Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Tristan Sloughter
    @tsloughter
    there is an error handling spec of some sort in the specification, but maybe it doesn't cover bootup?
    Daniel Dyla
    @dyladan
    The error handling doc suggests to catch errors globally and use sane defaults
    The error handler example in that doc is sdk-specific
    since most errors will happen in the SDK, this is fine for most things. But the problem we're specifically trying to solve is errors regarding setting up the SDK and registering it with the API.
    specifically, what should happen if you try to register a global tracer provider when one has already been registered?
    1. throw - may crash program
    2. nothing - may be confusing to users
    3. log - changes output of program, may break some things
    4. some new error handling interface
    5. replace old tracer provider with new - maybe confusing or unexpected behavior
    Bartlomiej Obecny
    @obecny
    with regards to point 3 I'm hesitant to say that logging should never break any program and we can't do anything especially in node.js, using console.log / error etc.
    John Watson
    @jkwatson
    Is Monday's meeting still on? It's a national holiday in the US, so I'm sure many of the folks in the US won't be there.
    Daniel Dyla
    @dyladan
    I apologize if this is a stupid question but is baggage part of the 1.0? If not, should it be moved to its own signal api package similar to metrics?
    1 reply
    Cijo Thomas
    @cijothomas
    https://github.com/open-telemetry/opentelemetry-specification/blob/master/specification/baggage/api.md#baggage-api Baggage spec has the status "Feature-freeze", means its a stable signal, no reason to not include in 1.0.0
    John Watson
    @jkwatson
    yes, baggage is intended to be part of 1.0, as far as I know
    Tristan Sloughter
    @tsloughter
    I hit an issue with linking a new span in a new process (erlang process, so think green thread or async operation) to the active span in the process that spawned the new process
    I think it could be an issue in any implementation/language
    even those not using automatic propagation, like threadlocals
    if there is a process A with an active span P in its context C1 and in A the process B is spawned, in which you want to create a new span C with a link to P. you also want to carry along the baggage. so you copy C1 to the new process B when creating it and attach it. now when creating the span C and passing the option for link to P you also end up with the parent being P because your context C2 has active span P
    it is the same issue if you create the new span C in process A with a link to P and then set it to active in B, the parent still is P when you really just want a link to P
    Tristan Sloughter
    @tsloughter
    hm, now that I think about it you probably don't want a copy of C1, you only want baggage... sinc eyou likely don't want the same deadline and what not in this separate process
    maybe I just have to document copying the baggage? is this something others have dealt with? assuming my attempt to explain it abstractly worked, hehe
    Daniel Dyla
    @dyladan
    JS here what's a thread /s
    Tristan Sloughter
    @tsloughter
    hehe
    Tristan Sloughter
    @tsloughter
    agh, missing the maintainers meeting yet again, wanted to bring this issue up there
    John Watson
    @jkwatson
    @tsloughter so. the question you're asking is: how do you propagate baggage (for, eg. for links) without propagating the rest of the context?
    Tristan Sloughter
    @tsloughter
    @jkwatson not exactly. I think its, how to propagate a context with an active span (not through a text map propagator) and create a linked child span, without parent being the active span from the context
    when propagating across services the new context has a NonRecordingSpan which then becomes the parent of the new span. But if it should be a link instead of a parent... i don't know aside from manual manipulation in every case
    Tristan Sloughter
    @tsloughter
    maybe a span StartOpt like "link_active" to link to the active (poassible NonRecording) span instead of being parent.. then it is at least clear how to do it in the spec/docs, i don't know
    John Watson
    @jkwatson
    yeah, seems like right now, it's a two-step operation, 1) get the active span/spancontext, and create a link from it, 2) set no parent on the new span.
    Tristan Sloughter
    @tsloughter
    right
    John Watson
    @jkwatson
    There definitely isn't a java API to do this in one step right now.
    Tristan Sloughter
    @tsloughter
    2 steps for something isn't bad, except when one of the steps is confusing (in my opinion) which is the need to set an undefined/null/whatever span as active in the current context at the beginning
    you can't "set no parent on the new span" since the parent option is gone
    it only takes a context. so you must have something like set_current_span(null)
    John Watson
    @jkwatson
    we have a "setNoParent" method on our span builder
    Tristan Sloughter
    @tsloughter
    oh
    you should have led with that ;)
    John Watson
    @jkwatson
    to explicitly veto anything that might creep in from the context.
    :)
    Tristan Sloughter
    @tsloughter
    that being a start opt may be good enough for making the link case not bad
    Tristan Sloughter
    @tsloughter
    with that I guess I can bring this up as a PR to add setNoParent to the list of start span options and maybe an alternative solution will come up.
    John Watson
    @jkwatson
    :tada:
    Bogdan Drutu
    @bogdandrutu
    @jkwatson looks like you miss the slack ability to have that emoticons on the comment, maybe comment open-telemetry/community#362
    2 replies
    Weyert de Boer
    @weyert
    Congratulations everyone!
    Cijo Thomas
    @cijothomas
    :clap:
    Tyler Yahn
    @MrAlias
    :tada:
    Bogdan Drutu
    @bogdandrutu
    For someone who was in the Monday meeting, did I asked everyone to wait the stable release until we finalize the release process in the Specs?
    And promised that the process will be published today (Wed)? Here it is open-telemetry/opentelemetry-specification#1418 :)
    Tristan Sloughter
    @tsloughter
    woooooo
    Bogdan Drutu
    @bogdandrutu
    there is a community meeting
    Ted Young
    @tedsuo

    @/all Hello maintainers! Couple announcements.

    I’ve created a new slack channel for us to inhabit. Rejoice and abandon ship!!! https://cloud-native.slack.com/archives/C01NJ7V1KRC

    I’ve also created an intial roadmap document for us to discuss on Monday. Please give it a read before then if you have a chance: https://docs.google.com/document/d/1jwtB0vFZ0uXOxHZJpuIaiFLnrFSiGDJRtX7VNI3DULc/edit#