summaryrefslogtreecommitdiffstats
path: root/docs
diff options
context:
space:
mode:
authorReshmasree <reshmasree.chamarthi2@wipro.com>2020-11-10 18:00:05 +0530
committerAhila <ahila.pandaram@wipro.com>2020-11-11 19:15:01 +0530
commit2cf33d0d1cecad5a309d428aa44f7740ca48032e (patch)
tree05c515ff22dfc0b3db84dd02b6ebaeecc5786290 /docs
parent7634901b6d877a6126978ee11ce489ebfc4e3669 (diff)
Add slice-analysis-ms documents
Application configurations and troubleshooting steps added Removed references to PNDA Issue-ID: DCAEGEN2-2515 Signed-off-by: Reshmasree <reshmasree.chamarthi2@wipro.com> Signed-off-by: Ahila P <ahila.pandaram@wipro.com> Change-Id: I941420b459b60e0e8f4c63463b16d7f6602cbfeb
Diffstat (limited to 'docs')
-rw-r--r--docs/sections/.architecture.rst.swobin12288 -> 0 bytes
-rw-r--r--docs/sections/architecture.rst223
-rw-r--r--docs/sections/services/serviceindex.rst53
-rw-r--r--docs/sections/services/slice-analysis-ms/index.rst51
-rw-r--r--docs/sections/services/slice-analysis-ms/installation.rst75
-rw-r--r--docs/sections/services/slice-analysis-ms/slice_analysis_ms_arch.jpgbin0 -> 58211 bytes
-rw-r--r--docs/sections/services/slice-analysis-ms/slice_analysis_ms_overview.rst65
-rw-r--r--docs/sections/services/slice-analysis-ms/slice_analysis_ms_troubleshooting.rst46
8 files changed, 377 insertions, 136 deletions
diff --git a/docs/sections/.architecture.rst.swo b/docs/sections/.architecture.rst.swo
deleted file mode 100644
index cc4da602..00000000
--- a/docs/sections/.architecture.rst.swo
+++ /dev/null
Binary files differ
diff --git a/docs/sections/architecture.rst b/docs/sections/architecture.rst
index 8231f44d..b5fe7dcb 100644
--- a/docs/sections/architecture.rst
+++ b/docs/sections/architecture.rst
@@ -1,110 +1,113 @@
-.. This work is licensed under a Creative Commons Attribution 4.0 International License.
-.. http://creativecommons.org/licenses/by/4.0
-.. _architecture:
-
-
-Architecture
-============
-
-Data Collection Analytics and Events (DCAE) is the primary data collection and analysis system of ONAP. DCAE architecture comprises of DCAE Platform and
-DCAE Service components making DCAE flexible, elastic, and expansive enough for supporting the potentially infinite number of ways of constructing intelligent
-and automated control loops on distributed and heterogeneous infrastructure.
-
-DCAE Platform supports the functions to deploy, host and perform LCM applications of Service components. DCAE Platform components enable model driven deployment of
-service components and middleware infrastructures that service components depend upon, such as special storage and computation platforms. When triggered by an
-invocation call (such as CLAMP or via DCAE Dashboard), DCAE Platform follows the TOSCA model of the control loop that is specified by the triggering call,
-interacts with the underlying networking and computing infrastructure such as OpenSatck installations and Kubernetes clusters to deploy and configure the virtual
-apparatus (i.e. the collectors, the analytics, and auxiliary microservices) that are needed to form the control loop, at locations that requested.
-DCAE Platform also provisions DMaaP topics and manages the distribution scopes of the topics following the prescription of the control loop model by interacting
-with controlling function of DMaaP.
-
-DCAE Service components are the functional entities that realize the collection and analytics needs of ONAP control loops. They include the collectors for various
-data collection needs, event processors for data standardization, analytics that assess collected data, and various auxiliary microservices that assist data
-collection and analytics, and support other ONAP functions. Service components and DMaaP buses form the "data plane" for DCAE, where DCAE collected data is
-transported among different DCAE service components.
