Skip to content

Commit d7f3275

Browse files
rjeberhardrjeberhard
committed
Merge remote-tracking branch 'origin/owls-73691' into release/2.2
2 parents 6cf06c3 + 64ac168 commit d7f3275

File tree

80 files changed

+5362
-228
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

80 files changed

+5362
-228
lines changed

‎docs/domains/Domain.json

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -407,7 +407,7 @@
407407
"type": "object",
408408
"properties": {
409409
"overallHealth": {
410-
"description": "Server health of this WebLogic server.",
410+
"description": "Server health of this WebLogic server. If the value is \"Not available\", the operator has failed to read the health. If the value is \"Not available (possibly overloaded)\", the operator has failed to read the health of the server possibly due to the server is in overloaded state",
411411
"type": "string"
412412
},
413413
"activationTime": {

‎docs/domains/Domain.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -173,7 +173,7 @@ Shutdown describes the configuration for shutting down a server instance.
173173
| Name | Type | Description |
174174
| --- | --- | --- |
175175
| `activationTime` | DateTime | RFC 3339 date and time at which the server started. |
176-
| `overallHealth` | string | Server health of this WebLogic server. |
176+
| `overallHealth` | string | Server health of this WebLogic server. If the value is "Not available", the operator has failed to read the health. If the value is "Not available (possibly overloaded)", the operator has failed to read the health of the server possibly due to the server is in overloaded state |
177177
| `subsystems` | array of [Subsystem Health](#subsystem-health) | Status of unhealthy subsystems, if any. |
178178

179179
### Channel

‎docs/domains/index.html

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1327,7 +1327,7 @@
13271327
"type": "object",
13281328
"properties": {
13291329
"overallHealth": {
1330-
"description": "Server health of this WebLogic server.",
1330+
"description": "Server health of this WebLogic server. If the value is \"Not available\", the operator has failed to read the health. If the value is \"Not available (possibly overloaded)\", the operator has failed to read the health of the server possibly due to the server is in overloaded state",
13311331
"type": "string"
13321332
},
13331333
"activationTime": {

