how to add flags to ifconfig at boot
Hi, How can I add flags to ifconfig at boot time, i.e. I want it to start with 'ifconfig em0 -tso' ? Thanks, ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: The fan is always on, even when the desktop is rather cool
Thanks David, I checked this page. now if I do: $ sysctl hw.acpi hw.acpi.supported_sleep_state: S1 S3 S4 S5 hw.acpi.power_button_state: S5 hw.acpi.sleep_button_state: S1 hw.acpi.lid_switch_state: NONE hw.acpi.standby_state: S1 hw.acpi.suspend_state: S3 hw.acpi.sleep_delay: 1 hw.acpi.s4bios: 0 hw.acpi.verbose: 0 hw.acpi.reset_video: 1 hw.acpi.cpu.cx_supported: C1/0 hw.acpi.cpu.cx_lowest: C1 hw.acpi.cpu.cx_usage: 100.00% and if i do #sudo acpiconf -s 5 OR #sudo acpiconf -s 1 nothing happens. Can I reinstall acpi, if yes, how can i do it? thanks. On 9/16/06, David J Brooks [EMAIL PROTECTED] wrote: On Friday 15 September 2006 18:10, Wei Hu wrote: I have 3 systems in my desktop: 1) When FreeBSD runs, my desktop fans are always running, and this make annoy noisy. 2) However when Debian runs, the fan eventually stops unless I am performing a load intensive task. 3) In Windows, the fan is almost always off. I tried to use acpi and apm, but they are for laptop.(?) In Freebsd, how can I control the cooling fans or how can the system turns the fans off when the load is not heavy. Any help would be greatly appreciated. It sounds like a broken ACPI code to me. Check the handbook chapter on debugging ACPI: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/acpi-debug.html David ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
FS size
Hi all I've read http://www.freebsd.org/projects/bigdisk/index.html and I want know actually on i386 arch is the limit of a fs is already 2 Tb What's the situation on amd64/EMT64, can we have big fs ? something like 10 or more TB ? Regards. -- Albert SHIH Universite de Paris 7 (Denis DIDEROT) U.F.R. de Mathematiques. 7 ième étage, plateau D, bureau 10 Heure local/Local time: Sat Sep 16 10:30:05 CEST 2006 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
PAY offered - sshd won't allow client from same domain
I will PAY someone who can either answer this question or who wants to log into my server and help me figure it out. I can pay an hourly rate, make a donation to your favorite project...whatever. This problem is killing my productivity I have a FreeBSD 6.1-p6 server running as server1.domain.com. sshd is allowing connections from any client except those which share the domain.com name..I can't be certain this is the problem, but after a month of debugging, its the only common factor I can find. My ssh client on server2.domain.com (also FreeBSD 6.1) returns with Read from socket failed: Connection reset by peer as output to my ssh client. On OS X the error message is Write failed: Broken pipe. ...So mac.domain.com and server2.domain.com which are on different networks from server1 (and from each other) are not allowed...I don't get any useful error messages. Even setting sshd_config LogLevel to DEBUG3 doesn't provide anything meaningful (to me) in auth.log or debug.log for server2.domain.com, I even have its ip as an A record in DNS and server1 can see this. mac.domain.com is not so lucky as it sits behind a DHCP NAT'ed structure. But this should hardly be a problem...PuTTY on Windows XP with no domain setting and behind a NAT'd DHCP structure CAN connect... Please allow me to offer some incentive this time around as this is my third post on this problem to this maillist. I have not received a single reply. Please get in touch. thanks ke han ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FS size
On Sep 16, 2006, at 5:33 PM, Albert Shih wrote: Hi all I've read http://www.freebsd.org/projects/bigdisk/index.html and I want know actually on i386 arch is the limit of a fs is already 2 Tb What's the situation on amd64/EMT64, can we have big fs ? something like 10 or more TB ? Regards. -- Albert SHIH Universite de Paris 7 (Denis DIDEROT) U.F.R. de Mathematiques. 7 ième étage, plateau D, bureau 10 Heure local/Local time: Sat Sep 16 10:30:05 CEST 2006 Think 4TB for 64-bit, if the limitations are truly integer based. 32 bit - 64 bit usually just increases the precision usable by 2 unless someone builds in increased number support. -Garrett___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: PAY offered - sshd won't allow client from same domain
On Sep 16, 2006, at 5:46 PM, ke han wrote: I will PAY someone who can either answer this question or who wants to log into my server and help me figure it out. I can pay an hourly rate, make a donation to your favorite project...whatever. This problem is killing my productivity I have a FreeBSD 6.1-p6 server running as server1.domain.com. sshd is allowing connections from any client except those which share the domain.com name..I can't be certain this is the problem, but after a month of debugging, its the only common factor I can find. My ssh client on server2.domain.com (also FreeBSD 6.1) returns with Read from socket failed: Connection reset by peer as output to my ssh client. On OS X the error message is Write failed: Broken pipe. ...So mac.domain.com and server2.domain.com which are on different networks from server1 (and from each other) are not allowed...I don't get any useful error messages. Even setting sshd_config LogLevel to DEBUG3 doesn't provide anything meaningful (to me) in auth.log or debug.log for server2.domain.com, I even have its ip as an A record in DNS and server1 can see this. mac.domain.com is not so lucky as it sits behind a DHCP NAT'ed structure. But this should hardly be a problem...PuTTY on Windows XP with no domain setting and behind a NAT'd DHCP structure CAN connect... Please allow me to offer some incentive this time around as this is my third post on this problem to this maillist. I have not received a single reply. Please get in touch. thanks ke han Do you have kerberos compiled and in use for authentication on the FreeBSD server and are you using it on the OSX client? ssh -vv server1.domain.com says? -Garrett ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: /usr/bin/ld: cannot find -lstdc++_p
No such file in /usr/lib. Came across libstdc++.a and libstdc++.so On 9/11/06, Lowell Gilbert [EMAIL PROTECTED] wrote: Viswas Nair [EMAIL PROTECTED] writes: I get the message /usr/bin/ld: cannot find -lstdc++_p while building the xfe X11 file manager. A google did not give any ideas. Need help. Well, start with whether libstdc++_p.a actually exists in /usr/lib. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Processors
On 9/15/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: To whom it may concern I have a computer with a dual-core processor. Will FreeBSD operate on this machine? Yes, of course. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: PAY offered - sshd won't allow client from same domain
On Sep 16, 2006, at 4:50 PM, Garrett Cooper wrote: ssh -vv server1.domain.com form OS X: (real domain name edited to domain.com) ssh -vv server1.domain.com OpenSSH_4.2p1, OpenSSL 0.9.7i 14 Oct 2005 debug1: Reading configuration data /etc/ssh_config debug2: ssh_connect: needpriv 0 debug1: Connecting to server1.domain.com [209.216.230.199] port 22. debug1: Connection established. debug1: identity file /Users/jhancock/.ssh/identity type -1 debug1: identity file /Users/jhancock/.ssh/id_rsa type -1 debug2: key_type_from_name: unknown key type '-BEGIN' debug2: key_type_from_name: unknown key type 'Proc-Type:' debug2: key_type_from_name: unknown key type 'DEK-Info:' debug2: key_type_from_name: unknown key type '-END' debug1: identity file /Users/jhancock/.ssh/id_dsa type 2 debug1: Remote protocol version 2.0, remote software version OpenSSH_4.2p1 FreeBSD-20050903 debug1: match: OpenSSH_4.2p1 FreeBSD-20050903 pat OpenSSH* debug1: Enabling compatibility mode for protocol 2.0 debug1: Local version string SSH-2.0-OpenSSH_4.2 debug2: fd 3 setting O_NONBLOCK debug1: Miscellaneous failure No credentials cache found debug1: Miscellaneous failure No credentials cache found debug1: SSH2_MSG_KEXINIT sent debug1: SSH2_MSG_KEXINIT received debug2: kex_parse_kexinit: diffie-hellman-group-exchange-sha1,diffie- hellman-group14-sha1,diffie-hellman-group1-sha1 debug2: kex_parse_kexinit: ssh-rsa,ssh-dss debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish-cbc,cast128- cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256-cbc,rijndael- [EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish-cbc,cast128- cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256-cbc,rijndael- [EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: none,[EMAIL PROTECTED],zlib debug2: kex_parse_kexinit: none,[EMAIL PROTECTED],zlib debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: first_kex_follows 0 debug2: kex_parse_kexinit: reserved 0 debug2: kex_parse_kexinit: diffie-hellman-group-exchange-sha1,diffie- hellman-group14-sha1,diffie-hellman-group1-sha1 debug2: kex_parse_kexinit: ssh-dss debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish-cbc,cast128- cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256-cbc,rijndael- [EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish-cbc,cast128- cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256-cbc,rijndael- [EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: none,[EMAIL PROTECTED] debug2: kex_parse_kexinit: none,[EMAIL PROTECTED] debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: first_kex_follows 0 debug2: kex_parse_kexinit: reserved 0 debug2: mac_init: found hmac-md5 debug1: kex: server-client aes128-cbc hmac-md5 none debug2: mac_init: found hmac-md5 debug1: kex: client-server aes128-cbc hmac-md5 none debug1: SSH2_MSG_KEX_DH_GEX_REQUEST(102410248192) sent debug1: expecting SSH2_MSG_KEX_DH_GEX_GROUP debug2: dh_gen_key: priv key bits set: 132/256 debug2: bits set: 523/1024 debug1: SSH2_MSG_KEX_DH_GEX_INIT sent debug1: expecting SSH2_MSG_KEX_DH_GEX_REPLY debug1: Host 'server1.domain.com' is known and matches the DSA host key. debug1: Found key in /Users/jhancock/.ssh/known_hosts:2 debug2: bits set: 527/1024 debug1: ssh_dss_verify: signature correct debug2: kex_derive_keys debug2: set_newkeys: mode 1 debug1: SSH2_MSG_NEWKEYS sent debug1: expecting SSH2_MSG_NEWKEYS debug2: set_newkeys: mode 0 debug1: SSH2_MSG_NEWKEYS received debug1: SSH2_MSG_SERVICE_REQUEST sent Read from socket failed: Connection reset by peer ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: PAY offered - sshd won't allow client from same domain
On Sep 16, 2006, at 4:50 PM, Garrett Cooper wrote: Do you have kerberos compiled and in use for authentication on the FreeBSD server and are you using it on the OSX client? server1 is the default from an original freeBSD 6.1 install and as of last week had a full cvsup and rebuild world (smae problem prior to the upgrade)...so its at 6.1-RELEASE--p6 now...I have not actively tried to enable or setup anything with kerberos on server or OS X client. My OS X client can connect fine to my other FreeBSD server2. server1 is the only server I can't connect to. The Windows XP client which can login to server1 can use either normal pam password or dsa key...very basic normal usage. The only line changed in sshd_config is UseDNS no. Changing it back to yes has no effect. ssh -vv server1.domain.com says? -Garrett___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions- [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FS size
Albert Shih wrote: Hi all I've read http://www.freebsd.org/projects/bigdisk/index.html and I want know actually on i386 arch is the limit of a fs is already 2 Tb What's the situation on amd64/EMT64, can we have big fs ? something like 10 or more TB ? There are people who use 8t partitions on i386. All you need to do is tweak some settings to allow fsck to assign enough memory for a file system check. This is possible because I think the file system is access block wise. With a block size of 4k an 8TB FS only requires an address space of 2g. I don't know anything about the UFS internals, so I cannot give you the real numbers, but they should that something aught to be technically possible. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
[CVSup]::[cvsup5.ru.freebsd.org]: No CVSROOT directory.
