[SIG-Network] Ingress NGINX meeting 20210511

Summary

The transcript captures a meeting of the Ingress Nginx subproject, focusing on addressing and cleaning up open issues, including stale and bug reports. New members introduce themselves, expressing their eagerness to contribute to the Kubernetes community. Discussions revolve around prioritizing issues, understanding the impact of breaking changes, and the need for better versioning and support strategies. The meeting also touches on the importance of adhering to the code of conduct and the collaborative effort to improve the project.

Takeaways

• 🗓️ The meeting took place on May 11th, 2021, focusing on the Ingress Nginx subproject within the Kubernetes community.
• 👥 New members introduced themselves, including those with experience in Kubernetes release engineering and others looking to contribute to the project.
• 📝 The meeting aimed to address stale issues and encourage movement on open issues to clean up the project's backlog.
• 👍 The importance of adhering to the code of conduct, which emphasizes being excellent with each other, was highlighted.
• 🔍 The discussion included reviewing and categorizing issues from the GitHub API to understand the scale of work ahead.
• 📉 There was an emphasis on investigating and potentially deprecating older versions of Nginx that are no longer supported.
• 📝 The process of labeling and managing issues, such as applying 'help wanted' labels, was discussed to encourage community involvement.
• 🔄 The team considered the implications of making breaking changes, such as changing annotations, and the need for clear communication in the changelog.
• 🤔 The challenge of supporting older issues with potentially outdated versions of software was acknowledged, and strategies for addressing these were explored.
• 📋 The need for better issue triage, including requesting more information or testing, was identified to move issues forward.
• 🚀 The meeting concluded with a commitment to continue working through issues and improving the project's development cycle.

Q & A

What was the main purpose of the meeting on May 11th, 2021?

The main purpose of the meeting was to discuss and clean up open issues in the NGINX Ingress subproject, particularly focusing on stale issues, and to encourage new members to introduce themselves and contribute to the project.

What is the expected behavior during the Kubernetes community meeting according to the code of conduct?

Participants are expected to be excellent with each other, adhering to the code of conduct, and any issues should be reported privately to the SIG chairs or meeting organizers.

What is the role of the person who introduced themselves as the 'release note shadow' for Kubernetes 1.22?

The person is responsible for shadowing or assisting with the release notes for the Kubernetes 1.22 version, contributing to the documentation process and ensuring that release information is accurate and complete.

Why did Carlos join the NGINX Ingress subproject meeting?

Carlos joined the meeting because he works on the secret release as a branch manager and also has involvement in Cluster API and Engine X Ingress. He has a space in his heart for the project and wants to contribute and help the community.

What is the primary concern of the user from the startup IoT company mentioned in the script?

The user is looking for options to replace Apache for web services and load balancing, as they are currently using Apache but are interested in exploring alternatives that might better suit their needs.

What is the process for handling issues that are considered stale or rotten in the NGINX Ingress project?

The process involves reviewing the stale or rotten issues to determine if they are still relevant or helpful. If they are not, the tag is removed, and the issues are either updated, closed, or moved forward based on their current relevance and the project's needs.

What is the significance of the 'good first issue' label in the context of the NGINX Ingress project?

The 'good first issue' label is used to identify issues that are suitable for new contributors to work on, providing them with an opportunity to get acquainted with the project and make their first contributions.

Why is the versioning and support of Kubernetes important in the context of the issues discussed in the meeting?

Versioning and support are important because many of the issues discussed are related to older versions of Kubernetes. Understanding the support status helps in determining whether issues are still relevant or if they have been resolved in newer versions.

What is the general approach to addressing issues that require testing with newer versions of Kubernetes?

The approach involves asking the issue submitters or the community to test the issues on newer versions of Kubernetes. This helps in validating whether the issues persist in the current or recent versions of the software.

What is the role of the steering committee in the context of the NGINX Ingress project?

The steering committee may be involved in providing resources such as funding or access to cloud services for testing and replicating issues, especially when such tasks require significant resources or infrastructure.

How does the NGINX Ingress project handle breaking changes in annotations or other features?

The project follows semantic versioning, and breaking changes typically involve revising the version number. It's important to communicate these changes clearly in the changelog to ensure users are aware of the impact and necessary actions, such as redeployment.

Outlines

00:00

🗓️ Welcome and Introductions at the NGINX Ingress Subproject Meeting

The script begins with a welcome to the NGINX Ingress subproject meeting on May 11th, 2021, which is part of the Kubernetes community meeting. The speaker emphasizes the importance of adhering to the code of conduct and encourages participants to report any issues privately to the SIG chairs. The main agenda includes discussing open topics, addressing stale issues, and potentially cleaning up the issue backlog. New members are invited to introduce themselves, with several individuals sharing their backgrounds and reasons for joining the meeting, ranging from contributing to Kubernetes release engineering to seeking options for web services and load balancing.

05:00

📝 Reviewing and Organizing Open Issues in the NGINX Ingress Project

The second paragraph focuses on the process of reviewing and organizing open issues within the NGINX Ingress project. The speaker mentions the use of an Excel spreadsheet to categorize and understand the scale of the issues. There is a discussion about stale or rotten tickets, and the need to determine their relevance and whether they should be deprecated. The speaker also addresses the challenge of assigning issues due to limited participants and the importance of walking through older issues to determine if they are still valid or need to be updated.

10:04

🛠️ Discussing Technical Debt and Versioning Challenges

In this paragraph, the discussion revolves around technical debt, specifically addressing older issues and determining whether they are still relevant or need to be deprecated. There is a mention of a potential breaking change due to a shift in annotation terminology from whitelist/blacklist to allowlisting, which raises concerns about versioning and the impact on users. The conversation also touches on the need for better processes for pushing out changes and the importance of testing with newer versions of Kubernetes.

15:06

🐛 Identifying and Prioritizing Bug Fixes in the NGINX Ingress Subproject

The fourth paragraph delves into the identification and prioritization of bug fixes within the NGINX Ingress subproject. The speaker highlights the need to test and validate issues, especially those related to older versions of the software. There is a call for community involvement in testing and a discussion about the process of closing issues if they are no longer relevant or if the submitter does not respond to inquiries about their relevance in newer versions.

20:08

🔍 Investigating and Addressing Long-standing Issues in the Project

This paragraph discusses the investigation and addressing of long-standing issues within the project. The speaker notes the difficulty in testing certain issues due to the need for specific environments and the reliance on submitter responses. There is also a mention of the process for requesting funding for replicating and debugging issues, as well as the need for better understanding and support for these processes within the project.

25:14

📝 Triage and Documentation: Navigating Project Issues and Features

The sixth paragraph centers on the triage process for issues and features within the project. The speaker discusses the need for more information on certain issues and the process for labeling them accordingly. There is a focus on determining whether certain behaviors are expected or not, and the importance of gathering more context to decide if they are supported features or if they need to be addressed as bugs.

30:19

🕵️‍♂️ Deep Dive into Specific Issues and Their Relevance to Current Versions

In this paragraph, the conversation takes a deeper dive into specific issues, examining their relevance to current and older versions of the software. The speaker discusses the importance of understanding whether issues are still present in newer versions and the need for community members to test these versions. There is also a mention of design questions related to expected behavior and the need for further investigation into certain features.

35:19

🏁 Wrapping Up the Meeting and Scheduling Future Work

The final paragraph wraps up the meeting with a summary of the issues discussed and an acknowledgment of the work that lies ahead. The speaker mentions scheduling time to continue working through the issues and encourages the group to keep moving things along. There is a note of thanks to the host and participants, and a reminder that meeting details and follow-up actions will be posted for reference.

Keywords

💡Nginx

Nginx is an open-source web server software that is widely used for serving static content, handling reverse proxying, and load balancing. In the context of the video, it is discussed as a subproject of Ingress, which is a Kubernetes API object that manages external access to the services in a cluster. The script mentions 'nginx ingress' and 'nginx controller', indicating that the conversation is centered around the use and development of Nginx within the Kubernetes ecosystem.

💡Kubernetes

Kubernetes is an open-source container-orchestration system for automating application deployment, scaling, and management. It is foundational to the video's theme as the meeting is a 'Kubernetes community meeting', and the participants discuss issues and features related to Kubernetes, such as 'cluster API' and 'Kubernetes release engineering'.

