The zabbix agent is responsible for collecting the system state and other
monitorable aspects. This include
- information from /proc
- read attributes of various files (tamper detection)
- connect to the ssh service (check if it is reachable)
- get file system information
- read login information
- ...
It should be noted that the agent can do a lot more, depending on the target
system (what is being monitored) and the running services. The allowed
privileges here will in the future expand more as more templates are
checked.
Update: follow styleguide
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
The zabbix server also connects to the agents (this is called "active
monitoring" in the zabbix terms). So we create a zabbix_agent_tcp_connect
interface, use it for the zabbix_t domain and, since zabbix can use
hostname-based connections, allow DNS resolving for the zabbix server.
Update: Follow styleguide
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
The zabbix agent has its own dedicated port (10050) on which it needs to
bind/listen.
Also, the agent connects to the server so we add the zabbix_tcp_connect
interface (shamelessly copied from mysql_tcp_connect) and use it for the
zabbix_agent_t domain.
Update: structure interface calls more closely to styleguide
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
The zabbix agent should be confined within its own domain. We start with the
definition of a small(er) skeleton to work from. This includes proper file
context definitions, standard interdomain privileges (which are quite
similar to those of the server) and the proper log- and pid access
privileges.
Update: attempt to follow styleguide more closely
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
The zabbix server uses a dedicated port (10051). We define it and allow the
zabbix server to bind/listen on it.
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
On Mon, Jun 13, 2011 at 10:28:15AM +0200, Sven Vermeulen wrote:
> Zabbix servers use shared memory to keep common information and structures.
> This is implemented on tmpfs. We support this by introducing a
> zabbix_tmpfs_t type and allow the server proper access to it.
After a small discussion and a few more tests, drop the "dir" in
fs_tmpfs_filetrans.
For posterity's sake, this is the denial one gets when no tmpfs_t related
privileges are given:
Jun 13 11:24:06 build kernel: [ 213.054230] type=1400
audit(1307957046.001:106): avc: denied { read write } for pid=3162
comm="zabbix_agentd" path=2F535953563663303132323534202864656C6574656429
dev=tmpfs ino=32768 scontext=system_u:system_r:zabbix_agent_t
tcontext=system_u:object_r:tmpfs_t tclass=file
With fs_tmpfs_filetrans(..., file) the same denial is given, but as
tcontext=zabbix_tmpfs_t. Hence the rw_files_pattern() enhancement.
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
The zabbix server process is a multi-process system.
In order to, for instance, shut it down, signalling within the domain is
necessary. Otherwise, the processes remain running.
Also, since there are multiple processes trying to use the same log file,
the zabbix server uses semaphores to ensure proper access to the log files
(concurrency).
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
On Tue, Mar 22, 2011 at 08:44:49AM -0400, Christopher J. PeBenito wrote:
> > +manage_dirs_pattern(courier_authdaemon_t, courier_var_lib_t, courier_var_lib_t)
>
> It sounds like this should be create_dirs_pattern instead.
Indeed, create_dirs_pattern is sufficient here. Retry ;-)
During startup, authdaemon creates /var/lib/courier/authdaemon and creates a
socket for communication with courier imapd and pop3d daemons.
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
The attached patch allows postgresql_t domain to read selabel definition files
(such as /etc/selinux/targeted/contexts/sepgsql_contexts).
The upcoming version (v9.1) uses selabel_lookup(3) to assign initial security context
of database objects, we need to allow this reference.
Thanks,
--
NEC Europe Ltd, SAP Global Competence Center
KaiGai Kohei <kohei.kaigai@eu.nec.com>
Updates on the file contexts, supporting AMD64 multilib environment
( Patch 10 has been revoked a-la-last-minute, needs further testing )
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
postalias should stay bin_t, is manually executed (no role executes
postfix_master_exec_t as it is only to be launched through init scripts).
The postalias command is used to regenerate the aliases.db file from the
mail aliases and as such is a system administrative activity. However, by
default, no role has execute rights on any postfix_master_exec_t domains as
the domain is apparently meant only to be started from the run_init_t
domain.
