Rebooted, permissive, setroubleshooter going nuts.
Gene Heskett
gene.heskett at verizon.net
Thu Mar 5 14:34:59 UTC 2009
On Thursday 05 March 2009, Daniel J Walsh wrote:
>Paul Howarth wrote:
>> Gene Heskett wrote:
>>> On Thursday 05 March 2009, Gene Heskett wrote:
>>>> On Wednesday 04 March 2009, Gene Heskett wrote:
>>>>> Greetings;
>>>>>
>>>>> And a portion of this lists archive on this box has gone missing to
>>>>> boot.
>>>>> So I can't look up the command to extract all these hits, about once
>>>>> every
>>>>> 2 minutes or so, to a logfile I can post. And when I click on the
>>>>> star,
>>>>> it tells me the connection has been lost to
>>>>> /var/run/setroubleshoot/setroubleshoot_server. But there is a zero
>>>>> length
>>>>> file there, generated when I rebooted to 2.6.29-rc7 5:18 ago WTH?
>>>>>
>>>>> And I just found a very short setroubleshooter.log which I will
>>>>> attach. It
>>>>> looks like it got a tummy ache just a few minutes ago.
>>>>>
>>>>> I think I will follow what I did with 29-rc7, and not build any sound
>>>>> modules for anything except the audigy2, cuz now I have sound, akonadi
>>>>> even starts!
>>>>>
>>>>> Help?
>>>>
>>>> No comment. Can anyone tell me why, when looking at the log
>>>> messages, and
>>>> it tells me to get the full report by running sealert with -l
>>>> hashnumber, I
>>>> as root am denied? From a root shell:
>>>> [root at coyote log]# sealert -l 1ed4cefd-aa3b-4727-b9ef-28b8e2cbb42c
>>>> failed to connect to server: Connection refused
>>>>
>>>> I am back on a 2.6.28.7 kernel now. And setroubleshooter's screen
>>>> alerts in
>>>> time with the kmail pongs of new mail coming are contributing to my
>>>> loss of
>>>> sanity or whatever. Somehow it has decided that fetchmail isn't
>>>> supposed to
>>>> be able to access its users directory/.f, uhh, I was gonna run it
>>>> and get
>>>> the exact file and the connection to its server has been lost, again. I
>>>> thought it was funny that the reject messages were going into the system
>>>> log...
>>>>
>>>> Uptodate Fedora 10. x86_64 running 32 bit.
>>>>
>>>> A 'service setroubleshoot restart' restarts it though. Anybody have
>>>> a clue,
>>>> I seem to be fresh out, and I'm about to compile it out.
>>>
>>> Ok, the restart allowed me to collect the most recent hit from sealert:
>>> ===============================
>>> [root at coyote init.d]# sealert -l
>>> 2ada4c61-64cb-40d7-8268-83488b12426e
>>>
>>> Summary:
>>>
>>> SELinux is preventing procmail (procmail_t) "append" to
>>> /var/log/fetchmail.log
>>> (var_log_t).
>>>
>>> Detailed Description:
>>>
>>> [SELinux is in permissive mode, the operation would have been denied
>>> but was
>>> permitted due to permissive
>>> mode.]
>>> SELinux is preventing procmail (procmail_t) "append" to
>>> /var/log/fetchmail.log
>>> (var_log_t). The SELinux type var_log_t, is a generic type for all
>>> files in the
>>> directory and very few processes (SELinux Domains) are allowed to
>>> write to this
>>> SELinux type. This type of denial usual indicates a mislabeled file.
>>> By default
>>> a file created in a directory has the gets the context of the parent
>>> directory,
>>> but SELinux policy has rules about the creation of directories, that
>>> say if a process running in one SELinux Domain (D1) creates a file in
>>> a directory with a
>>> particular SELinux File Context (F1) the file gets a different File
>>> Context (F2). The policy usually allows the SELinux Domain (D1) the
>>> ability to write, unlink, and append on (F2). But if for some reason
>>> a file (/var/log/fetchmail.log) was created with
>>> the wrong context, this domain will be
>>> denied. The usual solution to this problem is to reset the file
>>> context on the target file, restorecon -v '/var/log/fetchmail.log'.
>>> If the file context does not change from var_log_t, then this is
>>> probably a bug in policy. Please file a bug report
>>> (http://bugzilla.redhat.com/bugzilla/enter_bug.cgi) against the
>>> selinux-policy package. If it does change, you can try your
>>> application again to
>>> see if it works. The file context could have been mislabeled by
>>> editing the file
>>> or moving the file from a different directory, if the file keeps
>>> getting mislabeled, check the init scripts to see if they are
>>> doing something to mislabel the
>>> file.