-
-DCAE use Consul's distributed K-V store service to manage component configurations where each key is based on the unique identity of a DCAE component (identified by ServiceComponentName), and the value is the configuration for the corresponding component. The K-V store for each service components is created during deployment. DCAE platform creates and updates the K-V pairs based on information provided as part of the control loop blueprint deployment, or through a notification/trigger received from other ONAP components such as Policy Framework and CLAMP. Either through periodically polling or proactive pushing, the DCAE components get the configuration updates in realtime and apply the configuration updates. DCAE Platform also offers dynamic template resolution for configuration parameters that are dynamic and only known by the DCAE platform, such as dynamically provisioned DMaaP topics. This approach standardizes component deployment and configuration management for DCAE service components in multi-site deployment.
-
-
-DCAE Components
----------------
-
-The following lists the components included in ONAP DCAE . All DCAE components are offered as Docker containers. Following ONAP level deployment methods, these components can be deployed as Kubernetes Deployments and Services.
-
-- DCAE Platform
- - Core Platform
- - Cloudify Manager: TOSCA model executor. Materializes TOSCA models of control loop, or Blueprints, into properly configured and managed virtual DCAE functional components.
- - Plugins (K8S, Dmaap, Policy, Clamp, Postgres)
- - Extended Platform
- - Configuration Binding Service: Agent for service component configuration fetching; providing configuration parameter resolution.
- - Deployment Handler: API for triggering control loop deployment based on control loop's TOSCA model.
- - Policy Handler: Handler for fetching policy updates from Policy engine; and updating the configuration policies of KV entries in Consul cluster KV store for DCAE components.
- - Service Change Handler: Handler for interfacing with SDC; receiving new TOSCA models; and storing them in DCAE's own inventory.
- - DCAE Inventory-API: API for DCAE's TOSCA model store.
- - Platform services
- - Consul: Distributed service discovery service and KV store.
- - Postgres Database: DCAE's TOSCA model store.
- - Redis Database: DCAE's transactional state store, used by TCA for supporting persistence and seamless scaling.
-
-- DCAE Services
- - Collectors
- - Virtual Event Streaming (VES) collector
- - SNMP Trap collector
- - High-Volume VES collector (HV-VES)
- - DataFile collector
- - RESTConf collector
- - Analytics
- - Holmes correlation analytics
- - CDAP based Threshold Crosssing Analytics application (tca)
- - Docker based Threshold Crosssing Analytics
- - Heartbeat Services
- - SON-Handler Service
- - Event processors
- - PNF Registration Handler
- - VES Mapper Service
- - PM-Mapper Service
- - BBS-EventProcessor Service
- - PM Subcription Handler
- - DataLake Handlers
-
-
-The figure below shows the DCAE architecture and how the components work with each other. The components on the right constitute the Platform/controller components which are statically deployed. The components on the right represent the services which can be both deployed statically or dynamically (via CLAMP)
-
-.. image:: images/R6_architecture_diagram.png
-
-
-Deployment Scenarios
---------------------
-
-Because DCAE service components are deployed on-demand following the control loop needs for managing ONAP deployed services, DCAE must support dynamic and on-demand deployment of service components based on ONAP control loop demands. This is why all other ONAP components are launched from the ONAP level method, DCAE only deploys a subset of its components during this ONAP deployment process and rest of DCAE components will be deployed on-demand based on usecase needs triggered by control loop request originated from CLAMP, or even by operator manually invoking DCAE's deployment API call.
-
-ONAP supports deployment through OOM Helm Chart currently (Heat deployment support is discontinued since R3). Hence all DCAE Platform components are deployed via Helm charts - this includes Cloudify Manager, ConfigBinding service, ServiceChange Handler, Policy Handler, Dashboard and Inventory, each with corresponding Helm charts under OOM (https://git.onap.org/oom/tree/kubernetes/dcaegen2/components). Once DCAE platform components are up and running, rest of DCAE service components required for ONAP flow are deployed via bootstrap POD, which invokes Cloudify Manager API with Blueprints for various DCAE components that are needed for the built-in collections and control loops flow support.