Hello.
I've used cvsup application under NetBSD-CURRENT to get whole
CVS repositories from NetBSD, FreeBSD and OpenBSD projects.
The client software works with cvsup5.ru.freebsd.org mirror
flawlessly (no errors or somthing strange).
excess
I have issue with OpenBSD server (cvsup.no.openbsd.org) like:
src/sbin/isakmpd/isakmpd.conf.5,v: Checksum mismatch -- will transfer entire
file
/excess
All servers report the same version (SNAP_16_1h). The same client:
CVSup client, non-GUI version
Copyright 1996-2003 John D. Polstra
Software version: SNAP_16_1h
Protocol version: 17.0
Operating system: NetBSDi386
http://www.cvsup.org/
Report problems to [EMAIL PROTECTED]
CVSup is a registered trademark of John D. Polstra
I can't checkout any branch at all because CVS repository is not
consistent w/o CVSROOT directory. Thus my copy is useless.
Question is obvious: why?
No filter settings in sup.cfg files.
I can provide more info if you tell me what you want to get.
Please, CC me, I'm not on the list.
TIA,
--
= System Administrator, SAMTELECOM LLC, Samara, Russia =
= Alexey G. Khramkov (agkhram) @ Samtelecom agkhram{at}samtelecom{dot}ru =
= GPG fingerprint : 944D 0C8B 343B 6C8D 50A1 061A E2DA 3E11 7765 6B47 =
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Making simple colorful block diagrams for presentations
Hello, I want a simple tool that can be used for preparing block diagrams and arrows, that is all. I want to be able to use few colors, that is all. Please don't suggest openoffice or kde. I want something simple. Thanks. regards, Girish -- Whenever people agree with me I always feel I am wrong. - Oscar Wilde ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Making simple colorful block diagrams for presentations
On Sat, Sep 16, 2006 at 03:20:00PM +0530, Girish Venkatachalam wrote: Hello, I want a simple tool that can be used for preparing block diagrams and arrows, that is all. I want to be able to use few colors, that is all. Please don't suggest openoffice or kde. I want something simple. graphics/xfig might be worth looking at. -- Insert your favourite quote here. Erik Trulsson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
A question about programming RS-232
Andrew Falanga wrote:
I am by no means the worlds best serial programmer, but recently I have
done some work on this subject and I noticed one thing in the code sample
above that should be avoided. However, I'll give you what I saw in-line:
#include stdio.h
#include termios.h
#include unistd.h
#include fcntl.h
int main(void) {
int t = 0, num = 10, fd, iOut; char *ch;
struct termios my_termios;
ch = (char *)malloc(6);
memset(ch, 250, 6);
fd = open(/dev/cuad0, O_RDWR | O_NONBLOCK);
Ok, great, we've opened our serial device. Unless you need this to be a
controlling terminal, you should open with open( /dev/cuad0, O_RDWR |
O_NONBLOCK | O_NOCTTY ); Check with the open man page to make sure I've
given you the correct constant for opening as a non-controlling terminal.
printf(Opened com port\n);
if(fd 0) return 0;
// tcflush(fd, TCIFLUSH);
my_termios.c_cflag = CS8 | CLOCAL;
if(cfsetspeed(my_termios, B9600) 0) return 0;
if(tcsetattr(fd, TCSANOW, my_termios) 0) return 0;
You've set the attributes you want to use in the structure you defined,
my_termios. However, you should call tcgetattr() before changing what you
want to change (and make sure you always turn things on as you have done
above with bitwise or). So, your code should look something like,
// assume an open file descriptor named fd
struct termios my_termios;
if( tcgetattr( fd, my_termios ) 0 ) {
fprintf( stderr, error in getting termios properties\n );
return AN_ERROR;
}
// turn on what you want
my_termios.c_cflag = CS8 | CLOCAL;
if( tcsetattr( fd, my_termios ) 0 } {
fprintf( stderr, error in setting new properties to serial port\n );
return AN_ERROR;
}
I don't know if this will solve your problems but I do know I read that you
should always get the current settings because the serial driver may use
certain bits and you don't want to turn them off. Also, if you're going to
return the port settings to the state before you took hold of it, make two
termios structures and stuff the original settings away to be restored upon
exit or close of the port.
Lastly, here is a link to a serial programming guide that I found quite
helpful. The info is probably dated to some degree, but it is non the less
useful.
http://www.easysw.com/~mike/serial/serial.html
Andy
I have corrected, what you say, but it doesn't work at all!
The code I tryed:
#include stdio.h
#include termios.h
#include unistd.h
#include fcntl.h
int main(void) {
int t = 0, num = 10, fd, i, iOut; char *ch;
struct termios my_termios;
ch = (char *)malloc(6);
memset(ch, 50, 6);
fd = open(/dev/cuad0, O_RDWR | O_NOCTTY | O_NDELAY);
fcntl(fd, F_SETFL, 0);
printf(Opened com port\n);
if(fd 0) return 0;
if(tcgetattr(fd, my_termios) 0) return 0;
printf(Got my_termios struct\n);
if(cfsetspeed(my_termios, B9600) 0) return 0;
printf(Set speed B9600\n);
my_termios.c_cflag = ~PARENB;
my_termios.c_cflag = ~CSTOPB;
my_termios.c_cflag = ~CSIZE;
my_termios.c_cflag |= CS8;
my_termios.c_cflag |= CLOCAL;
my_termios.c_cflag |= CREAD;
my_termios.c_lflag = ~(ICANON | ECHO | ECHOE | ISIG);
my_termios.c_oflag = ~OPOST;
if(tcsetattr(fd, TCSANOW, my_termios) 0) return 0;
printf(Setting my_termios attr\n);
iOut = write(fd, ch, 6);
if(iOut 0) return 0;
printf(Number of bytes = %d\n, iOut);
printf(Writed %s!\n, ch);
close(fd);
printf(Closed!\n);
return 0;
}
After executing it writes:
Opened com port
Got my_termios struct
Set speed B9600
Setting my_termios attr
Number of bytes = 6
Writed 22!
Closed!
But! No effect ;(
I'm in shock! How can it work under Windows if it doesn't work in UNIX? May be
I'm a lamer or I'm doing stupid things? But such code in Delphi using
component TComPort works perfectly:
library cportio;
{ Important note about DLL memory management: ShareMem must be the
first unit in your library's USES clause AND your project's (select
Project-View Source) USES clause if your DLL exports any procedures or
functions that pass strings as parameters or function results. This
applies to all strings passed to and from your DLL--even those that
are nested in records and classes. ShareMem is the interface unit to
the BORLNDMM.DLL shared memory manager, which must be deployed along
with your DLL. To avoid using BORLNDMM.DLL, pass string information
using PChar or ShortString parameters. }
uses
SysUtils,
Classes,
CPort,
Forms,
Windows,
IniFiles;
var Com: TComPort;
const crlf: String = #10 + #13;
const Name: String = 'Com';
const xPort: String = 'COM1';
const xDataBits: TDataBits = dbEight;
const xStopBits: TStopBits = sbOneStopBit;
const xParityBits: TParityBits = prNone;
const xFlowControl: TFlowControl = fcNone;
const xBaudRate: TBaudRate = br9600;
{$R *.res}
function WriteComPort(xData: Integer): Integer; stdcall;
begin
Com := TComPort.Create(Application);
Com.Port := xPort;
Com.DataBits := xDataBits;
Com.StopBits := xStopBits;
Com.Parity.Bits
My USB modem works under Linux, but not under FreeBSD
Hi all, I have a hardware(controller-based) USB modem(Zoom 2985-00-00C) that is identified under Liunx by acm driver and can be accessed through /dev/ttyACM0 and with some tools like cu and AT command set. This means that it supports CDC/ACM. But under FreeBSD 6.1(which have the latest umodem driver) with loaded ucom and umodem driver into kernel, after attaching the modem to the USB port, these messages appear on screen: ucom0: data interface 1, has CM over data, has break ucom0: Could not find data bulk in device_attach: ucom0 attach returned 6 and there is no ucom0 or any other appropriate devices in /dev directory and the modem does not work. With due attention to successful working under Linux, is there any hope to solve this problem under FreeBSD? Is it necessary to update umodem driver or there is an easy way? Best regards ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: PAY offered - sshd won't allow client from same domain
ke han [EMAIL PROTECTED] wrote: I will PAY someone who can either answer this question or who wants to log into my server and help me figure it out. I can pay an hourly rate, make a donation to your favorite project...whatever. This problem is killing my productivity I have a FreeBSD 6.1-p6 server running as server1.domain.com. sshd is allowing connections from any client except those which share the domain.com name..I can't be certain this is the problem, but after a month of debugging, its the only common factor I can find. My ssh client on server2.domain.com (also FreeBSD 6.1) returns with Read from socket failed: Connection reset by peer as output to my ssh client. On OS X the error message is Write failed: Broken pipe. ...So mac.domain.com and server2.domain.com which are on different networks from server1 (and from each other) are not allowed...I don't get any useful error messages. Even setting sshd_config LogLevel to DEBUG3 doesn't provide anything meaningful (to me) in auth.log or debug.log for server2.domain.com, I even have its ip as an A record in DNS and server1 can see this. mac.domain.com is not so lucky as it sits behind a DHCP NAT'ed structure. But this should hardly be a problem...PuTTY on Windows XP with no domain setting and behind a NAT'd DHCP structure CAN connect... You've obscured a lot of information regarding DNS and other configs, so I can only make a guess, but my guess would be that the DNS for your domain is somehow configured incorrectly and the server is time out trying to resolve domain names. Log in to the server and verify (using host(1)) that domain names resolve for the client's you're having trouble with. If that fails, you have more information to trace the problem. If that doesn't indicate anything, log into the server and run a second sshd with -D and capture all of the output. You may also need to use -p to run it on another port to ensure it doesn't conflict with the system sshd. Try to log in via a failing host and see if the output gives you any clues. If not, post it to see if someone else can identify something wrong with the process. -- Bill Moran That's why I never kiss 'em on the mouth. Jayne Cobb ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
tunderbird with arts
Is it still possible?
I had modified /usr/X11/lib/thunderbird/run-mozilla.sh to read
artsdsp $prog ${1+$@}
instead of simply
$prog ${1+$@}
and it used to work.
It stopped recently after I don't know which upgrade.