💡SIG (Special Interest Group)

In the Kubernetes community, a SIG is a group of contributors who collaborate on a specific area of the project. The term is used in the script to refer to the group of people who are responsible for the Nginx subproject within the Kubernetes ecosystem, as indicated by 'sig netwicking' and 'sig chairs'.

💡Code of Conduct

A Code of Conduct is a set of guidelines that outline expectations for behavior within a community or project. In the script, it is mentioned as a rule that participants in the meeting are expected to follow, emphasizing the importance of a respectful and inclusive environment for collaboration.

💡Release Engineering

Release Engineering refers to the processes and practices involved in planning, scheduling, and managing software releases. The script mentions 'Kubernetes release engineering' and a participant who contributes to this area, indicating the importance of release management in the development lifecycle of Kubernetes.

💡Release Note Shadow

A Release Note Shadow is a person who reviews and provides feedback on release notes for a software project. In the script, a new member introduces themselves as the 'release note shadow for 1.22', indicating their role in the review process for a specific Kubernetes release.

💡IoT (Internet of Things)

IoT refers to the interconnection via the internet of computing devices embedded in everyday objects, enabling them to send and receive data. In the script, a participant mentions being from an IoT company, suggesting that their interest in the Nginx subproject is related to the needs of IoT applications.

💡Breaking Change

A breaking change in software development is a modification that makes the existing functionality behave differently in a way that is not backward compatible. In the script, there is a discussion about changing annotations from 'whitelist' to 'allowlist', which is considered a breaking change due to the potential impact on existing configurations.

💡Semantic Versioning

Semantic Versioning is a versioning scheme for software that uses a three-part number format (MAJOR.MINOR.PATCH) to convey the nature of changes made in each new release. The script mentions 'semantic versioning' in the context of deciding how to version a release that includes a breaking change.

💡Pull Request

A Pull Request is a proposal made by one contributor to a project to incorporate their changes into the project's repository. In the script, there is mention of a 'pull request' for mapping out the architecture, which is part of the collaborative process in open-source software development.

💡GitHub API

The GitHub API is a set of endpoints that allow developers to interact programmatically with GitHub's data. In the script, it is mentioned that issues were pulled from the GitHub API to understand the scope of work, indicating the use of automation in managing project issues.

💡Triage

In the context of software development, triage refers to the process of prioritizing and categorizing issues or bugs based on their severity and other factors. The script includes discussions about triaging issues, such as labeling them as 'needs more information' or considering them for closing based on their age and relevance.

Highlights

Introduction of the meeting and the importance of adhering to the code of conduct for a respectful and productive environment.

Welcome of new members and their introductions, bringing fresh perspectives and contributions to the project.

Discussion on the role of the release note shadow for Kubernetes 1.22 and the contribution to the community.

Carlos's introduction as a branch manager in the secret release and involvement in cluster API and Engine X Ingress.

Ajith's interest in exploring options for Apache and web services, and his role as a consumer rather than a developer in this area.

Christoph's introduction from Budapest, Hungary, and his company's heavy use of NGINX Ingress.

Jing Tao's desire to join the community and contribute to the project after previous involvement in a darker and kind project.

The introduction of a participant from the Apache PS6 community, interested in contributing to the project based on shared technology.

The process of addressing and cleaning up stale and rotten tickets to streamline the issue management.

The importance of understanding and categorizing issues to better manage the project's workload.

Investigation of deprecation versions versus supported NGINX versions to guide future development and support strategies.

The challenge of assigning issues due to limited resources and the need for community involvement.

Discussion on the use of labels such as 'help wanted' to encourage community members to contribute to specific issues.

The consideration of semantic versioning and the implications of making breaking changes in the project.

The need for better cycle management of pushing out changes and updates to the project.

The exploration of the process for requesting funding and resources for testing and replicating issues.

The importance of understanding whether certain behaviors are expected or bugs that need fixing.

The strategy for addressing issues related to older versions of the software and their relevance to current development.

The conclusion of the meeting with a summary of the next steps and the importance of continued progress on the identified issues.

Transcripts

hello everyone this is sig netwicking

ingress nginx sub project meeting today

is may 11th 2021 this is a kubernetes

community meeting

and we ask that you comply with the code

of conduct which is basically be

excellent with each other

if you have any issues please report it

to the sig chairs or anyone running this

meeting privately

thank you so

today i wanted to get started on the

conversation just moving through

i have the um i have some open topics

from the i want to look at some of the

open issues that are stale and then move

forward

the idea here is that i want to try to

start cleaning up some of the open

issues that we have and just

going through and just trying to get

movement through

the open issues we'll go ahead and start

with uh is there any

new members who want to introduce

themselves to the team

oh yeah yeah i'm new so this is my first

nginx uh

meeting um so i i contribute around

the kubernetes release engineering

release team meeting and um

i'm the release note shadow for 1.22

and i i seen this you know this this

uh email that the engineering group

needs some help so i'm here to maybe

uh contribute to some issues and help

the community out in general so

yeah excited to be here awesome thank

you for being here

i'm in you u2 as well but i did some

small contributions in the past for

nginx i think

was my first project that i did

something

at the kubernetes a long time ago

hi my name is carlos i work on the

secret release as a branch manager and

also in cluster api and

engine x ingress have some space in my

heart forever

and then i don't like to get this

project abundant

that's why i'm joining as well to see

where they can help

awesome thank you carlos thank you yeah

this

ajith here more a user

looking for options for apache and

web services basically and also load

balancing

i'm at a startup we are an iot company

with some framework and

we have the front end for our

visualization

end we use uh different databases and

we're trying to

present we're using apache so it's more

i'm going to be the consumer

rather than developing this area

so that was my reason my question about

performance

no problem you're welcome to hang out

and see how the sausage is made so

thank you for joining and being a user

jeff

anyone else hello my name is christoph

i'm from budapest hungary uh my company

uses also heavily uses

nginx ingress and this is my first time

at this meeting so yeah it's

good to be here and uh i try to uh

catch up with with the topics and

everything and

and get more involved as possible so

uh thank you guys for for having me here

thank you for being here christoph

hello uh hello everyone i'm jing tao

i want to spend more time to join the

community

i have been revealed here recently and

i was meaning activating darker and

kind project before thank

well thank you for joining us and thanks

for the help that you've uh already

been giving us so we'll be uh greatly

appreciated in the future as well

hello hello can you hear me hello yes

yeah yeah uh i mean one uh it's my

second time to

at least meeting um come from the apache

ps6 community api six is a api gateway

so

it's also based on engix and lula so i'm

interested to

contribute to this project

awesome thank you

anyone else

okay then go ahead and get started

running throughs

um i don't know why i just muted myself

so we've got some um previous action

items that we've been working through so

ricardo

has been helpful in the transition as

well so we've got a pull request out

there for

mapping out how the architecture works

from that perspective

we've categorized these issues really i

just pulled them from the github api to

give us an understanding an idea of

what we're looking at the you know the

molehill or mountain

so trying to understand how big some of

the issues are that we're working

through

i've got a couple of open things still

trying to understand investigating

decayed version versus nginx

um the case version and the supported

nginx version so

if somebody wants to look at that as

well that's definitely an open item that

we still need to

understand how we would start looking at

deprecation

but what i wanted to do and my thoughts

were from this perspective

is to start looking at the the older or

the stale of rotten

tickets understand if we still need

these if they're still helpful

if we need to remove that tag and then

start moving through

because about 25 of the things that are

out there are bugs or features so about

half of

everything that we've got our bugs are

featured so i want to get through some

of the older things

so we can move through that um remove

those and move that and move that label

and then go from there so i've got the

all of the issues in this git lab

are in this um excel spreadsheet that we

can slice and dice and look through

they give us a little bit more of an

insight but i pulled up the stale of

rotten tag so i don't know if anyone

has had a chance to look through those

if we wanted to spend the time to walk

through those

thoughts comments concerns

none okay yeah i think yeah i think if

you like planet

walk too we can do this work too as well

okay um so i was looking through this so

um alejandro obviously has made several

comments on some of these from the older

ones but

just wanted to go through because it

they do get stale and

i get through some issues on that so

this one i didn't get a chance to read

through

