Re: Init(8) cannot decrease securelevel

1999-09-07 Thread Dag-Erling Smorgrav
Matthew Dillon [EMAIL PROTECTED] writes: So making DDB 'secure-level friendly' would be a useful thing tgo do, I think. The idea is not to disable DDB, but to simply limit the actions that can be performed within it if the securelevel has been raised. The sysadmin would

Re: Init(8) cannot decrease securelevel

1999-09-07 Thread KATO Takenori
Dag-Erling Smorgrav [EMAIL PROTECTED] wrote: That's an excellent idea - it shouldn't be too hard to add a kernel option (say, DDB_RESTRICTED) and #ifndef the "dangerous" commands. To achieve both higher security and kenel hackers convenience, I'd submit following idea: - If securelevel 1,

Re: Init(8) cannot decrease securelevel

1999-09-07 Thread Matthew Dillon
: generated, DDB is the only way to figure out what is going on. : securelevel is a mechanism which attempts to guarentee data security, : at least to a degree. These two items do not clash. : : :Anyway, as soon as you can physically access the PC, youD loose anyway, :independent

Re: Init(8) cannot decrease securelevel

1999-09-07 Thread Nick Hibma
I disagree quite strongly. DDB provides a mechanism to allow a sysadmin to obtain a greater amount of information from a panic situation then he could get otherwise. Being able to obtain this information does not run counter to running with a raised securelevel. If

Re: Init(8) cannot decrease securelevel

1999-09-07 Thread Bill Fumerola
On Tue, 7 Sep 1999, Nick Hibma wrote: Anyway, as soon as you can physically access the PC, youD loose anyway, independent of whether you can go into DDB to do things. You can reboot, boot a floppy. Yes you can do something about those things, but only to a limited extent. Not without someone

Re: Init(8) cannot decrease securelevel

1999-09-07 Thread Dag-Erling Smorgrav
Matthew Dillon dil...@apollo.backplane.com writes: So making DDB 'secure-level friendly' would be a useful thing tgo do, I think. The idea is not to disable DDB, but to simply limit the actions that can be performed within it if the securelevel has been raised. The sysadmin

Re: Init(8) cannot decrease securelevel

1999-09-07 Thread KATO Takenori
Dag-Erling Smorgrav d...@flood.ping.uio.no wrote: That's an excellent idea - it shouldn't be too hard to add a kernel option (say, DDB_RESTRICTED) and #ifndef the dangerous commands. To achieve both higher security and kenel hackers convenience, I'd submit following idea: - If securelevel

Re: Init(8) cannot decrease securelevel

1999-09-07 Thread Matthew Dillon
: generated, DDB is the only way to figure out what is going on. : securelevel is a mechanism which attempts to guarentee data security, : at least to a degree. These two items do not clash. : : :Anyway, as soon as you can physically access the PC, youD loose anyway, :independent

Re: Init(8) cannot decrease securelevel

1999-09-07 Thread Poul-Henning Kamp
But, unfortunately, putting the console on a serial port creates vulnerabilities when DDB is enabled. You are, essentially, creating an unintentional backdoor into the system. Hence the problem. ports/*/conserver is your friend! -- Poul-Henning Kamp FreeBSD coreteam

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Bruce Evans
There used to be security holes that allowed root to lower `securelevel' using init. Rev.1.9 defends against any undiscovered holes. How about following change? OK. Bruce To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Bruce Evans
Once securelevel has been increased, no process can decrease it because kernel always refuse decreasing it. This is inconsistent with the manual page of init: The kernel runs with four different levels of security. Any super-user process can raise the security level, but only init

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread KATO Takenori
Bruce Evans [EMAIL PROTECTED] wrote: There used to be security holes that allowed root to lower `securelevel' using init. Rev.1.9 defends against any undiscovered holes. How about following change? -- *** init.8.ORIG Mon Sep 6 14:20:46 1999 --- init.8 Mon Sep 6 14:23:01 1999

Init(8) cannot decrease securelevel

1999-09-06 Thread KATO Takenori
Once securelevel has been increased, no process can decrease it because kernel always refuse decreasing it. This is inconsistent with the manual page of init: The kernel runs with four different levels of security. Any super-user process can raise the security level, but only init

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Dag-Erling Smorgrav
KATO Takenori [EMAIL PROTECTED] writes: The kernel runs with four different levels of security. ! Any super-user process can raise the security level, but no process can lower it. How about "The security level can only be raised by the super-user, and cannot be lowered by anyone."

