Re: best way to update ports
yeah, in that situation nfs mount will be easy. My servers are in different cities, and the ports are installed with different options on different servers, for example, some postfix use unix login accounts, some postfix use courier authentication with mysql database. So unfortunately I can't share the same ports among them. In a simelar situation I did this by setting up a centeral cvsup mirror (allows you to maintain patches between cvsupdates [i.e. if the file does not differ rcs stamp will not clobber it]). So you always have the most upto date to build against. I am not sure on how to handle the manual security patches on there first application but once applied until the effect files are updated you will not have to reapply the patches. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: rebranding a i386 binary to be a amd64 binary
Aryeh Friedman wrote: Even though I know this is asking for it I want to test the new nVidia driver on amd64 and the only issue with a hand compile (from nVidia's tar not the ports one) is src/nv-kernel.o is branded elf-i386-32 and amd64 wants it branded elf-amd64-64. This file comes from them as a precompiled object so rebranding seems to be my only option. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Uh it really is an i386 binary, rebranding won't magically change all the code. Kris ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: rebranding a i386 binary to be a amd64 binary
On 10/11/07, Kris Kennaway [EMAIL PROTECTED] wrote: Aryeh Friedman wrote: Even though I know this is asking for it I want to test the new nVidia driver on amd64 and the only issue with a hand compile (from nVidia's tar not the ports one) is src/nv-kernel.o is branded elf-i386-32 and amd64 wants it branded elf-amd64-64. This file comes from them as a precompiled object so rebranding seems to be my only option. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Uh it really is an i386 binary, rebranding won't magically change all the code. All I want to do is make it compile so I can test it (like I said I know it is inherently dangerous) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: rebranding a i386 binary to be a amd64 binary
On Oct 11, 2007, at 1:34 AM, Aryeh Friedman wrote: On 10/11/07, Kris Kennaway [EMAIL PROTECTED] wrote: Aryeh Friedman wrote: Even though I know this is asking for it I want to test the new nVidia driver on amd64 and the only issue with a hand compile (from nVidia's tar not the ports one) is src/nv-kernel.o is branded elf-i386-32 and amd64 wants it branded elf-amd64-64. This file comes from them as a precompiled object so rebranding seems to be my only option. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Uh it really is an i386 binary, rebranding won't magically change all the code. All I want to do is make it compile so I can test it (like I said I know it is inherently dangerous) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] It's not inherently dangerous, it's not possible. What you need to do is disassemble the code(probably illegal), port it to amd64, and reassemble it. It would require intimate knowledge of i386 asm and amd64 asm. The architectures are more different than you might expect. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: rebranding a i386 binary to be a amd64 binary
Aryeh Friedman wrote: On 10/11/07, Kris Kennaway [EMAIL PROTECTED] wrote: Aryeh Friedman wrote: Even though I know this is asking for it I want to test the new nVidia driver on amd64 and the only issue with a hand compile (from nVidia's tar not the ports one) is src/nv-kernel.o is branded elf-i386-32 and amd64 wants it branded elf-amd64-64. This file comes from them as a precompiled object so rebranding seems to be my only option. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Uh it really is an i386 binary, rebranding won't magically change all the code. All I want to do is make it compile so I can test it (like I said I know it is inherently dangerous) It aint gonna work and there is no tool to do it because it aint gonna work. Kris ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: rebranding a i386 binary to be a amd64 binary
Aryeh Friedman wrote: On 10/11/07, Kris Kennaway [EMAIL PROTECTED] wrote: Aryeh Friedman wrote: Even though I know this is asking for it I want to test the new nVidia driver on amd64 and the only issue with a hand compile (from nVidia's tar not the ports one) is src/nv-kernel.o is branded elf-i386-32 and amd64 wants it branded elf-amd64-64. This file comes from them as a precompiled object so rebranding seems to be my only option. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Uh it really is an i386 binary, rebranding won't magically change all the code. All I want to do is make it compile so I can test it (like I said I know it is inherently dangerous) There are several kernel features that need to be implemented before an amd64 nVidia driver will work - see http://wiki.freebsd.org/NvidiaFeatureRequests for more info. nVidia do want to create an amd64 driver, but they need the kernel work to be done first. -- Bruce ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: rebranding a i386 binary to be a amd64 binary
There are several kernel features that need to be implemented before an amd64 nVidia driver will work - see http://wiki.freebsd.org/NvidiaFeatureRequests for more info. nVidia do want to create an amd64 driver, but they need the kernel work to be done first. I just sent nVidia an offer to do the kernel work ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: shooting oneself in the foot with ldconfig -v
The previously configured directory list was fully populated, so effectively there should have been no change as the previously configured directories were untouched and I specified no additional pathnames. ... Are you saying that by specifying -v I no longer satisfied the no parameters are given clause and ended up in a default place in the logic? That wasn't actually what I was saying, but after checking the source code it turns out you are right and that is exactly what happens. ... IMHO a verbose switch shouldn't change behavior; it should just spam the console a lot. True. Current behavior sounds like, at best, a LOLA violation. Perhaps the OP would consider submitting a PR. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Nedit after xorg 7.3
On Wednesday 10 October 2007, [EMAIL PROTECTED] wrote: Hi list, After upgrading xorg to 7.3, nedit has started scrolling funnily on my system. It used to work just fine, but after the upgrade scrolling down (that is moving the text cursor up) one line at a time using the arrow keys replicates the same line on each text line in view. Scrolling down still works fine?!? Page-up and -down works fine as always. Has anyone else experienced this and perhaps even solved the issue?? This is a known issue. As a workaround you can disable the Composite extension in xorg.conf: Section Extensions Option Composite Disable EndSection HTH, Pieter de Goeje ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
FreeBSD Conferences 2008
Where can I get the FreeBSD conference/event schedule for 2008? I think I am most interested in BSDCan, EuroBSDCon - they seem to be the best. Can anybody recommend any other? Riaan Kruger PS. Sorry for asking 2 questions in one email. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: best way to update ports
On Thursday 11 October 2007 07:33:43 Bill Stwalley wrote: I need your advice on how to update security patches for ports on a dozen servers with minimal efforts. As I gathered, I should run portaudit in cron jobs and then manually update the ports with vulnerabilities after reading UPDATING. Is this the best way? Is this manual way feasible for managing a dozen servers? I used to run portupgrade in cron jobs, but that created too much nightmare. For example, imap-uw broke for a few days recently. Use a tinderbox buildbox, specifically read the part on `Customizing the Environment' and `configuring port OPTIONS' at http://tinderbox.marcuscom.com/README.html The only problem left then is that you still need to manually deploy the binary packages to the servers in case of UPDATING woes. However with a bit of scripting, you can batch this on a case-by-case base. The good part is that you have all things on one machine, know when builds are broken before they get deployed and can test packages to see if they break your applications in a test environment. As a side note: portaudit has a periodic script that installs in /usr/local/etc/periodic/security - you can enable it in /etc/periodic.conf so it's part of the daily security report (I think it's even on by default). -- Mel ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Booting a GELI encrypted hard disk
On Wed, 10 Oct 2007 23:09:10 +0200
Roland Smith [EMAIL PROTECTED] wrote:
On Wed, Oct 10, 2007 at 02:34:16PM -0400, Steve Bertrand wrote:
If you encrypted / and /usr, you might actually make the system more
vulnerable to a known-plaintext attack, because there are a lot of files
with well-known contents there.
I can get away with not having / encrypted, but I need /var encrypted
for databases and logs etc, /tmp so any temporary files are secured and
the swap file (swap very rarely gets used).
You can even encrypt /tmp with a one-time key (see 'geli onetime').
don't forget to do the same with your swap partition :) It may be a bit slower,
but your swap would have quite a bit of interesting info if your system used it.
In my rc.conf, i have :
# ENCRYPTED SWAP PARTITION - OPTIONS
geli_swap_flags=-e aes -l 192 -s 4096 -d
and /etc/fstab reads:
# DeviceMountpoint FStype Options DumpPass#
/dev/ad0s1b.eli noneswapsw 0 0
[]
However using a USB device presents it's own problems. If you plug-in a
USB stick there's no telling which device node it ends up with,
depending on how many other USB devices are on the bus. To make device
recognition easier, you should use a GEOM label on the USB stick, so
you'll know which /dev/label/* device node it gets. And you'd probably
have to hack an rc script to mount the USB stick _before_ the system
tries to attach the GELI device(s).
[...]
And remember that this USB stick is another thing you have to back-up
and store in a safe place. It would be bad if you lost your data because
your USB stick died or got lost.
hmm I find it much easier to have my normal partitions in clear text, and
then have big files (4 , 8 Gb) which I attach as a device node and mount as
part of my normal directory tree.[1]
Why do this? well, for a number of reason that work for me :
- my backups are VERY simple. Unmount my encrypted disks, back up the lot. How
do you, safely and with commonly available tools, backup a fully encrypted
partition? (yes, you could mount it, backup and encrypt the backup but
that's doubling up too much)
- I can take any of my disks and mount it in ANY other freebsd computer with
minimal fuss ( even a freebsd VM).
- i find the whole thing about having to have the USB @ boot time a bit of a
pain.
- same goes for keeping good, safe backups of USB keys...
- it works pretty well for specific work... eg, you want to have all your DB
data kept safely - make the postgresql script depend on your script and you
will be prompted for it on startup.
totally paranoid
BTW, is any work done on plausible deniability for geli? such as truecrypt's
feature :
http://www.truecrypt.org/docs/?s=plausible-deniability
/tp
thanks!!
[1]
-
#!/bin/sh
sudo mdconfig -d -u 11
sudo mdconfig -a -t vnode -f ~/blah.dsk -u 11 \
sudo geli attach /dev/md11 \
echo Running fsck... \
sudo fsck -p -t ufs /dev/md11.eli \
sudo geli detach -l /dev/md11 \
sudo mount -o noatime /dev/md11.eli ~/blah
sudo chown betom:betom ~/blah
--
and, the first time, to create blah.dsk:
dd if=/dev/random of=blah.dsk bs=1024 count=5
mdconfig -a -t vnode -f ./blah.dsk -u 13
sudo geli init -e AES -l 256 -s 4096 /dev/md13
sudo geli attach /dev/md13
sudo newfs -U /dev/md13.eli
_
{Beto|Norberto|Numard} Meijome
I've dirtied my hands writing poetry, for the sake of seduction; that is, for
the sake of a useful cause. Dostoevsky
I speak for myself, not my employer. Contents may be hot. Slippery when wet.
Reading disclaimers makes you go blind. Writing them is worse. You have been
Warned.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
How to create a user account with the same permission as root ?
Finally, I manage to setup X.org and then KDE 3.5.4 running on FreeBSD 6.2-Release. I created a user account named william and do not assign any group as I do not know what are the list of group name for me to select. To start KDE, i use command kdm but I can only logon using the newly created user name william, but it do not have same permission/access rights as root account. Please show on how to enable this user account, with the same permission as root ? Thank you. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
On 10/11/07, williamkow [EMAIL PROTECTED] wrote: Finally, I manage to setup X.org and then KDE 3.5.4 running on FreeBSD 6.2-Release. I created a user account named william and do not assign any group as I do not know what are the list of group name for me to select. To start KDE, i use command kdm but I can only logon using the newly created user name william, but it do not have same permission/access rights as root account. Please show on how to enable this user account, with the same permission as root ? run vipw from the command line and edit the entry's uid and gid (the 3rd and 4th fields) editing the password file directly is inherently so you should read all the related documentation and such (including the format and meaning of each field) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
On Thursday 11 October 2007 13:17:59 williamkow wrote: Finally, I manage to setup X.org and then KDE 3.5.4 running on FreeBSD 6.2-Release. I created a user account named william and do not assign any group as I do not know what are the list of group name for me to select. To start KDE, i use command kdm but I can only logon using the newly created user name william, No. Kdm only shows users for which $HOME/.kde exists in it's log on dialog. You *can* log on using 'root' if you simply type root as user name, rather then selecting it from the left pane of the log on dialog. It will then create a new .kde config for root. but it do not have same permission/access rights as root account. Real question is, do you need to? In the cases where you need to log on as root, use the root account, by all means do not create another user with user id 0. Normal operations should be done under normal user id, system administration with root. With KDE you can combine the two by using the 'root konsole' from your normal user if you need to do some administration - also there's various applications in the system menu from KDE that make you log on as root before it's started, for example KUser - User manager. -- Mel ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
Hi, FreeBSD is not Windows. You cannot have another root in the system. What you can do is the creation of the group wheel and put william into this group. Allow then all members of wheel to access the files needed by the group wheel. I would not do this as it creates many security wholes. If you just want to do something as root without being root, use su. Erich williamkow wrote: Finally, I manage to setup X.org and then KDE 3.5.4 running on FreeBSD 6.2-Release. I created a user account named william and do not assign any group as I do not know what are the list of group name for me to select. To start KDE, i use command kdm but I can only logon using the newly created user name william, but it do not have same permission/access rights as root account. Please show on how to enable this user account, with the same permission as root ? Thank you. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
can not load KDE in FreeBSD 6.2
After configuring to run X.org, and I manage to have KDE by running command kdm. Then I do some changes related to display, example : a) 1024x768 to 800x600 b) fonts size for menu, wallpaper ...etc but then after the computer have shutdown and restart and when I run the kdm command, and I unable to load KDE screen anymore, and it show the below error message, please advise what should I do to enabling the KDE. # kdm Updating KDM configuration Information: reading current kdmrc /usr/local/share/config/kdm/kdmrc (from kde = 2.2.x) Information: current kdmrc is from kde =3.1 (config version 2.3) # Oct 11 20:15:37 kdm-bin: :[692]: IO Error in XOpenDisplay Oct 11 20:15:37 kdm-bin[689]: Display :0 cannot be openned Oct 11 20:15:37 kdm-bin[689]: Unable to fire up local display :0; disabling. Please advise. Thank you. __ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: can not load KDE in FreeBSD 6.2
On 10/11/07, williamkow [EMAIL PROTECTED] wrote: After configuring to run X.org, and I manage to have KDE by running command kdm. Then I do some changes related to display, example : a) 1024x768 to 800x600 b) fonts size for menu, wallpaper ...etc but then after the computer have shutdown and restart and when I run the kdm command, and I unable to load KDE screen anymore, and it show the below error message, please advise what should I do to enabling the KDE. # kdm Updating KDM configuration Information: reading current kdmrc /usr/local/share/config/kdm/kdmrc (from kde = 2.2.x) Information: current kdmrc is from kde =3.1 (config version 2.3) # Oct 11 20:15:37 kdm-bin: :[692]: IO Error in XOpenDisplay Oct 11 20:15:37 kdm-bin[689]: Display :0 cannot be openned Oct 11 20:15:37 kdm-bin[689]: Unable to fire up local display :0; disabling. You have to run it from .xinitrc or use a login manager. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: can not load KDE in FreeBSD 6.2
On Thursday 11 October 2007 14:26:53 Aryeh Friedman wrote: On 10/11/07, williamkow [EMAIL PROTECTED] wrote: After configuring to run X.org, and I manage to have KDE by running command kdm. Then I do some changes related to display, example : a) 1024x768 to 800x600 b) fonts size for menu, wallpaper ...etc but then after the computer have shutdown and restart and when I run the kdm command, and I unable to load KDE screen anymore, and it show the below error message, please advise what should I do to enabling the KDE. # kdm Updating KDM configuration Information: reading current kdmrc /usr/local/share/config/kdm/kdmrc (from kde = 2.2.x) Information: current kdmrc is from kde =3.1 (config version 2.3) # Oct 11 20:15:37 kdm-bin: :[692]: IO Error in XOpenDisplay Oct 11 20:15:37 kdm-bin[689]: Display :0 cannot be openned Oct 11 20:15:37 kdm-bin[689]: Unable to fire up local display :0; disabling. You have to run it from .xinitrc or use a login manager. That's no help: - kdm *is* a login manager. - startkde is what you use from .xinitrc. - kdm can be run from command line without problems, or you can add it to /etc/ttys, but you need a working X config first or you'll create a nice loop. As for the problem, the X server doesn't start up correctly, view /var/log/Xorg.0.log for information. -- Mel ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: /usr/X11R6 before /usr/local in ldconfig?
Joshua Isom [EMAIL PROTECTED] writes: I noticed the other that that in -STABLE that /usr/X11R6 was in front of /usr/local for libraries. This results in any port that uses a library from another port to look for /usr/X11R6 first, and then /usr/local. I don't know if this would cause any real problems other than confusion for people, but with the Xorg upgrade that installs into /usr/local, is this just an oversight? I'm not too familiar with the details of how FreeBSD loads libraries(although I have noticed at least one peculiarity), so I don't know if any serious issues can happen(I renamed X11R6 and things worked ok). Preference between the two locations only matters if a particular library exists in different versions in both places. Once the two places have been consolidated, this is impossible. Otherwise, it's still not something you want to do without a really good reason. So I don't see any issue to solve here. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Samba and Swat are not restarting the daemons.
