tracing-subscriber: please log that there was an error serializing an event #1965
Assignees
Labels
crate/subscriber
Related to the `tracing-subscriber` crate
good first issue
Good for newcomers
kind/feature
New feature or request
Comments
hawkw
added
kind/feature
bryangarza
added a commit
to bryangarza/tracing
that referenced
this issue
Apr 29, 2022
This patch adds a single `eprintln` line to let the user know that we were unable to format a specific event. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves tokio-rs#1965.
bryangarza
added a commit
to bryangarza/tracing
that referenced
this issue
Apr 29, 2022
This patch adds a single `eprintln` line to let the user know that we were unable to format a specific event. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves tokio-rs#1965.
bryangarza
added a commit
to bryangarza/tracing
that referenced
this issue
Apr 29, 2022
When `Format_event::format_event(...)` returns an errror, we are currently silently dropping that Span/Event. tokio-rs/valuable#88 explains one such case in which this was encountered (due to a bug in valuable-serde). We want to be made aware whenever a Span/Event is dropped. This patch adds a single `eprintln` line to let the user know that we were unable to format a specific event. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves tokio-rs#1965.
bryangarza
added a commit
to bryangarza/tracing
that referenced
this issue
Apr 29, 2022
When `Format_event::format_event(...)` returns an error, we are currently silently dropping that Event. tokio-rs/valuable#88 explains one such case in which this was encountered (due to a bug in valuable-serde). We want to be made aware whenever an Event is dropped. This patch adds a single `eprintln` line to let the user know that we were unable to format a specific event. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves tokio-rs#1965.
bryangarza
added a commit
that referenced
this issue
Jul 25, 2022
…ter (#2102) Motivation: When `Format_event::format_event(...)` returns an error, we are currently silently dropping that Event. tokio-rs/valuable#88 explains one such case in which this was encountered (due to a bug in valuable-serde). We want to be made aware whenever an Event is dropped. Solution: Write to the Writer with an error message to let the user know that we were unable to format a specific event. If writing to the Writer fails, we fall back to writing to stderr. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves #1965. Co-authored-by: Eliza Weisman <eliza@buoyant.io> Co-authored-by: David Barsky <me@davidbarsky.com>
|
Resolved by #2102 |
davidbarsky
added a commit
that referenced
this issue
Aug 24, 2022
…ter (#2102) Motivation: When `Format_event::format_event(...)` returns an error, we are currently silently dropping that Event. tokio-rs/valuable#88 explains one such case in which this was encountered (due to a bug in valuable-serde). We want to be made aware whenever an Event is dropped. Solution: Write to the Writer with an error message to let the user know that we were unable to format a specific event. If writing to the Writer fails, we fall back to writing to stderr. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves #1965. Co-authored-by: Eliza Weisman <eliza@buoyant.io> Co-authored-by: David Barsky <me@davidbarsky.com>
davidbarsky
added a commit
that referenced
this issue
Aug 24, 2022
…ter (#2102) Motivation: When `Format_event::format_event(...)` returns an error, we are currently silently dropping that Event. tokio-rs/valuable#88 explains one such case in which this was encountered (due to a bug in valuable-serde). We want to be made aware whenever an Event is dropped. Solution: Write to the Writer with an error message to let the user know that we were unable to format a specific event. If writing to the Writer fails, we fall back to writing to stderr. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves #1965. Co-authored-by: Eliza Weisman <eliza@buoyant.io> Co-authored-by: David Barsky <me@davidbarsky.com>
davidbarsky
added a commit
that referenced
this issue
Aug 24, 2022
…ter (#2102) Motivation: When `Format_event::format_event(...)` returns an error, we are currently silently dropping that Event. tokio-rs/valuable#88 explains one such case in which this was encountered (due to a bug in valuable-serde). We want to be made aware whenever an Event is dropped. Solution: Write to the Writer with an error message to let the user know that we were unable to format a specific event. If writing to the Writer fails, we fall back to writing to stderr. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves #1965. Co-authored-by: Eliza Weisman <eliza@buoyant.io> Co-authored-by: David Barsky <me@davidbarsky.com>
