medianetlab / katana-slice_manager Goto Github PK

I used the command ./start.sh -g -m

Errors as following:

Traceback (most recent call last):
File "/usr/local/bin/ui", line 11, in
load_entry_point('UI-CLI', 'console_scripts', 'ui')()
File "/usr/local/lib/python3.7/site-packages/click/core.py", line 716, in call
return self.main(*args, **kwargs)
File "/usr/local/lib/python3.7/site-packages/click/core.py", line 696, in main
rv = self.invoke(ctx)
File "/usr/local/lib/python3.7/site-packages/click/core.py", line 1055, in invoke
cmd_name, cmd, args = self.resolve_command(ctx, args)
File "/usr/local/lib/python3.7/site-packages/click/core.py", line 1094, in resolve_comm and
cmd = self.get_command(ctx, cmd_name)
File "/app/cli/cli.py", line 41, in get_command
eval(code, ns, ns)
File "/app/cli/commands/cmd_db.py", line 3, in
from sqlalchemy_utils import database_exists, create_database
File "/usr/local/lib/python3.7/site-packages/sqlalchemy_utils/init.py", line 1, in
from .aggregates import aggregated # noqa
File "/usr/local/lib/python3.7/site-packages/sqlalchemy_utils/aggregates.py", line 372, in
from .functions.orm import get_column_key
File "/usr/local/lib/python3.7/site-packages/sqlalchemy_utils/functions/init.py", l ine 1, in
from .database import ( # noqa
File "/usr/local/lib/python3.7/site-packages/sqlalchemy_utils/functions/database.py", l ine 13, in
from ..expressions import explain_analyze
File "/usr/local/lib/python3.7/site-packages/sqlalchemy_utils/expressions.py", line 4, in
from sqlalchemy.sql.expression import (
ImportError: cannot import name '_literal_as_text' from 'sqlalchemy.sql.expression' (/usr /local/lib/python3.7/site-packages/sqlalchemy/sql/expression.py)
Traceback (most recent call last):
File "/usr/local/bin/ui", line 11, in
load_entry_point('UI-CLI', 'console_scripts', 'ui')()
File "/usr/local/lib/python3.7/site-packages/click/core.py", line 716, in call
return self.main(*args, **kwargs)
File "/usr/local/lib/python3.7/site-packages/click/core.py", line 696, in main
rv = self.invoke(ctx)
File "/usr/local/lib/python3.7/site-packages/click/core.py", line 1055, in invoke
cmd_name, cmd, args = self.resolve_command(ctx, args)
File "/usr/local/lib/python3.7/site-packages/click/core.py", line 1094, in resolve_comm and
cmd = self.get_command(ctx, cmd_name)
File "/app/cli/cli.py", line 41, in get_command
eval(code, ns, ns)
File "/app/cli/commands/cmd_db.py", line 3, in
from sqlalchemy_utils import database_exists, create_database
File "/usr/local/lib/python3.7/site-packages/sqlalchemy_utils/init.py", line 1, in
from .aggregates import aggregated # noqa
File "/usr/local/lib/python3.7/site-packages/sqlalchemy_utils/aggregates.py", line 372, in
from .functions.orm import get_column_key
File "/usr/local/lib/python3.7/site-packages/sqlalchemy_utils/functions/init.py", l ine 1, in
from .database import ( # noqa
File "/usr/local/lib/python3.7/site-packages/sqlalchemy_utils/functions/database.py", l ine 13, in
from ..expressions import explain_analyze
File "/usr/local/lib/python3.7/site-packages/sqlalchemy_utils/expressions.py", line 4, in
from sqlalchemy.sql.expression import (
ImportError: cannot import name '_literal_as_text' from 'sqlalchemy.sql.expression' (/usr /local/lib/python3.7/site-packages/sqlalchemy/sql/expression.py)

Thank you in advance!

Give a meaningful name to slices

Possible scenario: "${coverage_area}_${sst}"

• Split katana-mgr process
• Create Kafka producers - consumers

• Revise GST
• Change NEST Terminology
• Change Component DB
  • Add API for NS view
  • Make SST --> Component
  • Remove pdu
• Change Slice Mapping
• Testing
• Docs
  • Swagger
  • JSON Schemes
  • Wki

• Rename "Functions" to "sub-slice"
• Rename "Core-Radio" to "Core-Edge"
• Make all functions optional

Slice will not be terminated correctly when if you send request for deletion before it is done instatiated.

reason: Not all fields are correctly inserted in the DB

Bug reproduction:

katana slice add -f slice_nest.json #slice1
katana slice add -f slice_nest.json #slice2
katana slice rm -f {slice1_id}
katana slice add -f slice_nest.json #slice3

Slice 1 won't be terminated correctly

The first step for installation seems to be running the install.sh script but I can't find it in the repo.