Lisandro Grullon [EMAIL PROTECTED] writes: I have install samba version:3.0.26a from ports, the daemons appear to be working fine by enabling the apropiate parameters in rc.conf, yet I am speriencing the issue where SWAT is showing as the smbd and nmbd are not running nor will they restart. Can someone point me in the right direction of what is going on. Why is swat no allowing the process to be manipulated accordingly? Thanks in advance. Lisandro Grullon The actual problem is not really clear to me: Are you able to connect to the SMB shares? Are you able to connect to SWAT at all? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
On Thu, 11 Oct 2007, williamkow wrote: Finally, I manage to setup X.org and then KDE 3.5.4 running on FreeBSD 6.2-Release. I created a user account named william and do not assign any group as I do not know what are the list of group name for me to select. To start KDE, i use command kdm but I can only logon using the newly created user name william, but it do not have same permission/access rights as root account. Please show on how to enable this user account, with the same permission as root ? You cannot do exactly what you say you want to do. What you can do is create an alias for root that has its own home directory, choice of shells, etc. Look at how toor is set up in master.passwd. You can set up william like toor by editing master.passwd (always use vipw to edit master.passwd -- not a naked text editor -- but you can use the editor of your choice if you set the EDITOR environmental variable). But this makes william an alias of root, not another user with root permissions. (That is william must be UID=0, etc.) for example: william:encrypted password omitted:0:0::0:0:Bourne-again \ Superuser:/usr/home/william:/usr/local/bin/bash in master.passwd will take you to /usr/home/william when you log in as william, and your shell will be bash, but if you whoami, the answer is root. Everything that goes by UID will identify you as root. Basically because you are root. There are tons of reasons why this is a very bad idea, and you will probably hear most of them, but they boil down to this: You should not run as root. You should should be acutely aware of when you are doing something as root, and you should do as little as you possibly can as root. You can put an ordinary user william in the wheel group so he can assume root-like powers when necessary, but when unnecessary he shouldn't. There are also some okay reasons such as wanting a different shell or home directory for your root alias. -- Lars Eighner http://www.larseighner.com/index.html 8800 N IH35 APT 1191 AUSTIN TX 78753-5266 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
After seeing some of his other questions he should get a book on basic sysadmin also. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
On Thu, Oct 11, 2007 at 07:34:54PM +0800, Erich Dollansky wrote: Hi, FreeBSD is not Windows. You cannot have another root in the system. Yeah, you can. It's just a really bad idea. root and toor both have UID and GID of 0 - giving them both superuser privileges. There is nothing to prevent you from adding as many more UID/GID 0 users as your madness compels you to. The only stricture is that they must all have different names. What you can do is the creation of the group wheel and put william into this group. Group wheel already exists - it is root's (and toor's) primary group. William: log in as root and run this: # pw user mod -n william -G wheel william will now be a member of wheel, and able to su root. Allow then all members of wheel to access the files needed by the group wheel. This step shouldn't be necessary on a standard install, as membership of group wheel confers access rights to all files owned by wheel. I would not do this as it creates many security wholes. Er..? It is a standard technique for allowing certain users to su root to perform system maintenance tasks. If I misunderstand your point, Erich, please do explain. If you just want to do something as root without being root, use su. For which, in FreeBSD, you need to be a member of group wheel anyway... security/sudo doesn't have this prerequirement, and is a much more flexible tool. But, that flexibility comes with a cost - you must configure it correctly, or you could end up shooting yourself in the foot. Dan williamkow wrote: Finally, I manage to setup X.org and then KDE 3.5.4 running on FreeBSD 6.2-Release. I created a user account named william and do not assign any group as I do not know what are the list of group name for me to select. To start KDE, i use command kdm but I can only logon using the newly created user name william, but it do not have same permission/access rights as root account. Please show on how to enable this user account, with the same permission as root ? Thank you. -- Daniel Bye _ ASCII ribbon campaign ( ) - against HTML, vCards and X - proprietary attachments in e-mail / \ pgpYtHYFFK7aX.pgp Description: PGP signature
sysutils/lockdown
Hello All, Is this still a valid working port for a FreeBSD 6.2 box? It looks as if it has not been touched since 2005. Is there anything else that does this? Thanks in advance, ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
On Thu, 2007-10-11 at 19:17 +0800, williamkow wrote: Finally, I manage to setup X.org and then KDE 3.5.4 running on FreeBSD 6.2-Release. I created a user account named william and do not assign any group as I do not know what are the list of group name for me to select. To start KDE, i use command kdm but I can only logon using the newly created user name william, but it do not have same permission/access rights as root account. Please show on how to enable this user account, with the same permission as root ? In my case, william is better than root. I can never login to my GNOME desktop through GDM by root. A root account is absolutely useless to me. -- Byung-Hee HWANG [EMAIL PROTECTED] Will you send this to him? Sure, sure. You forget about Mikey, he no the man for you anymore. -- Kay Adams and Mrs. Corleone, Chapter 15, page 235 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
In my case, william is better than root. I can never login to my GNOME desktop through GDM by root. A root account is absolutely useless to me. I use gnome and there is not a single thing I can not do with a root account that I can't with one. If you mean opening GUI apps try adding the following (I am assuming your using tcsh... if your using something else look up how to do this): in ~/.cshrc add: xhost + in /root/.cshr add: setenv DISPLAY :0 When you su *ALWAYS* use the following syntax: su - *DO NOT* include any cmd arguement. Once your root you should be able to start any GUI based app from the command line... if you don't know what the command line is left click on the menu/icon and go to properties... if it requires arguements look it up in the man page or the GNOME help for it. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
On Thu, Oct 11, 2007 at 07:34:54PM +0800, Erich Dollansky wrote: Hi, FreeBSD is not Windows. True statement - thank heaven. You cannot have another root in the system. Unless I misunderstand what you are saying, this is NOT a true statement. You can create as many ids with a '0' UID as you want. It may not be a good idea, but it works just fine. Then, if you give it the same home directory and shell, it will be almost impossible to distinguish how it functions from how the 'root' account functions. Now, if you mean having two accounts named root, then you can't have that, but that isn't what you imply by your following statement about creating an account called 'william'. Having said all that, doing part of what follows is better -- create a regular user account with its own UID (eg not 0) and then add it to the 'wheel' group by editint /etc/group file. But, then, do not make all files have group wheel permission. Instead, when you want to work on those files or other things root might do, use su(1) to change your working UID to '0' temporarily. That way, files will have normal owner and group, user will have normal UID and GID, and everything will work nicely. What you can do is the creation of the group wheel and put william into this group. Allow then all members of wheel to access the files needed by the group wheel. Not the best idea. I would not do this as it creates many security wholes. If you just want to do something as root without being root, use su. Yes, do this. I guess you rethought what you wrote about the files. jerry Erich williamkow wrote: Finally, I manage to setup X.org and then KDE 3.5.4 running on FreeBSD 6.2-Release. I created a user account named william and do not assign any group as I do not know what are the list of group name for me to select. To start KDE, i use command kdm but I can only logon using the newly created user name william, but it do not have same permission/access rights as root account. Please show on how to enable this user account, with the same permission as root ? Thank you. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
On 10/11/07, Aryeh Friedman [EMAIL PROTECTED] wrote: In my case, william is better than root. I can never login to my GNOME desktop through GDM by root. A root account is absolutely useless to me. I use gnome and there is not a single thing I can not do with a root account that I can't with one. this also works for other accounts for example I am a one person company and thus have to maintain the software I sell as well as the web site for it. for this reason I keep two accounts one for development and one for the web page and I su to them (yes I do GUI developement). ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Installing Thawte Certificate on imap pop smtp
I currently have self-signed certificates on our mail server, but they are now expired. I have just received the CA-crt back from thawte. I have the webmail portion completed with installing the certificates, but I am having some issues with getting them installed on SMTP. I tried to put them in as the *.crt and *.key files, but sendmail didn't like that. The temporary certificates installed are: define(`confCACERT_PATH', `/etc/mail/certs')dnl define(`confCACERT', `/etc/mail/certs/mycert.pem')dnl define(`confSERVER_CERT', `/etc/mail/certs/mycert.pem')dnl define(`confSERVER_KEY', `/etc/mail/certs/mykey.pem')dnl define(`confCLIENT_CERT', `/etc/mail/certs/mycert.pem')dnl define(`confCLIENT_KEY', `/etc/mail/certs/mykey.pem')dnl and the certs i have generated and sent to thawte are: mail.server.name.crt (signed from thawte) mail.server.name.csr (what I generated and sent to them) mail.server.name.key Any help on how to get this converted from the files i have to *.pem files would be much appreciated! (our temporary certificates are now expired and I have to get these installed ASAP) THANKS! _ Help yourself to FREE treats served up daily at the Messenger Café. Stop by today. http://www.cafemessenger.com/info/info_sweetstuff2.html?ocid=TXT_TAGLM_OctWLtagline___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Different DNS responses depending on query source
The host that runs my internal DNS server is down for the count (I've already replaced the power supply on it once, and I don't feel like doing it again). Although I had other uses planned for that machine, the only useful thing it was doing was DNS for a local net and DHCP, the latter I've moved to my firewall box (running m0n0wall). So, until I build a replacement machine, I'd like to run the DNS service on 6.2-RELENG machine on my DMZ. However I have a conflict between providing IPs for the outside world to see, eg n114.ewd.goldmark.org172.64.118.114 versus what I want when querying from the local network, eg, n114.ewd.goldmark.org10.1.10.131 Also there are some internal names (eg, fluffy.ewd.goldmark.org) which shouldn't be advertised to the outside world at all. The obvious answer would be to run two instances of bind, listening on different IPs (possibly using jails). But I don't have an IP address to spare on the DMZ. So is there a way to have bind listening on the only interface and IP address the host can have give different answers depending on where the query comes from? Cheers, -j -- Jeffrey Goldberghttp://www.goldmark.org/jeff/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Different DNS responses depending on query source