davidbarsky
added a commit
that referenced
this issue
Aug 24, 2022
…ter (#2102) Motivation: When `Format_event::format_event(...)` returns an error, we are currently silently dropping that Event. tokio-rs/valuable#88 explains one such case in which this was encountered (due to a bug in valuable-serde). We want to be made aware whenever an Event is dropped. Solution: Write to the Writer with an error message to let the user know that we were unable to format a specific event. If writing to the Writer fails, we fall back to writing to stderr. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves #1965. Co-authored-by: Eliza Weisman <eliza@buoyant.io> Co-authored-by: David Barsky <me@davidbarsky.com>
davidbarsky
added a commit
that referenced
this issue
Aug 24, 2022
…ter (#2102) Motivation: When `Format_event::format_event(...)` returns an error, we are currently silently dropping that Event. tokio-rs/valuable#88 explains one such case in which this was encountered (due to a bug in valuable-serde). We want to be made aware whenever an Event is dropped. Solution: Write to the Writer with an error message to let the user know that we were unable to format a specific event. If writing to the Writer fails, we fall back to writing to stderr. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves #1965. Co-authored-by: Eliza Weisman <eliza@buoyant.io> Co-authored-by: David Barsky <me@davidbarsky.com>
davidbarsky
added a commit
that referenced
this issue
Aug 24, 2022
…ter (#2102) Motivation: When `Format_event::format_event(...)` returns an error, we are currently silently dropping that Event. tokio-rs/valuable#88 explains one such case in which this was encountered (due to a bug in valuable-serde). We want to be made aware whenever an Event is dropped. Solution: Write to the Writer with an error message to let the user know that we were unable to format a specific event. If writing to the Writer fails, we fall back to writing to stderr. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves #1965. Co-authored-by: Eliza Weisman <eliza@buoyant.io> Co-authored-by: David Barsky <me@davidbarsky.com>
davidbarsky
added a commit
that referenced
this issue
Aug 24, 2022
…ter (#2102) Motivation: When `Format_event::format_event(...)` returns an error, we are currently silently dropping that Event. tokio-rs/valuable#88 explains one such case in which this was encountered (due to a bug in valuable-serde). We want to be made aware whenever an Event is dropped. Solution: Write to the Writer with an error message to let the user know that we were unable to format a specific event. If writing to the Writer fails, we fall back to writing to stderr. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves #1965. Co-authored-by: Eliza Weisman <eliza@buoyant.io> Co-authored-by: David Barsky <me@davidbarsky.com>
hawkw
added a commit
that referenced
this issue
Aug 24, 2022
…ter (#2102) Motivation: When `Format_event::format_event(...)` returns an error, we are currently silently dropping that Event. tokio-rs/valuable#88 explains one such case in which this was encountered (due to a bug in valuable-serde). We want to be made aware whenever an Event is dropped. Solution: Write to the Writer with an error message to let the user know that we were unable to format a specific event. If writing to the Writer fails, we fall back to writing to stderr. We are not emitting an actual tracing Event, to avoid the risk of a cycle (the new Event could trigger the same formatting error again). Resolves #1965. Co-authored-by: Eliza Weisman <eliza@buoyant.io> Co-authored-by: David Barsky <me@davidbarsky.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Feature Request
tokio-rs/valuable#88 came because I realized that
tracing-subscriber, specifically the json formatter, missed some of the logged events.Crates
tracing-subscriberMotivation
Logs, and traces, carry information. It is good to know that some of it was lost. In my case what failed to serialize was the error which caused abnormal termination.
Proposal
If an event fails to serialize (whether in json or a different formatter), I'd like to see a new event generated, informing me about it. Best would be if this was optional, as I'm sure not everyone would want it.
Alternatives
The other alternatives are to either silently ignore the failed event (what happens now), or panic. Panic seems unreasonably harsh, especially since
tracingis used by long running services.The text was updated successfully, but these errors were encountered: