1
0
mirror of https://github.com/owncloud/ocis.git synced 2026-10-23 23:29:02 +03:00
Files
ocis/tests/acceptance/features/apiServiceAvailability/serviceAvailabilityCheck.feature
Sawjan Gurung 39192c7bea ci: use existing servers (#12100)
* ci: enable k8s setup for remaining suites

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* ci: use existing servers

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* ci: fix

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* test: fix tests

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* test: increase timeout

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* ci: fix mail setup

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* test: increase timeout

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* ci: fix tika

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* test: fix csp directives

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* ci: merge ocm provider configs

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* test: fix notification tests

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* ci: fix script file

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* ci: enable ocm server logs

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* ci: fix pattern

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* ci: fix cluster wait

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* ci: run auth-app only when required

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* test: fix sharing tests

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

* test: fix tests

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>

---------

Signed-off-by: Saw-jan <saw.jan.grg3e@gmail.com>
2026-03-12 10:46:13 +05:45

159 lines
9.1 KiB
Gherkin

Feature: service health check
Scenario: check default services health
When a user requests these URLs with "GET" and no authentication
| endpoint | service |
| http://%base_url_hostname%:9197/healthz | activitylog |
| http://%base_url_hostname%:9165/healthz | app-provider |
| http://%base_url_hostname%:9243/healthz | app-registry |
| http://%base_url_hostname%:9147/healthz | auth-basic |
| http://%base_url_hostname%:9167/healthz | auth-machine |
| http://%base_url_hostname%:9198/healthz | auth-service |
| http://%base_url_hostname%:9260/healthz | clientlog |
| http://%base_url_hostname%:9270/healthz | eventhistory |
| http://%base_url_hostname%:9141/healthz | frontend |
| http://%base_url_hostname%:9143/healthz | gateway |
| http://%base_url_hostname%:9124/healthz | graph |
| http://%base_url_hostname%:9161/healthz | groups |
| http://%base_url_hostname%:9239/healthz | idm |
| http://%base_url_hostname%:9134/healthz | idp |
| http://%base_url_hostname%:9234/healthz | nats |
| http://%base_url_hostname%:9163/healthz | ocdav |
| http://%base_url_hostname%:9281/healthz | ocm |
| http://%base_url_hostname%:9114/healthz | ocs |
| http://%base_url_hostname%:9255/healthz | postprocessing |
| http://%base_url_hostname%:9205/healthz | proxy |
| http://%base_url_hostname%:9224/healthz | search |
| http://%base_url_hostname%:9194/healthz | settings |
| http://%base_url_hostname%:9151/healthz | sharing |
| http://%base_url_hostname%:9139/healthz | sse |
| http://%base_url_hostname%:9179/healthz | storage-publiclink |
| http://%base_url_hostname%:9156/healthz | storage-shares |
| http://%base_url_hostname%:9217/healthz | storage-system |
| http://%base_url_hostname%:9159/healthz | storage-users |
| http://%base_url_hostname%:9189/healthz | thumbnails |
| http://%base_url_hostname%:9214/healthz | userlog |
| http://%base_url_hostname%:9145/healthz | users |
| http://%base_url_hostname%:9104/healthz | web |
| http://%base_url_hostname%:9119/healthz | webdav |
| http://%base_url_hostname%:9279/healthz | webfinger |
| http://%base_url_hostname%:9277/healthz | antivirus |
| http://%base_url_hostname%:9174/healthz | notification |
Then the HTTP status code of responses on all endpoints should be "200"
@env-config
Scenario: check extra services health
Given the following configs have been set:
| config | value |
| OCIS_ADD_RUN_SERVICES | auth-app,policies,invitations |
| AUTH_APP_DEBUG_ADDR | 0.0.0.0:9245 |
| POLICIES_DEBUG_ADDR | 0.0.0.0:9129 |
| INVITATIONS_DEBUG_ADDR | 0.0.0.0:9269 |
When a user requests these URLs with "GET" and no authentication
| endpoint | service |
| http://%base_url_hostname%:9245/healthz | auth-app |
| http://%base_url_hostname%:9269/healthz | invitations |
| http://%base_url_hostname%:9129/healthz | policies |