On Thursday 11 October 2007 17:55:20 Jeffrey Goldberg wrote: The obvious answer would be to run two instances of bind, listening on different IPs (possibly using jails). But I don't have an IP address to spare on the DMZ. So is there a way to have bind listening on the only interface and IP address the host can have give different answers depending on where the query comes from? http://www.isc.org/sw/bind/arm95/Bv9ARM.ch04.html#id2570613 -- Mel ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Different DNS responses depending on query source
Jeffrey Goldberg wrote: The host that runs my internal DNS server is down for the count (I've already replaced the power supply on it once, and I don't feel like doing it again). Although I had other uses planned for that machine, the only useful thing it was doing was DNS for a local net and DHCP, the latter I've moved to my firewall box (running m0n0wall). So, until I build a replacement machine, I'd like to run the DNS service on 6.2-RELENG machine on my DMZ. However I have a conflict between providing IPs for the outside world to see, eg n114.ewd.goldmark.org172.64.118.114 versus what I want when querying from the local network, eg, n114.ewd.goldmark.org10.1.10.131 Also there are some internal names (eg, fluffy.ewd.goldmark.org) which shouldn't be advertised to the outside world at all. The obvious answer would be to run two instances of bind, listening on different IPs (possibly using jails). But I don't have an IP address to spare on the DMZ. So is there a way to have bind listening on the only interface and IP address the host can have give different answers depending on where the query comes from? Cheers, -j You can use BIND's view statement: http://www.isc.org/sw/bind/arm94/Bv9ARM.ch06.html#view_statement_grammar HTH, Yuri ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Different DNS responses depending on query source
Hello Jeff: -Original Message- From: [EMAIL PROTECTED] [mailto:owner-freebsd- [EMAIL PROTECTED] On Behalf Of Jeffrey Goldberg Sent: Thursday, October 11, 2007 8:55 AM To: freebsd-questions@freebsd.org Subject: Different DNS responses depending on query source The host that runs my internal DNS server is down for the count (I've already replaced the power supply on it once, and I don't feel like doing it again). Although I had other uses planned for that machine, the only useful thing it was doing was DNS for a local net and DHCP, the latter I've moved to my firewall box (running m0n0wall). So, until I build a replacement machine, I'd like to run the DNS service on 6.2-RELENG machine on my DMZ. However I have a conflict between providing IPs for the outside world to see, eg n114.ewd.goldmark.org 172.64.118.114 versus what I want when querying from the local network, eg, n114.ewd.goldmark.org 10.1.10.131 Also there are some internal names (eg, fluffy.ewd.goldmark.org) which shouldn't be advertised to the outside world at all. The obvious answer would be to run two instances of bind, listening on different IPs (possibly using jails). But I don't have an IP address to spare on the DMZ. So is there a way to have bind listening on the only interface and IP address the host can have give different answers depending on where the query comes from? Cheers, -j -- Jeffrey Goldberghttp://www.goldmark.org/jeff/ I think what you're looking for is Bind Views. Check out: http://www.isc.org/sw/bind/arm93/Bv9ARM.ch06.html#id2562349 Regards, Mike ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Booting a GELI encrypted hard disk
That's a heck of a lot of trouble to go to, considering someone would have to steal your drive, alter it and put it back without you knowing it! Essentially, what I'm looking for is thus: - someone breaks into my always-locked equipment room - someone steals the box(es) in question, which obviously means shutting down the unit I don't want said thief to be able to retrieve the data after the box is stolen, which is why I'd like a passphrase, and a removable key. Even if the passphrase is captured, the data will still be protected because I have the only key to the system 35 miles away on my person. If the intruder has physical access to the machine, it would be much easier to put a keylogger device between the keyboard and the machine. There is no possible way this would go unnoticed. Anyone that could gain access to the already secured room would have a window of about 15 seconds to break into the building after hours (secured/alarmed), smash in the secured equipment room door, grab the box (out of about 40) and run. It's questionable though, whether you should leave your computer in an environment where this can happen undetected and probably better solved by increasing real life security. Like I said, it won't go undetected. The equipment is in a very secure equipment area, inside of a secured and alarmed building. All equipment is monitored 24/7, so if the box was physically altered, I would be alerted via SMS/email immediately. An important point that too many people forget. I agree, but this is not the case here. I just want the data protected if the box goes down, whether by physical intruder, or I force it down myself. Steve ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
canceling all jobs in cups
I am used to using cups via the web interface... well a user accidentally printed 1000 jobs and I have had to shut the printer down until they are all canceled... the question is how can I do this without clicking cancel job 1000 times ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Booting a GELI encrypted hard disk
As you can see only /home is encrypted because the rest doesn't hold data worth encrypting. Well, on mine it will. I was talking about my system. Yours will of course be different. :-) I know. I was not trying to be sarcastic in any way. Sorry if it seemed that way :) You can even encrypt /tmp with a one-time key (see 'geli onetime'). I will likely do this with /tmp and swap. Also have a look at the geli_* variables in /etc/defaults/rc.conf. Will do. It only needs to be present during creation of the GELI devices (geli attach). The rc scripts know they have to load GELI and attach the devices if they see an .eli device in /etc/fstab. Geli will ask for the passphrase(s) during boot-up if you're using them. You can specify which key-file to use in the geli_[devicename]_flags variable in /etc/rc.conf However using a USB device presents it's own problems. If you plug-in a USB stick there's no telling which device node it ends up with, depending on how many other USB devices are on the bus. To make device recognition easier, you should use a GEOM label on the USB stick, so you'll know which /dev/label/* device node it gets. And you'd probably have to hack an rc script to mount the USB stick _before_ the system tries to attach the GELI device(s). Getting around these issues is trivial. The only requirement is that my thumbdrive comes with me after the machine is reloaded. And remember that this USB stick is another thing you have to back-up and store in a safe place. It would be bad if you lost your data because your USB stick died or got lost. Understood. This has been considered, and it's exactly what I do with my TrueCrypt encrypted information on my Windows workstation. Steve ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Installing Thawte Certificate on imap pop smtp
brad davison wrote: I currently have self-signed certificates on our mail server, but they are now expired. I have just received the CA-crt back from thawte. I have the webmail portion completed with installing the certificates, but I am having some issues with getting them installed on SMTP. I tried to put them in as the *.crt and *.key files, but sendmail didn't like that. The temporary certificates installed are: define(`confCACERT_PATH', `/etc/mail/certs')dnl define(`confCACERT', `/etc/mail/certs/mycert.pem')dnl define(`confSERVER_CERT', `/etc/mail/certs/mycert.pem')dnl define(`confSERVER_KEY', `/etc/mail/certs/mykey.pem')dnl define(`confCLIENT_CERT', `/etc/mail/certs/mycert.pem')dnl define(`confCLIENT_KEY', `/etc/mail/certs/mykey.pem')dnl and the certs i have generated and sent to thawte are: mail.server.name.crt (signed from thawte) mail.server.name.csr (what I generated and sent to them) mail.server.name.key Any help on how to get this converted from the files i have to *.pem files would be much appreciated! (our temporary certificates are now expired and I have to get these installed ASAP) THANKS! Maybe you already got this solved? Mebbe `openssl x509 -inform der -in MYCERT.crt -out MYCERT.pem` ? IANAE, so I'd use make sure I had a backup copy of your *crt. HTH, Kevin Kinsey -- Most people have a mind that's open by appointment only. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: best way to update ports
On Thu, 11 Oct 2007 01:33:43 -0400 Bill Stwalley wrote: I need your advice on how to update security patches for ports on a dozen servers with minimal efforts. As I gathered, I should run portaudit in cron jobs and then manually update the ports with vulnerabilities after reading UPDATING. Is this the best way? Is this manual way feasible for managing a dozen servers? I used to run portupgrade in cron jobs, but that created too much nightmare. For example, imap-uw broke for a few days recently. Someone recommended http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/small-lan.html . It's great for maintaining machines with identical ports installed, but not good when ports are installed with different options on different servers. You may be interested in ports-mgmt/tinderbox. It let you package ports for different FreeBSD versions (jails in terms of tinderbox) and for different portstrees and options at a single machine. Then you may do a portupgrade -PP for the needed ports. WBR -- Boris Samorodov (bsam) Research Engineer, http://www.ipt.ru Telephone Internet SP FreeBSD committer, http://www.FreeBSD.org The Power To Serve ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: sysutils/lockdown
B. Cook wrote: Hello All, Is this still a valid working port for a FreeBSD 6.2 box? It looks as if it has not been touched since 2005. Check security/lockdown, and read the WWW sites mentioned in the Makefile there. The Makefile date is April 2007, but I've not done any other investigation. Kevin Kinsey -- As of next Thursday, UNIX will be flushed in favor of TOPS-10. Please update your programs. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: canceling all jobs in cups
On Thu, 11 Oct 2007 11:58:33 + Aryeh M. Friedman [EMAIL PROTECTED] wrote: I am used to using cups via the web interface... well a user accidentally printed 1000 jobs and I have had to shut the printer down until they are all canceled... the question is how can I do this without clicking cancel job 1000 times man cancel ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
On Thu, 2007-10-11 at 15:08 +, Aryeh Friedman wrote: In my case, william is better than root. I can never login to my GNOME desktop through GDM by root. A root account is absolutely useless to me. I use gnome and there is not a single thing I can not do with a root account that I can't with one. If you mean opening GUI apps try adding the following (I am assuming your using tcsh... if your using something else look up how to do this): in ~/.cshrc add: xhost + in /root/.cshr add: setenv DISPLAY :0 When you su *ALWAYS* use the following syntax: su - *DO NOT* include any cmd arguement. Once your root you should be able to start any GUI based app from the command line... if you don't know what the command line is left click on the menu/icon and go to properties... if it requires arguements look it up in the man page or the GNOME help for it. Oh... amazing... Thanks and really thanks... and really... Your the guidance has been useful and is greatly appreciated. -- Byung-Hee HWANG [EMAIL PROTECTED] You blaspheme. Resign yourself. -- Vito Corleone, Chapter 1, page 47 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: BSD Live?
