAIL PROTECTED]
I have a Freebsd 5.1 box connected to the internet.
It works. But I am now trying to network two other
Win XP machines as per the following network
hierarchy:
Setup
ISP- DSL Modem - FreeBSD box :
1) vr0 192.168.0.1 [Gateway machine address]
2) ed0 192.168.0.3 [Internal Network address]
connects to:-
4- port HUB -
1)WinXP machine #1 192.168.0.2
2)Freebsd Box 192.168.0.3
3)WinXP machine #2 192.168.0.4
Problem:
I cannot communicate to the Internet from WinXP #2
(Have not tried to config WinXP #1 yet).
Browser Config
IE Brower Settings for WinXP #2 {ToolsInternet
OptionsConnections)
-I set the browser so that it never dials a connection
because it is suppose to be networked right?
- in the LAN Settings option, I set the Proxyserver
option with the address of the gateway of 192.168.0.1
with Port 80
Dialouge
From Freebsd Machine
# ping 192.168.0.4
PING 192.168.0.4 (192.168.0.4): 56 data bytes
ping: sendto: Host is down
ping: sendto: Host is down
-at one point I was able to ping the freebsd machine
from WinXP #2 but then for some reason, I made a
change and cannot ping anymore...
RC.CONF
My rc.conf file looks like this:
font8x14=NO
font8x16=swiss-8x16
font8x8=swiss-8x8
inetd_enable=YES
linux_enable=YES
moused_enable=YES
moused_port=/dev/psm0
moused_type=auto
nfs_client_enable=YES
nfs_server_enable=YES
rpcbind_enable=YES
saver=rain
scrnmap=NO
usbd_enable=YES
ifconfig_vr0=DHCP
ifconfig_ed0=DHCP
##initialise NIC
network_interfaces=vr0 ed0 lo0 tun0
ifconfig tun0
ifconfig vr0= media 10baseT/UTP up
ifconfig_ed0=inet 192.168.0.3 netmask 255.255.0.0
ifconfig_vr0=inet 192.168.0.1 netmask 255.255.0.0
hostname=myserver
##User ppp configuration
ppp_enable=YES
ppp_mode=ddial
ppp_nat=NO
ppp_profile=bellnet
#ppp_user=root
## Firewall
gateway_enable=YES
firewall_enable=YES
firewall_type=OPEN
#firewall_quiet=NO
firewall_script=/etc/rc/firewall
natd_enable=YES
natd_interface=vr0
natd_flags=redirect_port tcp 192.168.0.3:80 80
rpc_statd_enable=YES
tcp_extensions=YES
## Mail
sendmail_enable=YES
HELP!
Thanks in advance.
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Mail(1) breaks when contrib/sendmail is replaced with postfix?
I use postfix for my mail servers, rather than sendmail. Postfix installs the sendmail-replacement programs under /usr/local. The programs from contrib/sendmail are under /usr, so you end up with two copies of these programs when postfix is installed. To avoid prevent security issues and other problems I remove the old contrib/sendmail binaries and define NO_SENDMAIL in /etc/make.conf. For better integration, postfix can be started through the mailer configuration in the base system by adding certain lines to /etc/rc.conf and /etc/mail/mailer.conf. This points sendmail, mailq, etc. to /usr/local/sbin/sendmail. This works very well, with one exception, the periodic scripts don't seem to run. I've been able to narrow this down to this error from /usr/bin/mail: /usr/bin/mail tries to run /usr/sbin/sendmail directly. This probably isn't a good idea, since IIRC sendmail can now be package-ized and removed from the base system as well as excluded from buildworld. This is what seems to be breaking the periodic script, since I get this error when running periodic: # periodic daily mail: /usr/sbin/sendmail: No such file or directory # mail someone Subject: foo foo . EOT # mail: /usr/sbin/sendmail: No such file or directory Why would /usr/bin/mail be doing this? I couldn't find any knobs for periodic.conf or mail.rc that would affect this. I've had this problem with other programs and shells where if you run a program without giving a path (forcing a path search), then move the program to another point in the path (say from /usr/bin to /usr/local/bin), you'll get similar file not found errors. Rebooting fixes the problem, but rebooting hasn't fixed this one. Also, I can fix the problem by making a symlink from the postfix sendmail binary to /usr/sbin/sendmail, but that's not really a workable solution. How do I fix this? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Freebsd 5.1 - Win XP Networking problems
From: freebsder [mailto:[EMAIL PROTECTED] Hi Dave, I've changed the rc.conf as per you suggestion see below. Do I also need to change the natd_flags=redirect_port tcp 192.168.0.3:80 80 to natd_flags=redirect_port tcp 192.168.1.0:80 80 ?? See below. You need to correctly renumber your LAN first. The redirect_port option is going to be the IP address you assign to the internal interface of myserver. Are there any other similar modifications that need to be made somewhere? That depends on what you changed and what you have running on myserver and how you configured them. Any configuration setting that has an IP address of one of your machines will need to be changed. ifconfig_ed0=inet 192.168.1.0/24 netmask 255.255.0.0 ifconfig_vr0=inet 192.168.0.0/24 netmask 255.255.0.0 192.168.1.0/24 and 192.168.0.0/24 are blocks of addresses in CIDR notation, not the actual addresses to be fed to ifconfig. You need to pick addresses within the netblock to use for myserver and all the other machines on your network. Since myserver can reach the internet just fine, you should keep the IP address for vr0 the same, just lengthen the netmask to allow the use of 192.168.1.0/24 on the LAN. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Mail(1) breaks when contrib/sendmail is replaced with postfix?