Then the HTTP status code of responses on all endpoints should be "200"
Scenario: check default services readiness
When a user requests these URLs with "GET" and no authentication
| endpoint | service |
| http://%base_url_hostname%:9197/readyz | activitylog |
| http://%base_url_hostname%:9165/readyz | app-provider |
| http://%base_url_hostname%:9243/readyz | app-registry |
| http://%base_url_hostname%:9147/readyz | auth-basic |
| http://%base_url_hostname%:9167/readyz | auth-machine |
| http://%base_url_hostname%:9198/readyz | auth-service |
| http://%base_url_hostname%:9260/readyz | clientlog |
| http://%base_url_hostname%:9270/readyz | eventhistory |
| http://%base_url_hostname%:9141/readyz | frontend |
| http://%base_url_hostname%:9143/readyz | gateway |
| http://%base_url_hostname%:9161/readyz | groups |
| http://%base_url_hostname%:9239/readyz | idm |
| http://%base_url_hostname%:9234/readyz | nats |
| http://%base_url_hostname%:9163/readyz | ocdav |
| http://%base_url_hostname%:9281/readyz | ocm |
| http://%base_url_hostname%:9114/readyz | ocs |
| http://%base_url_hostname%:9255/readyz | postprocessing |
| http://%base_url_hostname%:9224/readyz | search |
| http://%base_url_hostname%:9194/readyz | settings |
| http://%base_url_hostname%:9151/readyz | sharing |
| http://%base_url_hostname%:9139/readyz | sse |
| http://%base_url_hostname%:9179/readyz | storage-publiclink |
| http://%base_url_hostname%:9156/readyz | storage-shares |
| http://%base_url_hostname%:9217/readyz | storage-system |
| http://%base_url_hostname%:9159/readyz | storage-users |
| http://%base_url_hostname%:9189/readyz | thumbnails |
| http://%base_url_hostname%:9214/readyz | userlog |
| http://%base_url_hostname%:9145/readyz | users |
| http://%base_url_hostname%:9104/readyz | web |
| http://%base_url_hostname%:9119/readyz | webdav |
| http://%base_url_hostname%:9279/readyz | webfinger |
| http://%base_url_hostname%:9277/readyz | antivirus |
| http://%base_url_hostname%:9174/readyz | notification |
Then the HTTP status code of responses on all endpoints should be "200"
@env-config
Scenario: check extra services readiness
Given the following configs have been set:
| config | value |
| OCIS_ADD_RUN_SERVICES | auth-app,policies,invitations |
| AUTH_APP_DEBUG_ADDR | 0.0.0.0:9245 |
| POLICIES_DEBUG_ADDR | 0.0.0.0:9129 |
| INVITATIONS_DEBUG_ADDR | 0.0.0.0:9269 |
When a user requests these URLs with "GET" and no authentication
| endpoint | service |
| http://%base_url_hostname%:9245/readyz | auth-app |
| http://%base_url_hostname%:9269/readyz | invitations |
| http://%base_url_hostname%:9129/readyz | policies |
Then the HTTP status code of responses on all endpoints should be "200"
@issue-10661
Scenario: check default services readiness (graph, idp, proxy)
When a user requests these URLs with "GET" and no authentication
| endpoint | service |
| http://%base_url_hostname%:9124/readyz | graph |
| http://%base_url_hostname%:9134/readyz | idp |
| http://%base_url_hostname%:9205/readyz | proxy |
Then the HTTP status code of responses on all endpoints should be "200"
@env-config @issue-10661
Scenario: check auth-bearer service health
Given the following configs have been set:
| config | value |
| OCIS_ADD_RUN_SERVICES | auth-bearer |
| AUTH_BEARER_DEBUG_ADDR | 0.0.0.0:9149 |
When a user requests these URLs with "GET" and no authentication
| endpoint | service |
| http://%base_url_hostname%:9149/healthz | auth-bearer |
Then the HTTP status code should be "200"
@env-config @issue-10661
Scenario: check auth-bearer service readiness
Given the following configs have been set:
| config | value |
| OCIS_ADD_RUN_SERVICES | auth-bearer |
| AUTH_BEARER_DEBUG_ADDR | 0.0.0.0:9149 |
When a user requests these URLs with "GET" and no authentication
| endpoint | service |
| http://%base_url_hostname%:9149/readyz | auth-bearer |
Then the HTTP status code should be "200"
@env-config
Scenario: check services health and readiness while running separately
Given the following configs have been set:
| config | value |
| OCIS_EXCLUDE_RUN_SERVICES | audit |
And the administrator has started service "audit" separately with the following configs:
| config | value |
| OCIS_LOG_LEVEL | info |
| AUDIT_DEBUG_ADDR | 0.0.0.0:9229 |
When a user requests these URLs with "GET" and no authentication
| endpoint | service |
| http://%base_url_hostname%:9229/healthz | audit |
| http://%base_url_hostname%:9229/readyz | audit |
Then the HTTP status code of responses on all endpoints should be "200"