Replies: 1 comment 1 reply
-
Hey @ca-schue, context propagation is not limited to HTTP, it's just a case where propagation is "easy". Take a look at some of the high level documents on that topic:
For some languages we even have some docs around propagation (and some hints how you can do your own):
The challenge with lower protocols may be that you need to find a place where to put the properties you need for context propagation. |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
As mentioned in my other discussion thread (#1960), I am currently looking into the possibilities and limits of using OpenTelemetry in a constrained distributed system.
As far as I could find out, the main use case for OpenTelemetry spans are protocols using HTTP which make propergating context a feasible task.
However for my use case, OpenTelemetry would be the most helpful for providing TCP/IP observability. To pe more precise: My distributed system is more of a real-time and safety-critical cyber-physical system where most of the communication happens over protocols underneath HTTP in the OSI stack. (TCP/UDP, Socks, IP, IPSec ...)
Is OpenTelemetry still the right framework for this? How can OTLP context propagation be made possible for TCP or even IP?
Beta Was this translation helpful? Give feedback.
All reactions