Re: Release Apache APISIX 2.11.0 Round 2

[Zeping Bai]

+1 (non-binding)

I checked：
1. Download links are OK.
2. Checksum and signature are OK.
3. LICENSE and NOTICE files are exist.
4. Building on Gitpod (Ubuntu 20.04) was successful.


Zeping Bai  @bzp2010

孙毅  于2021年11月23日周二 下午8:48写道：

> Hello, Community,
> This is a call for the vote to release Apache APISIX version 2.11.0
>
> Release notes:
>
> https://github.com/apache/apisix/blob/release/2.11/CHANGELOG.md#2110
>
> The release candidates:
>
> https://dist.apache.org/repos/dist/dev/apisix/2.11.0/
>
> Release Commit ID:
>
>
> https://github.com/apache/apisix/commit/d39814f3b5291393effe9c258bcc4a7f77911334
>
> Keys to verify the Release Candidate:
>
> https://dist.apache.org/repos/dist/dev/apisix/KEYS
>
> Steps to validating the release:
>
> 1. Download the release
>
> wget
>
> https://dist.apache.org/repos/dist/dev/apisix/2.11.0/apache-apisix-2.11.0-src.tgz
>
> 2. Checksums and signatures
>
> wget https://dist.apache.org/repos/dist/dev/apisix/KEYS
>
> wget
>
> https://dist.apache.org/repos/dist/dev/apisix/2.11.0/apache-apisix-2.11.0-src.tgz.asc
>
> wget
>
> https://dist.apache.org/repos/dist/dev/apisix/2.11.0/apache-apisix-2.11.0-src.tgz.sha512
>
> gpg --import KEYS
>
> shasum -c apache-apisix-2.11.0-src.tgz.sha512
>
> gpg --verify apache-apisix-2.11.0-src.tgz.asc apache-apisix-2.11.0-src.tgz
>
> 3. Unzip and Check files
>
> tar zxvf apache-apisix-2.11.0-src.tgz
>
> 4. Build Apache APISIX:
>
>
> https://github.com/apache/apisix/blob/release/2.11/docs/en/latest/how-to-build.md#installation-via-source-release-package
>
> The vote will be open for at least 72 hours or until necessary number of
> votes are reached.
>
> Please vote accordingly:
>
> [ ] +1 approve
> [ ] +0 no opinion
> [ ] -1 disapprove with the reason
>

Re: Here are some feedbacks from our developers

[Zexuan Luo]

> Community maintainers should lead us to learn about them as well.

>From the above 3 PRs, I have led us to contribute in a good way.

For the first case:
In https://github.com/apache/apisix/pull/4892, I told the contributor:

> Smart people like you should focus on advanced issues.

See https://github.com/apache/apisix/pull/4892#issuecomment-906140179

For the second case, I told iamayushdas:
```
First of all, you should discuss this in the mail list
dev@apisix.apache.org before sending a bunch of PRs. It seems only a
few people know this purpose.

Secondly, please modify the doc in a repo first, and evaluate the
effect. Don't change a lot of things. Change it step by step.

Thirdly, it is not a good idea to put a visible mark on an obvious
place. I just spend some time on this issue and get an idea:

Since the doc is pulled from other repo before building it on the
website, we can introduce a preprocessor that converts the invisible
mark to the mark that can be recognized by the render.

I am not sure if it works but at least there is way to achieve the
goal with a better solution.
```

See https://github.com/apache/apisix-website/issues/363#issuecomment-867531022

For the third case, I told Xunzhuo privately that he should focus on
the good first issue.

In all three cases, I have told them how to contribute.

I am sure that I am not the only maintainer of APISIX, so where are
the other maintainers?
iamayushdas has worked for APISIX website for a time, haven't other
maintainers led them how to contribute in the correct way?

Where are the other maintainers? Only me be blamed for missing
leading? I have done it actually.