just yet um but this looks like this was

an open bug and i don't know if it still

is

um from that perspective so

i'm using an older version but it's

probably not been fixed

looks like there's just a regex issue

from the pathing do we have anybody who

wants to look and see

it be assigned up to this and see if

they can replicate the issue and see if

it's still an issue in the newest

version

you cannot say to me

don't know if i can technically assign

things out yet so that's the other thing

it's just me

being on-boarded so i'll go ahead can

you just copy the the issue link and

post in the channel then i can

assign yep i'll bring it out to you in

the channel and we'll go from there

thank you actually i'll put it in the

the meeting notes

it'll be easier for me to follow up with

yeah and if you can share this link

because i didn't find in my

notes here okay

the meeting notes

just not in their invite

should i get your name right

okay awesome that's that's pretty easy

from that perspective

um obviously we don't have enough people

to just keep assigning these out so it's

just

this is the just the conversation of

is this still an issue do we need to

continue to walk through but um this one

looks like it's a documentation question

from that perspective

clients are expected how to reproduce to

generate clients use it

yeah i think we can if this is

documentation we can maybe

add the help wanted label

but uh what's happening yeah we could do

that yeah

that uh that is that a kind label i'm

still trying to remember

yeah help dash wanted

no it's not a kind sorry yeah repository

doesn't have them

you can you can even add it to the gui i

guess yeah

yeah let me go look at the uh the labels

that are on for this one

this is the joys of learning as new

people right so if anyone's got any

suggestions on that let me know

ah i guess it's just help wanted the

slash command

slash health wanted okay yeah

and will that remove the um the

staleness

label or we have to do that yes and it

looks like we have to do that

let's see if we go there yeah

okay gotcha okay

this one looks a little bit more

complicated because

it would be easy to update the

documentation but looks like it's a

breaking change

because we want to change the um

the annotation instead of being um

a white uh whitelist or a blacklist

source ranging we want to change it to

allow listing the problem

is that it's a breaking change because

um the annotations won't be back

supported

so don't know what the proper way for us

to move forward with this i mean

if we we do it from a breaking change we

just rev the version up because right

now we're sitting at 46

um and this is all in terms of us moving

towards a more inclusive language so

we already have another pull request

open that's asking for

uh another annotation from a block list

and we're changing the verbiage so

i think if we were to just rev the

version

from that perspective making sure that

it's known in the change log that's a

breaking

change what do you guys think i mean

obviously we need to

do the update and the change but it's

going to be impactful

yeah hungry yeah

totally and i know we're using semantic

versioning so i don't know how

adolfo or alejandro or the team we're

using changes from that perspective or

to signal that it was a breaking change

so that's something too that we need to

discuss

um as a project so i haven't had a

chance to get that follow-up with

ricardo so

that can also be a point of discussion

as well

if we just rev it to 146.

i don't i don't know yeah if

maybe if it's a really breaking change

that we need like

make the users to redeploy the the

ingress then we should bump the major

version

maybe or the finer i also don't

yeah i also don't know how how used this

this feature and functionality is

i used to use the white list source

range

before but not using that anymore

let me i'll make sure to add that now

that was the other thing as part of the

discussion with everyone else as well is

just

how we're getting into a better cycle of

pushing out changes because

the only reason we had the last release

was because of the the

the issue with the local host problem

so we just need to get better

understanding of how we're going to get

into a really cycle

this looks like this is a good first one

too because there's no documentation on

it well now i keep reading

keep reading down

this also goes in this is part of the

larger discussion as well as

the the support the supported versions

with kubernetes

so it will be supported in 120 but it's

not going to be

then it won't be supported then it will

become the subproject issue

okay so we just need somebody to test

this with 120

and see if you only need to update the

service to support udp

and then we need to validate that would

validate that we wouldn't need to make

code changes

okay let me do what

65 73 let me delete the summary

okay

if it's easy to to like try to reproduce

or

yeah that sounds good i'm gonna assign

myself this one as well

you can

hey awesome

nginx controller can't reload custom

template changes

this sounds like a bug issue it also

says good first issue

so i'm just getting acquainted with the

labels as well so

i wonder if this is still a problem

yeah it is it looks this is a

configuration issue

how to reload custom template

yeah we need someone like to test this

and create a custom complaint and see

make a change and see if this is

reloading

[Music]