From: Matthew Seaman [mailto:[EMAIL PROTECTED] On Tue, Jul 13, 2004 at 12:57:21AM -0700, Darren Pilgrim wrote: /usr/bin/mail tries to run /usr/sbin/sendmail directly. This probably isn't a good idea, since IIRC sendmail can now be package-ized and removed from the base system as well as excluded from buildworld. This is what seems to be breaking the periodic script, since I get this error when running periodic: Piping a message into /usr/sbin/sendmail is the standard Unix API for sending e-mail. You shouldn't remove it just because of the name 'sendmail'. Despite the name, that is not the sendmail(8) binary on FreeBSD. It's a link to a wrapper program that substitutes which ever MTA (postfix, exim, qmail, sendmail, etc.) you wish to use. It's all explained in the mailwrapper(8) man page. I've read the man page and says sendmail, purgestat, etc. are supposed to be symlinks to mailwrapper. In this case, they were discrete files differing in size and content from mailwrapper and those installed by postfix. I reinstalled usr.bin and usr.sbin from a built world and everything works ok now. Could this have been the work of some trojan? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Can I access a USB device that has no driver attached?
From: Bernd Walter [mailto:[EMAIL PROTECTED] On Thu, Jun 03, 2004 at 06:22:54PM -0700, Darren Pilgrim wrote: I have a UPS with a USB interface. There isn't a specific UPS driver in FreeBSD beyond uhid. If I were to connect the UPS's interface port without having a driver attached, is it still possible to talk to the device in some way? I ask because I'm helping get nut (sysutils/nut) working for USB UPSes under FreeBSD. Maybe your question is outdated, but since noone answered yet. - You can enhance our uhid driver - You can write your own driver - You can do raw control transfers via /dev/usb* - You can also do raw control and pipe trandfers if ugen takes the device - You can use libusb for portable (Linux, ...) device access via ugen Wow, I'd given up on getting an answer for this one. Here's the situation right now: With some custom patches provided some helpful folks, I've been able to get nut's newhidups driver to compile, run, and talk to the device without crashing. As newhidups uses libusb, it is necessary to attach the USB with ugen. In debugging mode, the newhidups driver dumps the entire usage table data for the device. Unfortunately, APC irrationally believes the format of this data to be their intellectual property and thus refuses to make it available, even under NDA. APC will, however, send, at no cost to the customer, serial cables to be used with any UPS that comes with a USB cable. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
uuencode(1) doesn't work?
I'm trying to uuencode some data, but uuencode doesn't seem to work properly. I'm using FreeBSD 4.8-RELEASE-p22. Here are some examples: $ date | uuencode usage: uuencode [-m] [-o outfile] [infile] remotefile b64encode [-o outfile] [infile] remotefile $ cat /etc/rc.conf | uuencode usage: uuencode [-m] [-o outfile] [infile] remotefile b64encode [-o outfile] [infile] remotefile $ Specifying a filename, rather than using a pipe, doesn't seem to work either: $ uuencode file begin 644 file After printing the begin line it idles. A debug copy of uuencode run with gdb shows the program stopping on the read() call trying to get data (fread() called from the while loop in encode()). Anyone know why? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Can I access a USB device that has no driver attached?
I have a UPS with a USB interface. There isn't a specific UPS driver in FreeBSD beyond uhid. If I were to connect the UPS's interface port without having a driver attached, is it still possible to talk to the device in some way? I ask because I'm helping get nut (sysutils/nut) working for USB UPSes under FreeBSD. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Can not find libintl.so.4?