Zhiyuan Ju  于2021年11月25日周四 下午12:32写道：
>
> > I agree with zexuan's view that no meaningful document modification will
> not bring community activity.
>
> From the above 3 PRs, how to define a meaningful change?
>
> > Apache APISIX has such guidelines[1], we can improve this document
>
> For users who subscribed to this mailing list, how many of us read
> carefully or know exactly about this guideline? I agree to read it before
> involving Apache APISIX's Community, and Community maintainers should lead
> us to learn about them as well.
>
> Best Regards!
> @ Zhiyuan Ju 
>
>
> Ming Wen  于2021年11月25日周四 下午12:17写道：
>
> > Apache APISIX has such guidelines[1], we can improve this document
> >
> > [1] https://github.com/apache/apisix/blob/master/CONTRIBUTING.md
> >
> > Thanks,
> > Ming Wen, Apache APISIX PMC Chair
> > Twitter: _WenMing
> >
> >
> > Zhiyuan Ju  于2021年11月25日周四 下午12:11写道：
> >
> > > > If any PRs can be accepted, why requires reviewers?
> > >
> > > In my own opinion, lacking a "Contribution Chain/Workflow" is the key
> > point
> > > why Apache APISIX will receive those PRs. We think they know, actually
> > not.
> > >
> > > Best Regards!
> > > @ Zhiyuan Ju 
> > >
> > >
> > > Zexuan Luo  于2021年11月25日周四 下午12:08写道：
> > >
> > > > I think the contributor should learn the basic manner before
> > > contributing.
> > > > Eg, discuss in the community first, don't spam too many non-value
> > > > changes which don't bring a real contribution.
> > > >
> > > > Zexuan Luo  于2021年11月25日周四 下午12:06写道：
> > > > >
> > > > > > 2.3 In PR #5592, Xunzhuo wanted to translate docs from English to
> > > > Chinese.
> > > > > 2.3.1 From his changes, I also think those changes are not
> > > > MUST-to-CHANGE,
> > > > > but if we try to understand why he wanted to make those changes, in
> > my
> > > > own
> > > > > opinion, I think he just only wanted to contribute, to translate all
> > > > > English to Chinese to let more Chinese developers read more smoothly.
> > > His
> > > > > original intention was good and active :) Though this kind of change
> > > will
> > > > > increase the Reviewer's burden, we have 39 Apache APISIX members,
> > > right?
> > > > I
> > > > > sincerely hope we could take care of our community's activity.
> > > > >
> > > > > It's incorrect. In that PR he just translates English comments into
> > > > > Chinese doc. It is different from translating docs from English to
> > > > > Chinese.
> > > > >
> > > > > They are two different things. Why do need to change the English
> > > > > comments, because it is not pure Chinese? **Again, he didn't discuss
> > > > > this in the community.**
> > > > >
> > > > > And before he did this, he has been submitted several trivial PRs per
> > > > > day several times, which makes the behavior very skeptical.
> > > > >
> > > > > Zexuan Luo  于2021年11月25日周四 下午12:00写道：
> > > > > >
> > > > > > If any PRs can be accepted, why requires reviewers?
> > > > > >
> > > > > > Zexuan Luo  于2021年11月25日周四 上午11:59写道：
> > > > > > >
> > > > > > > The PR #4892 doesn't correct anything. It is retelling the same
> > > > thing in other words. The original statement is already correct.
> > > > > > >
> > > > > > > And if you look at the history carefully, that contributor
> > doesn't
> > > > have a good reputation to submit valua

Re: Here are some feedbacks from our developers

[Ming Wen]

Hi, zexuan,
I think this is a discussion, not blamed, please relax :)
We may have different views on PR, but we should respect each other.

In my opinion, low-quality PRs will indeed increase the workload of
maintainers,
so you can choose different PRs to review, and ignore PRs that you are not
interested in.

Thanks,
Ming Wen, Apache APISIX PMC Chair
Twitter: _WenMing


Zexuan Luo  于2021年11月25日周四 下午7:17写道：

> > Community maintainers should lead us to learn about them as well.
>
> From the above 3 PRs, I have led us to contribute in a good way.
>
> For the first case:
> In https://github.com/apache/apisix/pull/4892, I told the contributor:
>
> > Smart people like you should focus on advanced issues.
>
> See https://github.com/apache/apisix/pull/4892#issuecomment-906140179
>
> For the second case, I told iamayushdas:
> ```
> First of all, you should discuss this in the mail list
> dev@apisix.apache.org before sending a bunch of PRs. It seems only a
> few people know this purpose.
>
> Secondly, please modify the doc in a repo first, and evaluate the
> effect. Don't change a lot of things. Change it step by step.
>
> Thirdly, it is not a good idea to put a visible mark on an obvious
> place. I just spend some time on this issue and get an idea:
>
> Since the doc is pulled from other repo before building it on the
> website, we can introduce a preprocessor that converts the invisible
> mark to the mark that can be recognized by the render.
>
> I am not sure if it works but at least there is way to achieve the
> goal with a better solution.
> ```
>
> See
> https://github.com/apache/apisix-website/issues/363#issuecomment-867531022
>
> For the third case, I told Xunzhuo privately that he should focus on
> the good first issue.
>
> In all three cases, I have told them how to contribute.
>
> I am sure that I am not the only maintainer of APISIX, so where are
> the other maintainers?
> iamayushdas has worked for APISIX website for a time, haven't other
> maintainers led them how to contribute in the correct way?
>
> Where are the other maintainers? Only me be blamed for missing
> leading? I have done it actually.
>
> Zhiyuan Ju  于2021年11月25日周四 下午12:32写道：
> >
> > > I agree with zexuan's view that no meaningful document modification
> will
> > not bring community activity.
> >
> > From the above 3 PRs, how to define a meaningful change?
> >
> > > Apache APISIX has such guidelines[1], we can improve this document
> >
> > For users who subscribed to this mailing list, how many of us read
> > carefully or know exactly about this guideline? I agree to read it before
> > involving Apache APISIX's Community, and Community maintainers should
> lead
> > us to learn about them as well.
> >
> > Best Regards!
> > @ Zhiyuan Ju 
> >
> >
> > Ming Wen  于2021年11月25日周四 下午12:17写道：
> >
> > > Apache APISIX has such guidelines[1], we can improve this document
> > >
> > > [1] https://github.com/apache/apisix/blob/master/CONTRIBUTING.md
> > >
> > > Thanks,
> > > Ming Wen, Apache APISIX PMC Chair
> > > Twitter: _WenMing
> > >
> > >
> > > Zhiyuan Ju  于2021年11月25日周四 下午12:11写道：
> > >
> > > > > If any PRs can be accepted, why requires reviewers?
> > > >
> > > > In my own opinion, lacking a "Contribution Chain/Workflow" is the key
> > > point
> > > > why Apache APISIX will receive those PRs. We think they know,
> actually
> > > not.
> > > >
> > > > Best Regards!
> > > > @ Zhiyuan Ju 
> > > >
> > > >
> > > > Zexuan Luo  于2021年11月25日周四 下午12:08写道：
> > > >
> > > > > I think the contributor should learn the basic manner before
> > > > contributing.
> > > > > Eg, discuss in the community first, don't spam too many non-value
> > > > > changes which don't bring a real contribution.
> > > > >
> > > > > Zexuan Luo  于2021年11月25日周四 下午12:06写道：
> > > > > >
> > > > > > > 2.3 In PR #5592, Xunzhuo wanted to translate docs from English
> to
> > > > > Chinese.
> > > > > > 2.3.1 From his changes, I also think those changes are not
> > > > > MUST-to-CHANGE,
> > > > > > but if we try to understand why he wanted to make those changes,
> in
> > > my
> > > > > own
> > > > > > opinion, I think he just only wanted to contribute, to translate
> all
> > > > > > English to Chinese to let more Chinese developers read more
> smoothly.
> > > > His
> > > > > > original intention was good and active :) Though this kind of
> change
> > > > will
> > > > > > increase the Reviewer's burden, we have 39 Apache APISIX members,
> > > > right?
> > > > > I
> > > > > > sincerely hope we could take care of our community's activity.
> > > > > >
> > > > > > It's incorrect. In that PR he just translates English comments
> into
> > > > > > Chinese doc. It is different from translating docs from English
> to
> > > > > > Chinese.
> > > > > >
> > > > > > They are two different things. Why do need to change the English
> > > > > > comments, because it is not pure Chinese? **Again, he didn't
> discuss
> > > > > > this in the community.**
> > 