Any better way?
bye Thanks
av.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: PAY offered - sshd won't allow client from same domain
On Sep 16, 2006, at 6:05 PM, ke han wrote: On Sep 16, 2006, at 4:50 PM, Garrett Cooper wrote: ssh -vv server1.domain.com form OS X: (real domain name edited to domain.com) ssh -vv server1.domain.com OpenSSH_4.2p1, OpenSSL 0.9.7i 14 Oct 2005 debug1: Reading configuration data /etc/ssh_config debug2: ssh_connect: needpriv 0 debug1: Connecting to server1.domain.com [209.216.230.199] port 22. debug1: Connection established. debug1: identity file /Users/jhancock/.ssh/identity type -1 debug1: identity file /Users/jhancock/.ssh/id_rsa type -1 debug2: key_type_from_name: unknown key type '-BEGIN' debug2: key_type_from_name: unknown key type 'Proc-Type:' debug2: key_type_from_name: unknown key type 'DEK-Info:' debug2: key_type_from_name: unknown key type '-END' debug1: identity file /Users/jhancock/.ssh/id_dsa type 2 debug1: Remote protocol version 2.0, remote software version OpenSSH_4.2p1 FreeBSD-20050903 debug1: match: OpenSSH_4.2p1 FreeBSD-20050903 pat OpenSSH* debug1: Enabling compatibility mode for protocol 2.0 debug1: Local version string SSH-2.0-OpenSSH_4.2 debug2: fd 3 setting O_NONBLOCK debug1: Miscellaneous failure No credentials cache found debug1: Miscellaneous failure No credentials cache found debug1: SSH2_MSG_KEXINIT sent debug1: SSH2_MSG_KEXINIT received debug2: kex_parse_kexinit: diffie-hellman-group-exchange- sha1,diffie-hellman-group14-sha1,diffie-hellman-group1-sha1 debug2: kex_parse_kexinit: ssh-rsa,ssh-dss debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish-cbc,cast128- cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256-cbc,rijndael- [EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish-cbc,cast128- cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256-cbc,rijndael- [EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: none,[EMAIL PROTECTED],zlib debug2: kex_parse_kexinit: none,[EMAIL PROTECTED],zlib debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: first_kex_follows 0 debug2: kex_parse_kexinit: reserved 0 debug2: kex_parse_kexinit: diffie-hellman-group-exchange- sha1,diffie-hellman-group14-sha1,diffie-hellman-group1-sha1 debug2: kex_parse_kexinit: ssh-dss debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish-cbc,cast128- cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256-cbc,rijndael- [EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish-cbc,cast128- cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256-cbc,rijndael- [EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: none,[EMAIL PROTECTED] debug2: kex_parse_kexinit: none,[EMAIL PROTECTED] debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: first_kex_follows 0 debug2: kex_parse_kexinit: reserved 0 debug2: mac_init: found hmac-md5 debug1: kex: server-client aes128-cbc hmac-md5 none debug2: mac_init: found hmac-md5 debug1: kex: client-server aes128-cbc hmac-md5 none debug1: SSH2_MSG_KEX_DH_GEX_REQUEST(102410248192) sent debug1: expecting SSH2_MSG_KEX_DH_GEX_GROUP debug2: dh_gen_key: priv key bits set: 132/256 debug2: bits set: 523/1024 debug1: SSH2_MSG_KEX_DH_GEX_INIT sent debug1: expecting SSH2_MSG_KEX_DH_GEX_REPLY debug1: Host 'server1.domain.com' is known and matches the DSA host key. debug1: Found key in /Users/jhancock/.ssh/known_hosts:2 debug2: bits set: 527/1024 debug1: ssh_dss_verify: signature correct debug2: kex_derive_keys debug2: set_newkeys: mode 1 debug1: SSH2_MSG_NEWKEYS sent debug1: expecting SSH2_MSG_NEWKEYS debug2: set_newkeys: mode 0 debug1: SSH2_MSG_NEWKEYS received debug1: SSH2_MSG_SERVICE_REQUEST sent Read from socket failed: Connection reset by peer Your problem appears to be in how your user is being authenticated and not your DNS setup, I think. Example: shiina:~ gcooper$ uname -a Darwin shiina.local 8.7.0 Darwin Kernel Version 8.7.0: Fri May 26 15:20:53 PDT 2006; root:xnu-792.6.76.obj~1/RELEASE_PPC Power Macintosh powerpc shiina:~ gcooper$ ssh -vv tebo.cs.washington.edu OpenSSH_4.2p1, OpenSSL 0.9.7i 14 Oct 2005 debug1: Reading configuration data /etc/ssh_config debug2: ssh_connect: needpriv 0 debug1: Connecting to tebo.cs.washington.edu [128.208.6.74] port 22. debug1: Connection established. debug1: identity file /Users/gcooper/.ssh/identity type -1 debug2: key_type_from_name: unknown key type '-BEGIN' debug2: key_type_from_name: unknown key type 'Proc-Type:' debug2: key_type_from_name: unknown key type
Re: The fan is always on, even when the desktop is rather cool
On Saturday 16 September 2006 03:05, Wei Hu wrote: Thanks David, I checked this page. now if I do: $ sysctl hw.acpi hw.acpi.supported_sleep_state: S1 S3 S4 S5 hw.acpi.power_button_state: S5 hw.acpi.sleep_button_state: S1 hw.acpi.lid_switch_state: NONE hw.acpi.standby_state: S1 hw.acpi.suspend_state: S3 hw.acpi.sleep_delay: 1 hw.acpi.s4bios: 0 hw.acpi.verbose: 0 hw.acpi.reset_video: 1 hw.acpi.cpu.cx_supported: C1/0 hw.acpi.cpu.cx_lowest: C1 hw.acpi.cpu.cx_usage: 100.00% and if i do #sudo acpiconf -s 5 OR #sudo acpiconf -s 1 nothing happens. Can I reinstall acpi, if yes, how can i do it? thanks. On 9/16/06, David J Brooks [EMAIL PROTECTED] wrote: On Friday 15 September 2006 18:10, Wei Hu wrote: I have 3 systems in my desktop: 1) When FreeBSD runs, my desktop fans are always running, and this make annoy noisy. 2) However when Debian runs, the fan eventually stops unless I am performing a load intensive task. 3) In Windows, the fan is almost always off. I tried to use acpi and apm, but they are for laptop.(?) In Freebsd, how can I control the cooling fans or how can the system turns the fans off when the load is not heavy. Any help would be greatly appreciated. It sounds like a broken ACPI code to me. Check the handbook chapter on debugging ACPI: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/acpi-debug.html Its been a while since I messed with it, and I finally got my fan problem solved by replacing the computer with one that had a working ACPI bytecode. What The problem is that the ACPI code for your machine was probably compiled with the Microsoft compiler, which gives a clean compile on errors that would be caught by the Intel compiler. You can dump this bytecode to source and try to recompile it with the Intel compiler. That will likely show you what's broken. If you're lucky you may be able to recode to fix the errors and then load your new bytecode rather than the broken one that shipped with your machine. The best place to persue this question further is the freebsd-acpi mail-list. If Nate Lawson can't help you out, you're probably stuck with a noisy machine. At least you can be confident that it won't overheat. :) David -- Sure the Almighty created the world in only six days, but He didn't have an established user-base. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Making simple colorful block diagrams for presentations
On Sat, Sep 16, 2006 at 11:57:35AM +0200, Erik Trulsson wrote: On Sat, Sep 16, 2006 at 03:20:00PM +0530, Girish Venkatachalam wrote: Hello, I want a simple tool that can be used for preparing block diagrams and arrows, that is all. I want to be able to use few colors, that is all. Please don't suggest openoffice or kde. I want something simple. graphics/xfig might be worth looking at. -- Insert your favourite quote here. Erik Trulsson [EMAIL PROTECTED] try alsographics/tgif -- FreeBSD 6.0-RELEASE i386 GENERIC ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Crippled FreeBSD! Need help!
I have managed to scr** up my FBSD 6.1 installation. This is what happened: I had an installation of BSD with which I was experimenting and managed to get it to work to my taste. Call this installation A. This installation had a lot of unwanted ports installed so I decided to do a new installation. I installed BSD in another partition. Call this installation B. I wanted to custom build the kernel in B and hence created the config file needed for the custom build. I used sysinstall to copy the src from the 6.1 CD. When I marked Base inside sysinstall to be copied, it gave me an error: Write failure on transfer! (wrote -1 bytes of 1425408 bytes). I know this is not a problem with the CD because I was able to copy the base and src in another machine that I have. Src however got copied in B. I couldnt do a make buildworld because it gave me an error saying that it didnt know what buildworld was. I knew i needed the base files to get it to work. A friend of mine had helped me custom build the kernel in installation A. So I mounted the partition and copied the files in /usr/src (only files, excluded /usr/src/sys) from installation A to installation B. Then when I did a make buildworld it gave me an error that it was unable to cd into a directory by thename /somepath/usr.bin. I dont remember what somepath was. I couldn't note it down. Then I mounted the 6.1 CD and went into the 6.1Release folder and into the base directory and ran ./install.sh. It asked me if I wanted to copy files to / and i said yes. I attempted make buildworld again and it did not work. When this did not work, I decided to restart the machine and try again. However, after restarting, the login prompt does not accept my user id and password. And when i type root for user, it logs me in without asking for a password. I am clueless whats happening here. Please help. Thanks, Vishy p.s: i used make buildworld KERNCONF=CUSTOM. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: PAY offered - sshd won't allow client from same domain
On Sep 16, 2006, at 10:51 PM, Garrett Cooper wrote: On Sep 16, 2006, at 6:05 PM, ke han wrote: On Sep 16, 2006, at 4:50 PM, Garrett Cooper wrote: ssh -vv server1.domain.com form OS X: (real domain name edited to domain.com) ssh -vv server1.domain.com OpenSSH_4.2p1, OpenSSL 0.9.7i 14 Oct 2005 debug1: Reading configuration data /etc/ssh_config debug2: ssh_connect: needpriv 0 debug1: Connecting to server1.domain.com [209.216.230.199] port 22. debug1: Connection established. debug1: identity file /Users/jhancock/.ssh/identity type -1 debug1: identity file /Users/jhancock/.ssh/id_rsa type -1 debug2: key_type_from_name: unknown key type '-BEGIN' debug2: key_type_from_name: unknown key type 'Proc-Type:' debug2: key_type_from_name: unknown key type 'DEK-Info:' debug2: key_type_from_name: unknown key type '-END' debug1: identity file /Users/jhancock/.ssh/id_dsa type 2 debug1: Remote protocol version 2.0, remote software version OpenSSH_4.2p1 FreeBSD-20050903 debug1: match: OpenSSH_4.2p1 FreeBSD-20050903 pat OpenSSH* debug1: Enabling compatibility mode for protocol 2.0 debug1: Local version string SSH-2.0-OpenSSH_4.2 debug2: fd 3 setting O_NONBLOCK debug1: Miscellaneous failure No credentials cache found debug1: Miscellaneous failure No credentials cache found debug1: SSH2_MSG_KEXINIT sent debug1: SSH2_MSG_KEXINIT received debug2: kex_parse_kexinit: diffie-hellman-group-exchange- sha1,diffie-hellman-group14-sha1,diffie-hellman-group1-sha1 debug2: kex_parse_kexinit: ssh-rsa,ssh-dss debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish- cbc,cast128-cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256- cbc,[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish- cbc,cast128-cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256- cbc,[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: none,[EMAIL PROTECTED],zlib debug2: kex_parse_kexinit: none,[EMAIL PROTECTED],zlib debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: first_kex_follows 0 debug2: kex_parse_kexinit: reserved 0 debug2: kex_parse_kexinit: diffie-hellman-group-exchange- sha1,diffie-hellman-group14-sha1,diffie-hellman-group1-sha1 debug2: kex_parse_kexinit: ssh-dss debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish- cbc,cast128-cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256- cbc,[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: aes128-cbc,3des-cbc,blowfish- cbc,cast128-cbc,arcfour128,arcfour256,arcfour,aes192-cbc,aes256- cbc,[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,hmac- [EMAIL PROTECTED],hmac-sha1-96,hmac-md5-96 