Fix/exit pytest duthosts 202012#10263
Closed
ZhaohuiS wants to merge 816 commits intosonic-net:masterfrom
Closed
Conversation
… of a specific testbed (sonic-net#7578) What is the motivation for this PR? Add an ansible-playbook to collect show techsupport results on all DUTs of a specific testbed. How did you do it? In the ansible-playbook: 1. Run show techsupport on all the DUTs of a specific testbed. 2. Fetch the dumped files to localhost. 3. Delete the dumped files on DUT (to release disk space). I also integrated this utility to testbed-cli.sh. How did you verify/test it? Verified on physical testbeds. Signed-off-by: Zhijian Li <[email protected]>
…and t1 scenario (sonic-net#7581) (sonic-net#7585) What is the motivation for this PR? In current bgp test for m0, just simulate scenario that m0 device performs like t1. But m0 is like both t1 and t0. How did you do it? Add suport for m0 in bgp test to simulate both t0 and t1 scenario How did you verify/test it? Run tests on m0 topo and t0 topo Signed-off-by: Yaqiang Zhu <[email protected]>
…ew exabgp process naming convention) (sonic-net#7587) What is the motivation for this PR? While removing topo of a testbed, we have to ensure that all the exabgp processes are stopped before PTF container be removed: * For old naming convention, exabgp processes are named like exabgp-ARISTA01T0 and exabgp-ARISTA01T0-v6. * For new naming convention, exabgp processes are named like exabgpv4:exabgp-ARISTA01T0 and exabgpv6:exabgp-ARISTA01T0-v6. In this PR, I added some check to ensure both type of exabgp processes are stopped. How did you do it? 1. For the new naming convention, use ansible supervisor module to stop process group exabgpv4 and exabgpv6. 2. For the old naming convention, use exabgp.py library to stop processes one by one. How did you verify/test it? Verified on both type of PTF containers. Signed-off-by: Zhijian Li <[email protected]>
…ion (sonic-net#7541) * show counter for port, pg, queue, when ptftest enter, exit and exception * remove invalid delay
…#7621) Approach What is the motivation for this PR? in 202012 branch, there is an extra definition of function: _toggle_all_simulator_ports_to_target_dut. Which should be caused by the cherry-pick conflict resolvement. Let's remove it. Signed-off-by: Longxiang Lyu [email protected] How did you do it? Remove the deprecated second definition. How did you verify/test it?
…or ACL TABLE TYPE=MIRRORV6 (sonic-net#7589) Co-authored-by: bingwang-ms <[email protected]>
What is the motivation for this PR? test_stress_acl will fail on mx dut because mx doesn't have portchannel. Add support for that. How did you do it? For mx, get table_port by DATAACL How did you verify/test it? Run test. Signed-off-by: Yaqiang Zhu <[email protected]>
* Add data plane test in stress acl script What is the motivation for this PR? Stress acl test will continuously add/delete acl rules by writing config db, but the action is asynchronous with asic, that may cause some timing issue like sonic-net#7396 How did you do it? The purpose of this testcase is to test stability when add/delete acl rules, not to test responding speed, so add data plane test in this case. Data plane test will increase test time, so reduce loop times, will run confident in weekend nightly test. How did you verify/test it? Run tests Any platform specific information? Supported testbed topology if it's a new test case?