Re: Here are some feedbacks from our developers

[Zexuan Luo]

The word from Zhiyuan Ju is misleading. These words will hurt my
reputation if I don't explain myself clear. Unfortunately, I am a
person who sees his honor as very important. If I don't tell my story,
I can not go asleep.

Let me quote his sentences and **correct** them here:

> 2.1.3 This PR gets a "Requested Changes" finally :( I'm not familiar with
OpenTracing and Zipkin for now, and cannot see the difference between them,
so I search Google and here has a "VS" from StackShare[4], it seems that
the PR's change is reasonable. (also could see #4954)

The PR gets a "Requested Changes" because the author emphasizes it's
important to keep the title in the same case, which is trivial. The
author doesn't mention the difference between OpenTracing and Zipkin
at that time.

This is what the author says:

> This pr was intended to uniform the title naming style, maybe it's just a 
> kind of obsessive-compulsive disorder.

See https://github.com/apache/apisix/pull/4892#issuecomment-905362550

And since you have mentioned #4954, you should also mention
https://github.com/apache/apisix/pull/5196.
These two PRs do the same thing, but why #4954 is rejected but #5196 is merged?
Because the author of #5196 explains his idea clearly. Here is how the
author persuaded me:

```
At the same time, the minimum time window of the rate function in the
grafana dashboard provided is 30s.
sum(rate(apisix_bandwidth{instance=~"$instance"}[30s])) by (type)
Based on the above two points, if the scrape_interval in the user's
prometheus global configuration is greater than 15s, the rate function
may not calculate the result at some point in time, causing the
dashboard to be unavailable.

The ingress nginx project has a similar PR: kubernetes/ingress-nginx#2884
```

If you give enough reasons, a stubborn reviewer like me will accept
the same change, which is rejected previously. **It is a bad idea to
only think about the problem of reviewers when a PR is rejected.**

I think everyone can learn a lot from these different results.



> 2.2 In PR #4477, Ayush Das (from India) wanted to use Docusaurus's syntax
to have a better rendering effect on Apache APISIX's Docs website:
> 2.2.1 I agree with the reviewer's comment: this change will break the
reading experience by GitHub markdowns for most developers, and we could do
this after full discussions. But Ayush Das didn't get that point or didn't
notice/know the workflow IMO.

I was quite surprised that Ayush didn't notice the workflow. He has
been working at the APISIX website for a time. Doesn't the maintainer
from there teach him how to discuss in the mail list before submitting
a bunch of PRs?

> 2.2.4 Back to this issue, maybe our committers or PMCs can lead him to know
what's a proper workflow to corporate with the community? It will take much
time to lead our contributors, they grow, we grow, our community grows as
well :)

It is quite unfair to link this comment with a PR to APISIX. Actually,
it is the mistake of the maintainer of APISIX website who doesn't
teach Ayush from the beginning. No, if Ayush is my student I will tell
him to discuss it first in the mail list. Just like what I have told
Bisakh (my prospective student, a smart young man) to discuss the
Datadog plugin.

> 2.3 In PR #5592, Xunzhuo wanted to translate docs from English to Chinese.
2.3.1 From his changes, I also think those changes are not MUST-to-CHANGE,
but if we try to understand why he wanted to make those changes, in my own
opinion, I think he just only wanted to contribute, to translate all
English to Chinese to let more Chinese developers read more smoothly. His
the original intention was good and active

Someone already pointed out that #5592 doesn't translate English doc to Chinese.
In fact, the PR translates the English comments of Chinese doc to
Chinese. Each comment is very short, in one or two sentences. They are
even shorter than the variables in the example sometimes.
So there are two different things.