debug2: kex_parse_kexinit: none,[EMAIL PROTECTED] debug2: kex_parse_kexinit: none,[EMAIL PROTECTED] debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: first_kex_follows 0 debug2: kex_parse_kexinit: reserved 0 debug2: mac_init: found hmac-md5 debug1: kex: server-client aes128-cbc hmac-md5 none debug2: mac_init: found hmac-md5 debug1: kex: client-server aes128-cbc hmac-md5 none debug1: SSH2_MSG_KEX_DH_GEX_REQUEST(102410248192) sent debug1: expecting SSH2_MSG_KEX_DH_GEX_GROUP debug2: dh_gen_key: priv key bits set: 132/256 debug2: bits set: 523/1024 debug1: SSH2_MSG_KEX_DH_GEX_INIT sent debug1: expecting SSH2_MSG_KEX_DH_GEX_REPLY debug1: Host 'server1.domain.com' is known and matches the DSA host key. debug1: Found key in /Users/jhancock/.ssh/known_hosts:2 debug2: bits set: 527/1024 debug1: ssh_dss_verify: signature correct debug2: kex_derive_keys debug2: set_newkeys: mode 1 debug1: SSH2_MSG_NEWKEYS sent debug1: expecting SSH2_MSG_NEWKEYS debug2: set_newkeys: mode 0 debug1: SSH2_MSG_NEWKEYS received debug1: SSH2_MSG_SERVICE_REQUEST sent Read from socket failed: Connection reset by peer Your problem appears to be in how your user is being authenticated and not your DNS setup, I think. Example: shiina:~ gcooper$ uname -a Darwin shiina.local 8.7.0 Darwin Kernel Version 8.7.0: Fri May 26 15:20:53 PDT 2006; root:xnu-792.6.76.obj~1/RELEASE_PPC Power Macintosh powerpc shiina:~ gcooper$ ssh -vv tebo.cs.washington.edu OpenSSH_4.2p1, OpenSSL 0.9.7i 14 Oct 2005 debug1: Reading configuration data /etc/ssh_config debug2: ssh_connect: needpriv 0 debug1: Connecting to tebo.cs.washington.edu [128.208.6.74] port 22. debug1: Connection established. debug1: identity file /Users/gcooper/.ssh/identity type -1 debug2: key_type_from_name: unknown key type '-BEGIN' debug2: key_type_from_name: unknown key type 'Proc-Type:' debug2: key_type_from_name:
Re: Firefox+Flash works for sure
I use linux-opera and I have managed to get flash working like a charm. Just go to any website using flash and opera will ask you to download the plugin and automatically take you to the linux page of the flash plugin in the adobe website. Then download the flash plugin tar.gz and save it to some location. Extract the contents and copy the libflashplayer.so file to /usr/X11R6/share/linux-opera/plugins. Close opera and open again and enjoy the world of flash ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Firefox+Flash works for sure
On Sat, Sep 16, 2006 at 08:33:11PM +0530, Viswas Nair wrote: I use linux-opera and I have managed to get flash working like a charm. Just go to any website using flash and opera will ask you to download the plugin and automatically take you to the linux page of the flash plugin in the adobe website. Then download the flash plugin tar.gz and save it to some location. Extract the contents and copy the libflashplayer.so file to /usr/X11R6/share/linux-opera/plugins. Close opera and open again and enjoy the world of flash I also use linux-opera...but there some site that are running flash 8 that won't work with the most current adobe flash version 7. -- FreeBSD 6.0-RELEASE i386 GENERIC ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Firefox+Flash works for sure
On Saturday 16 September 2006 11:38, ajm wrote: I also use linux-opera...but there some site that are running flash 8 that won't work with the most current adobe flash version 7. A somewhat similar problem is starting to crop up with mplayer and the latest version of WMV codecs. I seems that a few sites are using the new code and mplayer is unable to interpret it correctly. I was just on the mplayer forum where it was being discussed along with the URLs of some of those sites. It is my belief though that mplayer will get the necessary changes make to remedy this problem faster than a fully functioning and current version of flash for FBSD is available. -- Gerard QUARK: The sound made by a well bred duck. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Firefox+Flash works for sure
At Sat, 16 Sep 2006 it looks like Viswas Nair composed: I use linux-opera and I have managed to get flash working like a charm. Just go to any website using flash and opera will ask you to download the plugin and automatically take you to the linux page of the flash plugin in the adobe website. Then download the flash plugin tar.gz and save it to some location. Extract the contents and copy the libflashplayer.so file to /usr/X11R6/share/linux-opera/plugins. Close opera and open again and enjoy the world of flash ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Hello Family, I'm running 6.1, installed linux-opera from ports in order to test the above, and the ports install seemed to go fine but I got this error when trying to start Opera, anyone seen this before? ## [EMAIL PROTECTED] ~]$ /usr/X11R6/share/linux-opera/bin/opera opera: Preference initialization failure. File not found or could not be opened (-7) ## TIA -- Bill Schoolcraft * http://wiliweld.com * If you turn your headlights on while going the speed of light, does anything happen? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Broadcomm NetXtreme BCM5708 NICs and 6.1 RELEASE
I'm posting this for documentary purposes in case someone has this problem and wants to find the answer. Under the 6.1 RELEASE, with all sources cvsup'd to current, both world and kernel rebuilt, the Broadcommm NetExtreme 5708 NICs will fall over under very light load when using a remote connection. (Console outbound connections work fine.) For example, trying to build apache22 from ports causes the NICs to fail, and only a reboot will fix the problem. The console error message is Error mapping mbuf into TX chain! Not good for servers. :-) The solution is to update the if_bce.c source to version 0.9.6 from the current 0.9.5, then rebuild world and kernel. Here's a webpage that has a brief explanation and a link to the updated source file: http://www.ifdnrg.com/freebsd_broadcom_dell_1950.htm Paul Schmehl ([EMAIL PROTECTED]) Adjunct Information Security Officer The University of Texas at Dallas http://www.utdallas.edu/ir/security/
Re: Firefox+Flash works for sure
On Saturday 16 September 2006 16:03, Viswas Nair wrote: I use linux-opera and I have managed to get flash working like a charm. For the third time, it has multiple critical vulnerabilities. If you use it your computer may work like a charm for someone else. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Video Device problems in new install by Newbie -- FreeBSD 6.1
Hello List, I took the plunge last night and installed FreeBSD6.1-STABLE. I found the experience rather exciting and I'm happy with the results, excep for one major challenge to overcome: no video support. I'm trying to set up X and it's failing. Prerequisites: FreeBSD6.1-STABLE (from the i386 CD iso's from freebsd.org) Dell Dimension 5100C Desktop WD SATA HD Intel 82945G Express Chipset Family on IRQ 16 My friend the computer professional tells me the video and a bunch of other devices are integrated into the motherboard to save space (the system unit is smaller than the printer, if that tells you anything). when I do: root-promptXorg -configure I get a quick list of all video devices and No Device Present The Xorg-configure-log is just a more verbose statement of the same list At boot, I get the following messages: ... acpi0:Dell 5100C on motherboard ... pci0:display at device 2.1 (no driver attached) pci0:multimedia at device 27.0 (no driver attached) ... the boot process then proceeds to recognize all my other devices and I get to my login prompt. I've tried booting with ACPI disabled (I'll admit I'm ignorant as to what that would do) and the boot process hangs up before grabbing my keyboard. The Handbook mentions when the visual kernel configuration option comes up choose it. Where and when does that happen? Is that still around in FreeBSD6.1? I've tried boot menu option 6 (the loader prompt) and I'll admit that totally confuses me. Supposing I have the driver, how do I attach it to resolve the no driver attached problem? Is that really the root of the problem? I know the VESA drivers are there... I'll add that Freesbie 1.1 and many Linux livecds boot just fine into X on this machine. I even got an Xf86config file from Freesbie with the monitor settings it uses. I'm writing this from a PCLinuxOS Livecd. Right now I've got a console, and that's cool in the words of Jeffrey Lebowski; however, for my wife to use this thing, and for me to use most of what I wanted (a desktop system), running X is essential ;) I appreciate any help. Thanks! I already love FreeBSD! Joel Joel J. Adamson Arlington, MA - Yahoo! Messenger with Voice. Make PC-to-Phone Calls to the US (and 30+ countries) for 2¢/min or less. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
sshd won't updates ( libsm.so.1 )
I have RELENG_6 ( just updated and installed kernel , 6_2 prerelease ) and by mistake ( or my stupid hands ) make : cd /usr/src/secure make install now sshd does not work. it tells: /libexec/ld-elf.so.1: Shared object libbsm.so.1 not found, required by sshd I tried to build the system again but the result is same. I did make cleandir ( twice ) in /usr/src and tried this solution: # cd /usr/src/secure/lib/libssh # make obj make depend make make install # cd /usr/src/secure/usr.sbin/sshd # make obj make depend make make install but build fails at: /usr/src/secure/usr.sbin/sshd/../../../crypto/openssh/audit-bsm.c:50:31: bsm/audit_uevents.h: No such file or directory /usr/src/secure/usr.sbin/sshd/../../../crypto/openssh/audit-bsm.c:51:30: bsm/audit_record.h: No such file or directory mkdep: compile failed *** Error code 1 But make buildworld did the job without error ( at the end of compilation ) If I trying to copy libbsm.so/a/1 to the /usr/lib - sshd launches, but does not allow to login. -- Pavel Sokolov ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Firefox+Flash works for sure
On Sat, Sep 16, 2006 at 06:45:56PM +0100, RW wrote: On Saturday 16 September 2006 16:03, Viswas Nair wrote: I use linux-opera and I have managed to get flash working like a charm. For the third time, it has multiple critical vulnerabilities. If you use it your computer may work like a charm for someone else. I heard you the first time. Adobe have released a patched version (7.0.68) that addresses these vulnerabilities. The Freshmeat notice went out yesterday, so I'm sure it is just a matter of time before the maintainer revs the FreeBSD port. Kurt ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
rebooting into single user mode on a remote server
Hello, could somebody help me to understand the best way to enter into a single user mode on a remote server. I need it for the moment, during rebuilding world, when I have to reboot into single user mode before 'mergemaster -p'. The only solution I found so far is to do 'shutdown -r now' and when the server boots to login with ssh and do 'shutdown now' - which should drop it to single user mode. I can ask the support at the hosting location to reboot in single user mode, but I do not know if I will have ssh then? Alternatively I can ask them to do the last few steps. Thank you for your advises, Iv. -- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
When is BuildWorld necessary?
Hi: I recently installed FreeBSD 6.1 over the net from sources. I am keeping things up-to-date using CVSup. When portaudit tells me I have a security issue; I update/re-install the affected port. When a kernel patch comes in, I re-compile the kernel; which now stands at FreeBSD 6.1-RELEASE-p6 #3. From what I can tell, buildworld re-builds the base system, something I have yet to do. My thought is to do a buildworld only when the OS version is updated to the next number above 6.1. I understand this happens at about 4 month intervals. My question is, is there a good reason to buildworld before a version change? I hate fixing something which is working perfectly, and this system has been stellar! Bob ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: When is BuildWorld necessary?
