conf/181681: periodic inline security output missing with daily_show_success=NO

[Jason Unovitch]

The following reply was made to PR conf/181681; it has been noted by GNATS.

From: Jason Unovitch <jason.unovitch at gmail.com>
To: "freebsd at omnilan.de" <freebsd at omnilan.de>
Cc: bug-followup at FreeBSD.org
Subject: Re: conf/181681: periodic inline security output missing with daily_show_success=NO
Date: Tue, 17 Sep 2013 20:59:55 -0400

 Harry,
 Conf/178611 fixed the issue and that is what is in 9.2.  I just tested=20
 each use case on 9.2-RC4 and they work as reading the=20
 450.status-security and /usr/sbin/periodic would suggest.  With this=20
 /etc/periodic.conf:
 
 daily_show_success=3D"NO"
 daily_status_security_inline=3D"YES"
 
 The "daily run output" periodic email shows the inline output (abbreviate=
 d)
 =2E..
 Security check:
 
 Checking setuid files and devices:
 =2E..continued
 
 It looks fine.  Both these /etc/periodic.conf examples do the exact same =
 
 thing.  The 450.status-security script sets a blank string right at the=20
 top because of inline being set and the rc=3D3 case results in it always =
 
 being show.  Outputting to a file or other user gets ignored because of=20
 inline.
 daily_show_success=3D"NO"
 daily_status_security_inline=3D"YES"
 daily_status_security_output=3D/tmp/file
 
 daily_show_success=3D"NO"
 daily_status_security_inline=3D"YES"
 daily_status_security_output=3Droot
 
 
 Without the inline, this one results in the security output being=20
 written to a file.
 daily_show_success=3D"NO"
 daily_status_security_output=3D/tmp/file
 
 This one results in the security output going to my user account.
 daily_show_success=3D"NO"
 daily_status_security_output=3Djason
 
 -Jason Unovitch
 
 On 09/16/2013 05:09 AM, freebsd at omnilan.de wrote:
 >   Bez=FCglich Jason Unovitch's Nachricht vom 04.09.2013 02:21 (localtim=
 e):
 >> Harald,
 >> By any chance do you have a work around for the original broken
 >> version of 450.status.security in 9.1-RELEASE? I submitted PR conf/178=
 611
 > Hello Jason,
 >
 > sorry for that late reply, just want to let you know that I don't have
 > anything further arround and unfortunately haven't looked into your
 > conf/178611 yet.
 >
 >> for a patch in 9.1-RELEASE and have been using the patched version
 >> successfully with the same options ever since. I have also tested
 >> 9.2-RC3 and do not see the issue described. Additionally, 6 lines
 >> below the suggested patch the "rc=3D1" used get overridden anyway by a=
 n
 > Couldn't you reproduce the problem from your PR (conf/178611) with
 > 9.2-RC3 or the probleme here (conf/181681)? The problem here is clearly=
 
 > =84by design=93 and should be seen unconditionally.
 >
 > Thanks,
 >
 > -Harry
 >