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
Cyrus sasl by default looks in /var/lib/sasl2 for its PID file, socket
creation and lock files.
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
Update on the file contexts for courier-imap. Also fixes a few context
directives which didn't update the directory itself.
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
When starting the X server from the console (using the startx script
that is being shipped with package xinit from X.Org), a few more
permissions are needed from the reference policy.
The label is for a file created by the startx script (from X.Org) and
the module being requested is ipv6 (which can be disabled by other
means).
Allow xserver_restricted_role domains to call/start Xorg (using startx), fixes
15-second lag/timeout (needs siginh permission as provided by
xserver_domtrans).
Apparently, the 15-second lag (or some other behavior) was already detected
in the past, giving rise to the SIGINH permission in the xserver_domtrans()
interface.
However, domains that are given the xserver_(restricted_)role do not call
the xserver_domtrans but rather the "standard" domtrans_pattern.
The new patch suggests to use xserver_domtrans in the
xserver_restricted_role, which automatically includes the siginh permission
then.
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
This patch adds a new interface to the cpufreqselector module
to allow dbus chat. It then uses such interface to allow dbus chat
with system_dbusd_t and xdm_t. This patch also adds some other
permissions needed to run cpufreqselector.
As one of entrypoint application, crond_t should have had the
files_polyinstantiate_all() interface called so that pam_namespace.so
could work well in crond_t. Otherwise the crond_t lacks the sys_admin
permission to make use of pam_namespace.so
BTW, the allow_polyinstantiation boolean need to be toggled true
accordingly.
Signed-off-by: Harry Ciao <qingtao.cao@windriver.com>
Entry point applications such as crond or atd use pam_loginuid.so for
the session phase of their PAM config files to set the process loginuid
attribute. Accordingly logging_set_loginuid interface should have been
called, otherwise we could run into below error message:
type=USER_START msg=audit(1296377641.212:213): user pid=2633 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:crond_t:s0-s15:c0.c1023 msg='op=PAM:session_open acct="root" exe="/usr/sbin/crond" (hostname=?, addr=?, terminal=cron res=failed)'
type=USER_END msg=audit(1296377641.220:214): user pid=2633 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:crond_t:s0-s15:c0.c1023 msg='op=PAM:session_close acct="root" exe="/usr/sbin/crond" (hostname=?, addr=?, terminal=cron res=failed)'
type=AVC msg=audit(1296377641.196:212): avc: denied { audit_control } for pid=2633 comm="crond" capability=30 scontext=system_u:system_r:crond_t:s0-s15:c0.c1023 tcontext=system_u:system_r:crond_t:s0-s15:c0.c1023 tclass=capability
BTW, other entrypoint applications such as sshd/login/remote have had
this interface called for their domains.
Signed-off-by: Harry Ciao <qingtao.cao@windriver.com>
On my system, I use XFCE and start X from the commandline (using "startx")
rather than through a graphical DM. During the start-up, XFCE4 creates
temporary ICE files in /tmp (like /tmp/.xfsm-ICE-ABCDEF) which are later
read in by iceauth and at some point X.
I'm not that good at the entire ICE stuff, but without this, I was unable to
shut down my session ("log off").
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
From 78d6e4acfc000b07dbf85b076fa523e95e72da3f Sun, 13 Feb 2011 18:55:53 +0100
From: Dominick Grift <domg472@gmail.com>
Date: Sun, 13 Feb 2011 18:55:09 +0100
Subject: [PATCH] Users calling apache_role were not able to manage httpd_user_content_t files, directories and symbolic links.
Users calling apache_role were not able to manage httpd_user_content_t files, directories and symbolic links.
Signed-off-by: Dominick Grift <domg472@gmail.com>
On an Xorg 1.9 system with evdev driver (for keyboard InputClass), the
xserver_t domain needs to be able to read from the proper device nodes as
well as query the udev_tbl_t directory and udev itself.
Signed-off-by: Sven Vermeulen <sven.vermeulen@siphos.be>
The attached patch adds a few database object classes, as follows:
* db_schema
------------
A schema object performs as a namespace in database; similar to
directories in filesystem.