I have a desktop running 5.1+KDE. Building a port (finance/gnucash) failed when the install of a fresh gettext build failed due to there being an older version already present. To fix this, I cd into devel/gettext, run make deinstall make reinstall. But now I have programs and other libs refusing to load (PHP, cyrus, sylpheed, et al). I get errors stating that libintl.so.4 could not be found. A check of/usr/local/lib shows a libintl.so.5 with libintl.so symlinked to it, but no libintl.so.4. Symlinking libintl.so.4 to .5 seems to have at least made it possible for everything to load again. I'm wondering, though, why this happened? Is there is a more correct way to fix the problem? Is this risk taken when one upgrades a massively interdependent set of installed ports? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
How to spin down a disk?
I have a machine that multi-boots with separate ata disks for each OS. I'd like to have FreeBSD (5.1) spin down the other disk to reduce heat and noise, since it doesn't use it at all. I haven't been able to find anything in the apm, ata, or acpi documentation I could find. How do I accomplish this? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
How do I get /dev entries for removeable media to appear wheninserted?
I have a zip drive (afd0). At boot, the only afd0* device is afd0, which makes it rather difficult to mount MSDOS formatted disks. I can get the device entries to appear by issuing `mount /dev/afd0 /mnt`, but it's annoying to have to do this. Is there a way to cause this update automatically, or to force the existance of device nodes for slices and partitions? This is on 5.0-R. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Secure tunnel: SSH or SSL or IPsec?
I need to create a secure connection over the inter between my workstation at home and a server I have elsewhere. My workstation is running RELENG_5_0 and the server runs RELENG_4_8, both up to date. I need the secure connection to occasionally access swat and VNC remotely. You can assume all the ports I'll be accessing are local to the server. My workstation is behind a Linksys BEFSR router doing NAT with an IPsec passthrough. What would work best in this situation? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Support for STV0602-AA based USB vidcams?
I have a Logitech Quickcam Express USB with the STV0602-AA chip. I know it's a piece of crap, but it was a free piece of crap. Is there any support for it? The chip is the typical cheap component put into a lot of cheap electronics meant for Windows. I didn't see a specific USB camera module and searching the archives didn't prove useful. Googling yielded source modifications for Linux' mod_quickcam, so I know it can be done. --- dmp To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: Why is there no JFS?
David Schultz wrote: Thus spake Terry Lambert [EMAIL PROTECTED]: David Schultz wrote: The easy way to fix this is to insert a new dependency for the completion of the allocation. Basically, this would put in a stall barrier that would cause the outstanding I/O to drain before the new I/O was attempted. All other operations behind the one that caused the stall would b held off, which would avoid the starvation deadlock you describe. Most likely, all this would require some minor code to maintain a running tally of virtual vs. real free block count. It really isn't a big deal. You're saying you can fix the problem where allocations can sometimes fail on a busy 99% full filesystem, but on such a filesystem, you're just as likely to hit it when it's 100% full. Kirk's solution is simple and has the advantage of not requiring additional dependency tracking for the common case. No, actually it should work for 100% full, as well, as long as that 100% full is the real disk vs. the real disk, after all pending updates have been applied. In other words, if it would have worked with soft updates turned off, then it will work with soft updates turned on. My point was that a busy disk that is nearly 100% full will probably experience intermitted ``disk full'' errors anyway, so it suffices to simply deal with cases such as 'rm -rf foo immediately create lots more files', which softupdates does handle in -CURRENT. IMO, this is not the reason for them being off on /; the real reason is as I've stated: sysinstall expects the common case to be an initial install, not operations after the initial install, and so does not turn it on by default. The original reason was due to the possibility of installworld failing, due to the case described above not being handled particularly well in FreeBSD 4.X. Sysinstall is perfectly happy with creating a root FS with softupdates enabled. If someone wants to bother changing the default for what little difference it might make in installworld/installkernel times, I would support it. For what its worth, I think all that's needed is to change line 339 in usr.sbin/sysinstall/label.c: --- label.c Mon Dec 30 21:19:15 2002 +++ label.c.new Thu Feb 13 11:50:44 2003 @@ -336,7 +336,7 @@ strcpy(pi-newfs_data.newfs_ufs.user_options, ); pi-newfs_data.newfs_ufs.acls = FALSE; pi-newfs_data.newfs_ufs.multilabel = FALSE; -pi-newfs_data.newfs_ufs.softupdates = strcmp(mpoint, /); +pi-newfs_data.newfs_ufs.softupdates = TRUE; pi-newfs_data.newfs_ufs.ufs2 = FALSE; return pi; The patch is against the 5.0-R tagged version, but it should still apply to the current version. I think softupdates is still (viewed as) riskier than synchronous writes, at least for large numbers of writes (like installworld) to a filesystem of limited size, so someone is going to inevitably ask if FreeBSD should be loading the bullets as well. Personally, if it's a matter of choosing overall safety or a performance gain for something you really shouldn't be doing to a live machine anyway, I'll take the safe route and option the performance gain. P.S., thanks everyone for the discussion, it was enlightening. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: Why is there no JFS?