• Add Components (VIM, NFVO, EMS, etc)
• Read Components (VIM, NFVO, EMS, etc)
• Update Components (VIM, NFVO, EMS, etc)
• Remove Componets (VIM, NFVO, EMS, etc)
• Add Slice
• Read Slice
• Remove Slice

Fix capitalization of the location field when using PUT method

I can't find “install.sh”.
I cannot pull katana-nbi when I run start.sh.

• Create the docker containers
• Design Kafka services & topics
• Implement Kafka topics

• Unit tests
• Integration tests

We need to have a specific slicing capabilities list beyond the generic features of the software platform.

In Release B (also for master branch) I ran sudo ./start.sh -p -m after building all docker images. I also initialised the database using docker container exec -it katana-ui bash , ui db init and ui db seed
I opened http://172.17.0.1:8080/login in my local. It gives a login option on the left and a register option on the right.
I tried login in with username:'admin' and password: 'password'. But it didn't work and it throws an error like this: "The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application." or sometimes this message when used some random text: "{"error":{"message":"Invalid identity or password"}}
I also tried the register option but it didn't change anything as evidently using the same username and password as used in register don't work on the login page.
Also, clicking on forget password is not doing any action.

Was it meant to be like this only or am I missing something?

• New Architecture

Tries to add multiple times the VIM account to OSM
Occurred during multi-NS system testing on issue #9

• Add Components (VIM, NFVO, EMS, etc)
• Read Components (VIM, NFVO, EMS, etc)
• Update Components (VIM, NFVO, EMS, etc)
• Remove Componets (VIM, NFVO, EMS, etc)
• Add Slice
• Read Slice
• Remove Slice

Use a K8s cluster as VIM

• Check implementation by OSM
• Create kubUtils.py under ./sharedUtils/vimUtils/

• PDU Wiki
• New NBI APIs
• New "how-to-use" for sst
• Templates
• Swagger External References
• CLI
• GUI
• Architecture
• GST

Deadline 8.03.2021

To be presented at the WP2 Meeting

I try to integrate Katana with Free5GC. I don't know if there any interfaces I can use to combine theses two things.

Thank you in advance!

Integration at infrastructure level may require some initial mock EMS functionality in order to test end-to-end slice creation, simulating possible RAN element configuration. In addition it will serve as example for developing actual EMS or a mediator tailored to the infrastructure capabilities

If a user creates a slice without specifying base_slice_des_id (optional in the nest), then GET /api/base_slice_des returns an error.

Create a function that will be able to retrieve information by a specified VIM component regarding the available and total resources. The function should be placed in ./shared_utils/monitoringUtils/ directory as a module to be used by both katana-mngr and katana-nbi containers.

Possible Solutions:

• Use prometheus/influx to query available/total resources
• Use API provided by the VIM (e.g, OpenStack)

Discussion (Requires Access): https://gitlab.fokus.fraunhofer.de/5genesis/slice-manager/issues/2

• Create the API for registering new policy management system
• Create the Utility
• Create the Data Model
• Create the cli command
• Send the NEAT to the policy management system
• Documentation
  • Swagger
  • Wiki

• Feature list on README
• Swagger type of the data

• Multi-NS Deployment
• Multi-Site Deployment
• Concurrent Slices
• Amarisoft vEPC

Extra tests for Release 2.0.0:

• Use base slice_des and add NSs
• Use base slice_des and modify parameters
• Error handling for missing SB components
• Not supported locations
• Slice-as-a-Service (No core services)

• Error handling from SB components (Example: Errored OSM NS)

• Shared NSSI

Functions

• Add Components (VIM, NFVO, EMS, etc)
• Read Components (VIM, NFVO, EMS, etc)
• Update Components (VIM, NFVO, EMS, etc)
• Remove Componets (VIM, NFVO, EMS, etc)
• Add Slice
• Read Slice
• Remove Slice

System Tests

• Multi-NS Deployment
• Multi-Site Deployment
• Concurrent Slices
• Amarisoft 4G vEPC Slice

Extra Tests

• Optional Extra NS
• Multi-Edge Extra NS

• Multi-NS Deployment
• Multi-Site Deployment
• Concurrent Slices
• Amarisoft 4G vEPC Slice

Video Cache Service

• Slice Descriptor

• Deployment at the edge

• Deployment at the central cloud (NCSRD)

• Is operational ?

Fails to remove tenant from more than 1 VIMs

Integrate OSM Release 7 to Slice Manager

Some possible solutions/steps:

• Add Prometheus and Grafana containers
• Create label per slice
• Create Prometheus exporter for ODL
• Collect metrics from OSM Prometheus
• Collect NS Status
• Collect OpenStack metrics from infra prometheus

• OpenStack version Stein
• OSM Release 8