Open Telemetry Uncategorized, 8 Jun 2022

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: 2022-06-08 meeting

Description

OpenTelemetry PHP SIG Meeting

A

Oh, I even got a mate today. That's super easy.

A

It's a happy wednesday. Indeed,.

B

B

It's all over here.

C

It's uh nearly bedtime for me.

A

Do you say something different at night time instead of g'day, or is it there's that just like this today.

B

A

That's like the standard g answer for uh australians for q for aussies.

B

I don't, I don't think we call ourselves aussies very often. Oh.

A

Really, that's surprising to me, I didn't know what so you just call yourselves australians tonight you haven't really it doesn't.

B

A

Is your school year over now.

C

uh No school holidays are coming up a couple of weeks. Nice.

A

I forget that your calendar is so much different than ours, because your weather is so much different. It's it's so nice here in the next couple of weeks, because, like um elementary middle and high school are finishing, so we don't have like the traffic gets exponentially better when school is over.

C

I don't think I realize that so your school year doesn't run to a calendar year.

A

No, no, no, not even close, we school start uh school starts in the fall in like mid-august and then runs through like pretty much through christmas, and then it starts back up in january and then it goes to like may and then yeah. We have summer break from may to august and then my other rents repeat.

C

Wow, I don't know how I didn't know that, given the amount of television I've watched.

A

It's um I mean it's all I've ever known, so I can't tell if it's good or not, but I mean.

C

B

The idea was that.

C

Everyone will be home in the summer to farm and bring in the crops right.

A

I think- and I think too most of the us is very hot in the summer. I think they also didn't want to make children go to school during the summer. If I remember correctly, I think that's part of the history.

A

Okay, I still think I think something we got to fix this whole five days of work. Two days of rest thing.

C

A

Other way around quite enough.

C

A

Think it should be two days of work five days on the best.

A

That would be nice.

C

I just need to be more efficient, hello, that's.

A

That's right, more efficient, more efficient, it would be good, it's always good, we'll wait until five after and we can get rolling. I.

A

Yeah two days of work would be pretty good.

C

I don't know what I would do with five days off. To be honest,.

A

You know it's f, you say that, but then I think that you would find stuff to do fred.

C

I think if I ever do that, I'd I'd be amazed if I got anything done in my two days of work having just completely unwound for five days. I'd take both of those days to get back into the swing of it just in time to to.

A

To do it all over again.

C

A

Well, maybe that's the idea.

C

A

We, a lot of companies, are experimenting with four day work weeks. Mailchimp has been doing uh summer hours for the last three years, which are really nice since every friday from memorial day, which is like the end of end of may to labor day, which is like the first week of september.

A

uh We get off on fridays at 1pm. So that's awesome, oh wow yeah. So that has been that's been really pleasant, but we, I would love to continue doing that. I think this is the last year we're gonna have that because of the intuit acquisition.

C

Okay, I um uh my university or my department anyway, has a thing called flex time when we have to work 36 hours in a week and as long as you work them anything above that you just go into credit and take a day off every fortnight or a half day as you need it. So it's.

A

C

It's it's pretty nice actually.

A

That is pretty good. There's nothing! There's nothing like getting the right amount of time off to keep. You focused.

C

Yeah, yes, I don't, I don't use it as well as I should, but um yeah it's definitely definitely convenient when you need to pop out and down errands or just take a day off.

A

Yeah, I totally don't use my vacation time appropriately, all right so we're at five after so we can get going today.

A

Let's so, the agenda, I think, is pretty short today because gret timo and I spent almost two hours just yesterday, parsing through the backlogs. We have a lot of really good, concentrated things to work on.

A

Oh, let's check the open, pull requests.

A

So we actually have eight pull requests open now, which is pretty large number for us uh w3c. The w3c test service that yuki's working on the b3 propagator from sean has almost completed um up an updated, layerville and symphony quick starts, and then we have dr luke's uh symphony. Quick start bundle, removal and then right. We have your. I saw your pr for the instrumentation library I'll review that today and emergent, if that's all kosher with you.