Matthew Emmerton wrote: Thus spake Daxbert [EMAIL PROTECTED]: The inspiration for this email was from a thread in -questions: Re: fsck takes very long after crash/reset Is anybody currently working on or does there exist a JFS for FreeBSD? Various people (including myself and Hiten Pandya) have done work to port the GPL'd JFS implementation, but there's one ugly problem -- the GPL. We can make JFS into a kernel module (avoiding the static-link policy of the GPL), but then it can only (legally) be used on non-root filesystems, as the code to read the root filesystem must be statically linked into the kernel. This in itself makes JFS support somewhat pointless. Not really. A properly laid-out filesystem hierarchy will result in no writes to / (except for installworld/kernel). That removes the problem that journalling addresses, and is probably why softupdates is disabled by default for /. For large, active filesystems, journalling would be a big improvement when you had to run a foreground fsck. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: Why is there no JFS?
Terry Lambert wrote: Darren Pilgrim wrote: Not really. A properly laid-out filesystem hierarchy will result in no writes to / (except for installworld/kernel). That removes the problem that journalling addresses, and is probably why softupdates is disabled by default for /. For large, active filesystems, journalling would be a big improvement when you had to run a foreground fsck. Soft updates are disable on / by default because of the chicken and egg problem of runing tunefs on /. If that's the problem, then why doesn't sysinstall enable it by default when partitioning for a new install? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: Which snd driver to use with Toshiba Satellite?
Laszlo Vagner wrote: Darren Pilgrim wrote: I have a Toshiba Satellite 1805-S207 notebook. According to the (limited) information I can find on this machine, I believe the audio hardware is integrated into the ALi CyberALADDiN-T M1535 chipset. The only thing I could find was the t4dwave driver, but that freezes the machine when kldloaded post-boot, and causes the kernel probe to hang if loaded via loader.conf. Is there a driver for this hardware? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message I have a similiar machine and just added device pcm to my kernel and it detects it fine. no need for drivers... it is based on the yamaha 754 chip. Which Satellite do you have, and what does the hardware probe as? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Why should I use `config;make depend;make...` instead of `make kernel`when building from a stock source tree? (ref. Handbook sec. 9.3)
There are two sets of commands you can use to build a kernel in FreeBSD: Procedure 1 is the old way: config, make depend, make, make install. Procedure 2 is the make kernel sequence from makeworld. Section 9.3 of the Handbook says I should use procedure 1 if I haven't updated my source tree. I can understand then need to use procedure 2 if I've updated my source tree, but why shouldn't I use it with an unmodified tree? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: Why should I use `config;make depend;make...` instead of `makekernel` when building from a stock source tree? (ref. Handbook sec. 9.3)
Lowell Gilbert wrote: Darren Pilgrim [EMAIL PROTECTED] writes: There are two sets of commands you can use to build a kernel in FreeBSD: Procedure 1 is the old way: config, make depend, make, make install. Procedure 2 is the make kernel sequence from makeworld. Section 9.3 of the Handbook says I should use procedure 1 if I haven't updated my source tree. I can understand then need to use procedure 2 if I've updated my source tree, but why shouldn't I use it with an unmodified tree? There's no reason you shouldn't. That section of the handbook explicitly says that you can use either procedure in that case. Hey you're right. In the bulleted list just before the two procedures are listed it says, #If you are building a new kernel without updating the source code...you can use either procedure. After the two procedures are listed, it says, If you have not upgraded your source tree in any way...you should use the config, make depend, make, make install sequence. That sounds contradictory to me. What do you think? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Why no /dev/one?
Why isn't there a /dev/one device to provide an infinite number of all-ones bytes? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: Why no /dev/one?