Bob wrote: Hi: I recently installed FreeBSD 6.1 over the net from sources. I am keeping things up-to-date using CVSup. When portaudit tells me I have a security issue; I update/re-install the affected port. When a kernel patch comes in, I re-compile the kernel; which now stands at FreeBSD 6.1-RELEASE-p6 #3. From what I can tell, buildworld re-builds the base system, something I have yet to do. My thought is to do a buildworld only when the OS version is updated to the next number above 6.1. I understand this happens at about 4 month intervals. My question is, is there a good reason to buildworld before a version change? I hate fixing something which is working perfectly, and this system has been stellar! Bob Hi Bob, I believe it is basically good to get the 'p' patches as they contain security fixes. My thinking is that if 'p' patch comes out - your system is, in some sense, not perfect anymore :) But I have one question - do you rebuild the world on a remote machine (without physical access) and if yes - how do you restart in single user mode. This is what I can't understand so far. Thanks, Iv -- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
how to make fixed Direct Access device (da) ID
It is awkward that dynamically and/or statically attaching SCSI hard drive and USB hard drive to the system will have different da IDs. For example, boot system with a SCSI drive (SCSI = 1), will have a da0 for this SCSI drive. Then plugging in a USB hard drive, which will be configured as da1. If boot system with both drives online, system will boot from SCSI drive fine till mounting root point. It fails because USB drive has da0 and SCSI drive has da1. Is there anyway to configure the system to have fixed da ID for SCSI drive or even for USB drive regardless if they are dynamically/statically attached to the system? -Jin ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: When is BuildWorld necessary?
On Saturday 16 September 2006 20:41, Bob wrote: Hi: I recently installed FreeBSD 6.1 over the net from sources. I am keeping things up-to-date using CVSup. When portaudit tells me I have a security issue; I update/re-install the affected port. When a kernel patch comes in, I re-compile the kernel; which now stands at FreeBSD 6.1-RELEASE-p6 #3. From what I can tell, buildworld re-builds the base system, something I have yet to do. My thought is to do a buildworld only when the OS version is updated to the next number above 6.1. I understand this happens at about 4 month intervals. My question is, is there a good reason to buildworld before a version change? I hate fixing something which is working perfectly, and this system has been stellar! Not all of the point releases are for the kernel, for example 6.1-RELEASE-p2 was a sendmail fix. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: When is BuildWorld necessary?
On Saturday 16 September 2006 16:13, RW wrote: Not all of the point releases are for the kernel, for example 6.1-RELEASE-p2 was a sendmail fix. Ok I see; just because my kernel is at p6, doesn't mean the base system is. I wasn't on FreeBSD when p2 was released. Would that p2 have triggered a portaudit warning? Assuming of course that p2 was a security related sendmail patch. What I am getting at is if, my sendmail were acting up, I would look for an update, and patch sendmail only. If the patch were security related I would patch it anyway, but I can't see why I would want to rebuild the entire system for a sendmail upgrade, or a kernel stability patch, when the individual broken/insecure pieces can be fixed with much less hassel, time, and risk. Is my logic flawed? Bob ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: When is BuildWorld necessary?
On Saturday 16 September 2006 15:52, [EMAIL PROTECTED] wrote: But I have one question - do you rebuild the world on a remote machine Sorry; I am a newbie at FreeBSD, and have never done a buildworld :-( I have spent lots of time on Linux, Solaris, and SCO, but this is my first cut at BSD. Just from past NIX experience though, I would never rebuild an entire OS remotely without having someone onsite to push the On/Off switch when the inevitable happens :-( Bob ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Is 6.1-RELEASE missing parts of X11?
People, I've done at least two cmplete make buildworlds (and all the rest: kernel/installworld/) and parts of mergemaster. I *am* missing /usr/X11R6/bin/xdm and more files in /usr/X11R6/lib/[*]. startx is also missing. I have read the UPDATING and README in /usr/src; Am I missing some knob[s]? I have gdm_enable=YES in /etc/rc.conf just now and have the exec gnome line in ~/.xinitrc. [~/.xinitrc is 0755 and chown'd kline:wheel] and so on. I do see ctwm installed as a backup, but without at least xdm, I'm wedged. Anybody know what's going on? thanks in adance, gary PS:: When I scp'd xdm from 5.5, it did exec, but the xlogin widget was White and logging in as anyone failed. I was always thrown back to the default white xlogin screen. -- Gary Kline [EMAIL PROTECTED] www.thought.org Public service Unix ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: When is BuildWorld necessary?
On Saturday 16 September 2006 15:52, [EMAIL PROTECTED] wrote: But I have one question - do you rebuild the world on a remote machine (without physical access) and if yes - how do you restart in single user mode. This is what I can't understand so far. I remembered something right after I sent the last post. I have done this before, years ago. Not with bsd, but with Linux. I was working on a small server farm, and cross-connected serial ports from one server to another. Made the serial port the console, and then I could telnet to the adjacent server, tip to the other one, and have the system console. From there you could pretty safely do whatever you wanted to do, if the kernel were to fail to boot, you would be left at the loader prompt, where you could boot the box into a known good kernel. I can't see why you couldn't do something like that with FreeBsd. All you need is a serial port you can control remotely, like an adjacent server, or a router set it all up beforehand, and you should be good to go. Bob ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: rebooting into single user mode on a remote server
[EMAIL PROTECTED] wrote: Hello, could somebody help me to understand the best way to enter into a single user mode on a remote server. I need it for the moment, during rebuilding world, when I have to reboot into single user mode before 'mergemaster -p'. The only solution I found so far is to do 'shutdown -r now' and when the server boots to login with ssh and do 'shutdown now' - which should drop it to single user mode. I can ask the support at the hosting location to reboot in single user mode, but I do not know if I will have ssh then? Alternatively I can ask them to do the last few steps. Yep. You've become the latest person to realise this perennial problem. In order to follow the upgrade instructions in the Handbook or /usr/src/UPDATING to the letter, you need console access to the machine being updated. That is no problem when the machine is on your desk, or probably not if it's just down the hall. But when it's in a hosting centre umpty dozen miles away and you can't actually get to it? There are essentially three possibilities. i) You've thought of this approach already: get someone local to the machine to do the bits requiring the console access. That works if the people at the other site are competent and trustworthy, and you can afford to pay for their time. ii) The next solution, and on the whole, probably the best solution available, is to arrange to get remote console access. That can be expensive if you go down the route of buying a dedicated console server. Or it can be very cheap indeed if you have another FreeBSD box close by the machine you're trying to update and you can string null modem cables between their serial ports. Then you configure your FreeBSD box requiring update to use ttya as its console and use tip(1) to get into it from the other machine. (Actually, you could probably make that approach work from any other unixoid OS or even from Windows so long as you can find the right serial console emulation software). If you're really lucky, you're running flashy new hardware with IPMI or similar lights out management capability and can get into the machine through that. It doesn't work in anything like the same way as a serial console, but the end result is just as good. iii) Finally, and not to be dismissed without due consideration, is the really quite simple approach of /not/ taking the machine down to single user mode. Most of the time, you can quite happily run 'make installworld' or 'make installkernel' or 'mergemaster' while the system is in multiuser mode. You should shutdown all active services except what you need to get in remotely and you should kick any other users off the machine as well as generally taking steps to ensure the machine is as quiescent as possible before trying that. You should also have a 'back to square one' plan for dealing with the eventuality that the machine does not come back after attempting to reboot into the new kernel -- you really absolutely will require someone quite FreeBSD savvy to get onto the console to unfuck things if so, and that illustrates the big drawback to this approach: if it goes wrong, you are truly left up a gum tree without a paddle. Don't try approach (iii) for an upgrade over too many version numbers at once. Jumping from, say 6.1-RELEASE to 6.1-RELEASE-p6 should be feasible, as should jumping from 6.0-RELEASE to 6.1-RELEASE. Going from say 5.5-RELEASE to 6.1-RELEASE is only for the brave or the most highly skilled, and anything more than that is only for the foolhardy. Neither is it a good idea to do method (iii) if you're making any major changes to the hardware on the system. Nor does approach (iii) mix at all well with the use of raised secure levels. Cheers, Matthew -- Dr Matthew J Seaman MA, D.Phil. 7 Priory Courtyard Flat 3 PGP: http://www.infracaninophile.co.uk/pgpkey Ramsgate Kent, CT11 9PW signature.asc Description: OpenPGP digital signature
Plesk and FreeBSD 6.1 64-bit getting frustrating
Hello I've got FreeBSD 6.1 installed on a Sun Fire X2100 server, and are trying to get plesk installed. But plesk isn't supported for FreeBSD 6.1 64-bit version yet, so it has been hard work to try trick it. Now I want to trick the uname command to show the version needed for plesk installation. Any one having an easy and pretty safe way to do this? This is the last try before I trash FreeBSD as it's pretty important that this server come up and running soon, though I love FreeBSD :( Also, I'm not member at this list so please mail / cc me directly. Thanks. Regards, Dan Schultzer ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
top(1) STATE column
Hi List, I'd like to know the meaning of the possible STATEs showing up in top. In the manual pages I found this: STATE is the current state (one of START, RUN (shown as CPUn on SMP systems), SLEEP, STOP, ZOMB, WAIT, LOCK or the event on which the process waits) Where can I found info about other possible states (nanslp, kserel, ttyin, ucond, sbwait, ...) that I usually see in top? I think these have to do with the the event on which the process waits part of the man page... isn't there any complete list on those? Thanx, regards -- Pietro Cerutti ICQ: 117293691 PGP: 0x9571F78E - ASCII Ribbon Campaign - against HTML e-mail and proprietary attachments www.asciiribbon.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: When is BuildWorld necessary?
Bob wrote: Hi: I recently installed FreeBSD 6.1 over the net from sources. I am keeping things up-to-date using CVSup. When portaudit tells me I have a security issue; I update/re-install the affected port. When a kernel patch comes in, I re-compile the kernel; which now stands at FreeBSD 6.1-RELEASE-p6 #3. From what I can tell, buildworld re-builds the base system, something I have yet to do. My thought is to do a buildworld only when the OS version is updated to the next number above 6.1. I understand this happens at about 4 month intervals. My question is, is there a good reason to buildworld before a version change? I hate fixing something which is working perfectly, and this system has been stellar! You can't assume that any patch release on a security branch is solely going to be to fix things in the kernel. More often than not, the upgrade is to fix things in the userland. That means you have to recompile and re-install the affected software. Gennerally security advisories will tell you how to patch and update the specifically affected stuff. On the whole though, it always works to apply a full buildworld cycle as described in /usr/ports/UPDATING, and for certain security problems it's the only way to be sure the base system is rendered invulnerable[*]. Also it means the system version number gets bumped making it easy to identify what machines have been patched weeks or months down the line. If you haven't been rebuilding and re-installing world along with kernel as part of the update cycle, then there is a distinct possibility that you are still exposed eg. to the sendmail vulnerabilities from SA-06:17 or the ypserv problems from SA-06:15 or to various others. You will find that running the full buildworld procedure is a pretty smooth operation and if applied with due care and attention it is not at all difficult to get the system successfully updated nor is it hard to avoid foot-shooting while doing so. Cheers, Matthew [*] Where there is significant change of a vulnerability from the base system affecting 3rd party software from the ports or wherever, that should be discussed in the security advisories that come out, as well as what measures are necessary to provide a fix. -- Dr Matthew J Seaman MA, D.Phil. 7 Priory Courtyard Flat 3 PGP: http://www.infracaninophile.co.uk/pgpkey Ramsgate Kent, CT11 9PW signature.asc Description: OpenPGP digital signature
Calling setxkbmap when starting X
Hello, I am from Bulgaria and I use Bulgarian language on my FreeBSD machine. I use the following command $ setxkbmap -model pc105 -layout us,bg -variant ,phonetic -option grp:alt_shift_toggle to enable both Bulgarian and English. However, I call this command every time I log in KDE. Is it possible to invoke it automatically when X is started? Thank you in advance for your answers. Regards Ivan -- Tangra Mega Rock: http://www.radiotangra.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Is 6.1-RELEASE missing parts of X11?