-
-To keep the ONAP footprint minimal, only minimal set of MS (required for ONAP Integration usecases) are deployed via bootstrap pod. Rest of service blueprints are available for operator to deploy on-demand as required.
-
-
-More details of the DCAE deployment can be found under Installation section.
-
-
-Usage Scenarios
----------------
-
-For ONAP DCAE participates in the following use cases.
-
-- vDNS: VES collector, TCA analytics
-
-- vFW: VES collector, TCA analytics
-
-- vCPE: VES collector, TCA analytics
-
-- vVoLTE: VES collector, Holmes analytics
-
-- CCVPN : RestConf Collector, Holmes
-
-- BBS : VES Collector, PRH, BBS-Event Processor, VES-Mapper, RESTConf Collector
-
-- 5g Bulk PM : DataFile Collector, PM-Mapper, HV-VES
-
-
-In addition, DCAE supports on-demand deployment and configuration of service components via CLAMP. In such case CLAMP invokes the deployment and configuration of additional TCA instances.
-
+.. This work is licensed under a Creative Commons Attribution 4.0 International License.
+.. http://creativecommons.org/licenses/by/4.0
+.. _architecture:
+
+
+Architecture
+============
+
+Data Collection Analytics and Events (DCAE) is the primary data collection and analysis system of ONAP. DCAE architecture comprises of DCAE Platform and
+DCAE Service components making DCAE flexible, elastic, and expansive enough for supporting the potentially infinite number of ways of constructing intelligent
+and automated control loops on distributed and heterogeneous infrastructure.
+
+DCAE Platform supports the functions to deploy, host and perform LCM applications of Service components. DCAE Platform components enable model driven deployment of
+service components and middleware infrastructures that service components depend upon, such as special storage and computation platforms. When triggered by an
+invocation call (such as CLAMP or via DCAE Dashboard), DCAE Platform follows the TOSCA model of the control loop that is specified by the triggering call,
+interacts with the underlying networking and computing infrastructure such as OpenSatck installations and Kubernetes clusters to deploy and configure the virtual
+apparatus (i.e. the collectors, the analytics, and auxiliary microservices) that are needed to form the control loop, at locations that requested.
+DCAE Platform also provisions DMaaP topics and manages the distribution scopes of the topics following the prescription of the control loop model by interacting
+with controlling function of DMaaP.
+
+DCAE Service components are the functional entities that realize the collection and analytics needs of ONAP control loops. They include the collectors for various
+data collection needs, event processors for data standardization, analytics that assess collected data, and various auxiliary microservices that assist data
+collection and analytics, and support other ONAP functions. Service components and DMaaP buses form the "data plane" for DCAE, where DCAE collected data is
+transported among different DCAE service components.
+
+DCAE use Consul's distributed K-V store service to manage component configurations where each key is based on the unique identity of a DCAE component (identified by ServiceComponentName), and the value is the configuration for the corresponding component. The K-V store for each service components is created during deployment. DCAE platform creates and updates the K-V pairs based on information provided as part of the control loop blueprint deployment, or through a notification/trigger received from other ONAP components such as Policy Framework and CLAMP. Either through periodically polling or proactive pushing, the DCAE components get the configuration updates in realtime and apply the configuration updates. DCAE Platform also offers dynamic template resolution for configuration parameters that are dynamic and only known by the DCAE platform, such as dynamically provisioned DMaaP topics. This approach standardizes component deployment and configuration management for DCAE service components in multi-site deployment.
+
+
+DCAE Components
+---------------
+
+The following lists the components included in ONAP DCAE . All DCAE components are offered as Docker containers. Following ONAP level deployment methods, these components can be deployed as Kubernetes Deployments and Services.