Hello, On 10/11/07, Timothy Klaver [EMAIL PROTECTED] wrote: Is there any BSD version that provides a LiveCD Yes, it is called Freesbie: http://www.freesbie.org/ Regards Rabius -- 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]
BSD Live?
Is there any BSD version that provides a LiveCD so that I can test the OS and see whether or not I will enjoy it before actually installing? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: kernel configuration failure
coriolinus wrote: I'm new to kernel building, so I followed the handbook's advice: cd to /usr/src, then make buildkernel KERNCONF=MYKERNEL. The kernel build process is failing mysteriously. The file it claims not to be able to find, /usr/src/sys/dev/aic7xxx/aicasm/aicasm.c, is sitting in my filesystem exactly where it should be. I'm not even sure why exactly it's trying to build that, as I'm pretty sure that I disabled the modules in the kernel configuration file which use that source. Actually, that's not the problem. The file which is not found is the compiler itself: gcc34:No such file or directory Maybe you've installed gcc 4.3 from ports, linked /usr/bin/cc to /usr/local/bin/gcc43 and then upgrade gcc? Any help getting this to work would be greatly appreciated. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] -- Pietro Cerutti PGP Public Key: http://gahr.ch/pgp signature.asc Description: OpenPGP digital signature
Re: Different DNS responses depending on query source
On Oct 11, 2007, at 11:10 AM, Yuri Pankov wrote: Jeffrey Goldberg wrote: So is there a way to have bind listening on the only interface and IP address the host can have give different answers depending on where the query comes from? You can use BIND's view statement: http://www.isc.org/sw/bind/arm94/ Bv9ARM.ch06.html#view_statement_grammar Thank you and others who have pointed out the the view statement in BIND 9 does exactly what I want. I (obviously) hadn't been aware of the view statement until now. I'd also like to thank Jonathan Horne who off-list pointed me to a detailed article with examples he wrote that covers precisely my case. -j -- Jeffrey Goldberghttp://www.goldmark.org/jeff/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: BSD Live?
Timothy Klaver wrote: Is there any BSD version that provides a LiveCD so that I can test the OS and see whether or not I will enjoy it before actually installing? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Look for FreeSBIE http://www.freesbie.org/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: canceling all jobs in cups
Aryeh M. Friedman wrote: I am used to using cups via the web interface... well a user accidentally printed 1000 jobs and I have had to shut the printer down until they are all canceled... the question is how can I do this without clicking cancel job 1000 times ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] If you look better cups web administration utility there is an option to purge all jobs from the printer. Last resort. You can simple deinstall and then install again your printer via CUPS admin utility. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: BSD Live?
On 2007-10-11 14:02, Timothy Klaver wrote: Is there any BSD version that provides a LiveCD so that I can test the OS and see whether or not I will enjoy it before actually installing? http://www.freesbie.org/ That's a LiveCD based on FreeBSD. It works quite well. -- Matthew Anthony Kolybabi (Mak) [EMAIL PROTECTED] () ASCII Ribbon Campaign | Against HTML e-mail /\ www.asciiribbon.org | Against proprietary extensions ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: [kde-freebsd] Problem compiling kdegraphics (exr problem?)
* Mel [ Oct 10, 2007 (18:37 )]: Well, it's weird that fixed it for you, because kdegraphics needs to be patched. Here's the work-around: [..] This issue should be fixed with my commit from yesterday. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
NFS export question diskless dirs
hi. I have a question regarding NFS-exports. /etc/exports /diskless/ro-ro -maproot=root leia /diskless/kernels leia /diskless/rw leia /usr-ro -alldirs leia /home -alldirs leia /etc-ro -alldirs -maproot=root leia but mountd only recognize /diskless/ro - /usr - /home/ and /etc (showmount -e). How can I export /diskless/ro -ro /diskless/kernels with rw /diskless/rw with rw tothe same host??? Also, I have another question. Is it possible to run a diskless system with /var and /tmp mounted on a NFS-mount instead of using memory-disks? If possible, I'd gladly appreciate some tips or links. Best regards, J ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: NFS export question diskless dirs
On Thu, 2007-10-11 at 21:23 +0200, mr. phreak wrote: hi. I have a question regarding NFS-exports. /etc/exports /diskless/ro-ro -maproot=root leia /diskless/kernels leia /diskless/rw leia /usr-ro -alldirs leia /home -alldirs leia /etc-ro -alldirs -maproot=root leia but mountd only recognize /diskless/ro - /usr - /home/ and /etc (showmount -e). How can I export /diskless/ro -ro /diskless/kernels with rw /diskless/rw with rw tothe same host??? Also, I have another question. Is it possible to run a diskless system with /var and /tmp mounted on a NFS-mount instead of using memory-disks? If possible, I'd gladly appreciate some tips or links. Best regards, J ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Check out the handbook page on NFS: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/network-nfs.html In /etc/exports, each line represents the export information for one file system to one host. A remote host can only be specified once per file system, and may only have one default entry. For example, assume that /usr is a single file system. The following /etc/exports would be invalid: # Invalid when /usr is one file system /usr/src client /usr/ports client One file system, /usr, has two lines specifying exports to the same host, client. The correct format for this situation is: /usr/src /usr/ports client The properties of one file system exported to a given host must all occur on one line. Lines without a client specified are treated as a single host. This limits how you can export file systems, but for most people this is not an issue. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
I performed an rm -r on /var/lib/pkg
Call it a moment of sheer stupidity, call it a misremembering, call it whatever you want (and I imagine I'll hear a few different ones), but I just did an rm -r /var/lib/pkg. Before I type anything to damage things further, does anyone have any suggestions as to how to recover from this? I have other FreeBSD boxes available to me, none with the same pkg list, though. I'll be reading man pkgdb in the meantime.. James ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: I performed an rm -r on /var/lib/pkg
James wrote: Call it a moment of sheer stupidity, call it a misremembering, call it whatever you want (and I imagine I'll hear a few different ones), but I just did an rm -r /var/lib/pkg. Before I type anything to damage things further, does anyone have any suggestions as to how to recover from this? I have other FreeBSD boxes available to me, none with the same pkg list, though. I'll be reading man pkgdb in the meantime.. I'm guessing you might be Real Tired(tm). Do you mean /var/db/pkg? $ ll /var/lib/pkg ls: /var/lib/pkg: No such file or directory Kevin Kinsey -- The proof of the pudding is in the eating. -- Miguel de Cervantes ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: I performed an rm -r on /var/lib/pkg
On Thu, 2007-10-11 at 15:53 -0500, Kevin Kinsey wrote: James wrote: Call it a moment of sheer stupidity, call it a misremembering, call it whatever you want (and I imagine I'll hear a few different ones), but I just did an rm -r /var/lib/pkg. Before I type anything to damage things further, does anyone have any suggestions as to how to recover from this? I have other FreeBSD boxes available to me, none with the same pkg list, though. I'll be reading man pkgdb in the meantime.. I'm guessing you might be Real Tired(tm). Do you mean /var/db/pkg? $ ll /var/lib/pkg ls: /var/lib/pkg: No such file or directory Kevin Kinsey Yes, you're right. On all counts, I'm afraid. But, yes, ultimately. And the more I'm reading man pages, the more I'm thinking that the only solution here will be to reinstall everything. I was wondering if portmaster or something similar might be able to solve this, but it looks like /var/db/pkg is what *everything* refers to. I'm feeling like the least competent user in the world right now. Though it *does* teach me a valuable lesson about backing up. James ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: BSD Live?