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Matthew Dillon
: :KATO Takenori [EMAIL PROTECTED] writes: : The kernel runs with four different levels of security. : ! Any super-user process can raise the security level, but no process : can lower it. : :How about "The security level can only be raised by the super-user, :and cannot be lowered by

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Greg Black
Matthew Dillon writes: Though, as a side note, it should be noted that if you have DDB enabled then lowering the secure level is pretty easy to do. If you have access to the console, of course. It should also be noted that it makes no sense to enable DDB on systems that need to

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Matthew Dillon
:On Mon, Sep 06, 1999 at 08:39:54AM -0700, a little birdie told me :that Matthew Dillon remarked : : Though, as a side note, it should be noted that if you have DDB : enabled then lowering the secure level is pretty easy to do. If you : have access to the console, of course. We

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Matthew Dillon
: Though, as a side note, it should be noted that if you have DDB : enabled then lowering the secure level is pretty easy to do. If you : have access to the console, of course. : :It should also be noted that it makes no sense to enable DDB on :systems that need to use elevated

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread KATO Takenori
Matthew Dillon [EMAIL PROTECTED] wrote: I disagree quite strongly. DDB provides a mechanism to allow a sysadmin to obtain a greater amount of information from a panic situation then he could get otherwise. Being able to obtain this information does not run counter to running

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Peter Jeremy
Matthew Dillon [EMAIL PROTECTED] wrote: If the system winds up in a state where a kernel core cannot be generated, DDB is the only way to figure out what is going on. securelevel is a mechanism which attempts to guarentee data security, at least to a degree. The problem is that DDB

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Bill Fumerola
On Tue, 7 Sep 1999, Nick Hibma wrote: Anyway, as soon as you can physically access the PC, youD loose anyway, independent of whether you can go into DDB to do things. You can reboot, boot a floppy. Yes you can do something about those things, but only to a limited extent. Not without

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Dag-Erling Smorgrav
KATO Takenori k...@ganko.eps.nagoya-u.ac.jp writes: The kernel runs with four different levels of security. ! Any super-user process can raise the security level, but no process can lower it. How about The security level can only be raised by the super-user, and cannot be lowered by

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Matthew Dillon
: :KATO Takenori k...@ganko.eps.nagoya-u.ac.jp writes: : The kernel runs with four different levels of security. : ! Any super-user process can raise the security level, but no process : can lower it. : :How about The security level can only be raised by the super-user, :and cannot be lowered

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Matthew D. Fuller
On Mon, Sep 06, 1999 at 08:39:54AM -0700, a little birdie told me that Matthew Dillon remarked Though, as a side note, it should be noted that if you have DDB enabled then lowering the secure level is pretty easy to do. If you have access to the console, of course. We used this

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Matthew Dillon
:On Mon, Sep 06, 1999 at 08:39:54AM -0700, a little birdie told me :that Matthew Dillon remarked : : Though, as a side note, it should be noted that if you have DDB : enabled then lowering the secure level is pretty easy to do. If you : have access to the console, of course. We used

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Greg Black
Matthew Dillon writes: Though, as a side note, it should be noted that if you have DDB enabled then lowering the secure level is pretty easy to do. If you have access to the console, of course. It should also be noted that it makes no sense to enable DDB on systems that need to

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread KATO Takenori
Matthew Dillon dil...@apollo.backplane.com wrote: Though, as a side note, it should be noted that if you have DDB enabled then lowering the secure level is pretty easy to do. If you have access to the console, of course. We used this trick at BEST a couple of times. Still,

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Matthew Dillon
: Though, as a side note, it should be noted that if you have DDB : enabled then lowering the secure level is pretty easy to do. If you : have access to the console, of course. : :It should also be noted that it makes no sense to enable DDB on :systems that need to use elevated

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread KATO Takenori
Matthew Dillon dil...@apollo.backplane.com wrote: I disagree quite strongly. DDB provides a mechanism to allow a sysadmin to obtain a greater amount of information from a panic situation then he could get otherwise. Being able to obtain this information does not run counter to

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread Peter Jeremy
Matthew Dillon dil...@apollo.backplane.com wrote: If the system winds up in a state where a kernel core cannot be generated, DDB is the only way to figure out what is going on. securelevel is a mechanism which attempts to guarentee data security, at least to a degree. The problem is

Re: Init(8) cannot decrease securelevel

1999-09-06 Thread David Scheidt
On Tue, 7 Sep 1999, KATO Takenori wrote: DDB does not provide enough security. Though securelevel cannot be changed, (1) Turn off power. (2) Boot as single-user mode. Setting the console as insecure should protect against this. or (1) Turn off power. (2)

Init(8) cannot decrease securelevel

1999-09-05 Thread KATO Takenori
Once securelevel has been increased, no process can decrease it because kernel always refuse decreasing it. This is inconsistent with the manual page of init: The kernel runs with four different levels of security. Any super-user process can raise the security level, but only init can

Re: Init(8) cannot decrease securelevel

1999-09-05 Thread Bruce Evans
Once securelevel has been increased, no process can decrease it because kernel always refuse decreasing it. This is inconsistent with the manual page of init: The kernel runs with four different levels of security. Any super-user process can raise the security level, but only init can

Re: Init(8) cannot decrease securelevel

1999-09-05 Thread KATO Takenori
Bruce Evans b...@zeta.org.au wrote: There used to be security holes that allowed root to lower `securelevel' using init. Rev.1.9 defends against any undiscovered holes. How about following change? -- *** init.8.ORIG Mon Sep 6 14:20:46 1999 --- init.8 Mon Sep 6 14:23:01 1999

Re: Init(8) cannot decrease securelevel

1999-09-05 Thread Bruce Evans
There used to be security holes that allowed root to lower `securelevel' using init. Rev.1.9 defends against any undiscovered holes. How about following change? OK. Bruce To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message