Setting Up Virtual Security Zones in a Linux Cluster
When DSM is loaded, it enforces default permissive security rules. To achieve cluster subdivision, we have to edit the DSP file ~/dsi/etc/SampleDSP.xml and replace all the existing security rules with our own.
PhoneMania's sockets are assigned ScID=10, and RingBell uses ScID=20. The following rule assigns ScID=10 to PhoneMania's entry point UDP socket (port 8800):
<class_SOCKET_INIT_rule> <protocol>UDP</protocol> <port>8800</port> <SnID>ALL</SnID> <ScID>10</ScID> </class_SOCKET_INIT_rule>
We need three other similar rules: one for PhoneMania's back-end server and two others for ScID=20 for RingBell.
Then, PhoneMania's processes (source ScID=10) are allowed to create, send or receive messages on sockets they own (that is, with target ScID=10):
<class_SOCKET_rule> <sScID>10</sScID> <sSnID>ALL</sSnID> <tScID>10</tScID> <tSnID>ALL</tSnID> <allow>CREATE CONNECT LISTEN RECEIVE SEND</allow> </class_SOCKET_rule>
Create a similar rule for RingBell's processes.
Of course, communication between ScID=10 and 20 must be denied. This is done simply by setting no socket permissions at all between those ScIDs:
<class_SOCKET_rule> <sScID>10</sScID> <sSnID>ALL</sSnID> <tScID>20</tScID> <tSnID>ALL</tSnID> <allow></allow> </class_SOCKET_rule>
Create a similar rule between source ScID=20 and target ScID=10.
Back-end and entry point servers of a given operator may be located on different nodes of a cluster; remember, we're sharing a cluster, not dedicating one node to RingBell and another to PhoneMania. Hence, processes of PhoneMania (source ScID=10) must be able to communicate with other PhoneMania processes (target ScID=10) through the network. The same holds true for RingBell:
<class_NETWORK_rule> <sScID>10</sScID> <sSnID>ALL</sSnID> <tScID>10</tScID> <tSnID>ALL</tSnID> <deny>NETWORK_RECEIVE</deny> </class_NETWORK_rule>
Finally, PhoneMania (ScID=10) and RingBell (ScID=10) processes usually are launched from a shell (default ScID=2). So, basically, we need to allow the shell to create a new process. This is done with a transition rule:
<class_TRANSITION_rule> <parent_ScID> 2 </parent_ScID> <SnID>ALL</SnID> <binary_ScID>10</binary_ScID> <new_ScID>10</new_ScID> </class_TRANSITION_rule>
The binary_ScID is the ScID explicitly assigned to the binary. Remember, we assigned ScIDs to PhoneManiaBE or PhoneManiaEP using SetSID. The new_ScID is the ScID assigned to the new process also created. As access to sockets 8800 and 8801 is granted only to ScID=10, for PhoneMania, the new process should be assigned ScID=10. A similar rule for RingBell should be created.
This is all we need in the DSP—12 security rules. Then, we update the security policy in the whole cluster by sending an update event to the security server:
[colby]$ cd ~/dsi/SS/test/demoSecOM [colby]$ ./dsiUpdatePolicy ~/dsi/etc/DSP.xml
The security server reads the updated DSP file (located in ~/dsi/etc/DSP.xml) and displays warnings if there are syntax errors. Finally, it automatically sends updates to each security manager; no need to log in to each machine to update the security policy manually or develop your own version of system management Perl-based software. This feature can be a benediction when you have a cluster of hundreds of nodes physically spread out around the world (think grid computing).
Now, it's time to try the case again in which PhoneMania forwards requests to RingBell's back-end server:
[colby demo]$ ./TelecomClient -h 10.1.1.1 -p 8800 Requesting quotation for Ericsson Quote Ericsson ... [colby demo]$ ./PhoneManiaEP -h 10.1.1.1 -p 8800 -b 10.1.1.2 -r 9001 PHONEMANIA: bind/connect on 10.1.1.1:8800 = 0 PHONEMANIA: bind/connect on 10.1.1.2:9001 = 0 Quote Ericsson Quotation request received ... [munster demo]$ ./RingBellBE -h 10.1.1.2 -p 9001 RINGBELL: bind on 10.1.1.2:9001 ...
On the other node (munster), we notice that RingBell's back-end server is no longer handling PhoneMania's requests, although PhoneMania illicitly redirects them to RingBell. You can use logs generated by DSI, located in /var/log/messages to trace illicit requests:
May 6 07:47:31 munster kernel: DSI-LSM MODULE - dsi_sock_rcv_skb check permission sscid 10 ssnid 1 tscid 20 May 6 07:47:31 munster kernel: DSI-LSM MODULE Error - dsi_sock_rcv_skb - No Permission
|Happy Birthday Linux||Aug 25, 2016|
|ContainerCon Vendors Offer Flexible Solutions for Managing All Your New Micro-VMs||Aug 24, 2016|
|Updates from LinuxCon and ContainerCon, Toronto, August 2016||Aug 23, 2016|
|NVMe over Fabrics Support Coming to the Linux 4.8 Kernel||Aug 22, 2016|
|What I Wish I’d Known When I Was an Embedded Linux Newbie||Aug 18, 2016|
|Pandas||Aug 17, 2016|
- Happy Birthday Linux
- Download "Linux Management with Red Hat Satellite: Measuring Business Impact and ROI"
- ContainerCon Vendors Offer Flexible Solutions for Managing All Your New Micro-VMs
- What I Wish I’d Known When I Was an Embedded Linux Newbie
- Updates from LinuxCon and ContainerCon, Toronto, August 2016
- New Version of GParted
- NVMe over Fabrics Support Coming to the Linux 4.8 Kernel
- All about printf
- Tor 0.2.8.6 Is Released
- Blender for Visual Effects
With all the industry talk about the benefits of Linux on Power and all the performance advantages offered by its open architecture, you may be considering a move in that direction. If you are thinking about analytics, big data and cloud computing, you would be right to evaluate Power. The idea of using commodity x86 hardware and replacing it every three years is an outdated cost model. It doesn’t consider the total cost of ownership, and it doesn’t consider the advantage of real processing power, high-availability and multithreading like a demon.
This ebook takes a look at some of the practical applications of the Linux on Power platform and ways you might bring all the performance power of this open architecture to bear for your organization. There are no smoke and mirrors here—just hard, cold, empirical evidence provided by independent sources. I also consider some innovative ways Linux on Power will be used in the future.Get the Guide