14th International Conference on Accelerator and Large Experimental Physics Control Systems 06-11 October 2013, The Hyatt Embarcadero Center, San Francisco, California
Challenge to providing a successful Central Configuration Service to support CERN’s
new Controls Diagnostic and Monitoring System
Z. Makonnen, M. Buttner, Z. Zaharieva, CERN, Geneva, Switzerland
Abstract
The Controls Configuration Service (CCS) provides the Configuration Management facilities for the Controls System for all CERN accelerators. It complies with Config-uration Management standards, tracking the life of configuration items and their relationships by allowing identification and triggering change management processes. Data stored in the CCS is extracted and propagated to the controls hardware for remote configuration. The article will present the ability of the CCS to audit items and verify conformance to specification with the implementation of on-line feedback focusing on Front-End Computers (FEC) configurations. Long-standing problems existed in this area such as discrepancies between the actual state of the FEC and the configuration sent to it at reboot. This resulted in difficult-to-diagnose behavior and disturbance for the Operations team. The article will discuss the solution architecture (tailored processes and tools), the development and implementation challenges, as well as the advantages of this ap-proach and the benefits to the user groups – from equipment specialists and controls systems experts to the operators in the Accelerators Controls Centre.
The Controls Configuration Service provides the complete set of configuration manage-ment functionalities needed for the new Diagnostics and Monitoring System despite the im-portant constraints that needed to be respected while transitioning from the old to the new DIAMON System. The main goals, to centralize the DIAMON configurations, to improve the scalability and
the robustness of the configuration processes as well as the responsiveness to changes as configuration data can be loaded into the DIAMON server in real-time have been achieved. Other important advantages are the possibility to integrate the DIAMON data with the con-figuration data for other systems (e.g. FECs configurations, CMW, RBAC, etc.) in order to obtain a global and consistent configuration across the entire Controls System, to eliminate the redundancy of the data between the old DIAMON DB schema and the CCDB and thus improve the quality of the data thereby minimizing the errors and down time due to errone-ous configurations. From the users perspective new and improved tools have been made available to provide the configuration data, a notification system has been put in place, processes, auditing and reporting capabilities have been set up in order to ensure a coherent data flow from the CCS to the DIAMON system and to check the quality of the data and its consistency.
ICALEPCS'13
References
Conclusion
TUPPC024
[1] W. Buczak et al., “DIAMON2-Improved monitoring of CERN’s Accelerator control infrastructure”, ICALEPCS’13, San Francisco, USA, Oct-2013
[2] Z. Zaharieva et al., “Database Foundation for the configuration management of the CERN Accelerator Control System”, ICALEPCS’11, Grenoble, France, Oct-2011.
[3] M. Buttner et al., “Diagnostic and Monitoring CERN Accelerator infrastructure”, ICALEPCS’09, Kobe, Japan, Oct- 2009. [4] ITIL: www.itil-officialsite.com [5] IEEE Std. 828-2012: standards.ieee.org [6] COBIT: www.isaca.org/COBIT [7] Z. Zaharieva, R. Billen, “Rapid Development of Database Interfaces with Oracle APEX, used for the Controls Systems at
CERN”, ICALEPCS’09, Kobe, Japan, Oct-2009 [8] J. Nguyen Xuan, V. Baggiolini, “A Testbed for validating the LHC controls system core before deployment”, ICALEPCS’11, Gre-
noble, France, Oct-2011
Controls Configuration Logical Domains
Central Controls Configuration Service
Automatic Generation of configuration Tag - The generation of configuration tags is regenerated every time that the
corresponding configuration item is updated. - It simplifies the interface between the CCDB and the DIAMON server
- The configuration tags have predefined types, e.g. Create Computer, Remove Equipment, Load Limit, Update Equipment Rule, etc
Inheritance of the configuration items
- An inheritance mechanism, based on the hierarchy depicted, is used to
monitor consistently the configuration items.
- The inheritance of the configuration items is used not only during the generation of the configuration tags, but at run time as well.
Data Browsing Interfaces
160 reports
Audit tools
300 users
Oracle APEX technology
Development Environments to run unit
Testing Environments to run functional test
NEXT Controls Configuration Environment part of the
Controls Test Bed since 2010 to run continuous integra-
tion test.
Browsing Tools
Quality Assurance
Operations User Applications
Data Editing Interfaces
Configuration items Design
Enable monitoring
Configuration Upload
Strict authorization
Parallel operation of the old and new
DIAMON systems
Diamon2 Diamon1
DIAMON CCDB
Challenges
Editing Tools
Database Statistics
Components
monitored
~3000
Equipment > 6,000
Metrics 96,000
Commands 24,000
Rules 54,000
Scalability
Functionalities
Data model complexity
DIAMON Cache DB DIAMON Short Term
Log DB
DIAgnostic and MONitoring system
Diamon2 Server
CCDB
XML
2 - Optimization
1 - Reorganization
Configuration data for
the new DIAMON
Devices (Virtual..)
Property 1
Property 2
Property n
... .
CMW RBA OASIS …
Controls Configuration model
Equipment Metric
Rule Command
DIAMON model
?
Complete integration
with configuration model
How to fit the DIAMON model into the Device-Property
model?
Implementing coherent data flows and configuration
management processes
Mechanism to ensure exact TAG
sequence generation.
Processes to guaranty coherent
state of the configuration items.
DIAMON TAG Configuration
TAG ID
2– TAG Generation
4– Upload Configuration
1– Enable monitoring 3– Notification
Database Statistics
Tables 960
Constraints 5,571
Lines PL/SQL code 70,000
Volume 100 GB
Controls Configuration DB
Top Related