On 10/11/07, Timothy Klaver [EMAIL PROTECTED] wrote: Is there any BSD version that provides a LiveCD so that I can test the OS and see whether or not I will enjoy it before actually installing? Last year I played with RoFreeSBIE, and was quite impressed. See http://www.rofreesbie.org/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: I performed an rm -r on /var/lib/pkg
On Thu, Oct 11, 2007 at 03:07:37PM -0600, James wrote: On Thu, 2007-10-11 at 15:53 -0500, Kevin Kinsey wrote: James wrote: Call it a moment of sheer stupidity, call it a misremembering, call it whatever you want (and I imagine I'll hear a few different ones), but I just did an rm -r /var/lib/pkg. Before I type anything to damage things further, does anyone have any suggestions as to how to recover from this? I have other FreeBSD boxes available to me, none with the same pkg list, though. I'll be reading man pkgdb in the meantime.. I'm guessing you might be Real Tired(tm). Do you mean /var/db/pkg? $ ll /var/lib/pkg ls: /var/lib/pkg: No such file or directory Kevin Kinsey Yes, you're right. On all counts, I'm afraid. But, yes, ultimately. And the more I'm reading man pages, the more I'm thinking that the only solution here will be to reinstall everything. I was wondering if portmaster or something similar might be able to solve this, but it looks like /var/db/pkg is what *everything* refers to. Yes, /var/db/pkg/ is where all the information about installed ports/packages is stored. To recreate that information you will have to reinstall everything. I'm feeling like the least competent user in the world right now. Though it *does* teach me a valuable lesson about backing up. Backups are good, yes. Regular, up-to-date, backups are even better. -- 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]
Re: BSD Live?
J65nko wrote: On 10/11/07, Timothy Klaver [EMAIL PROTECTED] wrote: Is there any BSD version that provides a LiveCD so that I can test the OS and see whether or not I will enjoy it before actually installing? You can also set up a virtual machine and test it that way (on windows you can get a trial copy of vmware to do this) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: I performed an rm -r on /var/lib/pkg
James wrote: Yes, you're right. On all counts, I'm afraid. But, yes, ultimately. And the more I'm reading man pages, the more I'm thinking that the only solution here will be to reinstall everything. I was wondering if portmaster or something similar might be able to solve this, but it looks like /var/db/pkg is what *everything* refers to. I'm feeling like the least competent user in the world right now. Though it *does* teach me a valuable lesson about backing up. Well, first off, be glad you weren't in / with your rm. :-) I'll go out on a limb (IANAE), and suggest to you that /var/db/pkg is very important when installing, removing, and upgrading ports (or 3rd party software), but it's not critical to the moment by moment operating of such ports. So, it's quite possible that everything can wait until you get some sleep. However, you are probably right about reinstall everything being the course of action to take. Another possibility: get another box, install everything on that, and copy /var/db/pkg over. You will probably face some issues with consistency in the package database as a result. This will cause a few problems when you get ready to update in the future; however, you can't really get stuck too badly as a deinstall/reinstall will usually fix such things. ... then, of course, you have the possibility that a dependency will not work with the new program. This sort of thing bites in any number of ways, especially after a reboot. I'd probably try to wait for a period of relatively low demand on the box, then do the reinstalling. And get some sleep first ;-) But, as I said, IANAE. Kevin Kinsey -- One way to make your old car run better is to look up the price of a new model. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: I performed an rm -r on /var/lib/pkg
On Thu, 2007-10-11 at 23:13 +0200, Erik Trulsson wrote: On Thu, Oct 11, 2007 at 03:07:37PM -0600, James wrote: On Thu, 2007-10-11 at 15:53 -0500, Kevin Kinsey wrote: James wrote: Call it a moment of sheer stupidity, call it a misremembering, call it whatever you want (and I imagine I'll hear a few different ones), but I just did an rm -r /var/lib/pkg. Before I type anything to damage things further, does anyone have any suggestions as to how to recover from this? I have other FreeBSD boxes available to me, none with the same pkg list, though. I'll be reading man pkgdb in the meantime.. I'm guessing you might be Real Tired(tm). Do you mean /var/db/pkg? $ ll /var/lib/pkg ls: /var/lib/pkg: No such file or directory Kevin Kinsey Yes, you're right. On all counts, I'm afraid. But, yes, ultimately. And the more I'm reading man pages, the more I'm thinking that the only solution here will be to reinstall everything. I was wondering if portmaster or something similar might be able to solve this, but it looks like /var/db/pkg is what *everything* refers to. Yes, /var/db/pkg/ is where all the information about installed ports/packages is stored. To recreate that information you will have to reinstall everything. I'm feeling like the least competent user in the world right now. Though it *does* teach me a valuable lesson about backing up. Backups are good, yes. Regular, up-to-date, backups are even better. Alas, though, regular, up-to-date backups ain't happened here. What has happened, though, is I've never ran rm in /usr/ports/distfiles. I'm going to think for a little bit about a script that can move through /usr/ports/distfiles and reinstall everything that exists there. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: BSD Live?