+
+- DCAE Platform
+ - Core Platform
+ - Cloudify Manager: TOSCA model executor. Materializes TOSCA models of control loop, or Blueprints, into properly configured and managed virtual DCAE functional components.
+ - Plugins (K8S, Dmaap, Policy, Clamp, Postgres)
+ - Extended Platform
+ - Configuration Binding Service: Agent for service component configuration fetching; providing configuration parameter resolution.
+ - Deployment Handler: API for triggering control loop deployment based on control loop's TOSCA model.
+ - Policy Handler: Handler for fetching policy updates from Policy engine; and updating the configuration policies of KV entries in Consul cluster KV store for DCAE components.
+ - Service Change Handler: Handler for interfacing with SDC; receiving new TOSCA models; and storing them in DCAE's own inventory.
+ - DCAE Inventory-API: API for DCAE's TOSCA model store.
+ - Platform services
+ - Consul: Distributed service discovery service and KV store.
+ - Postgres Database: DCAE's TOSCA model store.
+ - Redis Database: DCAE's transactional state store, used by TCA for supporting persistence and seamless scaling.
+
+- DCAE Services
+ - Collectors
+ - Virtual Event Streaming (VES) collector
+ - SNMP Trap collector
+ - High-Volume VES collector (HV-VES)
+ - DataFile collector
+ - RESTConf collector
+ - Analytics
+ - Holmes correlation analytics
+ - CDAP based Threshold Crosssing Analytics application (tca)
+ - Docker based Threshold Crosssing Analytics
+ - Heartbeat Services
+ - SON-Handler Service
+ - Slice Analysis
+ - Event processors
+ - PNF Registration Handler
+ - VES Mapper Service
+ - PM-Mapper Service
+ - BBS-EventProcessor Service
+ - PM Subcription Handler
+ - DataLake Handlers
+
+
+The figure below shows the DCAE architecture and how the components work with each other. The components on the right constitute the Platform/controller components which are statically deployed. The components on the right represent the services which can be both deployed statically or dynamically (via CLAMP)
+
+.. image:: images/R6_architecture_diagram.png
+
+
+Deployment Scenarios
+--------------------
+
+Because DCAE service components are deployed on-demand following the control loop needs for managing ONAP deployed services, DCAE must support dynamic and on-demand deployment of service components based on ONAP control loop demands. This is why all other ONAP components are launched from the ONAP level method, DCAE only deploys a subset of its components during this ONAP deployment process and rest of DCAE components will be deployed on-demand based on usecase needs triggered by control loop request originated from CLAMP, or even by operator manually invoking DCAE's deployment API call.
+
+ONAP supports deployment through OOM Helm Chart currently (Heat deployment support is discontinued since R3). Hence all DCAE Platform components are deployed via Helm charts - this includes Cloudify Manager, ConfigBinding service, ServiceChange Handler, Policy Handler, Dashboard and Inventory, each with corresponding Helm charts under OOM (https://git.onap.org/oom/tree/kubernetes/dcaegen2/components). Once DCAE platform components are up and running, rest of DCAE service components required for ONAP flow are deployed via bootstrap POD, which invokes Cloudify Manager API with Blueprints for various DCAE components that are needed for the built-in collections and control loops flow support.
+
+To keep the ONAP footprint minimal, only minimal set of MS (required for ONAP Integration usecases) are deployed via bootstrap pod. Rest of service blueprints are available for operator to deploy on-demand as required.
+
+More details of the DCAE deployment can be found under Installation section.
+
+
+Usage Scenarios
+---------------
+
+For ONAP DCAE participates in the following use cases.