C

uh It is, but let's get timo to look at it as well, because I thought it was right before I yeah.

B

I think it's right now, but but.

C

I I don't think I'm the right person to to judge that.

A

Guess what all right and then nothing crazy and the contribution, um no special open bugs um yeah I'll put a note on the agenda that we have. We fixed up the backlog yesterday and we have a lot. I think we have what we decide. Yesterday, 11 yeah 11 required for beta tickets and we're going to try and prioritize those. So if new people come and they want to help or if people are looking for work to do, that's the right thing to do.

A

So I'm going to put it back in the agenda today.

A

Your tickets are prioritized here and then just for the recording after we fin after we get uh beta, live and rolling for tracing, then we can start looking at some of the other signals most, namely metrics and then followed by logging and now profiling is a new signal, but we are far and far away from that. But that's okay.

A

The nice thing is know about profiling. Oh yeah, there's um profiling is a new signal. That's being developed the specification for it's being developed right now, uh I'll post the link in in the uh agenda. It's called uh there's a room called hotel profiles.

A

That's why we have this meeting.

C

A

Hotel profiles has been opened, uh we're also participating in the demo web store for those who don't know which is the conglomeration of most, if not all, of the libraries we are doing, the authentication service timo is taking that on. I will post a ticket about that in here too, because that's a very relevant conversation for people to see how open telemetry php is working as intended.

A

It's really neat because then we'd see all of the different libraries and how they work and stuff so.

A

That's what I had on the agenda amber is there anything else. You'd like y'all would like to discuss in our meeting today.

C

uh I was just going to point out that um psr 22 sort of went quiet for for a couple of weeks, but is getting a bit active again now. So um I'm doing a bit of work on that which is psr. 22 is a tracing um yeah, a set of tracing interfaces for a php.

A

um Yeah is there, um is there like a draft spec, or is it just all in that discord now.

C

No there's a no it's in discord and discord. Links off to. uh I think that uh adam has uh a github where he's sort of sort of theory crafting what it might look like and uh okay.

A

Good journey crafting that that has gone in my vernacular for sure.

C

I like that one, yes,.

B

A

Very crafting cool- that's uh that's great news that just shows how we're getting more tied in with the php community at large, especially with specification, which is top-notch. That is I'm glad you mentioned that brett, because the nice thing about the other signals is, there are very good active, live instances of all of those things we have.

A

There are lots of prometheus libraries for php, vlogging obviously has monologue and all the psrs that are related to logging and profiling has xh prof and tideways, and a bunch of these other things. So at least we'll have things to garner inspiration from, rather than trying to write things from scratch like we've had to with tracy. So that's nice.

C

Yes, yeah I've been doing a bit of um sort of uh a survey, I suppose of of what um you know, what the other uh tracing implementations in php are and how they handle transactions and traces uh spans. Sorry just to try and make the psr standard.

C

You know so that so that it's nice nice for us and not too onerous for us to implement, but.

A

Potentially useful.

C

To other implementations.

A

The long term goal of open telemetry is to be is to have backwards compatibility with the with open tracing and open census, and I think that um I think the spec just like leans us in that direction. But that's always a good thing to uh walk forward. Yeah, wonderful,.

A

Amber do you have anything that you'd like to bring up in today's meeting.

D

Oh hi bob uh nothing much just that, I'm starting to work in this uh issue 704, so just I'm reviewing it. I will start today, probably.

A

Yeah uh that issue 704. Thank you for letting us know. Please let us uh please ask in the uh please ask in the slack channel, if you have any questions or concerns or if you yes,.

D

Sure yeah I'm reviewing right now, I'm going through the code and I'll ask if there is any thing to ask careful all right.

A

All right, I think, that's uh that'll- do us for today uh we'll see you on there and I'll catch y'all next week.

C

Thanks all bye.

D