And most importantly, in all these three cases, I tried to point out
the correct way to do contribution, like discussing first, don't
submit trivial PR unless you have a strong reason, focus on the good
first issue. I have explained them in my previous mail, so I won't
repeat them here.

Therefore, once you know the context of each conflict, you will know
why the "Request Change" is given.  Although I am a strict and
stubborn man, in fact, I don't request change because of irrational
reasons. I tried to follow the rule, gave a detailed review. And most
of all, I review the PR as early as I can, so the developer can get
the feedback in time.

I have worked with Open Source for years, contributed to many famous
projects. Being friendly is good, but being quick and being careful is
more important. An easy-going teacher can make you happy, but only a
teacher who provides timely feedback and detailed review can make you
grow. If we spoil the contributor - accept trivial PR from non-newbie,
it will always ke

Consumer identification from OpenId/Keycloak token

[Joga, singh]

Hi Guys,
I was advised on the discussions group to post my query here to see if somebody 
has already similar in mind.

I have my consumers already defined in Keycloak. Consumers get the token using 
Client Credentials flow i.e. using ClientId and ClientSecret. Using 
‘openid-connect’ plugin I am able successfully authenticate/authorize the 
consumers. However, I would like to implement rate limiting based on consumer 
name, because due to some reason I cannot do it based on consumer ip address.
Therefore, I need a way to identify the consumer from the token. I know that 
APISIX does not support this now. But I think with a small change in the 
‘openid-connect’ plugin, it should be possible to do, because the response from 
‘introspection_endpoint’ already contains the attributes (e.g. username, 
clientId etc.) needed to identify the consumer.

I think ‘ladp_auth’ plugin already does this.

What do you think? Is it possible to do? Are there any challenges in it?
If yes, I would like to try my hands on this. Therefore, would need getting 
started tips/hints for building, compiling and other related processes.

Best Regards,

Joga


Hyundai AutoEver Europe GmbH
Kaiserleistr. 8A, 63067 Offenbach a.M., Deutschland
Geschäftsführer/CEO: Jong-Il Yun   Registergericht/registration court: 
Amtsgericht Offenbach   Registernummer/registration number: HRB 42684   
USt-IdNr./tax ID-no.: DE252841722

[Proposal] AWS lambda serverless plugin into Apache APISIX

[Bisakh Mondal]

Hi folks,

This is a proposal to integrate AWS Lambda serverless support into Apache
APISIX. We just have integrated the azure functions faas plugin into the
ecosystem and the addition of lambda from aws will extend the use case of
apisix in diverse scenarios.

Proposed plugin name: `aws-lambda`

The plugin takes the AWS API Gateway URI as a dynamic upstream and any
request to the particular apisix gateway URI where the plugin is enabled
gets proxied to the lambda function through the AWS gateway.

Before discussing any plugin-specific information, I want to talk about the
refactoring on #5616 [1]. If we notice closely, all the serverless plugins
from different vendors do one thing in common - terminate the incoming
request and initiate another request to the modified upstream and return
the response body with the headers. Also except for the authorization
information, the attributes of the plugin schema are exactly similar.
That's why #5616 abstracts away the generic code into the
`generic-upstream.lua` module. So any new plugin can make use of this
package module -

How developers should write new (faas) plugins:

1. They should define plugin-specific authorization schema & metadata
schema(if required)
2. Write a request_processor function on the plugin lua file that takes
`plugin conf, ngx context and the URL params (that contains headers, body,
urlpath, query params etc. etc.)". The function can perform a series of
operations including updating/adding the necessary authorization headers,
performing encoding/decoding (base64 or anything) on the body, request
signing etc.
3. And yes, that's it. (see the azure-functions and aws-lambda Lua code for
example)

-- Now again coming back to the aws-lambda plugin, for the initial version
we are going to support authorization via apikey and AWS IAM request
signing.
The authz schema looks like

```
{
type = "object",
properties = {
-- API Key based authorization
apikey = {type = "string"},
-- IAM role based authorization, works via aws v4 request signing
-- more at
https://docs.aws.amazon.com/general/latest/gr/sigv4_signing.html
iam = {
type = "object",
properties = {
accesskey = {
type = "string",
description = "access key id from from aws iam console"
},
secretkey = {
type = "string",
description = "secret access key from from aws iam
console"
},
aws_region = {
type = "string",
default = "us-east-1",
description = "the aws region that is receiving the
request"
},
service = {
type = "string",
default = "execute-api",
description = "the service that is receiving the
request"
}
},
required = {"accesskey", "secretkey"}
}
}
}
```
I have opened a PR [2] implementing the same including the support for
authorization via request signing with AWS signature v4. So end-users can
choose either one of the two ways to authorize invocations into AWS lambda.
No worries, next week I am going to write a blog discussing the same. Stay
tuned.


Thank you! Feel free to share your views, suggestions or anything.

Best regards,
Bisakh 

 [1]: https://github.com/apache/apisix/pull/5616
 [2]: https://github.com/apache/apisix/pull/5594

Re: Consumer identification from OpenId/Keycloak token

[ZhengSong Tu]

1. Did you mean to put the consumer_name to the ctx in open-connect plugin?
2. Then we can limit by consumer_name in limit-* plugins.