Gary Kline wrote: People, I've done at least two cmplete make buildworlds (and all the rest: kernel/installworld/) and parts of mergemaster. I *am* missing /usr/X11R6/bin/xdm and more files in /usr/X11R6/lib/[*]. startx is also missing. I have read the UPDATING and README in /usr/src; Am I missing some knob[s]? I have gdm_enable=YES in /etc/rc.conf just now and have the exec gnome line in ~/.xinitrc. [~/.xinitrc is 0755 and chown'd kline:wheel] and so on. I do see ctwm installed as a backup, but without at least xdm, I'm wedged. Anybody know what's going on? thanks in adance, gary PS:: When I scp'd xdm from 5.5, it did exec, but the xlogin widget was White and logging in as anyone failed. I was always thrown back to the default white xlogin screen. I think the howls of protest would have been audible from the moons of Jupiter had 6.1-RELEASE shipped without a complete set of workable X windows ports / packages. No, you are definitely experiencing a problem with your own machine and not with the FreeBSD release. However important such software may be, it is not actually a part of the base system. portupgrade(1) is your friend in this case, not 'make buildworld'. If you are updating from 5.x to 6.x then you should be sure to reinstall all your ports / packages. A command line of the form: portupgrade -Niaf will get that job done. There are various alternative options you might want to consider employing, such as telling portupgrade to use packages rather than re-compiling everything from source: the man page for portupgrade will elucidate. You need to do this not because 5.x programs won't work on a 6.x system (they manifestly will run if the compat5x shlibs are installed) but because any future software update runs the risk of different parts of the same program being linked against different versions of a shared library and consequently failing to work. Cheers, Matthew -- Dr Matthew J Seaman MA, D.Phil. 7 Priory Courtyard Flat 3 PGP: http://www.infracaninophile.co.uk/pgpkey Ramsgate Kent, CT11 9PW signature.asc Description: OpenPGP digital signature
Re: When is BuildWorld necessary?
On Saturday 16 September 2006 21:34, Bob wrote: On Saturday 16 September 2006 16:13, RW wrote: Not all of the point releases are for the kernel, for example 6.1-RELEASE-p2 was a sendmail fix. Ok I see; just because my kernel is at p6, doesn't mean the base system is. I wasn't on FreeBSD when p2 was released. Would that p2 have triggered a portaudit warning? Assuming of course that p2 was a security related sendmail patch. What I am getting at is if, my sendmail were acting up, I would look for an update, and patch sendmail only. If the patch were security related I would patch it anyway, but I can't see why I would want to rebuild the entire system for a sendmail upgrade, or a kernel stability patch, when the individual broken/insecure pieces can be fixed with much less hassel, time, and risk. In FreeBSD the most conservative approach is to rebuild both world and kernel, they are more of a matched pair than in Linux. Since I don't bother to drop into single-user mode, or do the extra reboot for point releases, I just run a single script that does the whole thing (including cvsup), then reboot at my convenience. Having said that, I know some people that run STABLE will just rebuild individual parts of world. IMHO this is a lot more hassle than typing the name of a script, and letting the hardware take the strain. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Is 6.1-RELEASE missing parts of X11?
On Sat, Sep 16, 2006 at 10:35:42PM +0100, Matthew Seaman wrote: Gary Kline wrote: People, I've done at least two cmplete make buildworlds (and all the rest: kernel/installworld/) and parts of mergemaster. I *am* missing /usr/X11R6/bin/xdm and more files in /usr/X11R6/lib/[*]. startx is also missing. I have read the UPDATING and README in /usr/src; Am I missing some knob[s]? I have gdm_enable=YES in /etc/rc.conf just now and have the exec gnome line in ~/.xinitrc. [~/.xinitrc is 0755 and chown'd kline:wheel] and so on. I do see ctwm installed as a backup, but without at least xdm, I'm wedged. Anybody know what's going on? thanks in adance, gary PS:: When I scp'd xdm from 5.5, it did exec, but the xlogin widget was White and logging in as anyone failed. I was always thrown back to the default white xlogin screen. I think the howls of protest would have been audible from the moons of Jupiter had 6.1-RELEASE shipped without a complete set of workable X windows ports / packages. No, you are definitely experiencing a problem with your own machine and not with the FreeBSD release. However important such software may be, it is not actually a part of the base system. portupgrade(1) is your friend in this case, not 'make buildworld'. If you are updating from 5.x to 6.x then you should be sure to reinstall all your ports / packages. A command line of the form: portupgrade -Niaf will get that job done. There are various alternative options you might want to consider employing, such as telling portupgrade to use packages rather than re-compiling everything from source: the man page for portupgrade will elucidate. You need to do this not because 5.x programs won't work on a 6.x system (they manifestly will run if the compat5x shlibs are installed) but because any future software update runs the risk of different parts of the same program being linked against different versions of a shared library and consequently failing to work. Thanks for several clues! When I chose the packages from /stand/sysinstall maybbe [[ evidently?!]] I only selected base. I've rebuilt several hundred ports from src. (Some packages were marked missing when I tried to install [gnome|kde]-lite so have to hand-built:) ... But this was with 6.1-STABLE rather than -RELEASE.) take care, gary PS: This is obv'ly a local fault; if there were a better OS I would be using it. Cheers, Matthew -- Dr Matthew J Seaman MA, D.Phil. 7 Priory Courtyard Flat 3 PGP: http://www.infracaninophile.co.uk/pgpkey Ramsgate Kent, CT11 9PW -- Gary Kline [EMAIL PROTECTED] www.thought.org Public service Unix ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
ipfw and temporary port access
Hi there, I am trying to figure out how to open a port temporarily for a specific IP who is able to provide a proper username and password on the website of the box. After authentication is verified then the IP address is cached and temporarily allowed to access a specific port on the server. This temporary firewall changes would be handled by ipfw. Any clues if a system like this is a already coded and out there somewhere? Cheers, Noah ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
28.8kbs/56kbs modems
To whom it may concern, Currently, I am using Ubuntu Linux 6.06 and it is really a let-down after I got it when I realized that Ubuntu does not do well with 28.8kbs/56kbs modems. It will not let me use my modem. I was wondering how Free-bsd does with dial-up modems (2 year old computer) and highspeed interenet, (I might get high speed soon). I was thinking that if FreeBSD worked better for going online using dial-up modems? If it worked well, then I was thinking of setting up a partion for both OS's to run. Would I be able to send files between them, over the partion? Thanks in advance. Sincerly, David Fontenot P.S. If my family did share a high speed internet connection, could I still connect to their network and share the internet, even if they are both using Windows XP? :-) - Get your email and more, right on the new Yahoo.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: When is BuildWorld necessary?
[EMAIL PROTECTED] wrote: But I have one question - do you rebuild the world on a remote machine (without physical access) and if yes - how do you restart in single user mode. This is what I can't understand so far. Thanks, Iv In 6 years, I've never dropped any machine to single user to do any part of a buildworld upgrade. I've stopped many running services, but never gone to single user. The only time I had any problems with this approach was when I blindly flubbed versions in my supfile and cvsup'd a 6 system with 4 source. That wasn't pretty. But it would have been not pretty in single user mode as well. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Calling setxkbmap when starting X
On Sun, Sep 17, 2006 at 12:30:37AM +0300, Ivan Rambius Ivanov wrote: I am from Bulgaria and I use Bulgarian language on my FreeBSD machine. I use the following command $ setxkbmap -model pc105 -layout us,bg -variant ,phonetic -option grp:alt_shift_toggle to enable both Bulgarian and English. However, I call this command every time I log in KDE. Is it possible to invoke it automatically when X is started? I think what you're looking for is xinit(1). My own: $ cat ~/.xinitrc #!/bin/sh xmodmap .xmodmaprc xsetroot -solid dimgray xgamma -gamma 0.8 exec /usr/X11R6/bin/gnome-session Note also you can also define keyboard settings in rc.conf: $ grep keymap /etc/rc.conf keymap=us.iso.kbd.custom My own custom keymap is a quick hack to swap the Caps_Lock key with Escape for non-X uses (something that only vi users would appreciate). Alternatively, KDE, like Gnome, etc. most likely offers a mechanism to execute scripts at startup, but I'd advise against that approach. Hope that helped. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Freebsd, Suse Linux dual booting
FreeBSD and Linux people, I have a PC which I want to boot as windows, FreeBSD, and Suse 10.1 Linux. Currently, FreeBSD boot0 menu shows both Windows and FreeBSD as boot-able. The FreeBSD boot0 menu does not show the Linux OS (which I just installed). So, I did some reading of the FreeBSD handbook: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/boot-blocks.html It suggests that I have 2 ways to solve this problem: 1. Configure the FreeBSD boot0 menu so that it can boot Windows, FreeBSD, and Linux Or, 2. Replace The FreeBSD boot0 menu with LILO Boot Manager I like option 1. Q1: How do I add Suse 10.1 Linux to the FreeBSD boot0 menu? As for option 2, if I want to try LILO, I'll need to toss my FreeBSD boot0 menu in the trash. Q2: If I cannot get LILO to boot FreeBSD, how do I boot get FreeBSD to boot and then how do I restore my old FreeBSD boot0 menu? Thanks, -Dan ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Freebsd, Suse Linux dual booting
On Saturday, 16 September 2006 at 16:49:17 -0700, Dan Bikle wrote: FreeBSD and Linux people, I have a PC which I want to boot as windows, FreeBSD, and Suse 10.1 Linux. Currently, FreeBSD boot0 menu shows both Windows and FreeBSD as boot-able. The FreeBSD boot0 menu does not show the Linux OS (which I just installed). So, I did some reading of the FreeBSD handbook: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/boot-blocks.html It suggests that I have 2 ways to solve this problem: 1. Configure the FreeBSD boot0 menu so that it can boot Windows, FreeBSD, and Linux Or, 2. Replace The FreeBSD boot0 menu with LILO Boot Manager I like option 1. Q1: How do I add Suse 10.1 Linux to the FreeBSD boot0 menu? That depends on how you have laid out your Linux partition. Given that you have three systems on the disk, you have almost certainly put Linux in a BIOS extended partition. If that's the case, you can't use the FreeBSD boot manager, because it doesn't handle extended partitions. As for option 2, if I want to try LILO, I'll need to toss my FreeBSD boot0 menu in the trash. You also have the option of GRUB, which is what I used in this situation. See http://www.lemis.com/grog/diary-apr2006.html#21 for further details. Q2: If I cannot get LILO to boot FreeBSD, how do I boot get FreeBSD to boot and then how do I restore my old FreeBSD boot0 menu? Save the very first sector of the disk somewhere: # dd if=/dev/ad0 of=bootsector count=1 To restore it, you'll need to somehow boot, of course (I'd recommend FreesBIE (http://www.freesbie.org/), and copy it back: # dd if=bootsector of=/dev/ad0 count=1 Greg -- When replying to this message, please copy the original recipients. If you don't, I may ignore the reply or reply to the original recipients. For more information, see http://www.lemis.com/questions.html See complete headers for address and phone numbers. pgpz34x19QJOY.pgp Description: PGP signature
Re: Freebsd, Suse Linux dual booting
On 17/09/06, Dan Bikle [EMAIL PROTECTED] wrote:
FreeBSD and Linux people,
I have a PC which I want to boot as windows, FreeBSD, and Suse 10.1 Linux.