On Thu, 11 Oct 2007 at 17:12 -, [EMAIL PROTECTED] confabulated: You can also set up a virtual machine and test it that way (on windows you can get a trial copy of vmware to do this) VMWare Server is free and has the same functionality. However, it is over a 200 meg download. -- _|_ (_| | ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: NFS export question diskless dirs
On Thu, 2007-10-11 at 23:03 +0200, mr. phreak wrote: James wrote: On Thu, 2007-10-11 at 22:36 +0200, mr. phreak wrote: James wrote: On Thu, 2007-10-11 at 21:23 +0200, mr. phreak wrote: hi. I have a question regarding NFS-exports. /etc/exports /diskless/ro-ro -maproot=root leia /diskless/kernels leia /diskless/rw leia /usr-ro -alldirs leia /home -alldirs leia /etc-ro -alldirs -maproot=root leia but mountd only recognize /diskless/ro - /usr - /home/ and /etc (showmount -e). How can I export /diskless/ro -ro /diskless/kernels with rw /diskless/rw with rw tothe same host??? Also, I have another question. Is it possible to run a diskless system with /var and /tmp mounted on a NFS-mount instead of using memory-disks? If possible, I'd gladly appreciate some tips or links. Best regards, J ___ freebsd-questions@freebsd.org mailto:freebsd-questions@freebsd.org mailto:freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] mailto:[EMAIL PROTECTED] Check out the handbook page on NFS: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/network-nfs.html In /etc/exports, each line represents the export information for one file system to one host. A remote host can only be specified once per file system, and may only have one default entry. For example, assume that /usr is a single file system. The following /etc/exports would be invalid: # Invalid when /usr is one file system /usr/src client /usr/ports client One file system, /usr, has two lines specifying exports to the same host, client. The correct format for this situation is: /usr/src /usr/ports client The properties of one file system exported to a given host must all occur on one line. Lines without a client specified are treated as a single host. This limits how you can export file systems, but for most people this is not an issue. Yeah, I know. I read it as well. I guess I'll have to create separate filesystems if I want to export them with different permissions to the same host... I thought the mountd -r was a sollution to skip that. But I guess I was wrong. Reading on in the handbook has this: The following is an example of a valid export list, where /usr and /exports are local file systems: # Export src and ports to client01 and client02, but only # client01 has root privileges on it /usr/src /usr/ports -maproot=rootclient01 /usr/src /usr/ports client02 # The client machines have root and can mount anywhere # on /exports. Anyone in the world can mount /exports/obj read-only /exports -alldirs -maproot=root client01 client02 /exports/obj -ro The handbook claims that /exports is a single file system, yet it treats /exports/obj as a separate entity on a separate line. I was wondering if this was a result of /exports being exported in its entirety. You might want to goof around with that, see if it'll let you do what you want to do. I've played around, and my conclusion is that you can have the same dirs/filesystems on seperate lines, IF the host entry isn't the same. i.e if the example above would have client01 and client02 on /export/obj it would be a erratic entry. However it's a shame because it means you cannot export subdirs of a same filesystem with different permissions to the same host(s). Well, now, that gives some hope. Open up /etc/hosts and make a couple of junk entries, like: 192.168.1.77 junkhost.host 192.168.1.78 junkhost2.host And use *them* in your exports to differentiate each nfs line. Even better, don't use hostnames in /etc/exports, just use junk ip addresses. That *surely* can't be the only way to permanently differentiate the lines, but it might be a good way to start. James ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
NAT Question
I have a question regarding ipf and ipnat. I have a firewall with two public IP addresses. One of the IP addresses is for incoming Internet traffic only and the other is for incoming e-mail. I'm not sure why my ISP has done, this, but they have. In otherwords, all incoming http traffic (port 80) will be going to the address 1.2.3.4 and all incoming smtp traffic (port 25) will be going to 1.2.3.5. The internal address of the firewall is 10.129.10.40/24. The webserver has an internal address of 10.129.10.49 and a default gateway of 10.129.10.40 (the firewall). If I use rdr on an incoming connection, will repsonses exit the network on the same interface they entered the firewall on? Following are the rules I would use. ipnat.rules rdr em1 1.2.3.4/32 port 80 - 10.129.10.49 port 80 tcp ipf.rules pass in on em1 from any to 1.2.3.4 port = 80 keep state pass out on em1 from 1.2.3.4 port = 80 to any keep state Does this solution make sense, or is there a better way to accomplish the same thing? Thanks for your help. Jay ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: I performed an rm -r on /var/lib/pkg
James writes: What has happened, though, is I've never ran rm in /usr/ports/distfiles. I'm going to think for a little bit about a script that can move through /usr/ports/distfiles and reinstall everything that exists there. Having been in almost the identical situation for different rasons, I sympathize. Yes, this will involve a sweep through /usr/ports distfiles. If you haven't ever deleted anything, I suggest a prelimiary manual run deleting everything but the most recent version. This has a down-side, but it will prevent cluttering the rebuilt system with unused ports. The other thing you want do is start with something /big/ - OpenOffice, FireFox, GIMP, Java, Apache. While rebuilding it will take time (possibly days) it will automatically suck in the dependencies. Send the output of the rebuild to a file; have a cron job e-mail you the last 50 lines every hour. Robert Huff ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Installing textproc/py-libxml2 port for python 2.5 while keeping 2.4?
Hello, When I install textproc/py-libxml2 from the port tree, it installs py24-libxml2. I want to install py25-libxml2 while keeping py24-libxml2. lib/%%PYTHON_VERSION%%/site-packages/drv_libxml2.py lib/%%PYTHON_VERSION%%/site-packages/drv_libxml2.pyc lib/%%PYTHON_VERSION%%/site-packages/drv_libxml2.pyo lib/%%PYTHON_VERSION%%/site-packages/libxml2.py lib/%%PYTHON_VERSION%%/site-packages/libxml2.pyc lib/%%PYTHON_VERSION%%/site-packages/libxml2.pyo lib/%%PYTHON_VERSION%%/site-packages/libxml2mod.a lib/%%PYTHON_VERSION%%/site-packages/libxml2mod.la lib/%%PYTHON_VERSION%%/site-packages/libxml2mod.so These files are per python version while docs are not. Is there anyway to do this? Should I just do force install? Any other better way? Is it possible for one port to be installed to multiple versions of python? Thanks. -- Naoyuki Tai Tai, ntai a t smartfruit d o t com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: I performed an rm -r on /var/lib/pkg
On Thursday 11 October 2007 23:29:05 James wrote: What has happened, though, is I've never ran rm in /usr/ports/distfiles. I'm going to think for a little bit about a script that can move through /usr/ports/distfiles and reinstall everything that exists there. Well, if you figure out what ports you have installed, you can regenerate the pkgdb using: make -DNO_BUILD -DNO_INSTALL generate-plist fake-pkg for each port. I just tested that using a temporary PKG_DBDIR. In case you wanna see what happens, here's what I did: mkdir -p /tmp/var/db/pkg cd /usr/ports/shells/bash env PKG_DBDIR=/tmp/var/db/pkg make -DNO_BUILD -DNO_INSTALL \ generate-plist fake-pkg # ls /tmp/var/db/pkg/bash-3.2.25/ +COMMENT+DEINSTALL +INSTALL +CONTENTS +DESC +MTREE_DIRS It's missing +REQUIRED_BY and @pkgdep lines in +CONTENTS, but haven't been able to figure out yet why that is. I hope this gets you a bit closer. -- Mel ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: I performed an rm -r on /var/lib/pkg
On Thu, 2007-10-11 at 18:14 -0400, Robert Huff wrote: James writes: What has happened, though, is I've never ran rm in /usr/ports/distfiles. I'm going to think for a little bit about a script that can move through /usr/ports/distfiles and reinstall everything that exists there. Having been in almost the identical situation for different rasons, I sympathize. Yes, this will involve a sweep through /usr/ports distfiles. If you haven't ever deleted anything, I suggest a prelimiary manual run deleting everything but the most recent version. This has a down-side, but it will prevent cluttering the rebuilt system with unused ports. /usr/ports/distfiles is definitely looking promising. awk is too damn painful to work with, so I'm going to dust off my perl skills. Hell, this could actually turn out to be fun. And if I write the script properly, it might make a nice disaster recovery tool for /usr/ports/ports-mgmt - it can be called WhenYou'reAnIdiotLikeJamesWasOnFreeBSDQuestions Well, if you figure out what ports you have installed, you can regenerate the pkgdb using: make -DNO_BUILD -DNO_INSTALL generate-plist fake-pkg for each port. I just tested that using a temporary PKG_DBDIR. In case you wanna see what happens, here's what I did: mkdir -p /tmp/var/db/pkg cd /usr/ports/shells/bash env PKG_DBDIR=/tmp/var/db/pkg make -DNO_BUILD -DNO_INSTALL \ generate-plist fake-pkg Wow, that's great! I understand that it has the caveats that you mentioned, but it's *at least* a fantastic start. James ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: How to create a user account with the same permission as root ?
Hi, Jerry McAllister wrote: On Thu, Oct 11, 2007 at 07:34:54PM +0800, Erich Dollansky wrote: FreeBSD is not Windows. True statement - thank heaven. You cannot have another root in the system. Unless I misunderstand what you are saying, this is NOT a true statement. You can create as many ids with a '0' UID as you want. It may not be But they are the same as it is still the same UID. Under WIndows, you can create as many 'root' accounts you want. root is special. Allow then all members of wheel to access the files needed by the group wheel. Not the best idea. Really not. But at least better than to work as root. I would not do this as it creates many security wholes. Erich ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
question about Intel PRO/1000 GT dual port
Does anyone out there have one? If so, which connector is identified as em0 - upper or lower? Robert Huff ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: system admin question...
((A parenthetical note): In prep for this posting I finished (or expanded) my mail-strip program that eliminates most of the cruft and leaves the body. ) So I'll look at bigsister, conky, nagios, monit, and Ksysguard. (Mel, if you have a cheatsheet for Ksysguard, that would be a big win.) The more I can automate, the better. thanks to everybody who emailed me, onlist and off; if I can turn this into a how-to article, i'll publish it on my bsd pages. gary -- Gary Kline [EMAIL PROTECTED] www.thought.org Public Service Unix http://jottings.thought.org http://transfinite.thought.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: question about Intel PRO/1000 GT dual port
On Thu, Oct 11, 2007 at 08:32:36PM -0400, Robert Huff wrote: Does anyone out there have one? If so, which connector is identified as em0 - upper or lower? Robert Huff We have a fair number of those. The upper port registers as em0. Josef -- FreeBSD 6.2 | Josef Grosch| You can't expect to wield supreme executive power [EMAIL PROTECTED] | just 'cause some watery tart threw a sword at you! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Installing textproc/py-libxml2 port for python 2.5 while keeping 2.4?
On Thu, 11 Oct 2007 17:55:00 -0400
Naoyuki Tai [EMAIL PROTECTED] wrote:
Hello,
When I install textproc/py-libxml2 from the port tree, it installs
py24-libxml2.