yeah i think this one this is probably

like i like the

label does read this is a good first

issue to test

because it looks like he just had a

misconfiguration so this is a possible

candidate to close

when that issue when that tissue was

open

um so that's the thing too is that these

are these are a little old as well so

december 2020 yeah

maybe we can like comment and see if the

the speed meter

have this is still an issue for him

or for this person yeah yeah otherwise

we close we gave like some days if they

if they don't reply with to close the

issue

yeah so this is a possible candidate for

closing um kristoff if you wanted to

test out something

as a new user this seems like a pretty

straightforward setup

not trying to put you on a spot sorry

okay and this one looks like a feature

request that's just

old this is actually really old

three years old

yeah this may be like

we need to discuss this if it's we're

gonna add in the road map or not

maybe like per alejandro

comments he said maybe it's not possible

yeah i just come

yeah this one might need to be a

discussion point of like if we need if

we want to put it in

really if we want to put it in the road

map

that's a bigger conversation than just

trying to remove a label

okay

um so that was the um

that was all the either sterile or the

rotten tickets does anyone have before

we move into bugs let me just check the

mixture

um are we good on time because i know we

scheduled this for an

hour but the agenda only has it really

set up for a half hour

so we've still got time we can start

moving into do we

does anyone have any bugs that we need

to discuss from this perspective

no

i didn't really see any open issues um

from that perspective that were

at least i again there's 250 issues and

i've

not gone through all of them and there

doesn't seem to be

no one's screaming about which ones we

should be taking which ones we shouldn't

be

so i just need to figure out from that

perspective

with you know getting more support and

trying to figure out which ones

we should be looking at from the

community perspective

i guess we can just go straight to the

open issues and start looking at the

oldest bugs what do you guys think

[Music]

agreed

looks like this one's got a lot of top

conversation

it's a 37-17 if anybody wants to look at

it from their

own perspective

this is a really old version and again

this

ticket's all almost three years old

it's also on one four so wondering if

this is still

[Music]

this is one of those things where it's

do we is it still an issue

we scroll down to where we're at oh

yesterday

this might have been one that ricardo

was looking into

about to get some follow-up with uh with

him on this

to see if he's actually looking into

this one

we have about 10 minutes to go through

some of the older ones

not working properly with proxy protocol

again a lot of these and this i think

was why we had the conversation

around versioning and support

because a lot of these issues are old

and using older versions

this is 25 as well whether these got

fixed or not that's the real story

the real question

yeah there are some cases that is

hard for us to test because we need to

like

set up some environments and that's

quite hard we need to like

depends on the submitter

responses because this can turn out

a little bit expensive for us

yeah and i don't know how that how we do

that from a support perspective from a

project

so that's one of the things i need to

look into if we need to replicate issues

like this

um just to understand how we can support

this and being able to

have a cluster to replicate this issue

yeah this we can

we can if we need that to replicate and

debugging issues we need to request from

the steering committee

like a founding and maybe a project in

gcp

let me find

this is probably more it this one's

probably more for me on my education

to make sure i understand the process as

well

now let me find here the repo

okay just give me a second

[Music]

yeah this is the project we need to open

an issue

to request founding

[Music]