Giorgos Keramidas wrote: On 2003-01-30 00:25, Darren Pilgrim wrote: Why isn't there a /dev/one device to provide an infinite number of all-ones bytes? Because it's easy to get any sequence of equal bytes by using just /dev/zero and tr(1). Try this command and check the output of hd(1) :-) $ dd if=/dev/zero bs=512 count=1 | tr '\0' '\777' | hd What I was trying to get at was more a question of if there's some deep technical reason for the lack of a /dev/one beyond the triviality of flipping the bits in a pipe. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Filesystem tuning for lots of small files (a Maildir)?
I'm currently facing a problem of having used Netscape (now Mozilla) for years in Windows and now trying to find something I can regularly use in FreeBSD without losing Mozilla in Windows. I've mostly settled on IMAP (courier) with procmail filters, but that raises the issue of filesystem performance for directories with large numbers of files/subdirectories in them. I have more than 32,000 emails stored. How do I calculate/see the number of available inodes? The existing filesystem was newfs'd with the sysinstall defaults. Should I re-newfs it with different values? What would I want to set them at? I know I'd need to adjust things to make sure I have enough inodes for 40,000+ files, but what about the block and fragment size? Should I use smaller values like 8192/1024 or 4096/512 or is the default 16384/2048 best? Higher values would just increase slack space, right? What are the impacts of lower values? Some folders, like the one for the postfix-users list, can have 3000-4000 messages in them. For growth, we'll say 5000 messages. The IMAP layout with Courier means all the folders sit all on one level under ~/Maildir, which means I'd have 200 or so subdirectories in one place. I have the UFS_DIRHASH option enabled for the my MP3 collection, but that's as case of 300 subdirecories in one directory, not 5000 files. What else can I do to tune for this kind of (ab)use? P.S. I really would like to stick with Maildirs and Courier-IMAP for this. I know CIMAP well and it has proven very fast and stable for what I do with it. However, if these demands are just too much to expect from an IMAP-accessed Maildir, Courier, or FreeBSD, what are my alternatives? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: Filesystem tuning for lots of small files (a Maildir)?
Craig Reyenga wrote: - Original Message - From: Darren Pilgrim To: Sent: Friday, January 24, 2003 07:13 Subject: Filesystem tuning for lots of small files (a Maildir)? I'm currently facing a problem of having used Netscape (now Mozilla) for years in Windows and now trying to find something I can regularly use in FreeBSD without losing Mozilla in Windows. I've mostly settled on IMAP (courier) with procmail filters, but that raises the issue of filesystem performance for directories with large numbers of files/subdirectories in them. I have more than 32,000 emails stored. How do I calculate/see the number of available inodes?^^^ ^ df -i /filesystem-in-question The filesystem has about 1.4m free inodes, so I guess that's not really going to be a problem. What's the max. number of inodes I can have? 2^32? The existing filesystem was newfs'd with the sysinstall defaults. Should I re-newfs it with different values? What would I want to set them at? I know I'd need to adjust things to make sure I have enough inodes for 40,000+ files, but what about the block and fragment size? Should I use smaller values like 8192/1024 or 4096/512 or is the default 16384/2048 best? Higher values would just increase slack space, right? What are the impacts of lower values? The number of inodes varies with the filesystem size and bytes per inode. So if you're talking about a huge filesystem, you're probably all set as it is. However, I needed a /usr that has many inodes, so I doubled the default by doing this: newfs -b 16384 -f 2048 -i 4096 /usr -i 4096 is half as many bytes per inode compared to the default 8192, therefore, I have 2X as many inodes. See newfs(8) for more info. tuning(7) also. I know that one inode is used for every file (for arguement's sake we'll say everything that uses an inode is a file) in a filesystem. So the number of inodes is the number of files you can have. But what happens when the file is larger than the inode size? It still uses one inode, but the filesystem has to allocate space in blocks. I'm trying to determine the size of that block, if it's adjustable, and if I even should be adjusting it. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
How to safely unmount a filesystem mounted async?