+
+- vDNS: VES collector, TCA analytics
+
+- vFW: VES collector, TCA analytics
+
+- vCPE: VES collector, TCA analytics
+
+- vVoLTE: VES collector, Holmes analytics
+
+- CCVPN : RestConf Collector, Holmes
+
+- BBS : VES Collector, PRH, BBS-Event Processor, VES-Mapper, RESTConf Collector
+
+- 5G Bulk PM : DataFile Collector, PM-Mapper, HV-VES
+
+- 5G OOF SON: VES collector, SON-Handler
+
+- 5G E2E Network Slicing: VES collector, Slice Analysis, DES, PM-Mapper, DFC, Datalake feeder
+
+
+In addition, DCAE supports on-demand deployment and configuration of service components via CLAMP. In such case CLAMP invokes the deployment and configuration of additional TCA instances.
diff --git a/docs/sections/services/serviceindex.rst b/docs/sections/services/serviceindex.rst
index 2f8dd61e..0a968c91 100644
--- a/docs/sections/services/serviceindex.rst
+++ b/docs/sections/services/serviceindex.rst
@@ -1,26 +1,27 @@
-.. This work is licensed under a Creative Commons Attribution 4.0 International License.
-.. http://creativecommons.org/licenses/by/4.0
-
-
-DCAE Service components
-=======================
-
-.. Add or remove sections below as appropriate for the platform component.
-
-.. toctree::
- :maxdepth: 1
-
- ./snmptrap/index.rst
- ./ves-http/index.rst
- ./ves-hv/index.rst
- ./mapper/index.rst
- ./prh/index.rst
- ./dfc/index.rst
- ./heartbeat-ms/index.rst
- ./pm-mapper/index.rst
- ./pm-subscription-handler/index.rst
- ./bbs-event-processor/index.rst
- ./son-handler/index.rst
- ./restconf/index.rst
- ./tcagen2-docker/index.rst
- ./datalake-handler/index.rst \ No newline at end of file
+.. This work is licensed under a Creative Commons Attribution 4.0 International License.
+.. http://creativecommons.org/licenses/by/4.0
+
+
+DCAE Service components
+=======================
+
+.. Add or remove sections below as appropriate for the platform component.
+
+.. toctree::
+ :maxdepth: 1
+
+ ./snmptrap/index.rst
+ ./ves-http/index.rst
+ ./ves-hv/index.rst
+ ./mapper/index.rst
+ ./prh/index.rst
+ ./dfc/index.rst
+ ./heartbeat-ms/index.rst
+ ./pm-mapper/index.rst
+ ./pm-subscription-handler/index.rst
+ ./bbs-event-processor/index.rst
+ ./son-handler/index.rst
+ ./restconf/index.rst
+ ./tcagen2-docker/index.rst
+ ./datalake-handler/index.rst
+ ./slice-analysis-ms/index.rst \ No newline at end of file
diff --git a/docs/sections/services/slice-analysis-ms/index.rst b/docs/sections/services/slice-analysis-ms/index.rst
new file mode 100644
index 00000000..252a2f34
--- /dev/null
+++ b/docs/sections/services/slice-analysis-ms/index.rst
@@ -0,0 +1,51 @@
+.. This work is licensed under a Creative Commons Attribution 4.0 International License.
+.. http://creativecommons.org/licenses/by/4.0
+
+
+Slice Analysis MS
+=================
+
+.. Add or remove sections below as appropriate for the platform component.
+
+
+**Slice Analysis MS** is introduced in ONAP for:
+(a) Analyzing the FM/PM data (reported from the xNFs) and KPI data (computed from PM data) related to various slice instances (NSIs), slice sub-net instances (NSSIs) and services catered to by the slices (S-NSSAIs).
+(b) Determining and triggering appropriate Control Loop actions based on the analysis above
+(c) Receiving recommendations for closed loop actions from ML or Analytics engines, performing validity checks, etc. to determine if the actions can be carried out, and then triggering the appropriate Control Loop
+
+In Guilin, this MS:
+- Performs simple Closed Loop control action for the RAN slice sub-net instances based on simple analysis of a set of RAN PM data
+- Initiates simple control loop actions in the RAN based on recommendation from an ML engine for RAN slice sub-net instance re-configuration
+
+For the Control loops, SO, VES Collector, Policy, DMaaP and CCSDK/SDN-R, AAI, PM-mapper and DFC are involved apart from this MS.