What is the motivation for this PR? The current default configuration of DUTs does not support using lab tacacs server for authentication. This change added a new variable tacacs_enabled_by_default in group_vars file to control enablement of lab tacacs server. How did you do it? The ansible playbook for deploy minigraph was updated to support configuring lab tacacs server based on value of this variable. The tacacs test scripts are updated to support recover the lab tacacs server in a clean way after test is done. How did you verify/test it? Test run the tacacs scripts. Signed-off-by: Xin Wang <[email protected]>
Approach What is the motivation for this PR? The tacacs authorization test has a case to verify that authorization is still working if one of the tacacs servers is down. In case the previous authorization method is local, after it is switched to tacacs+, the tacacs client will start to contact the tacacs servers for authorization. If we immediately run a command after the authorization method is changed, this command may fail with authorization because the client is still trying the invalid tacacs server. How did you do it? The fix is to add a delay which is longer than the configured tacacs timeout after authorization method is changed from local to tacacs+. Extra improvements: Fixed issues detected by pre-commit. Created new test fixtures to configure and restore tacacs authorization configuration. Improved the code for starting and stopping tacacs server. The existing method of stopping tacacs server takes up to 40 seconds. Overall testing time of test_authorization.py takes up to 270 seconds. After the improvement, overall test time of this script is around 70 seconds. Co-authorized by: [email protected]
What is the motivation for this PR? Update port_utils.py to support generating name/alias mapping for DellEMC-Z9332f-C32. How did you do it? Update port_utils.py according to port_config.ini How did you verify/test it? Verified on Dell Z9332f switches with 100G cables connected. Signed-off-by: Zhijian Li <[email protected]>
…sonic-net#7682) Approach What is the motivation for this PR? As mux is stopped in heartbeat failure testcases, the error from monit about mux is not running is expected. Signed-off-by: Longxiang Lyu [email protected] How did you do it? Add the error msg to the loganalyzer ignore list. How did you verify/test it? Run test_active_tor_heartbeat_failure_upstream
PFC tests fail on Sonic fanout because it has Python 3 env but PFC gen script is written for Python 2. Adapted PFC gen script to work both on Python 2 and 3.
…ic dut.(202012) (sonic-net#6756) (sonic-net#7726) In the existing fwutil test implement, user can only define the component(BIOS, ONIE, CPLD) based on the platform type, if for the same platform, it require to define different components for different dut(such as some setup are respined), them the origin implementation dose not support. modify the script to support such scenario. The fwutil test case should not be skipped, since the [test_fwutil]fixture 'fw_pkg_name' not found sonic-net#6489 is not a real issue. Fix some pep8 issue
* added t0 lag topo filwith 8 LAG uplinks and eos leaf template
* added warm-up traffic to all pfcwd tests
…onic-net#7737) What is the motivation for this PR? After config reload in one tor of dual-tor, mux simulator will be inconsistent for several seconds, that's because mux simulator needs some time to toggle to correct status. How did you do it? Add wait time in sanity check to verify mux status, and check mux status later.
…SICs (sonic-net#7718) remove skip_pfcwd_test for broadcom devices since all pfcwd tests have warm up traffic now
…ts (sonic-net#7761) What is the motivation for this PR? During upgrade path, regex picking should be done based on the version currently running on the device. This bug has created issues in 201911 to 202205 upgrade path, where for post upgrade timing collection the test picks 201911 regex. How did you do it? Fix this by checking the current running version to select that version's regex for timing data collection. How did you verify/test it? Tested on a physical device.
…f DUT VLAN interface (sonic-net#7801) * before arp refresh, dst mac address keep original value instead of default vlan * add some comments for background to avoid break it again
What is the motivation for this PR? Fix setup error in platform_tests/fwutil/test_fwutil.py when test suite is executed without argument --fw-pkg How did you do it? How did you verify/test it? Run platform_tests/fwutil/test_fwutil.py without passing --fw-pkg option: PASSED platform_tests/fwutil/test_fwutil.py::test_fwutil_show PASSED platform_tests/fwutil/test_fwutil.py::test_fwutil_install_bad_name SKIPPED [2] platform_tests/fwutil/test_fwutil.py:123: Command not yet merged into sonic-utilites SKIPPED [6] /var/user/jenkins/sonic-mgmt/tests/platform_tests/fwutil/conftest.py:35: No fw package specified. Signed-off-by: Andrii-Yosafat Lozovyi <[email protected]>
* remove polynomial query in codeql.
What is the motivation for this PR? There is a need to add tacacs account to fanout without affecting current fanout devices that still uses local credentials. How did you do it? Add fanout_tacacs_sonic_user/password that can override fanout_sonic_user/password, and also a fanout_tacacs_user/password that can override fanout_tacacs_sonic_user/password. How did you verify/test it? Verified in lab. Signed-off-by: Xichen Lin <[email protected]>
…-net#7795) Approach What is the motivation for this PR? Add missing topo mark util. Signed-off-by: Longxiang Lyu [email protected] How did you do it? How did you verify/test it?