For performance reasons, I have a filesystem mounted with the async option. The FS isn't used for anything I really value, just assorted object files and other temp data. My question, though, is, how do I safely unmount an active async filesystem? Does the unmount process automatically force a complete flush to disk? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Maildirs and their filesystem overhead/impact
I'm currently facing a problem of having used Netscape (now Mozilla) for years in Windows and now trying to find something I can regularly use in FreeBSD without losing Mozilla in Windows. I was thinking of something involving IMAP and NFS. I already have Courier-IMAP running, so I can use IMAP with Mozilla in Windows, and use either IMAP or NFS-mount my Maildir in FreeBSD. That part is almost a no-brainer. Moving my Mozilla mail store poses a number of technical questions and problems. I have more than 32,000 emails saved up in 200 folders. Some folders, like the one for the postfix-users list, can have 3000-4000 messages in them. For growth, we'll say 5000 messages. How well would FreeBSD, Courier-IMAP, and postfix be able to handle a directory with 5000 files in it? How about a directory with 200 subdirectories in it? I know that the parameters given to newfs can greatly impact the suitablility of a filesystem for a given task. How do I get that information from an existing filesystem? Dumpfs? Dumpfs spews HUGE amounts of information, but I don't know enough to make sense of the output. I really would like to stick with Maildirs and Courier-IMAP for this. I know CIMAP well and is very fast and stable. However, if these demands are just too much to expect, then can someone please tell me where I can find Cyrus for Dummies? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Support for USB cable modems?
I have an RCA cable modem provided to me by ATT Broadband and out of the same curiosity clinically proven lethal to the average domestic feline, I was wondering if I can use the USB interface with FreeBSD. ugen picks up the device as, Thomson Consumer Electronics Thomson RCM245 Cable Modem, rev 1.00/26.00. The device doesn't seem to be attached by if_aue, if_cue, or if_kue. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: Support for USB cable modems?
Mike Meyer wrote: In [EMAIL PROTECTED], Darren Pilgrim [EMAIL PROTECTED] typed: I have an RCA cable modem provided to me by ATT Broadband and out of the same curiosity clinically proven lethal to the average domestic feline, I was wondering if I can use the USB interface with FreeBSD. ugen picks up the device as, Thomson Consumer Electronics Thomson RCM245 Cable Modem, rev 1.00/26.00. The device doesn't seem to be attached by if_aue, if_cue, or if_kue. If it only shows up as ugen, then the answer is no. Not without more software than ships with the kernel, anyway. It might be possible that all is needed is some tweak or bit of information added to one of the existing USB ethernet drivers. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
USB hub detach causing panic in 4.7p3?!
I have a USB hub that's built into my Viewsonic PT775 monitor. The hub works fine, and has worked fine for years in Windows, Linux, and FreeBSD. The USB hub is only attached to the rest of the USB bus when the monitor is on, so turning the monitor on or off produces the expected uhub attach/detach noticed. On Friday (Jan 10), I cvsupped and recompiled to upgrade from 4.7p2 to 4.7p3. Prior to this upgrade, I know that the attach/dettach process worked without problems. As of about 5pm today, turning the monitor off causes this result (please forgive typos, this is a transcription, though I'm over 99.9% sure the numbers are correct): uhub1: at uhub0 port 2 (addr 2) disconnected uhub1: detached Fatal trap 12: page fault while in kernel mode fault virtual addres= 0x3 fault code = supervisor read, page not present instruction pointer = 0x8:0xc031fe04 stack pointer = 0x10:0xc0250fb0 frame pointer = 0x10:0xc0250fc4 code segment= base 0x0, limit 0xf, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags= interrupt enabled, resume, IOPL = 0 current process = Idle interrupt mask = bio trap number = 12 panic: page fault syncing disks... Fatal trap 12: page fault while in kernel mode fault virtual address = 0x30 fault code = supervisor read, page not present instruction pointer = 0x8:0xc01c2498 stack pointer = 0x10:0xc0250d98 frame pointer = 0x10:0xc0250da0 code segment= base 0x0, limit 0xf, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags= interrupt enabled, resume, IOPL = 0 current process = Idle interrupt mask = bio trap number = 12 panic: page fault Uptime: 3m57s -- Press a key on the console to reboot, -- or switch off the system now. Nothing is plugged into the hub at the time. This seems to be a reliable panic in that I got the exact same panic when I switch the monitor off again later after having warm rebooted. I can't say beyond now vs. before Friday if the upgrade to 4.7p3 caused the problem or not. Nor can I say for sure that the problem existed between Friday and today, as I can't recall if I had turned off the monitor while running FreeBSD between now and then (I reboot between FreeBSD and Windows multiple times per day). I made sure to delete the contens of /usr/obj/usr and run make cleandir twice before compiling 4.7p3, and I've made no other changes to the system configuration beyond adding a couple of X programs (x11-fm/asfiles and x11-fm/endeavour). I'm going to try a few more things, like plugging and unplugging the hub with the monitor on, as well as plugging and unplugging devices from both the root hub and the hub in the monitor to see if it's more general, or if it's just something wrong with the uhub detach routine. Is this a known problem or does anyone have any suggestions for tracking down this problem? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: USB hub detach causing panic in 4.7p3?! (more information)