+
+Flow diagrams are available at:
+https://wiki.onap.org/display/DW/Closed+Loop
+https://wiki.onap.org/display/DW/Intelligent+Slicing+flow
+
+
+Slice Analysis MS overview and functions
+----------------------------------------
+
+.. toctree::
+ :maxdepth: 1
+
+ ./slice_analysis_ms_overview.rst
+
+
+Slice Analysis MS Installation Steps
+------------------------------------
+
+.. toctree::
+ :maxdepth: 1
+
+ ./installation.rst
+
+
+Slice Analysis MS Troubleshooting Steps
+---------------------------------------
+
+.. toctree::
+ :maxdepth: 1
+
+ ./slice_analysis_ms_troubleshooting.rst \ No newline at end of file
diff --git a/docs/sections/services/slice-analysis-ms/installation.rst b/docs/sections/services/slice-analysis-ms/installation.rst
new file mode 100644
index 00000000..ebe2b782
--- /dev/null
+++ b/docs/sections/services/slice-analysis-ms/installation.rst
@@ -0,0 +1,75 @@
+Installation
+============
+
+Slice Analysis MS can be deployed using cloudify blueprint using bootstrap container of an existing DCAE deployment.
+
+Deployment Pre-requisites
+~~~~~~~~~~~~~~~~~~~~~~~~~
+- DCAE and DMaaP pods should be up and running.
+
+- DMaaP Bus Controller PostInstalls job should have completed successfully (executed as part of an OOM install).
+
+- PM mapper service should be running.
+
+- Config DB service should be running.
+
+- Make sure that cfy is installed and configured to work with the Cloudify deployment.
+
+
+Deployment steps
+~~~~~~~~~~~~~~~~
+1. Execute bash on the bootstrap Kubernetes pod.
+
+ kubectl -n onap exec -it <dcaegen2-dcae-bootstrap> bash
+
+2. Go to the /blueprints directory.
+
+ Check that the tag_version in the slice-analysis-ms blueprint is correct for the release
+ of ONAP that it is being installed on see Nexus link below for slice-analysis-ms for tag_versions.
+ Nexus link: https://nexus3.onap.org/#browse/browse:docker.public:v2%2Fonap%2Forg.onap.dcaegen2.services.components.slice-analysis-ms%2Ftags
+
+3. Create an input file.
+
+4. Run the Cloudify install command to install the slice-analysis-ms with the blueprint and the newly created input file k8s-slice-input.yaml.
+
+ $ cfy install k8s-slice-analysis-ms.yaml -i k8s-slice-input.yaml --blueprint-id sliceanalysisms
+
+ Details of the sample output are available at:
+ https://wiki.onap.org/pages/viewpage.action?pageId=92998809.