It seems some of (but not all) database objects are stored within
a certain schema logically. We can qualify these objects using
schema name. For example, a table: "my_tbl" within a schema: "my_scm"
is identified by "my_scm.my_tbl". This table is completely different
from "your_scm.my_tbl" that it a table within a schema: "your_scm".
Its characteristics is similar to a directory in filesystem, so
it has similar permissions.
The 'search' controls to resolve object name within a schema.
The 'add_name' and 'remove_name' controls to add/remove an object
to/from a schema.
See also,
http://developer.postgresql.org/pgdocs/postgres/sql-createschema.html
In the past discussion, a rubix folks concerned about no object
class definition for schema and catalog which is an upper level
namespace. Since I'm not certain whether we have a disadvantage
when 'db_schema' class is applied on catalog class, I don't add
this definition yet.
Default security context of 'db_table' and 'db_procedure' classes
get being computed using type_transition with 'db_schema' class,
instead of 'db_database' class. It reflects logical hierarchy of
database object more correctly.
* db_view
----------
A view object performs as a virtual table. We can run SELECT
statement on views, although it has no physical entities.
The definition of views are expanded in run-time, so it allows
us to describe complex queries with keeping readability.
This object class uniquely provides 'expand' permission that
controls whether user can expand this view, or not.
The default security context shall be computed by type transition
rule with a schema object that owning the view.
See also,
http://developer.postgresql.org/pgdocs/postgres/sql-createview.html
* db_sequence
--------------
A sequence object is a sequential number generator.
This object class uniquely provides 'get_value', 'next_value' and
'set_value' permissions. The 'get_value' controls to reference the
sequence object. The 'next_value' controls to fetch and increment
the value of sequence object. The 'set_value' controls to set
an arbitrary value.
The default security context shall be computed by type transition
rule with a schema object that owning the sequence.
See also,
http://developer.postgresql.org/pgdocs/postgres/sql-createsequence.html
* db_language
--------------
A language object is an installed engine to execute procedures.
PostgreSQL supports to define SQL procedures using regular script
languages; such as Perl, Tcl, not only SQL or binary modules.
In addition, v9.0 or later supports DO statement. It allows us to
execute a script statement on server side without defining a SQL
procedure. It requires to control whether user can execute DO
statement on this language, or not.
This object class uniquely provides 'implement' and 'execute'
permissions. The 'implement' controls whether a procedure can
be implemented with this language, or not. So, it takes security
context of the procedure as subject. The 'execute' controls to
execute code block using DO statement.
The default security context shall be computed by type transition
rule with a database object, because it is not owned by a certain
schema.
In the default policy, we provide two types: 'sepgsql_lang_t' and
'sepgsql_safe_lang_t' that allows unpriv users to execute DO
statement. The default is 'sepgsql_leng_t'.
We assume newly installed language may be harm, so DBA has to relabel
it explicitly, if he want user defined procedures using the language.
See also,
http://developer.postgresql.org/pgdocs/postgres/sql-createlanguage.htmlhttp://developer.postgresql.org/pgdocs/postgres/sql-do.html
P.S)
I found a bug in MCS. It didn't constraint 'relabelfrom' permission
of 'db_procedure' class. IIRC, I fixed it before, but it might be
only MLS side. Sorry.
Thanks,
--
KaiGai Kohei <kaigai@ak.jp.nec.com>
policy/flask/access_vectors | 29 ++++++++
policy/flask/security_classes | 6 ++
policy/mcs | 16 ++++-
policy/mls | 58 ++++++++++++++-
policy/modules/kernel/kernel.if | 8 ++
policy/modules/services/postgresql.if | 125 +++++++++++++++++++++++++++++++--
policy/modules/services/postgresql.te | 116 +++++++++++++++++++++++++++++-
7 files changed, 342 insertions(+), 16 deletions(-)
On 01/05/2011 08:48 AM, Christopher J. PeBenito wrote:
> On 12/16/10 12:32, Paul Nuzzi wrote:
>> On 12/15/2010 03:54 PM, Christopher J. PeBenito wrote:
>>> On 12/10/10 18:22, Paul Nuzzi wrote:
>>>> Added labeled IPSec support to hadoop. SELinux will be able to enforce what services are allowed to
>>>> connect to. Labeled IPSec can enforce the range of services they can receive from. This enforces
>>>> the architecture of Hadoop without having to modify any of the code. This adds a level of
>>>> confidentiality, integrity, and authentication provided outside the software stack.