*ZhengSong Tu*
My GitHub: https://github.com/tzssangglass
Apache APISIX: https://github.com/apache/apisix

Joga, singh  于2021年11月26日周五 上午12:26写道：
>
> Hi Guys,
> I was advised on the discussions group to post my query here to see if 
> somebody has already similar in mind.
>
> I have my consumers already defined in Keycloak. Consumers get the token 
> using Client Credentials flow i.e. using ClientId and ClientSecret. Using 
> ‘openid-connect’ plugin I am able successfully authenticate/authorize the 
> consumers. However, I would like to implement rate limiting based on consumer 
> name, because due to some reason I cannot do it based on consumer ip address.
> Therefore, I need a way to identify the consumer from the token. I know that 
> APISIX does not support this now. But I think with a small change in the 
> ‘openid-connect’ plugin, it should be possible to do, because the response 
> from ‘introspection_endpoint’ already contains the attributes (e.g. username, 
> clientId etc.) needed to identify the consumer.
>
> I think ‘ladp_auth’ plugin already does this.
>
> What do you think? Is it possible to do? Are there any challenges in it?
> If yes, I would like to try my hands on this. Therefore, would need getting 
> started tips/hints for building, compiling and other related processes.
>
> Best Regards,
>
> Joga
>
>
> Hyundai AutoEver Europe GmbH
> Kaiserleistr. 8A, 63067 Offenbach a.M., Deutschland
> Geschäftsführer/CEO: Jong-Il Yun   Registergericht/registration court: 
> Amtsgericht Offenbach   Registernummer/registration number: HRB 42684   
> USt-IdNr./tax ID-no.: DE252841722

[Proposal] Add blog contribution guidelines to apisix-website repository

[Yilin Zeng]

Hi community,


Currently apisix-website repository[1] does not have a set of blog contribution 
guidelines. It could block potential contributors who want to write blogs but 
cannot find the proper way to do it. As I was reaching out to other open-source 
communities in the last two weeks, I went through some of theirs website and 
blog repositories, and found some valuable guidelines that could be implemented 
in apisix-website repository[1], such as TiDB Documentation Contributing 
Guide[2].


I would like to propose adding contribution guidelines to apisix-website 
repository[1], which will cover the following aspects:


1 What can you contribute
1.1 Add a new blog
1.2 Fix typos or formats

2 Contribute style guides
2.1 Commit message style
2.1 Pull request title style

3 How to contribute properly via git command line


Feel free to share your views, suggestions, or anything else. Thank you!


Regards,

Yilin



[1]https://github.com/apache/apisix-website

[2]https://github.com/pingcap/docs/blob/master/CONTRIBUTING.md

Re: [Proposal] Add blog contribution guidelines to apisix-website repository

[Sylvia]

Hi Yilin,

I think this proposal is very good, and it is more convenient to help
contributors who want to share content instantly.

Look forward to the guideline!

Best Regards,
Sylvia

Yilin Zeng  于2021年11月26日周五 下午2:20写道：

> Hi community,
>
>
> Currently apisix-website repository[1] does not have a set of blog
> contribution guidelines. It could block potential contributors who want to
> write blogs but cannot find the proper way to do it. As I was reaching out
> to other open-source communities in the last two weeks, I went through some
> of theirs website and blog repositories, and found some valuable guidelines
> that could be implemented in apisix-website repository[1], such as TiDB
> Documentation Contributing Guide[2].
>
>
> I would like to propose adding contribution guidelines to apisix-website
> repository[1], which will cover the following aspects:
>
>
> 1 What can you contribute
> 1.1 Add a new blog
> 1.2 Fix typos or formats
>
> 2 Contribute style guides
> 2.1 Commit message style
> 2.1 Pull request title style
>
> 3 How to contribute properly via git command line
>
>
> Feel free to share your views, suggestions, or anything else. Thank you!
>
>
> Regards,
>
> Yilin
>
>
>
> [1]https://github.com/apache/apisix-website
>
> [2]https://github.com/pingcap/docs/blob/master/CONTRIBUTING.md
>
>

Re: [Proposal] Add blog contribution guidelines to apisix-website repository

[jing li]

Nice proposal +1~
There are indeed many people who have written  blogs related to APISIX ,
but don’t know how to submit to our website.

On Fri, Nov 26, 2021 at 2:27 PM Sylvia  wrote:

> Hi Yilin,
>
> I think this proposal is very good, and it is more convenient to help
> contributors who want to share content instantly.
>
> Look forward to the guideline!
>
> Best Regards,
> Sylvia
>
> Yilin Zeng  于2021年11月26日周五 下午2:20写道：
>
> > Hi community,
> >
> >
> > Currently apisix-website repository[1] does not have a set of blog
> > contribution guidelines. It could block potential contributors who want
> to
> > write blogs but cannot find the proper way to do it. As I was reaching
> out
> > to other open-source communities in the last two weeks, I went through
> some
> > of theirs website and blog repositories, and found some valuable
> guidelines
> > that could be implemented in apisix-website repository[1], such as TiDB
> > Documentation Contributing Guide[2].
> >
> >
> > I would like to propose adding contribution guidelines to apisix-website
> > repository[1], which will cover the following aspects:
> >
> >
> > 1 What can you contribute
> > 1.1 Add a new blog
> > 1.2 Fix typos or formats
> >
> > 2 Contribute style guides
> > 2.1 Commit message style
> > 2.1 Pull request title style
> >
> > 3 How to contribute properly via git command line
> >
> >
> > Feel free to share your views, suggestions, or anything else. Thank you!
> >
> >
> > Regards,
> >
> > Yilin
> >
> >
> >
> > [1]https://github.com/apache/apisix-website
> >
> > [2]https://github.com/pingcap/docs/blob/master/CONTRIBUTING.md
> >
> >
>

Re: [Proposal] Add blog contribution guidelines to apisix-website repository

[Serendipity 96]

Hi Yilin,

Great idea!  I also think we should add contribution guidelines, it will
help new partners.