How did you do it? Skip test_fwutil for other platforms. How did you verify/test it? Run test_fwutil.py on Broadcom platform devices. Signed-off-by: Zhaohui Sun <[email protected]>
What is the motivation for this PR? To skip fast and warm reboot for non t1 topos How did you do it? To refine the mark conditional file, which use wrong variable for the skip. How did you verify/test it? Manually run test.
…unning 'show aaa' (sonic-net#7834) What is the motivation for this PR? Fix test_authorization does not wait long enough before running show aaa. How did you do it? Use wait_until to run show aaa, remove time.sleep before checking it. How did you verify/test it? Run TC and TC passed. Signed-off-by: Chun'ang Li <[email protected]>
…t#9639) Retry reboot in ro-disk UT when DUT not reachable. ### Description of PR Retry reboot in ro-disk UT when DUT not reachable. ### Type of change <!-- - Fill x for your type of change. - e.g. - [x] Bug fix --> - [ ] Bug fix - [ ] Testbed and Framework(new/improvement) - [x] Test case(new/improvement) ### Back port request - [ ] 201911 - [ ] 202012 - [x] 202205 ### Approach #### What is the motivation for this PR? Device unreachable after TACACS ro-disk UT failed. #### How did you do it? Add retry when ro-disk DUT not reachable before reboot. #### How did you verify/test it? Pass all UT #### Any platform specific information? #### Supported testbed topology if it's a new test case? ### Documentation <!-- (If it's a new feature, new test case) Did you update documentation/Wiki relevant to your implementation? Link to the wiki page? -->
Host dir like "/data/ceos/ceos_vms6-1_VM0100" is mounted to cEOS container as flash disk. This dir contains all kinds of configuration files that may affect cEOS behaviors. For example, older version of cEOS has file ".arista_archive_config" which enables archiving of configs every 1 minute. On newer version of cEOS, this config file is removed by default. However, after cEOS is upgraded, the left over ".arista_archive_config" file will cause newer cEOS to archive configs every 1 minute by default. To ensure that the cEOS container is always "cleanly" started, this change added code to always cleanup the mount dir before creating cEOS. Signed-off-by: Xin Wang <[email protected]>
The sonic-mgmt docker image has been updated to globally install all the packages in env-python3. Now, in the global python3 environment, we have all the required packages for running tests. Then it is unnecessary to have the env-python3 virtual environment. Before "env-python3" is deprecated in sonic-mgmt docker image, we can skip creating "env-python3" in the setup-container.sh script which is used for setting up sonic-mgmt container for user different than "AzDevOps" to solve permission issues of mapped directory. This change added code to check if pytest is globally installed in python3 environment. If yes, then we can assume that all the env-python3 packages are also globally installed. Then we can skip creating env-python3 in setup-container.sh script. Signed-off-by: Xin Wang <[email protected]>
…_po_update case (sonic-net#10044) What is the motivation for this PR? Cherry pick sonic-net#10031 to 202012. Signed-off-by: Zhaohui Sun <[email protected]>
…restore (sonic-net#10083) * Add testcase for pfcwd multi port storm trigger and restore Backport sonic-net#8709 Signed-off-by: Neetha John <[email protected]>
…t#10125) Signed-off-by: Zhaohui Sun <[email protected]>
Cherry pick sonic-net#10135 Signed-off-by: Zhaohui Sun <[email protected]>
ZhaohuiS
requested review from
StormLiangMS,
XuChen-MSFT,
jcaiMR,
lipxu,
lolyu,
neethajohn,
prgeor,
wangxin,
wsycqyz and
yxieca
as code owners
Contributor
Author
|
commit into wrong branch. |
ZhaohuiS
removed request for
StormLiangMS,
XuChen-MSFT,
jcaiMR,
lipxu,
lolyu,
neethajohn,
prgeor,
wangxin,
wsycqyz and
yxieca
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
20 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description of PR
Summary:
Fixes # (issue)
Cherry pick #10243 into 202012.
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?
Any platform specific information?
Supported testbed topology if it's a new test case?
Documentation