>>>
>>> A few things.
>>>
>>> The verb used in Reference Policy interfaces for peer recv is recvfrom
>>> (a holdover from previous labeled networking implementations). So the
>>> interfaces are like hadoop_recvfrom_datanode().
>>
>> Easy change.
>>
>>> It seems like setkey should be able to setcontext any type used on ipsec
>>> associations. I think the best thing would be to add additional support
>>> to either the ipsec or corenetwork modules (I haven't decided which one
>>> yet) for associations. So, say we have an interface called
>>> ipsec_spd_type() which adds the parameter type to the attribute
>>> ipsec_spd_types. Then we can have an allow setkey_t
>>> ipsec_spd_types:association setkey; rule and we don't have to update it
>>> every time more labeled network is added.
>>
>> That seems a lot less clunky than updating setkey every time we add a new association.
>>
>>> This is definitely wrong since its not a file:
>>> +files_type(hadoop_lan_t)
>>
>> Let me know how you would like to handle associations and I could update the
>> patch.
>
> Lets go with putting the associations in corenetwork.
>
>> Will the files_type error be cleared up when we re-engineer this?
>
> I'm not sure what you mean. The incorrect rule was added in your patch.
>
Adds labeled IPSec policy to hadoop to control the remote processes that are allowed to connect to the cloud's services.
Signed-off-by: Paul Nuzzi <pjnuzzi@tycho.ncsc.mil>
On 12/15/2010 03:17 PM, Christopher J. PeBenito wrote:
> On 12/13/10 10:39, Paul Nuzzi wrote:
>> On 12/11/2010 04:01 AM, Dominick Grift wrote:
>> On 12/11/2010 12:22 AM, Paul Nuzzi wrote:
>>
>> Does hadoop depend on kerberos? If no then kerberos_use should probably
>> be optional.
>>
>>
>>> The new version of hadoop added Kerberos for authentication.
>
> So, to be explicit, its an unconditional requirement?
Yes. I think all future versions of hadoop will be kerberos enabled.
> It seems like there should be a hadoop_home_t that is
> userdom_user_home_content()
Updated.
Signed-off-by: Paul Nuzzi <pjnuzzi@tycho.ncsc.mil>
* Rearrange rules in the template.
* Remove samhain_etc_t:dir perms since there are no such dirs.
* Add extra docs in samhain_domtrans().
* Include samhaind_t in admin interface process perms.
Note, extra privileges may need to be granted to the samhain domain
if its default configuration file(/etc/samhainrc) is changed.
The samhain program could be used in the following way:
(In secadm_r role)
1. Initialize filesystem signature database:
newrole -l s15:c0.c1023 -p -- -c "samhain -t init"
(Note, the current secadm console will be blocked until
the database is completed)
2. Start samhain deamon to check filesystem integrity
newrole -l s15:c0.c1023 -p -- -c "samhain -t check -D"
3. Update filesystem signature database:
newrole -l s15:c0.c1023 -p -- -c "samhain -t update"
(In sysadm_r role)
1. Start samhain in daemon mode:
run_init /etc/init.d/samhain start
2. Stop samhain daemon:
run_init /etc/init.d/samhain stop
3. Check samhain daemon status:
run_init /etc/init.d/samhain status
4. Read/write samhain log files:
newrole -l s15:c0.c1023 -p -- -c "cat /var/log/samhain_log"
5. Remove samhain database files
newrole -l s15:c0.c1023 -p -- -c "rm /var/lib/samhain/samhain_file"
Note:
1. Stop samhain daemon before updating signature database.
2. Don't try to start samhain daemon twice.
3. Need to toggle SELinux into the Permissive mode in order to remove
the samhain_log files from /var/log/.