‎integration-tests/JRFUSECASES.MD

Lines changed: 61 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,61 @@
1+
# Use Cases for FMW Infra Domain in the Operator
2+
3+
Java integration tests cover the following use cases:
4+
5+
## Basic test Configuration & Use Cases
6+
7+
| | |
8+
| --- | --- |
9+
| Operator Configuration | operator1 deployed in `weblogic-operator` namespace and manages domains in `default` and `jrfdomains` namespaces |
10+
| Domain Configuration | Domain on PV using WLST, Traefik load balancer |
11+
12+
**Basic Use Cases**
13+
14+
1. Create operator `weblogic-operator` which manages `default` and `jrfdomains` namespaces, verify it's deployed successfully, pods created, operator ready and verify external REST service, if configured.
15+
2. Create domain `jrfdomain` in `jrfdomains` namespace and verify the pods, services are created and servers are in ready state.
16+
3. Verify the admin external service by accessing the admin REST endpoint with `nodeport` in URL.
17+
4. Verify exec into the admin pod and deploying webapp using the admin port with WLST.
18+
5. Verify web app load balancing by accessing the webapp using `loadBalancerWebPort`.
19+
20+
**Advanced Use Cases**
21+
22+
6. Verify domain life cycle (destroy and create) should not have any impact on operator managing the domain and web app load balancing and admin external service.
23+
7. Cluster scale up/down using operator REST endpoint, webapp load balancing should adjust accordingly.
24+
8. Operator life cycle (destroy and create) should not impact the running domain.
25+
26+
Also the below use cases are covered for each test:
27+
28+
9. Verify the liveness probe by killing managed server 1 process 3 times to kick pod auto-restart.
29+
10. Shutdown the domain by changing domain `serverStartPolicy` to `NEVER`.
30+
31+
32+
## Full test Configuration & Use Cases
33+
34+
Additional Operator Configuration:
35+
36+
| | |
37+
| --- | --- |
38+
| Operator Configuration | `operator2` deployed in `weblogic-operator2` namespace and manages domains in `test2` namespace |
39+
40+
Basic Use Cases described above are verified in all the domain configurations. Also the below use cases are covered:
41+
42+
| Domain | Use Case |
43+
| --- | --- |
44+
| Domain on PV using WLST | as above in basic use cases |
45+
| Domain with ADMIN_ONLY | verify only admin server is started and managed servers are not started. Shutdown domain by deleting domain CRD. Create domain on existing PV dir, pv is already populated by a shutdown domain. |
46+
| Domain with situational config | create domain with listen address not set for admin server and t3 channel/NAP and incorrect file for admin server log location. Introspector should override these with sit-config automatically. Also, with some junk value for t3 channel public address and using custom situational config override replace with valid public address using secret. Also, on Jenkins this domain uses NFS instead of HOSTPATH PV storage |
47+
| Two domains managed by two operators | verify scaling and restart of one domain doesn't impact another domain. Delete domain resources using delete script from samples. |
48+
| Two domains in the same namespace managed by one operator | create two FMW Infra Domains in the same namespace managed by one operator. Verify scaling and restart of one domain doesn't impact another domain. |
49+
| Two domains in the different namespaces managed by one operator | create two FMW Infra Domains in the different namespaces managed by one operator. Domain1 uses VOYAGER load balancer. Domain2 uses TRAEFIK load balancer. Verify scaling and restart of one domain doesn't impact another domain. |
50+
| Domain with Recycle policy | create domain with pvReclaimPolicy="Recycle" and using Configured cluster. Verify that the PV is deleted once the domain and PVC are deleted |
51+
| Domain with default sample values | create domain using mostly default values for inputs |
52+
53+
Test cases cover the bugs found:
54+
55+
| Domain | Use Case |
56+
| --- | --- |
57+
| Domain on PV using WLST | cluster rolling restart with restartVersion and maxUnavailable set to 2, verify the not-ready server counts can not exceed maxUnavailable value |
58+
| Domain on PV using WLST | in create domain input file, set exposeAdminNodePort to false and exposeAdminT3Channel to true, verify the managed server pods are created |
59+
| Domain on PV using WLST | in create domain input file, set createDomainScriptsMountPath to non-default value, verify the create domain sample script works |
60+
| Domain on PV using WLST | in createFMWDomain.py file, set administration_port_enabled to true, verify the admin sever pod is running and ready and all managed server pods are created |
61+
| Domain on PV using WLST | in create domain input file, set exposeAdminT3Channel to true, verify the admin t3 channel is exposed |

‎integration-tests/README.md

Lines changed: 45 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -16,6 +16,8 @@ Shared cluster runs only Quick test use cases, Jenkins runs both Quick and Full
1616

1717
Use Cases covered in integration tests for the operator is available [here](USECASES.MD)
1818

19+
JRF Use Cases are covered [here](JRFUSECASES.MD)
20+
1921
# Directory Configuration and Structure
2022

2123
Directory structure of source code:
@@ -209,7 +211,7 @@ The tests accepts optional env var overrides:
209211
| RESULT_ROOT | The root directory to use for the tests temporary files. See "Directory Configuration and Structure" for defaults and a detailed description of test directories. |
210212
| PV_ROOT | The root directory on the kubernetes cluster used for persistent volumes. See "Directory Configuration and Structure" for defaults and a detailed description of test directories. |
211213
| QUICKTEST | When set to "true", limits testing to a subset of the tests. |
212-
| LB_TYPE | The default value is "TRAEFIK". Set to "VOYAGER" if you want to use it as LB. |
214+
| LB_TYPE | The default value is "TRAEFIK". Set to "VOYAGER" if you want to use it as LB. Using "VOYAGER" requires unique "voyagerWebPort"|
213215
| INGRESSPERDOMAIN | The defult value is true. If you want to test creating TRAEFIK LB by kubectl yaml for multiple domains, set it to false. |
214216
| SHARED_CLUSTER | Set to true if invoking on shared cluster, set to false or "" if running stand-alone or from Jenkins. Default is "". |
215217
| JENKINS | Set to true if invoking from Jenkins, set to false or "" if running stand-alone or on shared cluster. Default is "". |
@@ -226,8 +228,20 @@ SHARED_CLUSTER=true:
226228
| IMAGE_NAME_OPERATOR | Docker image name for operator. Default is weblogic-kubernetes-operator |
227229
| IMAGE_PULL_POLICY_OPERATOR | Default 'Never'. |
228230
| IMAGE_PULL_SECRET_OPERATOR | Default ''. |
229-
| IMAGE_PULL_SECRET_WEBLOGIC | Default ''.
231+
| IMAGE_PULL_SECRET_WEBLOGIC | Default ''. |
230232