What do you think to add a blog template? just like:

1 What can you contribute
1.1 the way to add a new blog
1.2 Fix typos or formats
1.3 blog template


Looking forward++!

Best Regards,
@Serendipity96

Sylvia  于2021年11月26日周五 14:27写道：

> Hi Yilin,
>
> I think this proposal is very good, and it is more convenient to help
> contributors who want to share content instantly.
>
> Look forward to the guideline!
>
> Best Regards,
> Sylvia
>
> Yilin Zeng  于2021年11月26日周五 下午2:20写道：
>
> > Hi community,
> >
> >
> > Currently apisix-website repository[1] does not have a set of blog
> > contribution guidelines. It could block potential contributors who want
> to
> > write blogs but cannot find the proper way to do it. As I was reaching
> out
> > to other open-source communities in the last two weeks, I went through
> some
> > of theirs website and blog repositories, and found some valuable
> guidelines
> > that could be implemented in apisix-website repository[1], such as TiDB
> > Documentation Contributing Guide[2].
> >
> >
> > I would like to propose adding contribution guidelines to apisix-website
> > repository[1], which will cover the following aspects:
> >
> >
> > 1 What can you contribute
> > 1.1 Add a new blog
> > 1.2 Fix typos or formats
> >
> > 2 Contribute style guides
> > 2.1 Commit message style
> > 2.1 Pull request title style
> >
> > 3 How to contribute properly via git command line
> >
> >
> > Feel free to share your views, suggestions, or anything else. Thank you!
> >
> >
> > Regards,
> >
> > Yilin
> >
> >
> >
> > [1]https://github.com/apache/apisix-website
> >
> > [2]https://github.com/pingcap/docs/blob/master/CONTRIBUTING.md
> >
> >
>

Re: Release Apache APISIX 2.11.0 Round 2

[JinChao Shuai]

+1 (binding) I checked： 1. Download links are OK. 2. Checksum and signature
are OK. 3. LICENSE and NOTICE have existed. 4. Building on Ubuntu 20.04 was
successful.

Zeping Bai  于2021年11月25日周四 下午5:44写道：

> +1 (non-binding)
>
> I checked：
> 1. Download links are OK.
> 2. Checksum and signature are OK.
> 3. LICENSE and NOTICE files are exist.
> 4. Building on Gitpod (Ubuntu 20.04) was successful.
>
>
> Zeping Bai  @bzp2010
>
> 孙毅  于2021年11月23日周二 下午8:48写道：
>
> > Hello, Community,
> > This is a call for the vote to release Apache APISIX version 2.11.0
> >
> > Release notes:
> >
> > https://github.com/apache/apisix/blob/release/2.11/CHANGELOG.md#2110
> >
> > The release candidates:
> >
> > https://dist.apache.org/repos/dist/dev/apisix/2.11.0/
> >
> > Release Commit ID:
> >
> >
> >
> https://github.com/apache/apisix/commit/d39814f3b5291393effe9c258bcc4a7f77911334
> >
> > Keys to verify the Release Candidate:
> >
> > https://dist.apache.org/repos/dist/dev/apisix/KEYS
> >
> > Steps to validating the release:
> >
> > 1. Download the release
> >
> > wget
> >
> >
> https://dist.apache.org/repos/dist/dev/apisix/2.11.0/apache-apisix-2.11.0-src.tgz
> >
> > 2. Checksums and signatures
> >
> > wget https://dist.apache.org/repos/dist/dev/apisix/KEYS
> >
> > wget
> >
> >
> https://dist.apache.org/repos/dist/dev/apisix/2.11.0/apache-apisix-2.11.0-src.tgz.asc
> >
> > wget
> >
> >
> https://dist.apache.org/repos/dist/dev/apisix/2.11.0/apache-apisix-2.11.0-src.tgz.sha512
> >
> > gpg --import KEYS
> >
> > shasum -c apache-apisix-2.11.0-src.tgz.sha512
> >
> > gpg --verify apache-apisix-2.11.0-src.tgz.asc
> apache-apisix-2.11.0-src.tgz
> >
> > 3. Unzip and Check files
> >
> > tar zxvf apache-apisix-2.11.0-src.tgz
> >
> > 4. Build Apache APISIX:
> >
> >
> >
> https://github.com/apache/apisix/blob/release/2.11/docs/en/latest/how-to-build.md#installation-via-source-release-package
> >
> > The vote will be open for at least 72 hours or until necessary number of
> > votes are reached.
> >
> > Please vote accordingly:
> >
> > [ ] +1 approve
> > [ ] +0 no opinion
> > [ ] -1 disapprove with the reason
> >
>


-- 
Thanks,
Janko

Re: Here are some feedbacks from our developers

[Serendipity 96]

> In my own opinion, lacking a "Contribution Chain/Workflow" is the key
point
why Apache APISIX will receive those PRs. We think they know, actually not.