Darren Pilgrim wrote: I'm going to try a few more things, like plugging and unplugging the hub with the monitor on, as well as plugging and unplugging devices from both the root hub and the hub in the monitor to see if it's more general, or if it's just something wrong with the uhub detach routine. I tried various patterns of plugging and unplugging devices and the hub, as well as booting with and without the hub connected and with and without additional devices connected to the hub. The additional device was a Microsoft Wireless Intellimouse Explorer. I should note that I have /etc/usbd.conf modified to not run moused when a ums device is attached. In all cases, panics only occured when the hub was detached without anything attached to it. In all other cases the hub and the ums device attached to the hub it detached and reattached repeatedly without problems. While doing this testing, the fault information was different in two places: the stack and frame pointer values in the second (after the syncing disks message) fault information block were different, but consistent for all panics: stack pointer = 0x10:0xc0250dd0 frame pointer = 0x10:0xc0250dd8 The first fault's information block and the remaining information from the second block was the same as previously reported. I also looked through my logs and found that, until this afternoon, the ums device was always attached when I had turned the monitor off. My logs cover 4.7-R, 4.7p2, and 4.7p3, with over 20 reboots between 1am Friday morning and 5pm today, the time of the first panic. I suppose the question now is: why does uhub cause a page fault when an unused hub is detached? Unused meaning there are no devices hanging off it. I'd also like to take this moment to thank all the hackers who made FreeBSD so damned robust that I could deliberately panic the machine over and over and have all the filesystems come up fine every time. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Is ldd recursive?
When I run ldd on a given program, does ldd check the dependencies of the libaries as well? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: Is ldd recursive? (nevermind)
Darren Pilgrim wrote: When I run ldd on a given program, does ldd check the dependencies of the libaries as well? Nevermind, I found the answer myself (it does). I don't know why I missed it on the man page. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: FTP installation from the floppies through ADSL modem with PPPoEor PPTP protocol.
Andrew wrote: On Fri, 20 Dec 2002, Asker wrote: The modem can be configured to use PPPoE or PPTP protocol for making the connection with my Internet Servise Provider. Well if the modem does PPPoE itself (and preusmably NAT) then you need no speical support from the OS. From its poitn of view you are just conencted via ethernet. If you need the machine to do PPPoE, ppp supports PPPoE. For this to work, though, you need netgraph, which isn't in GENERIC. You will need to make a custom kernel and build your own set of custom floppies. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: ATA errors
Vladislav V. Zhuk wrote: On Fri, Dec 06, 2002 at 09:34:09AM -0800, Darren Pilgrim wrote: This is almost always the sign of a bad cable, but it can also be the logic board on the drive dying (though much rarer). Check your cables. Better yet, go to your local hardware store and buy a new ATA/100-spec cable, flat, not rounded, preferably with pull-loops. I don't think like you. I check my hardware and I consider that problem in new ATA driver. Under FreeBSD 4.1.1 my hardware work excellent. After 4.5 release I get more troubles with IDE devices. There was some pretty major changes in 4.5. It really should have been a .0 release. It was the same situation when 4.0 came out and they ditched the old ata code. After reboot my system work excellent 2-5 days, than I get read timeout problem with my CDROM and all system hang. There are a LOT of CDROM drives that don't work properly. I had a Toshiba XM-6402B drive that didn't work half the time, and a cheap 50X drive from Acer that works perfectly. About the only way to get a guarantee of having a good-quality CD/DVD drive is to buy SCSI. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: ATA errors
Thomas T. Veldhouse wrote: Can anybody explain what has happened here? My machine seems to be functioning normally. ad0: READ command timeout tag=0 serv=0 - resetting ata0: resetting devices .. ata0-slave: ATA identify retries exceeded done This is almost always the sign of a bad cable, but it can also be the logic board on the drive dying (though much rarer). Check your cables. Better yet, go to your local hardware store and buy a new ATA/100-spec cable, flat, not rounded, preferably with pull-loops. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: ATA errors
Tenebrae wrote: BTW, the dead drive is an IBM Deskstar 75GXP (DTLA-307060). I miss it. I wish there was some way to recover it. 30GB of data gone. Maybe I'll try putting it in the freezer and drop it into a different machine and see if I can mount it... Funny, I had *exactly* the same thing happen to me with an IBM DTLA-307030, also 30GB. Awesome drive, screaming transfers, even with seeking involved. Then one day it started throwing the same errors you mention. Later that day it died. Called up IBM, replaced it for the cost of me shipping the dead one back. What hurts is that the 25GB of data on it is perfectly fine. I just don't have any means of getting the data off without paying a recovery company, unless someone knows I trick I don't. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
A program to list off packet target/source addrs as they pass?