Signed-off-by: Harry Ciao <qingtao.cao@windriver.com>
* a pass cleaning up the style.
* adjusted some regular expressions in the file contexts: .* is the same as (.*)? since * means 0 or more matches.
* renamed a few interfaces
* two rules that I dropped as they require further explanation
> +files_read_all_files(hadoop_t)
A very big privilege.
and
> +fs_associate(hadoop_tasktracker_t)
This is a domain, so the only files with this type should be the /proc/pid ones, which don't require associate permissions.
On 10/04/2010 02:18 PM, Christopher J. PeBenito wrote:
> On 10/04/10 13:15, Paul Nuzzi wrote:
>> On 10/01/2010 01:56 PM, Christopher J. PeBenito wrote:
>>> On 10/01/10 11:17, Paul Nuzzi wrote:
>>>> On 10/01/2010 08:02 AM, Dominick Grift wrote:
>>>>> On Thu, Sep 30, 2010 at 03:39:40PM -0400, Paul Nuzzi wrote:
>>>>>> I updated the patch based on recommendations from the mailing list.
>>>>>> All of hadoop's services are included in one module instead of
>>>>>> individual ones. Unconfined and sysadm roles are given access to
>>>>>> hadoop and zookeeper client domain transitions. The services are started
>>>>>> using run_init. Let me know what you think.
>>>>>
>>>>> Why do some hadoop domain need to manage generic tmp?
>>>>>
>>>>> files_manage_generic_tmp_dirs(zookeeper_t)
>>>>> files_manage_generic_tmp_dirs(hadoop_t)
>>>>> files_manage_generic_tmp_dirs(hadoop_$1_initrc_t)
>>>>> files_manage_generic_tmp_files(hadoop_$1_initrc_t)
>>>>> files_manage_generic_tmp_files(hadoop_$1_t)
>>>>> files_manage_generic_tmp_dirs(hadoop_$1_t)
>>>>
>>>> This has to be done for Java JMX to work. All of the files are written to
>>>> /tmp/hsperfdata_(hadoop/zookeeper). /tmp/hsperfdata_ is labeled tmp_t while
>>>> all the files for each service are labeled with hadoop_*_tmp_t. The first service
>>>> will end up owning the directory if it is not labeled tmp_t.
>>>
>>> The hsperfdata dir in /tmp certainly the bane of policy writers. Based on a quick look through the policy, it looks like the only dir they create in /tmp is this hsperfdata dir. I suggest you do something like
>>>
>>> files_tmp_filetrans(hadoop_t, hadoop_hsperfdata_t, dir)
>>> files_tmp_filetrans(zookeeper_t, hadoop_hsperfdata_t, dir)
>>>
>>> filetrans_pattern(hadoop_t, hadoop_hsperfdata_t, hadoop_tmp_t, file)
>>> filetrans_pattern(zookeeper_t, hadoop_hsperfdata_t, zookeeper_tmp_t, file)
>>>
>>
>> That looks like a better way to handle the tmp_t problem.
>>
>> I changed the patch with your comments. Hopefully this will be one of the last updates.
>> Tested on a CDH3 cluster as a module without any problems.
>
> There are several little issues with style, but it'll be easier just to fix them when its committed.
>
> Other comments inline.
>
I did my best locking down the ports hadoop uses. Unfortunately the services use high, randomized ports making
tcp_connect_generic_port a must have. Hopefully one day hadoop will settle on static ports. I added hadoop_datanode port 50010 since it is important to lock down that service. I changed the patch based on the rest of the comments.
Signed-off-by: Paul Nuzzi <pjnuzzi@tycho.ncsc.mil>
This is based on Fedoras' miscfiles_cert_type implementation.
The idea was that openvpn needs to be able read home certificates (home_cert_t) which is not implemented in refpolicy yet, as well as generic cert_t certificates.
