Update test_po_cleanup.py#5453
Update test_po_cleanup.py#5453wenyiz2021 merged 3 commits intosonic-net:202012from wenyiz2021:202012
Conversation
[test_po_cleanup] Change timeout length to 800 instead of 300 to allow more time for multi-asic asics to come back.
|
Same issue as: #5294 |
|
pc/test_po_cleanup.py::test_po_cleanup[str-multi-asic-acs-2-0] PASSED [ 14%] ============================================================================================================== 7 passed in 1630.24 seconds |
1. Execute an "config_reload" at the end of test_po_cleanup, to start up not only all critical services, but also wait for all port, interfaces, bgp neighbors to come up.
2. Call duthost.stop_service("swss") directly from duthost, instead of passing each asics inside the test_po_cleanup. to avoid execution sequence of test_po_cleanup[asic0], test_po_cleanup_after_reload[duthost], test_po_cleanup[asic1].
|
New change commited: ---------------------------------------------------------------------------------------------- generated xml file: /var/src/sonic-mgmt-int/tests/logs/tr.xml ---------------------------------------------------------------------------------------------- |
|
New change description:
|
|
@wenyiz2021 Can you port this PR to master branch also ? |
* Update test_po_cleanup.py Update same changes in master branch from PR: (#5453)
* Update test_po_cleanup.py Update same changes in master branch from PR: (#5453) * Update test_fan_drawer_fans.py * Update test_fan_drawer_fans.py * Update test_fan_drawer_fans.py Disable thermalctld when running test_fan_drawer_fans.py
Description of PR
For a multi-asic device, restart of swss and teamd is done per asic, noticed that for a 6-asic device, asic3 and asic4 are experiencing more time during restart. Also, interfaces and bgp neighbors need more minutes to come back. Execute a config_reload at the end of test_po_cleanup helps to bring back processes in a safe manner.
duthost.asic_instance(enum_asic_index).start_service("swss")
assert wait_until(300, 20, 0, duthost.critical_services_fully_started),\
E AssertionError: Not all critical services are fully started
duthost =
duthosts = []
enum_asic_index = 3
enum_rand_one_per_hwsku_frontend_hostname = 'xxx'
tbinfo = {'auto_recover': 'True', 'comment': 'Arvind', 'conf-name': 'xxx', 'duts': ['xxx'], ...}
[pc/test_po_cleanup.py:66: AssertionError]
Summary:
Fixes # (issue)
#5294
#4887
Type of change
Back port request
Approach
What is the motivation for this PR?
How did you do it?
How did you verify/test it?
pc/test_po_cleanup.py::test_po_cleanup[str-masic-acs-2] PASSED [100%]
---------------------------------------------------------------------------------------------- generated xml file: /var/src/sonic-mgmt-int/tests/logs/tr.xml ----------------------------------------------------------------------------------------------
=============================================================================================================== 1 passed in 489.35 seconds ================================================================================================================
Any platform specific information?
Supported testbed topology if it's a new test case?
Documentation