I'm in need of a program that can show me just a list of packets as they pass through a given interface. The only information I need are the transport protocol (TCP, UDP, etc) and the source and destination addresses and ports. I don't want to know any other information for various privacy/legal reasons, this is for debugging and tuning routers and firewalls. The idea is to run this on an interface to show what's actually passing through. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Secure tunneling of remote-access Windows sessions?
I want to setup VNC on some Windows machines so I can access them over the internet, but I need to secure the connection in a way that will work with NAT'ing firewalls on both ends of the connection. How can I do this? I was thinking of setting up a tunnel between the two firewalls. On the local end, the tunnel starts at a given port on the firewall, which is connected to a port on the remote firewall that forwards to the VNC port on the remote machine. How would I go about doing this? Is there a better option? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: Secure tunneling of remote-access Windows sessions?
Doug Poland wrote: Darren Pilgrim said: I want to setup VNC on some Windows machines so I can access them over the internet, but I need to secure the connection in a way that will work with NAT'ing firewalls on both ends of the connection. How can I do this? I was thinking of setting up a tunnel between the two firewalls. On the local end, the tunnel starts at a given port on the firewall, which is connected to a port on the remote firewall that forwards to the VNC port on the remote machine. How would I go about doing this? Is there a better option? I recommend you use the TightVNC form of VNC. Read the info on this link: http://www.uk.research.att.com/vnc/sshvnc.html then read the ssd man page paying close attention to the -L switch. If you have particular problems after this leg work, then ask again. Okay, I see how I can use ssh/sshd running on the FreeBSD gateways on each end of the connection to make the remote VNC port accessible via a port on the local gateway. However, their setup requires that the remote machine have a routable IP address, doesn't it? Modifying the model on the page you sent me: local machine (me) - gateway1 10.2.3.4/24`ssh -g -L 5900:10.1.2.3:5900 gateway2` runs vncviewer| internet | gateway2 - remote machine running sshd 10.1.2.3/24 running vnc server on port 5900 Since the IP address I'm forwarding is non-routable, what happens? What happens to the source IP address, which is also non-routable and, to gateway2, non-local? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: Secure tunneling of remote-access Windows sessions?
Murat Bicer wrote: If remote address is not routable you will not be able to access it anyways. So you have to either open port 22 on the firewall of the remote machine which will be natted to the internal ip:port or you have to open port 5900 ( which is not secure). Either way you have to punch a hole on the firewall if you need to access non-routable addresses. Maybe I'm not understanding what you're trying to explain, or maybe I'm not explaining myself well enough. I know this is possible when public IPs are used. What I'm trying to determine (before I spend the time and money to reconfigure gateway2) is if this is possible when the VNC client and server machines aren't directly accessible from the public internet because they're behind NAT'ing gateways. The SSH tunnel gets me through the firewall via the ssh port on gateway2. Local only sees and uses the faked VNC port on gateway1. Assuming local can reach the faked port on gateway1 and gateway2 can reach the actual port on remote, do the IP addresses used even matter? Darren Pilgrim wrote: Doug Poland wrote: Darren Pilgrim said: I want to setup VNC on some Windows machines so I can access them over the internet, but I need to secure the connection in a way that will work with NAT'ing firewalls on both ends of the connection. How can I do this? I was thinking of setting up a tunnel between the two firewalls. On the local end, the tunnel starts at a given port on the firewall, which is connected to a port on the remote firewall that forwards to the VNC port on the remote machine. How would I go about doing this? Is there a better option? I recommend you use the TightVNC form of VNC. Read the info on this link: http://www.uk.research.att.com/vnc/sshvnc.html then read the ssd man page paying close attention to the -L switch. If you have particular problems after this leg work, then ask again. Okay, I see how I can use ssh/sshd running on the FreeBSD gateways on each end of the connection to make the remote VNC port accessible via a port on the local gateway. However, their setup requires that the remote machine have a routable IP address, doesn't it? Modifying the model on the page you sent me: local machine (me) - gateway1 10.2.3.4/24`ssh -g -L 5900:10.1.2.3:5900 gateway2` runs vncviewer| internet | gateway2 - remote machine running sshd 10.1.2.3/24 running vnc server on port 5900 Since the IP address I'm forwarding is non-routable, what happens? What happens to the source IP address, which is also non-routable and, to gateway2, non-local? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
