Open Telemetry Uncategorized, 1 Aug 2022

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: 2022-08-01 meeting

Description

cncf-opentelemetry meeting-2's Personal Meeting Room

A

A

A

A

A

Foreign foreign.

A

A

A

Is out today, is there anyone else, who's prepared to run this meeting.

B

Well good point: I was I was not aware. Sorry um not really prepared, but I I think it doesn't need need a whole lot of preparation. We already have the agenda fill. um Please make sure that you add yourself to the attendee list and then we can get started um in the specification.

B

Currently, the semantic conventions are extended and refined, so HTTP is extended and there are some is refined, and there are some extensions like process, parent pit, for example, and some updates around tooling um one is for splitting metrics and then also the group semantic convention. Group type was added everywhere so that it doesn't complain about falling back to spans.

B

So in case you use code generation to generate code constants, then this might have an impact if you're we're running on um on assumptions there, but I think it should be fine in general, um since it it just added span and resource where it was Spain and resource before other than that, I'm not sure. If, if reading out the um the seek check-in is, is beneficial but I'll do it. Nevertheless, um so for PHP, two omniscious Labs before I.

A

Beat it share your screen? Would that be helpful.

B

ah I can do so.

A

Perfect thanks.

B

A

All right, foreign.

B

Tracing released- and this seems like some left over um Java reduced capacity down to one part time- one full-time maintainer um so I think a call for maintenance.

B

um It's supposed to be, if, if you have any uh any person power to dedicate to Java, please consider doing so for JS 1.5 released and the next release should include the metrics GA for python. Last. Eight issues for metric stability in review course approaching the metrics SDK release 76 percent complete and release 1.9 in progress without Matrix SDK.

B

At this point, then C plus plus half of the Matrix sdkrc, is done and release 1.5 as well, and then let's hand over to Emily and Bruno, and the pr on the SDK enabled environment variable go ahead. Please.

C

uh Okay, I will start probably Bruno. You will come in uh so if you click on the link, so basically uh the currently there's the environmental variable called the audio experimental enabled there is um basically is you know the answer: the detailed chat with the JavaScript, so we want I mean we have this name experimental removed from the stream from the other in the Bruno, when I had to do a PR, so basically the same click on a link on the pr.

C

So basically the channel we asked to have this more discussion here to see I mean. Are there any other languages also use this um variable to support awareness. So basically the the context is the micro profile Community. We want to consume a doctor, open, Telemetry and actually some because it's also the medical profile is a the standard for cloud native application development. So we asked to the community want the experiment will removed from the environmental variable name. So basically, hopefully you need to call so we agree on.

C

So this is a good thing to do and then then you can have this PR marks.

D

Okay, thanks Emily. Let me just add that this property is uh originally in the Java SDK I I, don't know I'm, not aware if anyone else anyone else is using it, and we would like to to make it basically official uh on the spec and have it having the ability to disable the SDK by configuration.

D

um And the pr uh discusses the the last Point discusses the nature of the property, so I originally Define it as a Boolean, but uh tigrant uh mentioned that we really don't have that concept on the on the SD on the specification. Apparently so it can be I I, don't know if you guys have any thoughts about this foreign.

D

If you see any usefulness on the property or if the, if it's okay, to have a bullying in there.

B

So far, it seems like only um Java is making use of this, and what you in particular, are seeking for is having it um standardized so that it's non-experimental in Java, if there is no need for the for the other sdks to make use of it, is at this point, then it would be fine too edit, as is, and then Java can go. Non-Experimental and the others could add support later. If they want to and then it would be would be aligned and if there is no need for it.

B

Since it's an optional feature, it will just not not going to the other sdks. So maybe that's why the the other maintainers don't have any opinions there. At this point.

A

Yeah I mean as JS I didn't I, don't have an opinion I'd book we're probably not going to implement this because you can't just enable the SDK it needs to be bundled, and everything like that. um This.

B

A

Seems more like a topic for the specification meeting.

B

Actually then, this is not about. um It's actually meant the other way around. It's only intended for disabling and otherwise enabled SDK. So only if you have an SDK, that's well configured with everything that you need, then you can set enabled two fours and this will disable it. So um then you can run if I understood it correctly. Benchmarking on it.

A

Got it that's correct.

C

Yeah, that's my understanding as well. So in this case, is that possible to have well like uh I think either way, there's a two two kind of approvals. You know right, so we need another one.

C

He looked at and get him merged.

B

Yeah I think by bringing it up here, you have brought it to most of the maintenance attention, um as Dan said, I think the the spec meeting also makes sense. So if you can make it there, you could add it to the agenda there or you just edit there and it's discussed in in your absence. But if enough people have have had their eyes on it, then it should be fine. um It doesn't seem to be to be too controversial.

B

It should just be well understood what the the consequences for other sdks would be if they were to implement it as well.

C

Right, the second meeting in the tomorrow right.

B

Yeah, that's tomorrow, one hour earlier than this one right.

C

So, uh okay, so Bruno yeah sounds like uh could I go to tomorrow as a meeting to talk about her as well right.

D

Yeah I can go there as well.

C

Okay, cool all right thanks, so I have one more question so once he's emerged once it is API emerged also in can we get a release out.

B

And that's something you'd need to to discuss with the Java maintainers um or your I.

C

Think it's as a spec I need to reduce the first right and then Java team can uh consume. That's that's a stack.

B

A release of the spectruming yeah, um yes, I- think a release should be due um early August. um So if.

A

B

You raise your need for it tomorrow. um Okay, then, it then depending on when it's it should go in there pretty soon. um The current spec release was blocked on the semantic convention tooling, for this uh split split metric operation, but that one was merged, so um we would be able to to draft a new release anytime.

C

Okay, that's a good enough! That's awesome! Thank you! So much thanks sure, okay, I think at this point our job yeah. Thank you sure. Thank.

B

You all right and the next one is brought up by churrosy, but he's not on the call and I don't have context on what this is about. Is there anyone who who has an idea or did he send anyone to discuss it.

B

All right doesn't seem seem like it, um then I'll I'll get back to shuracy and let him know that we were not able to discuss it. So we can bring it up either in the maintenance leg core on the next chord.

B

All right that would be then thank you, everyone and have a nice rest of your day. Bye.

D

Thanks everyone, bye-bye.