I agree with this view. I didn't know what to do before I contributed code
for the first time. I consulted my friend who had contributed to open
source and he told me to start a discussion, show my technical solution,
discuss in the community, and then start writing the code.

Maybe we should write a contribution guide to help others contribute.

Best Regards!
@Serendipity96


Zexuan Luo  于2021年11月25日周四 22:58写道：

> The word from Zhiyuan Ju is misleading. These words will hurt my
> reputation if I don't explain myself clear. Unfortunately, I am a
> person who sees his honor as very important. If I don't tell my story,
> I can not go asleep.
>
> Let me quote his sentences and **correct** them here:
>
> > 2.1.3 This PR gets a "Requested Changes" finally :( I'm not familiar with
> OpenTracing and Zipkin for now, and cannot see the difference between them,
> so I search Google and here has a "VS" from StackShare[4], it seems that
> the PR's change is reasonable. (also could see #4954)
>
> The PR gets a "Requested Changes" because the author emphasizes it's
> important to keep the title in the same case, which is trivial. The
> author doesn't mention the difference between OpenTracing and Zipkin
> at that time.
>
> This is what the author says:
>
> > This pr was intended to uniform the title naming style, maybe it's just
> a kind of obsessive-compulsive disorder.
>
> See https://github.com/apache/apisix/pull/4892#issuecomment-905362550
>
> And since you have mentioned #4954, you should also mention
> https://github.com/apache/apisix/pull/5196.
> These two PRs do the same thing, but why #4954 is rejected but #5196 is
> merged?
> Because the author of #5196 explains his idea clearly. Here is how the
> author persuaded me:
>
> ```
> At the same time, the minimum time window of the rate function in the
> grafana dashboard provided is 30s.
> sum(rate(apisix_bandwidth{instance=~"$instance"}[30s])) by (type)
> Based on the above two points, if the scrape_interval in the user's
> prometheus global configuration is greater than 15s, the rate function
> may not calculate the result at some point in time, causing the
> dashboard to be unavailable.
>
> The ingress nginx project has a similar PR: kubernetes/ingress-nginx#2884
> ```
>
> If you give enough reasons, a stubborn reviewer like me will accept
> the same change, which is rejected previously. **It is a bad idea to
> only think about the problem of reviewers when a PR is rejected.**
>
> I think everyone can learn a lot from these different results.
>
>
>
> > 2.2 In PR #4477, Ayush Das (from India) wanted to use Docusaurus's syntax
> to have a better rendering effect on Apache APISIX's Docs website:
> > 2.2.1 I agree with the reviewer's comment: this change will break the
> reading experience by GitHub markdowns for most developers, and we could do
> this after full discussions. But Ayush Das didn't get that point or didn't
> notice/know the workflow IMO.
>
> I was quite surprised that Ayush didn't notice the workflow. He has
> been working at the APISIX website for a time. Doesn't the maintainer
> from there teach him how to discuss in the mail list before submitting
> a bunch of PRs?
>
> > 2.2.4 Back to this issue, maybe our committers or PMCs can lead him to
> know
> what's a proper workflow to corporate with the community? It will take much
> time to lead our contributors, they grow, we grow, our community grows as
> well :)
>
> It is quite unfair to link this comment with a PR to APISIX. Actually,
> it is the mistake of the maintainer of APISIX website who doesn't
> teach Ayush from the beginning. No, if Ayush is my student I will tell
> him to discuss it first in the mail list. Just like what I have told
> Bisakh (my prospective student, a smart young man) to discuss the
> Datadog plugin.
>
> > 2.3 In PR #5592, Xunzhuo wanted to translate docs from English to
> Chinese.
> 2.3.1 From his changes, I also think those changes are not MUST-to-CHANGE,
> but if we try to understand why he wanted to make those changes, in my own
> opinion, I think he just only wanted to contribute, to translate all
> English to Chinese to let more Chinese developers read more smoothly. His
> the original intention was good and active
>
> Someone already pointed out that #5592 doesn't translate English doc to
> Chinese.
> In fact, the PR translates the English comments of Chinese doc to
> Chinese. Each comment is very short, in one or two sentences. They are
> even shorter than the variables in the example sometimes.
> So there are two different things.
>
> And most importantly, in all these three cases, I tried to point out
> the correct way to do contribution, like discussing first, don't
> submit trivial PR unless you have a strong reason, focus on the good
> first issue. I have explained them in my 

Re: [ANNOUNCE] New committer Shivam Singh

[Serendipity 96]

Congratulations!