233+
The below env variables are required for SHARED_CLUSTER=true:
234+
235+
| Variable | Description |
236+
| --- | --- |
237+
| REPO_REGISTRY | OCIR Server to push/pull the Operator image |
238+
| REPO_USERNAME | OCIR Username |
239+
| REPO_PASSWORD | OCIR token |
240+
| REPO_EMAIL | OCIR email |
241+
| DOCKER_USERNAME | Docker username to pull the Weblogic image |
242+
| DOCKER_PASSWORD | Docker password |
243+
| DOCKER_EMAIL | Docker email |
244+
| K8S_NODEPORT_HOST | DNS name of a Kubernetes worker node. |
231245

232246
Successful run will have the output like below:
233247
```
@@ -282,6 +296,35 @@ Failed run will have the output like
282296
```
283297
JUnit test results can be seen at "integration-tests/target/failsafe-reports/TEST-oracle.kubernetes.operator.ITOperator.xml". This file shows how much time each test case took to run and the failed test results if any.
284298

299+
# How to run JRF domain In Operator related tests
300+
* Setup docker access to FMW Infrastructure 12c Image and Oracle Database 12c Image
301+
302+
Method 1
303+
- Setup a personal account on hub.docker.com
304+
- Then sign in to hub.docker.com and signup for access to Oracle Database 12c Images via https://hub.docker.com/_/oracle-database-enterprise-edition
305+
- Then export the following before running the tests:
306+
```
307+
export DOCKER_USERNAME=<docker_username>
308+
export DOCKER_PASSWORD=<docker_password>
309+
export DOCKER_EMAIL=<docker_email>
310+
```
311+
- Setup an account in phx.ocir.io
312+
- Then sign in to phx.ocir.io to get access to FMW Infrastructure 12c Image: **_phx.ocir.io/weblogick8s/oracle/fmw-infrastructure:12.2.1.3_**
313+
- export the following before running the tests:
314+
```
315+
export REPO_USERNAME=<ocir_username>
316+
export REPO_PASSWORD=<ocir_password>
317+
export REPO_EMAIL=<ocir_email>
318+
```
319+
320+
Method 2
321+
- Make sure the FMW Infrastructure image i.e. **_phx.ocir.io/weblogick8s/oracle/fmw-infrastructure:12.2.1.3_** and the Oracle database image i.e. **_store/oracle/database-enterprise:12.2.0.1_** already exist locally in a docker repository the k8s cluster can access
322+
323+
* Command to run the tests:
324+
```
325+
mvn clean verify -P jrf-integration-tests 2>&1 | tee log.txt
326+
```
327+
285328
# How to run a single test
286329
287330
mvn -Dit.test="ITOperator#testDomainOnPVUsingWLST" -DfailIfNoTests=false integration-test -P java-integration-tests

‎integration-tests/USECASES.MD

Lines changed: 28 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@
22

33
Java integration tests cover the below use cases:
44

5-
## Quick test Configuration & Use Cases -
5+
## Quick test Configuration & Use Cases -
66

77
| | |
88
| --- | --- |
@@ -34,8 +34,7 @@ Also the below use cases are covered for Quick test:
3434
| | |
3535
| --- | --- |
3636
| Operator Configuration | operator2 deployed in weblogic-operator2 namespace and manages domains test2 namespace |
37-
| Domain Configuration | Domain on PV using WDT <p> Domain home in image using WLST <p> Domain home in image using WDT <p> Domain with serverStartPolicy ADMIN_ONLY <p> Domain with auto and custom situational configuration <p> Two domains managed by two operators <p> Domain with Recycle weblogicDomainStorageReclaimPolicy <p> Domain with default sample values |
38-
37+
| Domain Configuration | Domain on PV using WDT <p> Domain home in image using WLST <p> Domain home in image using WDT <p> Domain with serverStartPolicy ADMIN_ONLY <p> Domain with auto and custom situational configuration <p> Two domains managed by two operators <p> Domain with Recycle weblogicDomainStorageReclaimPolicy <p> Domain with default sample values <p> Domain on PV with two configured clusters <p> Domain on PV with one configured cluster and one dynamic cluster |
3938