Currently, FreeBSD boot0 menu shows both Windows and FreeBSD as boot-able.
The FreeBSD boot0 menu does not show the Linux OS (which I just
installed).
So, I did some reading of the FreeBSD handbook:
http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/boot-blocks.html
It suggests that I have 2 ways to solve this problem:
1. Configure the FreeBSD boot0 menu so that it can boot
Windows, FreeBSD, and Linux
Or,
2. Replace The FreeBSD boot0 menu with LILO Boot Manager
I like option 1.
Q1: How do I add Suse 10.1 Linux to the FreeBSD boot0 menu?
As for option 2,
if I want to try LILO, I'll need to toss my FreeBSD boot0 menu in the
trash.
Q2: If I cannot get LILO to boot FreeBSD, how do I boot get
FreeBSD to boot and then how do I restore my old FreeBSD boot0 menu?
AFAIK, FreeBSD's boot loader cannot be configured, but merely loads the OSes
detected when it is run. If it does not detect something, you're out of
luck.
A better option than LILO is GRUB, which is installed by default by SUSE
10.x. XP will probably be detected by the installlation program, but if not,
here's how to add both XP and FreeBSD to the menu:
Edit the file /boot/grub/menu.1st. Create new entries as follows.
# The following entries assume that Windows XP is on drive 0, partition 0
(/dev/hda1 in Linux, /dev/ad0s1 in FBSD), with SuSE Linux on drive 0,
partition 1 (/dev/hda2 or /dev/ad0s2), and FreeBSD on drive 1, partition 0
(/dev/hdb1, /dev/ad1s0a)
title=WindowsXP
root (hd0,0)
chainloader +1
title=SuSE Linux 10.1
root (hd0,1)
kernel={the correct parameters should already be here}
title=FreeBSD 6.1
root (hd1,0,a)
chainloader +1
# menu.1st ends here
HTH,
Jeff.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: rebooting into single user mode on a remote server
On Sep 17, 2006, at 2:47 AM, [EMAIL PROTECTED] wrote: Hello, could somebody help me to understand the best way to enter into a single user mode on a remote server. I need it for the moment, during rebuilding world, when I have to reboot into single user mode before 'mergemaster -p'. I had this same issue last week... fortunately, my hosting provider had a remote KVM solution and hooked it up to my server while I got the job done. btw, that provider was m5hosting.com. I originally found them from the freebsd.org community page and have been very happy with their knowledge and support. good luck, ke han The only solution I found so far is to do 'shutdown -r now' and when the server boots to login with ssh and do 'shutdown now' - which should drop it to single user mode. I can ask the support at the hosting location to reboot in single user mode, but I do not know if I will have ssh then? Alternatively I can ask them to do the last few steps. Thank you for your advises, Iv. -- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions- [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Firefox and flash revisited. Don't shoot me for this post.
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 OK, let me preface this by saying flash 7 was marked broken for a reason (a very good one at that). I'm probably going to be stepping on some toes here and I'd like to apologize in advance. I've managed to install the broken port by changing a couple of files and using macromedia's own package. It was a no-brainer. I don't recommend doing this by any means. I've carried out this installation for two reasons, one was for development on an internal machine that never touches the net. Second because of the sheer volume of people asking about how to do this, I was curious to see if it could be done. I've detailed the process here: http://altbit.org/?p=207 Let me say again, that this is a VERY bad idea. There are serious implications in installing this port. Modification of the port's files could probably break something that I don't completely understand as well. Use this at your own risk. One last time, I'd like to apologize for stepping on any toes, but I understand the annoyance of this port not functioning. Thank you. cmh - -- C.M. Hobbs, KD5RYO http://altbit.org -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFFDJjv+HhXKrh8irARAl5rAKDNAtAcmboCLhBawun3yZWn/2hkxQCeJjp6 jmoeWa0aHyrH8rdNYBz46Y8= =MAb0 -END PGP SIGNATURE- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Freebsd, Suse Linux dual booting
People,
this is great info; thanks for taking time to type it up.
I'm now convinced that Grub is good.
On my FreeBSD box I see this:
bash moibsd maco /usr/home/maco 3 $
bash moibsd maco /usr/home/maco 3 $
bash moibsd maco /usr/home/maco 3 $
bash moibsd maco /usr/home/maco 3 $
bash moibsd maco /usr/home/maco 3 $ cat /etc/fstab
# DeviceMountpoint FStype Options Dump
Pass#
/dev/ad8s3b noneswapsw 0 0
/dev/ad8s3a / ufs rw 1 1
##/dev/ad8s4a /u1 ufs rw 1 1
/dev/acd0 /dvd1 cd9660 ro,noauto 0 0
/dev/acd1 /dvd2 cd9660 ro,noauto 0 0
linprocfs /compat/linux/proc linprocfs rw 0 0
bash moibsd maco /usr/home/maco 4 $
bash moibsd maco /usr/home/maco 4 $
bash moibsd maco /usr/home/maco 4 $
bash moibsd maco /usr/home/maco 4 $
bash moibsd maco /usr/home/maco 4 $ df
bash moibsd maco /usr/home/maco 4 $ df
bash moibsd maco /usr/home/maco 4 $ df
bash moibsd maco /usr/home/maco 4 $ df
Filesystem 1K-blocks UsedAvail Capacity Mounted on
/dev/ad8s3a 91913630 37443012 4711752844%/
devfs 110 100%/dev
linprocfs 440 100%/usr/compat/linux/proc
bash moibsd maco /usr/home/maco 5 $
bash moibsd maco /usr/home/maco 5 $
Comparing that with the information in the mail list
and this page:
http://www.lemis.com/grog/diary-apr2006.html#21
suggests to me,
that this Grub entry would be appropriate:
title FreeBSD 5.5
root (hd8,2,a)
kernel /boot/loader
Anyone care to confirm before I pull the plug on my FreeBSD boot0 menu?
-Dan
On 9/16/06, Jeff Rollin [EMAIL PROTECTED] wrote:
On 17/09/06, Dan Bikle [EMAIL PROTECTED] wrote:
FreeBSD and Linux people,
I have a PC which I want to boot as windows, FreeBSD, and Suse 10.1Linux.
Currently, FreeBSD boot0 menu shows both Windows and FreeBSD as
boot-able.
The FreeBSD boot0 menu does not show the Linux OS (which I just
installed).
So, I did some reading of the FreeBSD handbook:
http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/boot-blocks.html
It suggests that I have 2 ways to solve this problem:
1. Configure the FreeBSD boot0 menu so that it can boot
Windows, FreeBSD, and Linux
Or,
2. Replace The FreeBSD boot0 menu with LILO Boot Manager
I like option 1.
Q1: How do I add Suse 10.1 Linux to the FreeBSD boot0 menu?
As for option 2,
if I want to try LILO, I'll need to toss my FreeBSD boot0 menu in the
trash.
Q2: If I cannot get LILO to boot FreeBSD, how do I boot get
FreeBSD to boot and then how do I restore my old FreeBSD boot0 menu?
AFAIK, FreeBSD's boot loader cannot be configured, but merely loads the
OSes detected when it is run. If it does not detect something, you're out of
luck.
A better option than LILO is GRUB, which is installed by default by SUSE
10.x. XP will probably be detected by the installlation program, but if
not, here's how to add both XP and FreeBSD to the menu:
Edit the file /boot/grub/menu.1st. Create new entries as follows.
# The following entries assume that Windows XP is on drive 0, partition 0
(/dev/hda1 in Linux, /dev/ad0s1 in FBSD), with SuSE Linux on drive 0,
partition 1 (/dev/hda2 or /dev/ad0s2), and FreeBSD on drive 1, partition 0
(/dev/hdb1, /dev/ad1s0a)
title=WindowsXP
root (hd0,0)
chainloader +1
title=SuSE Linux 10.1
root (hd0,1)
kernel={the correct parameters should already be here}
title=FreeBSD 6.1
root (hd1,0,a)
chainloader +1
# menu.1st ends here
HTH,
Jeff.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: 28.8kbs/56kbs modems
On Sep 16, 2006, at 3:29 PM, David Fontenot wrote: To whom it may concern, Currently, I am using Ubuntu Linux 6.06 and it is really a let- down after I got it when I realized that Ubuntu does not do well with 28.8kbs/56kbs modems. It will not let me use my modem. Cheap is the buzzword for internal modems. Windows only is a common way to make cheap modems, aka, winmodem. I have never used FreeBSD with a winmodem but understand there is a way to use some models. Has been many years since, but have used FreeBSD over dialup external modem with many years of success. Generally one finds better support for Windows-specific hardware with Linux than FreeBSD. Linux seems to want badly to supplant Microsoft Windows and to that goal developers will work to equal every minutia. FreeBSD says, Bill who?, Bill Joy? I was wondering how Free-bsd does with dial-up modems (2 year old computer) and highspeed interenet, (I might get high speed soon). Unless things have changed, FreeBSD works perfectly with external modems using PPP protocol to your ISP. P.S. If my family did share a high speed internet connection, could I still connect to their network and share the internet, even if they are both using Windows XP? Yes. Either an XP machine can share its internet connection (presumably you will use ethernet) or your FreeBSD system can do the same for the others. Internet is not yet a Microsoft-proprietary protocol, quite the opposite as Unix shares its internet protocols with Microsoft. Sent from MacOS X thru a shared network using a FreeBSD gateway. -- David Kelly N4HHE, [EMAIL PROTECTED] Whom computers would destroy, they must first drive mad. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Freebsd, Suse Linux dual booting
On 17/09/06, Dan Bikle [EMAIL PROTECTED] wrote: People, this is great info; thanks for taking time to type it up. I'm now convinced that Grub is good. On my FreeBSD box I see this: bash moibsd maco /usr/home/maco 3 $ bash moibsd maco /usr/home/maco 3 $ bash moibsd maco /usr/home/maco 3 $ bash moibsd maco /usr/home/maco 3 $ bash moibsd maco /usr/home/maco 3 $ cat /etc/fstab # DeviceMountpoint FStype Options Dump Pass# /dev/ad8s3b noneswapsw 0 0 /dev/ad8s3a / ufs rw 1 1 ##/dev/ad8s4a /u1 ufs rw 1 1 /dev/acd0 /dvd1 cd9660 ro,noauto 0 0 /dev/acd1 /dvd2 cd9660 ro,noauto 0 0 linprocfs /compat/linux/proc linprocfs rw 0 0 bash moibsd maco /usr/home/maco 4 $ bash moibsd maco /usr/home/maco 4 $ bash moibsd maco /usr/home/maco 4 $ bash moibsd maco /usr/home/maco 4 $ bash moibsd maco /usr/home/maco 4 $ df bash moibsd maco /usr/home/maco 4 $ df bash moibsd maco /usr/home/maco 4 $ df bash moibsd maco /usr/home/maco 4 $ df Filesystem 1K-blocks UsedAvail Capacity Mounted on /dev/ad8s3a 91913630 37443012 4711752844%/ devfs 110 100%/dev linprocfs 440 100%/usr/compat/linux/proc bash moibsd maco /usr/home/maco 5 $ bash moibsd maco /usr/home/maco 5 $ Comparing that with the information in the mail list and this page: http://www.lemis.com/grog/diary-apr2006.html#21 suggests to me, that this Grub entry would be appropriate: title FreeBSD 5.5 root (hd8,2,a) kernel /boot/loader Dang, I always mix up XP and FBSD syntax. Yes, that looks fine. Good luck! Jeff ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Firefox and flash revisited. Don't shoot me for this post.