I want to install py25-libxml2 while keeping py24-libxml2.
lib/%%PYTHON_VERSION%%/site-packages/drv_libxml2.py
lib/%%PYTHON_VERSION%%/site-packages/drv_libxml2.pyc
lib/%%PYTHON_VERSION%%/site-packages/drv_libxml2.pyo
lib/%%PYTHON_VERSION%%/site-packages/libxml2.py
lib/%%PYTHON_VERSION%%/site-packages/libxml2.pyc
lib/%%PYTHON_VERSION%%/site-packages/libxml2.pyo
lib/%%PYTHON_VERSION%%/site-packages/libxml2mod.a
lib/%%PYTHON_VERSION%%/site-packages/libxml2mod.la
lib/%%PYTHON_VERSION%%/site-packages/libxml2mod.so
These files are per python version while docs are not.
Is there anyway to do this?
Should I just do force install?
Any other better way?
Is it possible for one port to be installed to multiple versions of
python?
I've had exactly the same problem trying to install www/py-turbogears
in a python2.5 world (it needs python2.4, and a lot of ports that
must of course be installed in the python2.4 instance as well) :-(
This is what I've discovered while doing so:
1.) If you want to install a py24-* in addition to py25-* of the
same port, you'll often need to switch between two versions of
easy_install:
So before you want to install a py24-* package, do something
like this:
# cd /usr/local/bin
# mv easy_install easy_install.orig
# cp easy_install-2.4 easy_install
now force install the py24-* package
# mv easy_install.org easy_install
You need this, because you'll get some errors about .pth files
not being supported etc...
2.) When compiling the py24-* version of an existing port,
you'll often need to define PYTHON_DEFAULT_VERSION to be
python2.4 in /etc/make.conf. For example:
.if ${.CURDIR:M*/devel/py-setuptools}
PYTHON_DEFAULT_VERSION=python2.4
.endif
Then go into the ports directory, recompile with 'make', and
then force install by defining FORCE_PKG_REGISTER. Then you
my comment out those lines in /etc/make.conf again.
3.) It is safe to have py24-* and py25-* of the same port
simultaneously on your system, as long as they only install
files in the respective python trees, a.k.a.:
/usr/local/lib/python2.4/site-packages/
/usr/local/lib/python2.5/site-packages/
But not all py- ports are so well behaved! Some ports also
install stuff in /usr/local/bin (e.g. devel/py-twistedCore,
which adds stuff like /usr/local/bin/twistd. This 'twistd'
is tied to the python version you've last used to install
the port (look at its first line).
So what I'm doing is:
a.) first install all py24-* packages with easy_install set to
easy_install-2.4, and PYTHON_DEFAULT_VERSION set to python2.4
in /etc/make.conf
b.) reinstall all those py24-* packages as py25-* (of course
with easy_install set to easy_install-2.5), with
PYTHON_DEFAULT_VERSION set to python2.5 in /etc/make.conf
and using FORCE_PKG_REGISTER, and overwriting stuff that's
outside site-packages/...
It's not ideal (we really need a better way to simultaneously
install the same ports for different versions of python!), but
it's good enough for me: once the packages are installed, the
python installations are nicely self-contained and working
(except for the common binaries in /usr/local/bin etc..., which
would need a -2.4 or -2.5 suffix IMHO). Updating all those
py24-* packages is not as nice though..., cause you'll have to
redo it manually (at least for the py24-*... the py25-* will update
themselves automatically).
Naoyuki Tai Tai, ntai a t smartfruit d o t com
Regards,
-cpghost.
--
Cordula's Web. http://www.cordula.ws/
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Ports vs Pkgsrc
Would anyone be able to either offer a link, or explain the differences between NetBSD/DragonflyBSD's pkgsrc and FreeBSD/OpenBSD's ports systems? Google searches such as 'pkgsrc vs ports' have yielded nothing satisfying. -- Matthew Anthony Kolybabi (Mak) [EMAIL PROTECTED] () ASCII Ribbon Campaign | Against HTML e-mail /\ www.asciiribbon.org | Against proprietary extensions ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: question about Intel PRO/1000 GT dual port
Thanks to the many who have responded. Robert Huff ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
BitTorrent configuration in FreeBSD-6.2 -for Large file downloads uploads
Hi all, I had a freebsd 6.2server , I want this m/c to be serve iso images , so users of my webportal have the the facility to download iso images , each iso about 650 MB size , I heard about Bittorrent so I installed this through FreeBSd ports tree , the installation is complete .( I can see the commands bittoreent-curses, bittorrent-console, bittorrent-tracker etc .. by typing them in shell) I dont know where to put these iso images (these images stored in this servers one of the normal users eg : /home /myname /1.iso, 2.iso,3.iso etc ..) so bittorent can serve these files . I want the users of my webportal can down load these images (by clicking a link in the website that I hosted in this server machine ) I have apache2.2 installed and a plone /zope installation for my webportal and all working in this box But how I can use Bittorrent to serve these big files to the remote users of my website (so that I can save a lot of bandwidth of my network connection ) The Bittorent is installed in this box was( py24-BitTorrent-4.20.2_1,1 ) . I have the ISO images , but how can I put these ISO's to be served via Bittorrent how others can accesss these iso's from my webserver through bittorrent pls provide your tips to do this Thanks in advance kk _ Search from any Web page with powerful protection. Get the FREE Windows Live Toolbar Today! http://toolbar.live.com/?mkt=en-in___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
uid 80: exited on signal 6
Hello, With each (daily) log rotation I get this in /var/log/messages file. szalbot.homedns.org kernel log messages: +++ /tmp/security.BfIqepKO Fri Oct 12 03:08:35 2007 +pid 82543 (httpd), uid 80: exited on signal 6 +pid 82542 (httpd), uid 80: exited on signal 6 +pid 82541 (httpd), uid 80: exited on signal 6 +pid 82537 (httpd), uid 80: exited on signal 6 +pid 82533 (httpd), uid 80: exited on signal 6 +pid 82536 (httpd), uid 80: exited on signal 6 +pid 82535 (httpd), uid 80: exited on signal 6 +pid 82534 (httpd), uid 80: exited on signal 6 +pid 3653 (httpd), uid 80: exited on signal 6 This seems to be saying to me that httpd died but it works and I do not have to start apache by hand. I know this may be caused by extensions. Currently I have a few of them commented out. When I enable them, apache crashes for good. $ cat /usr/local/etc/php/extensions.ini extension=calendar.so extension=ctype.so extension=zlib.so extension=imap.so extension=xml.so extension=exif.so extension=bcmath.so extension=sockets.so #extension=ftp.so extension=mbstring.so extension=mcrypt.so #extension=posix.so extension=dbase.so extension=gettext.so extension=xmlrpc.so extension=simplexml.so extension=session.so extension=pdo.so extension=pdo_sqlite.so extension=hash.so extension=gmp.so extension=gd.so extension=bz2.so #extension=pspell.so extension=openssl.so #extension=pdf.so extension=pcre.so extension=filter.so extension=curl.so extension=fileinfo.so #extension=zip.so extension=mhash.so extension=iconv.so extension=mysql.so extension=spl.so extension=sqlite.so extension=dom.so extension=xmlwriter.so #extension=json.so extension=mysqli.so extension=soap.so extension=tokenizer.so extension=xmlreader.so For example, if I uncomment json.so, apache will crash when restarted. I do not need this extension so I keep it commented out. My question is a bit general. Should I worry about the warning of apache exiting on signal 6? How best to debug it further? Thanks! -- Zbigniew Szalbot ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: question about Intel PRO/1000 GT dual port
Quoting Robert Huff [EMAIL PROTECTED]: Does anyone out there have one? If so, which connector is identified as em0 - upper or lower? Robert Huff ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] on mine, em0 is the upper. cheers, -- Jonathan Horne http://www.dfwlp.org [EMAIL PROTECTED] This message was sent using IMP, the Internet Messaging Program. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Ports vs Pkgsrc
2007/10/12, Mak Kolybabi [EMAIL PROTECTED]: Would anyone be able to either offer a link, or explain the differences between NetBSD/DragonflyBSD's pkgsrc and FreeBSD/OpenBSD's ports systems? Google searches such as 'pkgsrc vs ports' have yielded nothing satisfying. -- Matthew Anthony Kolybabi (Mak) [EMAIL PROTECTED] The only diff for me is: pkgsrc is workable on other than BSD systems. For example you can use it on Solaris or Linux - I say that it(the portability) is typical for a third party software package system from the NetBSD project - . So far I know is the ports tree not (at least not without pain) usable on other systems. And so far I read is pkgsrc a derivative of ports - but this can be very well bull -. The technical differences (configure files, standard dirs etc. etc.) you have look up part for part in the docs. I hope I could help you. regards Gueven ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