and can you drop that in the action

okay thank you i i post in the chat but

can you just let give me the link for

the

the meeting notes then i can put there

right away

slack wants to load

recording on zoom running in chrome and

opening up slack we'll see how uh

oh you can send in the chat in the zone

yeah so it's going to drop yeah there

you go i dropped an engine next down as

well

i'm gonna save this

yeah so this one looks like it will

require funding to replicate and test so

nothing we can really do about that from

that perspective right now

a yeah bit newer so we're getting so it

looks like we've got about two or three

that are just on

really old versions so this is getting

getting a little bit better

the ls certificate is not updated when

the secret is updated

a swarm cluster

stored in secret

does the this is just a question and

probably

i don't know if anyone on it doesn't

automatically reload if the secret gets

updated

if you've got something provisioned if

you've got a service provision

or that's using a secret using a cert

from a secret does it automatically

provision it

one if you update the secret

i'm guessing that's the assumption here

he's assuming that there's no real

trigger as i would think

yeah it doesn't trigger

there's the this certificate is for the

the the dns you really want to

to get the certificate right it's not on

yeah that's the stuff and it needs to be

done

it always reloads in the config map if

the config map changes

yeah but the config yeah the config map

will point to where the secret is

and it's mounted so i'm wondering no

no this is the deployment

i don't i don't remember in the config

map that's a

field that points to us particular

secret i can be wrong

well this is a newer comment but yeah so

this would just require investigation on

if

it actually does that because i mean

this would make sense if he's got newer

pods that

get the new cert and the old ones that

haven't been

recycled yet oh

he did delete everything and redeploy

this is going to require someone to do

some testing

it looks like it's so he went from 30

and it looks like someone

43 is rather recent so it looks like

it's still an issue

okay so the label is not there anymore

okay so this does look like an actual

bug that we're gonna have to investigate

but this is a bug or this is just not a

feature that was not

implemented yeah that's right sorry

thank you um

this yeah i'll put that in the notes

um so if we're doing that i guess triage

needs more information would be a proper

label in that bug

it just slash triage

needs more information

just one second bro

[Music]

open up and then

triage should do the work

come on i'm trying to copy this

being lazy

no no it's all together

more information yeah no more

no no yeah like this but there's no more

i guess

just needs information okay yeah

yes

there we go

okay and i've got it in the in the notes

on that one how are we doing in time

about 35

i don't think there's really anything

else do we have

we can just keep going through triaging

for

the next 10 to 15 minutes unless there's

anything else that the

he needs to discuss anyone else that

wants to discuss anything on here

you know okay let me keep working

through the issues then

okay

again much older version this is

it keeps popping up so it's like that's

about the third what the

fourth time now i've said that from an

older version perspective

yeah i i would say like if is all the

version just a recommend uh

mentioning that meter and asking like if

this is still an issue for the

newer versions yeah ask them to

ask them to test on the newer version

yeah

i'll go through the issues and the six

comments and see if they've

someone else has tried to test it on a

newer version yeah because

like a cousin maybe uh 41 yeah

yeah 41's recent rather recent so yeah

then looks like it's

actually issue

24

turns out the ingress yellow from my

previous comment didn't work as well

let me go back up to the top so we can

read through the actual issue

it's fun to watch people read isn't it

uh

okay so yeah the the match seems correct

right they

but they're expecting it to send

so when ingress proxies everything they

expect the the same

headers to be sent as well to the

backend service

folks

this is this is really a design question

whether we

expect it to be to pass those through

we're not

probably going to need to get a little

bit more backstory on whether this was a

supported feature

expectation i don't know

no i don't know what you do

you follow up with uh ricardo anyone and

see if we can

again this is one of the things where is

it expected behavior or not

and so it is still an issue with 4 44 41

sorry

the same thing 31 so it's

if someone's tried to do this with a

newer version

still relatively relatively old as well

[Music]

we can disable this

oh so he's just trying to push

the logging from my security to standard

out

looks like this might be a configuration

[Music]

um

it looks like this is still an open

issue

this this is going to require

development time for something to fix

this

um

so this has got to fix but it's not

a desirable one

i'd rather be able to do this globally

versus having to set it in the

configuration file

uh i think from that perspective then

this is just

there's a fix but there's uh

that's not a real um

i don't want to say a real issue it is

an issue but this is one

i'm just going to say we help wanted

it's just help just help thank you

okay

but it's 5 21

oh wow 21-6 yeah

james uh i need to yes

okay all right this is about that this

is 12.

yeah we i mean we've been through a good

number of issues and we just got to keep

working through them

um so i've got time scheduled for myself

throughout the week

um but yeah we just want to make sure we

keep moving things along

so yeah we can go ahead and stop now for

everyone because that

considerable amount of people have

already dropped off

sounds good to me thanks james for

hosting yesterday

awesome thanks everyone for joining um

i'll post that

all the meetings are posted and i'll get

the ticket to sign

out and update some of the other things

as well so thanks everyone

thank you bye bye