4039
Basic Use Cases described above are verified in all the domain configurations. Also the below use cases are covered:
4140

@@ -96,5 +95,29 @@ Configuration Overrides Usecases
9695

9796
| Server Discovery | Use Case |
9897
| --- | --- |
99-
| Discover a newly started server | stop operator and apply the modified domain.yaml with replicas count increated, restart Operator and verify that the cluster is scaled up accordingly |
100-
| Discover dead weblogic servers | stop Operator, kill admin server and all managed servers in the cluster, restart Operator and verify that it restarts all dead servers |
98+
| Discover a newly started server | Stop operator and apply the modified domain.yaml with replicas count increased, restart Operator and verify that the cluster is scaled up accordingly |
99+
| Discover dead weblogic servers | Stop Operator, kill admin server and all managed servers in the cluster, restart Operator and verify that it restarts all dead servers |
100+
101+
| Sticky Session | Use Case |
102+
| --- | --- |
103+
| Server affinity | Use a web application deployed on Weblogic cluster to track HTTP session. Test server-affinity by sending two HTTP requests to Weblogic and verify that all requests are directed to same Weblogic server |
104+
| Session state isolation | Verify that values saved in a client session state are not visible to another client |
105+
106+
| Monitoring Exporter | Use Case |
107+
| --- | --- |
108+
| Check Metrics via Prometheus | build, deploy webapp for Monitoring Exporter, start Prometheus and verify the metrics was produced by using Prometheus APIs |
109+
| Replace Configuration via exporter console| Verify that configuration for monitoring exporter can be replaced during runtime, check applied metrics via Prometheus APIs|
110+
| Append Configuration via exporter console| Verify that configuration for monitoring exporter can be appended during runtime, check applied metrics via Prometheus APIs|
111+
| Append Configuration with varios combinations of attributes via exporter console| Append monitoring exporter configuration [a] to new config [a,b] and verify it was applied |
112+
| Replace Configuration with only one attribute as array via exporter console| Replace monitoring exporter configuration [a,b,c] attributes with new config [a] and verify it was applied |
113+
| Replace Configuration with empty config file via exporter console| Replace monitoring exporter configuration with empty config file, verify it was applied |
114+
| Replace/Append Configuration with config file written in non yml format via exporter console| Try to replace/append monitoring exporter configuration with config file written in non yml format, verify configuration has not changed |
115+
| Replace/Append Configuration with corrupted yml file via exporter console| Try to replace/append monitoring exporter configuration with config file written in corrupted yml format, verify configuration has not changed |
116+
| Replace/Append Configuration with dublicated values in the config file via exporter console| Try to replace/append monitoring exporter configuration with dublicated values in the config file, verify configuration has not changed |
117+
| Replace/Append Configuration with invalid credentials via exporter console| Try to replace/append monitoring exporter configuration with varios combos for invalid credentials, verify configuration has not changed and `401 Unauthorized` exception was thrown |
118+
119+
| Logging with Elastic Stack | Use Case |
120+
| --- | --- |
121+
| Search log level | Use Elasticsearch Count API to query logs of level=INFO and verify that total number of logs for level=INFO is not zero and failed count is zero |
122+
| Search Operator log | Use Elasticsearch Search APIs to query Operator log info and verify that log hits for type=weblogic-operator are not empty |
123+
| Search Weblogic log | Use Elasticsearch Search APIs to query Weblogic log info and verify that log hits for Weblogic servers are not empty |

‎integration-tests/pom.xml

Lines changed: 6 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -113,6 +113,11 @@
113113
<artifactId>jaxb-api</artifactId>
114114
<version>2.3.1</version>
115115
</dependency>
116+
<dependency>
117+
<groupId>net.sourceforge.htmlunit</groupId>
118+
<artifactId>htmlunit</artifactId>
119+
<version>2.32</version>
120+
</dependency>
116121
</dependencies>
117122

118123
<build>
@@ -352,7 +357,7 @@
352357
<id>jrf-integration-tests</id>
353358
<properties>
354359
<skipITs>false</skipITs>
355-
<includes-failsafe>**/JrfInOperatorTest.java</includes-failsafe>
360+
<includes-failsafe>**/JrfInOperator*Test.java</includes-failsafe>
356361
<jrf_enabled>true</jrf_enabled>
357362
</properties>
358363
</profile>

0 commit comments

Comments
 (0)