[Buffer manager][201911]Regression test case for traditional buffer manager in 201911

[stephenxs]

Description of PR

Summary:
Add test case for traditional buffer manager for fundamental functionalities and reclaiming unused buffer.

Signed-off-by: Stephen Sun stephens@nvidia.com

Type of change

• Bug fix
• Testbed and Framework(new/improvement)
• Test case(new/improvement)

Back port request

• 201911

Approach

What is the motivation for this PR?

Add test case for traditional buffer manager for fundamental functionalities and reclaiming unused buffer.

How did you do it?

1. For all ports in the config_db,
   • Check whether there is no lossless buffer PG configured on an admin-down port
   • Check whether the lossless PG aligns with the port's speed and cable length
   • If the name-to-OID maps exist for port and PG, check whether the information in ASIC_DB aligns with that in CONFIG_DB
   • If a lossless profile hasn't been checked, check whether the lossless profile in CONFIG_DB aligns with
     • pg_profile_lookup.ini according to speed and cable length
     • information in ASIC_DB
2. Shutdown a port and check whether the lossless buffer PG has been removed
3. Startup the port and check whether the lossless PG has been readded.

How did you verify/test it?

Run regression test.

Any platform specific information?

Supported testbed topology if it's a new test case?

Documentation

[neethajohn]

@stephenxs , isn't this testcases only applicable to 201911? Then we also need to add a version check in the beginning and skip if there is a version mismatch

[stephenxs]

@stephenxs , isn't this testcases only applicable to 201911? Then we also need to add a version check in the beginning and skip if there is a version mismatch

Fixed.