Best Regards
@Serendipity96

Sylvia  于2021年11月25日周四 09:46写道：

> Hi Shivam,
>
> I'm Sylvia. After our discussion, I think you should go to the official
> website(https://github.com/apache/apisix-website)  and submit PR for this
> article(
> https://gist.github.com/1502shivam-singh/e63089c19ae5cc29aef96e7ae7069373)
> yourself, then you can add me as a reviewer. Just put the English version
> (if needed Chinese version I will translate it afterward).
>
> If you don't know the *path of publishing an article* on the Apache
> APISIX website, you can refer to this one:
> https://github.com/apache/apisix-website/pull/760.
>
> If you have any questions about it, please let me know or talk to me on
> Slack.
>
> Best Regards,
> Sylvia
>
> - https://github.com/SylviaBABY
>
>
> yeliang wang  于2021年11月24日周三 上午9:45写道：
>
> > Hi  Shivam,
> >
> > You can write the article once and submit PR directly on the blog of
> Apache
> > APISIX Website project.
> >
> > Shivam Singh  于2021年11月23日周二 下午8:37写道：
> >
> > > Hi Yeliang,
> > >
> > > Sure, I would be happy to share my journey here.
> > >
> > > Shivam Singh
> > > 
> > > 
> > >
> > > On Tue, Nov 23, 2021 at 11:08 AM yeliang wang <
> > > wangyeliang19820...@gmail.com> wrote:
> > >
> > >> Hi, Shivam
> > >>
> > >> My name is Yeliang Wang, From my previous experience, the experience
> of
> > >> growing into a committer is very valuable.
> > >> Are you interested in writing down this experience and sharing it with
> > >> more people?
> > >>
> > >> Look forward to your reply.😁
> > >>
> > >> Shivam Singh  于2021年11月22日周一 下午6:29写道：
> > >>
> > >>> Thank you team !
> > >>> Best regards,
> > >>>
> > >>> Shivam Singh
> > >>> 
> > >>> 
> > >>>
> > >>> On Mon, Nov 22, 2021 at 12:30 PM YuanSheng Wang  >
> > >>> wrote:
> > >>>
> > >>> > Hi, community,
> > >>> >
> > >>> > The Project Management Committee(PMC) of Apache APISIX has voted
> > >>> > and invited Shivam Singh to be a committer. We are pleased to
> > announce
> > >>> > that he has accepted.
> > >>> >
> > >>> >
> > >>> > --
> > >>> >
> > >>> > *MembPhis*
> > >>> > My GitHub: https://github.com/membphis
> > >>> > Apache APISIX: https://github.com/apache/apisix
> > >>> >
> > >>>
> > >>
> >
>

Re: [ANNOUNCE] New committer Shivam Singh

[Bozhong Yu]

Congratulations!

Serendipity 96  于2021年11月26日周五 下午3:22写道：

> Congratulations!
>
> Best Regards
> @Serendipity96
>
> Sylvia  于2021年11月25日周四 09:46写道：
>
> > Hi Shivam,
> >
> > I'm Sylvia. After our discussion, I think you should go to the official
> > website(https://github.com/apache/apisix-website)  and submit PR for
> this
> > article(
> >
> https://gist.github.com/1502shivam-singh/e63089c19ae5cc29aef96e7ae7069373)
> > yourself, then you can add me as a reviewer. Just put the English version
> > (if needed Chinese version I will translate it afterward).
> >
> > If you don't know the *path of publishing an article* on the Apache
> > APISIX website, you can refer to this one:
> > https://github.com/apache/apisix-website/pull/760.
> >
> > If you have any questions about it, please let me know or talk to me on
> > Slack.
> >
> > Best Regards,
> > Sylvia
> >
> > - https://github.com/SylviaBABY
> >
> >
> > yeliang wang  于2021年11月24日周三 上午9:45写道：
> >
> > > Hi  Shivam,
> > >
> > > You can write the article once and submit PR directly on the blog of
> > Apache
> > > APISIX Website project.
> > >
> > > Shivam Singh  于2021年11月23日周二 下午8:37写道：
> > >
> > > > Hi Yeliang,
> > > >
> > > > Sure, I would be happy to share my journey here.
> > > >
> > > > Shivam Singh
> > > > 
> > > > 
> > > >
> > > > On Tue, Nov 23, 2021 at 11:08 AM yeliang wang <
> > > > wangyeliang19820...@gmail.com> wrote:
> > > >
> > > >> Hi, Shivam
> > > >>
> > > >> My name is Yeliang Wang, From my previous experience, the experience
> > of
> > > >> growing into a committer is very valuable.
> > > >> Are you interested in writing down this experience and sharing it
> with
> > > >> more people?
> > > >>
> > > >> Look forward to your reply.😁
> > > >>
> > > >> Shivam Singh  于2021年11月22日周一 下午6:29写道：
> > > >>
> > > >>> Thank you team !
> > > >>> Best regards,
> > > >>>
> > > >>> Shivam Singh
> > > >>> 
> > > >>> 
> > > >>>
> > > >>> On Mon, Nov 22, 2021 at 12:30 PM YuanSheng Wang <
> membp...@apache.org
> > >
> > > >>> wrote:
> > > >>>
> > > >>> > Hi, community,
> > > >>> >
> > > >>> > The Project Management Committee(PMC) of Apache APISIX has voted
> > > >>> > and invited Shivam Singh to be a committer. We are pleased to
> > > announce
> > > >>> > that he has accepted.
> > > >>> >
> > > >>> >
> > > >>> > --
> > > >>> >
> > > >>> > *MembPhis*
> > > >>> > My GitHub: https://github.com/membphis
> > > >>> > Apache APISIX: https://github.com/apache/apisix
> > > >>> >
> > > >>>
> > > >>
> > >
> >
>