Note that openvpn is allowed to read all cert_types, as i know that it needs access to both generic cert_t as well as (future) home_cert_t. Dwalsh noted that other domains may need this as well but because i do not know exactly which domains i will not changes any other domains call to generic cert type interfaces.
Signed-off-by: Dominick Grift <domg472@gmail.com>
Retry: forgot to include attribute mmap_low_domain_type attribute to domain_mmap_low() :
Inspired by similar implementation in Fedora.
Wine and vbetool do not always actually need the ability to mmap a low area of the address space.
In some cases this can be silently denied.
Therefore introduce an interface that facilitates "mmap low" conditionally, and the corresponding boolean.
Also implement booleans for wine and vbetool that enables the ability to not audit attempts by wine and vbetool to mmap a low area of the address space.
Rename domain_mmap_low interface to domain_mmap_low_uncond.
Change call to domain_mmap_low to domain_mmap_low_uncond for xserver_t. Also move this call to distro redhat ifndef block because Redhat does not need this ability.
Signed-off-by: Dominick Grift <domg472@gmail.com>
Libcgroup moved cgclear to /sbin.
Confine it so that initrc_t can domain transition to the cgclear_t domain. That way we do not have to extend the initrc_t domains policy.
We might want to add cgroup_run_cgclear to sysadm module.
Signed-off-by: Dominick Grift <domg472@gmail.com>
When cgroup policy was merged, some changes were made. One of these changes was the renaming of the type for cgroup rules engine daemon configuration file. The cgroup_admin interface was not modified to reflect this change.
Signed-off-by: Dominick Grift <domg472@gmail.com>
Signed-off-by: Chris PeBenito <cpebenito@tresys.com>
Clean up network control section.
Implement hddtemp_etc_t for /etc/sysconfig/hddtemp. The advantages are:
- hddtemp_t no longer needs access to read all generic etc_t files.
- allows us to implement a meaningful hddtemp_admin()
Signed-off-by: Dominick Grift <domg472@gmail.com>
Signed-off-by: Chris PeBenito <cpebenito@tresys.com>
Removed 'SELinux policy for' from policy summaries
Removed rgmanager interface for semaphores (doesn't appear to be needed or used)
Removed redundant calls to libs_use_ld_so and libs_use_shared_libs
Fixed rhcs interface names to match naming rules
Merged tmpfs and semaphore/shm interfaces
Edits:
- Style and whitespace fixes
- Removed interfaces for default_t from ricci.te - this didn't seem right
- Removed link files from rgmanager_manage_tmpfs_files
- Removed rdisc.if patch. it was previously committed
- Not including kernel_kill interface call for rgmanager
- Not including ldap interfaces in rgmanager.te (currently not in refpolicy)
- Not including files_create_var_run_dirs call for rgmanager (not in refpolicy)
Edits:
- Style and whitespace fixes
- Removed read_lnk_files_pattern from nx_read_home_files
- Delete declaration of nx_server_home_ssh_t and files_type since the template already does this
asterisk_manage_lib_files(logrotate_t)
asterisk_exec(logrotate_t)
Needs net_admin
Drops capabilities
connects to unix_stream
execs itself
Requests kernel load modules
Execs shells
Connects to postgresql and snmp ports
Reads urand and generic usb devices
Has mysql and postgresql back ends
sends mail
- signal interfaces
- fusefs support
- bug 566984: getattrs on all blk and chr files
Did not include:
- changes related to samba_unconfined_script_t and samba_unconfined_net_t
- samba_helper_template (didn't appear to be used)
- manage_lnk_files_pattern in samba_manage_var_files
- signal allow rule in samba_domtrans_winbind_helper
- samba_role_notrans
- userdom_manage_user_home_content
Some style and spacing fixes
Removed manage_var_run and manage_var_lib interfaces
Added missing requires to admin interface
Removed permissive line
Fixed some spacing / style issues
Matthew Ife
Chris PeBenito
Sven Vermeulen
Chris PeBenito
Kohei Kaigai
Elia Pinto
Guido Trentalancia
Harry Ciao
Dominick Grift
Stephen Smalley
KaiGai Kohei
Paul Nuzzi
Jeremy Solt