does not remove capabilities

2005-06-09 Thread martin f krafft
Note: I am not filing this as a bug for reasons of responsible disclosure. Maybe I am just being too paranoid. Let me know if I should file the bug, or just forward my mail... Package: kernel-source-2.6.11 Version: 2.6.11-5 Severity: grave Tags: security Note that the flags are correctly manipula

Re: does not remove capabilities

2005-06-09 Thread martin f krafft
also sprach martin f krafft <[EMAIL PROTECTED]> [2005.06.09.1106 +0200]: > Note: I am not filing this as a bug for reasons of responsible > disclosure. Maybe I am just being too paranoid. Let me know if > I should file the bug, or just forward my mail... Ha! I am a dork. So I guess it's now full-d

Re: does not remove capabilities

2005-06-09 Thread Humberto Massa GuimarĂ£es
Just curious, what does bit 8 (SYS_SETPCAP) means? Why is it off? Does it have anything to do with your bug? -- []s, Massa -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: does not remove capabilities

2005-06-09 Thread martin f krafft
also sprach Humberto Massa GuimarĂ£es <[EMAIL PROTECTED]> [2005.06.09.1718 +0200]: > Just curious, what does bit 8 (SYS_SETPCAP) means? Why is it off? > Does it have anything to do with your bug? On all the machines I've tested, this bit was always off right after boot. SETPCAP deals with process

Re: does not remove capabilities

2005-06-09 Thread Jurij Smakov
On Thu, 9 Jun 2005, martin f krafft wrote: Note: I am not filing this as a bug for reasons of responsible disclosure. Maybe I am just being too paranoid. Let me know if I should file the bug, or just forward my mail... Hi Martin, In order for the capability stuff to function the capability.ko

Re: does not remove capabilities

2005-06-10 Thread martin f krafft
also sprach Jurij Smakov <[EMAIL PROTECTED]> [2005.06.10.0258 +0200]: > In order for the capability stuff to function the capability.ko > module should be loaded. The situation you describe indeed occurs > when capability.ko is not loaded into the kernel. So I would say > that this is lcap bug, as