Christopher Hobbs wrote: OK, let me preface this by saying flash 7 was marked broken for a reason (a very good one at that). I'm probably going to be stepping on some toes here and I'd like to apologize in advance. I've managed to install the broken port by changing a couple of files and using macromedia's own package. It was a no-brainer. I don't recommend doing this by any means. I've carried out this installation for two reasons, one was for development on an internal machine that never touches the net. Second because of the sheer volume of people asking about how to do this, I was curious to see if it could be done. I've detailed the process here: http://altbit.org/?p=207 To get the distinfo file, just run 'make makesum', it's a lot easier. Commenting out 'RESTRICTED' is not required. Let me say again, that this is a VERY bad idea. If it's a bad idea then why do you tell people how to do it? You could perhaps just send in a PR with your changes and the maintainer and the committers will take a look at it. --jona ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Firefox and flash revisited. Don't shoot me for this post.
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Jona Joachim wrote: Christopher Hobbs wrote: OK, let me preface this by saying flash 7 was marked broken for a reason (a very good one at that). I'm probably going to be stepping on some toes here and I'd like to apologize in advance. I've managed to install the broken port by changing a couple of files and using macromedia's own package. It was a no-brainer. I don't recommend doing this by any means. I've carried out this installation for two reasons, one was for development on an internal machine that never touches the net. Second because of the sheer volume of people asking about how to do this, I was curious to see if it could be done. I've detailed the process here: http://altbit.org/?p=207 To get the distinfo file, just run 'make makesum', it's a lot easier. Commenting out 'RESTRICTED' is not required. Let me say again, that this is a VERY bad idea. If it's a bad idea then why do you tell people how to do it? You could perhaps just send in a PR with your changes and the maintainer and the committers will take a look at it. --jona Thanks for the 'make makesum' and RESTRICTED tips, while I know my way around the ports system, there are still a few odds and ends that I don't completely understand. I need to read more about it. As for the PR, I didn't actually change anything but the Makefile and distinfo. If I could, I'd do something useful with flash but it's not open source, after all. I posted it because I know the feeling of frustration when something won't work. I prefaced it with the whole bad idea spill because it can still be used for internal purposes like I had previously mentioned (i.e. the development box at my office that never touches the net). I'm really sorry if I'm coming off wrong here. I'm probably not following etiquette too well. I'll look into a PR, though I don't know that it'll do much good. I'm not looking to start any wars, I just wanted to help out. cmh - -- C.M. Hobbs, KD5RYO http://altbit.org -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFFDK+g+HhXKrh8irARAm13AKCqtbZiw9uoNabRt8sSmZpdDT4SkACgoWtv S3FvOl9OlxKO1IZgp6IxShU= =9stO -END PGP SIGNATURE- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
wg311t and via chipset
I've run onto a problem with a netgear wg311t and a msi k8 neo with a via chip. I've placed the card into all the pci slots removed all other pci cards and still it refuses to work for 6.1 bsd or m$. I've placed the card into an older pc and it works for 6.1 bsd with no problems. pciconf -lv shows that the card is not there, it lights one of the tx led in the back but thats the only life it shows in the msi neo. I've searched the netgear site and found an article about needing the via 4-in-1 drivers for the via chip, installed them but no joy even in windows. Are there any hints for the sys.ctl file to aid the use if this card?? Someone else must be running this combo, if so how?? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: rebooting into single user mode on a remote server
In order to follow the upgrade instructions in the Handbook or /usr/src/UPDATING to the letter, you need console access to the machine being updated. That is [a] problem ... when it's in a hosting centre umpty dozen miles away ... There are essentially three possibilities. i) get someone local to the machine to do the bits requiring the console access ... ii) arrange to get remote console access. That can be expensive if you go down the route of buying a dedicated console server. Or it can be very cheap indeed if you have another FreeBSD box close by the machine you're trying to update and you can string null modem cables between their serial ports ... iii) Finally, and not to be dismissed without due consideration, is the really quite simple approach of /not/ taking the machine down to single user mode ... iv) (actually a variant of ii, but different enough to warrant separate mention IMO) Put a PC Weasel or similar in any machine that is going to be located remotely. This card looks like a VGA to the machine, but allows for remote access. The simple ones support only text mode via a serial port; some of the fancier ones act as X11 clients so as to also support graphics modes. This gives you access not only to the FreeBSD console, but to the BIOS. And no, I do not work for any manufacturer or supplier of such. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: top(1) STATE column
In the last episode (Sep 16), Pietro Cerutti said: I'd like to know the meaning of the possible STATEs showing up in top. In the manual pages I found this: STATE is the current state (one of START, RUN (shown as CPUn on SMP systems), SLEEP, STOP, ZOMB, WAIT, LOCK or the event on which the process waits) Where can I found info about other possible states (nanslp, kserel, ttyin, ucond, sbwait, ...) that I usually see in top? I think these have to do with the the event on which the process waits part of the man page... isn't there any complete list on those? They're only documented in the source, as far as I know. A quick grep comes up with around 300 different unique waits and mutexes in the kernel: find /sys -name *.c | xargs grep 'sleep(.*.*' | sed -e 's/^.*\\(.*\)\.*$/\1/' | sort -u | wc -l -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: When is BuildWorld necessary?
Bob wrote: On Saturday 16 September 2006 15:52, [EMAIL PROTECTED] wrote: But I have one question - do you rebuild the world on a remote machine Sorry; I am a newbie at FreeBSD, and have never done a buildworld :-( I have spent lots of time on Linux, Solaris, and SCO, but this is my first cut at BSD. Just from past NIX experience though, I would never rebuild an entire OS remotely without having someone onsite to push the On/Off switch when the inevitable happens :-( We have someone to push the switch. I just thought if it is possible to be done without engaging the support. Iv. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: rebooting into single user mode on a remote server
Matthew Seaman wrote: [EMAIL PROTECTED] wrote: Hello, could somebody help me to understand the best way to enter into a single user mode on a remote server. I need it for the moment, during rebuilding world, when I have to reboot into single user mode before 'mergemaster -p'. The only solution I found so far is to do 'shutdown -r now' and when the server boots to login with ssh and do 'shutdown now' - which should drop it to single user mode. I can ask the support at the hosting location to reboot in single user mode, but I do not know if I will have ssh then? Alternatively I can ask them to do the last few steps. Yep. You've become the latest person to realise this perennial problem. In order to follow the upgrade instructions in the Handbook or /usr/src/UPDATING to the letter, you need console access to the machine being updated. That is no problem when the machine is on your desk, or probably not if it's just down the hall. But when it's in a hosting centre umpty dozen miles away and you can't actually get to it? There are essentially three possibilities. i) You've thought of this approach already: get someone local to the machine to do the bits requiring the console access. That works if the people at the other site are competent and trustworthy, and you can afford to pay for their time. ii) The next solution, and on the whole, probably the best solution available, is to arrange to get remote console access. That can be expensive if you go down the route of buying a dedicated console server. Or it can be very cheap indeed if you have another FreeBSD box close by the machine you're trying to update and you can string null modem cables between their serial ports. Then you configure your FreeBSD box requiring update to use ttya as its console and use tip(1) to get into it from the other machine. (Actually, you could probably make that approach work from any other unixoid OS or even from Windows so long as you can find the right serial console emulation software). If you're really lucky, you're running flashy new hardware with IPMI or similar lights out management capability and can get into the machine through that. It doesn't work in anything like the same way as a serial console, but the end result is just as good. iii) Finally, and not to be dismissed without due consideration, is the really quite simple approach of /not/ taking the machine down to single user mode. Most of the time, you can quite happily run 'make installworld' or 'make installkernel' or 'mergemaster' while the system is in multiuser mode. You should shutdown all active services except what you need to get in remotely and you should kick any other users off the machine as well as generally taking steps to ensure the machine is as quiescent as possible before trying that. You should also have a 'back to square one' plan for dealing with the eventuality that the machine does not come back after attempting to reboot into the new kernel -- you really absolutely will require someone quite FreeBSD savvy to get onto the console to unfuck things if so, and that illustrates the big drawback to this approach: if it goes wrong, you are truly left up a gum tree without a paddle. Don't try approach (iii) for an upgrade over too many version numbers at once. Jumping from, say 6.1-RELEASE to 6.1-RELEASE-p6 should be feasible, as should jumping from 6.0-RELEASE to 6.1-RELEASE. Going from say 5.5-RELEASE to 6.1-RELEASE is only for the brave or the most highly skilled, and anything more than that is only for the foolhardy. Neither is it a good idea to do method (iii) if you're making any major changes to the hardware on the system. Nor does approach (iii) mix at all well with the use of raised secure levels. Cheers, Matthew Matthew, thanks (and all others) for the detailed reply. The possibilities are now kind of clear to me and I'll have to work out which one I can implement best. Thanks a lot again, Iv ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: When is BuildWorld necessary?
Laurence Sanford wrote: [EMAIL PROTECTED] wrote: But I have one question - do you rebuild the world on a remote machine (without physical access) and if yes - how do you restart in single user mode. This is what I can't understand so far. Thanks, Iv In 6 years, I've never dropped any machine to single user to do any part of a buildworld upgrade. I've stopped many running services, but never gone to single user. The only time I had any problems with this approach was when I blindly flubbed versions in my supfile and cvsup'd a 6 system with 4 source. That wasn't pretty. But it would have been not pretty in single user mode as well. I heard this from another place as well. It just sounds too scary for me at the moment... But may be when I feel more comfortable with the things and/or there is no other way. Thanks anyway for pointing that out! Iv. -- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: When is BuildWorld necessary?
--On September 17, 2006 6:18:24 AM +0200 [EMAIL PROTECTED] wrote: Bob wrote: On Saturday 16 September 2006 15:52, [EMAIL PROTECTED] wrote: But I have one question - do you rebuild the world on a remote machine Sorry; I am a newbie at FreeBSD, and have never done a buildworld :-( I have spent lots of time on Linux, Solaris, and SCO, but this is my first cut at BSD. Just from past NIX experience though, I would never rebuild an entire OS remotely without having someone onsite to push the On/Off switch when the inevitable happens :-( We have someone to push the switch. I just thought if it is possible to be done without engaging the support. No one has mentioned the security/freebsd-update port. With that you can apply updates to the kernel and world without having to build them *if* (and only if!) you are running a GENERIC kernel. For remote administration, this may be a good option for some. I've done a number of build world and kernel routines without a problem. make buildworld make buildkernel make installkernel reboot mergemaster -p make install world mergemaster reboot This has worked for me on three different systems, all of which are easily accessible if something goes wrong. I have one server that's about 20 miles away and much more critical than the others (in terms of uptime and accessibility) *and* I don't have remote access to the server through a KVM or similar. For that one I use freebsd-update, because I don't want to have to suddenly jump in the car and drive 30 minutes (while the server is down) to fix a problem. Paul Schmehl ([EMAIL PROTECTED]) Adjunct Information Security Officer The University of Texas at Dallas http://www.utdallas.edu/ir/security/