>>> Allowing Access:
>>>
>>> You can attempt to fix file context by executing restorecon -v
>>> '/var/log/fetchmail.log'
>>> Fix Command:
>>>
>>> restorecon '/var/log/fetchmail.log'
>>>
>>> Additional Information:
>>>
>>> Source Context system_u:system_r:procmail_t:s0
>>> Target Context system_u:object_r:var_log_t:s0
>>> Target Objects /var/log/fetchmail.log [ file ]
>>> Source procmail
>>> Source Path /usr/bin/procmail
>>> Port <Unknown>
>>> Host coyote.coyote.den
>>> Source RPM Packages procmail-3.22-22.fc10
>>> Target RPM Packages
>>> Policy RPM selinux-policy-3.5.13-46.fc10
>>> Selinux Enabled True
>>> Policy Type targeted
>>> MLS Enabled True
>>> Enforcing Mode Permissive
>>> Plugin Name mislabeled_file
>>> Host Name coyote.coyote.den
>>> Platform Linux coyote.coyote.den 2.6.28.7 #6 SMP
>>> PREEMPT
>>> Wed Mar 4 23:08:30 EST 2009 i686 athlon
>>> Alert Count 63
>>> First Seen Sat Feb 28 16:34:21 2009
>>> Last Seen Thu Mar 5 02:20:43 2009
>>> Local ID 2ada4c61-64cb-40d7-8268-83488b12426e
>>> Line Numbers
>>>
>>> Raw Audit Messages
>>>
>>> node=coyote.coyote.den type=AVC msg=audit(1236237643.658:745): avc:
>>> denied { append } for pid=8712 comm="procmail"
>>> path="/var/log/fetchmail.log" dev=sda3 ino=23527557
>>> scontext=system_u:system_r:procmail_t:s0
>>> tcontext=system_u:object_r:var_log_t:s0 tclass=file
>>>
>>> node=coyote.coyote.den type=SYSCALL msg=audit(1236237643.658:745):
>>> arch=40000003 syscall=11 success=yes exit=0 a0=8941670 a1=8941748
>>> a2=8940af8 a3=0 items=0 ppid=2784 pid=8712 auid=4294967295 uid=501
>>> gid=501 euid=501 suid=501 fsuid=501 egid=501 sgid=501 fsgid=501
>>> tty=(none) ses=4294967295 comm="procmail" exe="/usr/bin/procmail"
>>> subj=system_u:system_r:procmail_t:s0 key=(null)
>>>
>>> Thanks Guys.
>>
>> Is this a fetchmail log or a procmail log? What do you expect to get
>> logged here?
>>
>> I guess you're running fetchmail in daemon mode with procmail as local
>> delivery agent?
>>
>> See if this helps:
>>
>> # semanage fcontext -a -t procmail_log_t '/var/log/fetchmail\.log'
>> # restorecon -v /var/log/fetchmail.log
>>
>> Paul.
>>
>>
>>
>> --
>> fedora-selinux-list mailing list
>> fedora-selinux-list at redhat.com
>> https://www.redhat.com/mailman/listinfo/fedora-selinux-list
>
>Currently f10 policy has
>
>./policy/modules/services/mta.te:logging_append_all_logs(system_mail_t)
>./policy/modules/system/init.te:logging_append_all_logs(initrc_t)
>./policy/modules/system/init.te:logging_append_all_logs(daemon)
>
>I think it could be argued that we should allow all confined domains to
>append to any log file, since simple redirection of stdout causes the
>AVC in question. Being able to write to a log file allows a cracked
>program to erase the log contents. Being able to append to a log file
>means you could fill up the system with garbage or write something to a
>log file that would cause some other app or Human to do something bad.
>
>Fetchmail policy does not allow for the creation of a logfile right now.
fetchmail itself cannot create the file either, it must exist.
Which is why my logrotate "mail" script, posted in another message, uses the
copytruncate method of pruning the file(s). I had to touch these originally,
and chown etc them. And I have now added the restorecon -v directive also.
However that is running as root I assume, so should I wrap those two lines
in an "su gene -c"?
> I guess the default is to write to syslog.
I believe the default only writes a log if a logfile is defined in the
matching ~/.fetchmailrc or ~/.procmailrc, otherwise they are silent.
> We need to add a mechansim
>for fetchmail to create a fetchmail_log_t and allow procmail_t to append
>to it.
I think so, Daniel. Something along those lines anyway would certainly reduce
the noise. OTOH, maybe my solution in the logrotate.d/mail file is
sufficient.
But I have NDI exactly when it will trigger a logrotation again. ATM, trying
to debug rules, I have procmail set verbose so its noisy as can be with one
incoming mail generating log entries larger than the usual 1 or 2 paragraph
mail itself.
--
Cheers, Gene
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
The sooner our happiness together begins, the longer it will last.
-- Miramanee, "The Paradise Syndrome", stardate 4842.6
More information about the fedora-selinux-list
mailing list