+
+5. To un-deploy
+
+ $ cfy uninstall sliceanalysisms
+
+
+Application configurations
+~~~~~~~~~~~~~~~~~~~~~~~~~~
++-------------------------------+------------------------------------------------+
+|Configuration | Description |
++===============================+================================================+
+|samples | Minimum number of samples to be present |
+| | for analysis |
++-------------------------------+------------------------------------------------+
+|minimumPercentageChange | Minimum percentage of configuration change |
+| | above which control loop should be triggered |
++-------------------------------+------------------------------------------------+
+|initialDelaySeconds | Initial delay in milliseconds for the consumer |
+| | thread to start after the application startup |
++-------------------------------+------------------------------------------------+
+|config_db | Host where the config DB application is running|
++-------------------------------+------------------------------------------------+
+|performance_management_topicurl| Dmaap Topic URL to which PM data are posted |
+| | by network functions |
++-------------------------------+------------------------------------------------+
+|dcae_cl_topic_url | Dmaap topic to which onset message to trigger |
+| | the control loop are posted |
++-------------------------------+------------------------------------------------+
+|dcae_cl_response_topic_url | Dmaap topic URL to which Policy posts the |
+| | message after successful control loop trigger |
++-------------------------------+------------------------------------------------+
+|intelligent_slicing_topic_url | Dmaap topic URL to which ML MS posts the |
+| | messages |
++-------------------------------+------------------------------------------------+
+|dmaap_polling_interval | Dmaap Polling interval in milliseconds |
++-------------------------------+------------------------------------------------+
+
diff --git a/docs/sections/services/slice-analysis-ms/slice_analysis_ms_arch.jpg b/docs/sections/services/slice-analysis-ms/slice_analysis_ms_arch.jpg
new file mode 100644
index 00000000..64ead8ab
--- /dev/null
+++ b/docs/sections/services/slice-analysis-ms/slice_analysis_ms_arch.jpg
Binary files differ
diff --git a/docs/sections/services/slice-analysis-ms/slice_analysis_ms_overview.rst b/docs/sections/services/slice-analysis-ms/slice_analysis_ms_overview.rst
new file mode 100644
index 00000000..05cb2383
--- /dev/null
+++ b/docs/sections/services/slice-analysis-ms/slice_analysis_ms_overview.rst
@@ -0,0 +1,65 @@
+.. This work is licensed under a Creative Commons Attribution 4.0
+ International License. http://creativecommons.org/licenses/by/4.0
+
+.. _docs_slice_analysis_ms_overview:
+
+Architecture
+------------
+The internal architecture of Slice Analysis MS is shown below.
+
+.. image:: ./slice_analysis_ms_arch.jpg
+
+The Slice Analysis MS has a DMaaP interface towards towards Policy and VES-Collector, and a REST
+interface towards Config DB. It also has a DMaaP interface to receive any recommendations for
+Closed Loop updates from an ML engine, which is then used to trigger a control loop message to
+Policy.
+
+- **DMAAP Client** creates a thread pool for every DMaaP topic consumer. The thread
+ polls the DMaaP topic for every configured time interval and whenever a message is
+ received it stores that message in the Postgres DB.
+
+- **PM Thread** reads the PM event from the database and puts the PM sample in the
+ internal queue in the format which is needed for further processing.
+
+- **Consumer Thread** consumes PM samples from the internal queue and make all the
+ required Config DB calls, perform the analysis, and puts the onset message to the DMaaP topic.
+
+- **Database** is a PG DB.
+
+Detailed flow diagrams are available at:
+
+Closed Loop: https://wiki.onap.org/display/DW/Closed+Loop
+
+Intelligent Slicing: https://wiki.onap.org/display/DW/Intelligent+Slicing+flow
+
+
+Functional Description
+----------------------
+- Slice Analysis ms consumes PM messages from PERFORMANCE_MEASUREMENTS topic.
+
+- For analysis Slice Analysis MS consumes various data from Config DB including List of Network
+ Functions which serves the S-NSSAI, List of Near-RT RICs and the corresponding cell mappings of the
+ S-NSSAI, Current Configuration of the Near-RT RICs, Slice Profile associated with the S-NSSAI and
+ Subscriber details of the S-NSSAI (for sending the onset message to policy).
+
+- Based on the collected PM data, Slice Analysis MS computes the DLThptPerSlice and ULThptPerSlice
+ for the Near-RT RICs relevant for the S-NSSAI, and the computed value is compared with the current
+ configuration of the Near-RT RICs. If the change in configuration exceeds the minimum percentage
+ value which is kept as a configuration parameter, then the closed-loop will be triggered by posting
+ the onset message to DMaaP.
+
+- Upon reception of recommendation to update the configuration of RAN from e.g., an ML MS, the Slice
+ Analysis MS prepares and sends a control loop onset message.
+
+
+Deployment aspects
+------------------
+The SON-Handler MS will be deployed on DCAE as an on-demand component. Details of the installation
+steps are available at ./installation.rst. Further details can be obtained from:
+https://wiki.onap.org/pages/viewpage.action?pageId=92998809
+
+
+Known Issues and Resolutions
+----------------------------
+The assumptions of functionality in Guilin release is documented in:
+https://wiki.onap.org/display/DW/Assumptions+for+Guilin+release
diff --git a/docs/sections/services/slice-analysis-ms/slice_analysis_ms_troubleshooting.rst b/docs/sections/services/slice-analysis-ms/slice_analysis_ms_troubleshooting.rst
new file mode 100644
index 00000000..66dee8c2
--- /dev/null
+++ b/docs/sections/services/slice-analysis-ms/slice_analysis_ms_troubleshooting.rst
@@ -0,0 +1,46 @@
+Trouble shooting steps
+----------------------
+1. **Microservice stops and restarts during startup**
+
+ Possible reason & Solution: Microservice is not registered with the consul
+ - Check the consul if the microservice is registered with it and the MS is able to fetch the app config from the CBS. Check if CBS and consul are deployed properly and try to redeploy the MS
+ The below logs will be seen if CBS is not reachable by the MS
+
+ 15:14:13.861 [main] WARN org.postgresql.Driver - JDBC URL port: 0 not valid (1:65535)
+ 15:14:13.862 [main] WARN o.s.b.w.s.c.AnnotationConfigServletWebServerApplicationContext -
+ Exception encountered during context initialization - cancelling refresh attempt:
+ org.springframework.beans.factory.UnsatisfiedDependencyException: Error creating bean with name
+ 'org.springframework.boot.autoconfigure.orm.jpa.HibernateJpaConfiguration': Unsatisfied
+ dependency expressed through constructor parameter 0; nested exception is
+ org.springframework.beans.factory.BeanCreationException: Error creating bean with name 'dataSource'
+ defined in org.onap.dcaegen2.services. sliceanalysisms.Application: Initialization of bean failed;
+ nested exception is org.springframework.beans.factory.BeanCreationException: Error creating bean with name
+ 'org.springframework.boot.autoconfigure.jdbc.DataSourceInitializerInvoker': Invocation of init method failed; nested exception is
+ org.springframework.jdbc.datasource.init.UncategorizedScriptException: Failed to execute database script; nested exception is
+ java.lang.RuntimeException: Driver org.postgresql.Driver claims to not accept jdbcUrl, jdbc:postgresql://null:0/sliceanalysisms
+ 15:14:13.865 [main] INFO o.a.catalina.core.StandardService - Stopping service [Tomcat]
+ 15:14:13.877 [main] INFO o.s.b.a.l.ConditionEvaluationReportLoggingListener - Error starting ApplicationContext.
+ To display the conditions report re-run your application with 'debug' enabled.
+ 15:14:13.880 [main] ERROR o.s.boot.SpringApplication - Application run failed
+
+2. **No PostgreSQL clusters have been deployed on this manager**
+
+ Solution:
+
+ kubectl exec -ti -n onap dev-dcaemod-db-primary-56ff585cf7-dxkkx bash
+ psql
+ ALTER ROLE "postgres" WITH PASSWORD 'onapdemodb';
+ \q
+
+ kubectl exec -ti -n onap dev-dcae-bootstrap-b47854569-dnrqf bash
+ cfy blueprints upload -b pgaas_initdb /blueprints/k8s-pgaas-initdb.yaml
+ cfy deployments create -b pgaas_initdb -i k8s-pgaas-initdb-inputs.yaml pgaas_initdb
+ cfy executions start -d pgaas_initdb install
+
+
+Logging
+-------
+Since the Slice Analysis MS is deployed as a pod in the kubernetes, we can check the logs by
+using the following command:
+
+ $ kubectl logs <pod-name> –namespace onap \ No newline at end of file