feat: support etcd prefix as apisix does

[liuyang211]

Please answer these questions before submitting a pull request

• Why submit this pull request?
• New feature provided
• Related issues
  Fixes feat: support etcd prefix as apisix does #1456

---

New feature or improvement

• Describe the details and related test reports.
  support etcd prefix as apisix does

---

Please add the corresponding test cases if necessary.

---

Backport patches

• Why need to backport?

• Source branch

• Related commits and pull requests

• Target branch

[juzhiyuan]

Hi, I see there have 3 checks failed:

• Backend CLI & Backend E2E: @nic-chen will help you to check what happened.
• Semantic Pull Request: You could ignore it safely.

[starsz]

Can you write some test cases at "/api/test/shell/cli_test.sh" to test this?

[juzhiyuan]

Hi, please sync your codes with the master branch to fix the CI failure.

[codecov-io]

Codecov Report

Merging #1477 (901b399) into master (bd5ed03) will decrease coverage by 17.32%.
The diff coverage is 0.00%.

@@             Coverage Diff             @@
##           master    #1477       +/-   ##
===========================================
- Coverage   68.95%   51.63%   -17.33%     
===========================================
  Files          48       39        -9     
  Lines        3038     2597      -441     
===========================================
- Hits         2095     1341      -754     
- Misses        705     1082      +377     
+ Partials      238      174       -64     

Impacted Files	Coverage Δ
api/internal/core/store/storehub.go	0.00% <0.00%> (-74.49%)	⬇️
api/internal/utils/version.go	0.00% <0.00%> (-100.00%)	⬇️
api/internal/filter/request_id.go	0.00% <0.00%> (-100.00%)	⬇️
api/internal/core/entity/entity.go	0.00% <0.00%> (-100.00%)	⬇️
api/internal/filter/cors.go	0.00% <0.00%> (-66.67%)	⬇️
api/internal/filter/schema.go	0.00% <0.00%> (-55.47%)	⬇️
api/internal/utils/consts/api_error.go	0.00% <0.00%> (-50.00%)	⬇️
api/internal/handler/data_loader/route_import.go	27.41% <0.00%> (-37.50%)	⬇️
api/internal/handler/server_info/server_info.go	57.14% <0.00%> (-33.34%)	⬇️
... and 29 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update bd5ed03...901b399. Read the comment docs.

[nic-chen]

@liuyang211 need some test cases. please refer to https://github.com/apache/apisix-dashboard/blob/master/api/test/shell/cli_test.sh#L300-L327
thanks.

[tokers]

Use short assignment is better:

prefix := "/apisix"

[liuyang211]

OK, I'll change it this way.

[liuyang211]

@liuyang211 need some test cases. please refer to https://github.com/apache/apisix-dashboard/blob/master/api/test/shell/cli_test.sh#L300-L327
thanks.

I run these cmds to test it locally,

cd apisix-dashboard/api
./test/shell/cli_test.sh

but fail at
./test/shell/cli_test.sh: line 347: ./etcd-v3.4.14-linux-amd64/etcd: No such file or directory

Do I miss something?

BTW

1. /tmp/manager-api is used as both directory and file.
   https://github.com/apache/apisix-dashboard/blob/master/api/test/shell/cli_test.sh#L143-L144
   https://github.com/apache/apisix-dashboard/blob/master/api/test/shell/cli_test.sh#L166
   This seems to raise this error if rm /tmp/manager-api is not manually excuted before next test.
   cp: cannot overwrite non-directory '/tmp/manager-api' with directory '/home/newbie/dev/apisix-dashboard/api'

2. These two lines should swap to avoid error message when stopping manager-api
   https://github.com/apache/apisix-dashboard/blob/master/api/test/shell/cli_test.sh#L385-L387

[nic-chen]

@liuyang211
please have a look at https://github.com/apache/apisix-dashboard/blob/master/.github/workflows/backend-cli-test.yml#L27-L31

If there is something not good, you could fix it. thanks.

[liuyang211]

@liuyang211
please have a look at https://github.com/apache/apisix-dashboard/blob/master/.github/workflows/backend-cli-test.yml#L27-L31

If there is something not good, you could fix it. thanks.

Hi, the testcase runs OK in my local machine, but fails in the github workflow, could you review my testcase and give some advice? Thanks.

[juzhiyuan]

cc @nic-chen to take a look

[nic-chen]

@liuyang211
The reason for the failure is that there is no sleep after the previous stop, which causes the Manager API to not really restart successfully:
https://github.com/apache/apisix-dashboard/pull/1477/checks?check_run_id=1946779863#step:5:530

[liuyang211]

@liuyang211
The reason for the failure is that there is no sleep after the previous stop, which causes the Manager API to not really restart successfully:
https://github.com/apache/apisix-dashboard/pull/1477/checks?check_run_id=1946779863#step:5:530

Thanks, the testcase is OK now.
I'll add more testcases for routes, services, etc.

[nic-chen]

@liuyang211
merged. thanks a lot !

[liuyang211]

@liuyang211
merged. thanks a lot !

Thank you for your help!
@nic-chen @juzhiyuan @starsz @tokers

[nic-chen]

@liuyang211
More contributions are welcome ! : )