Thanks to @AdamFrey for reporting this issue! Ref. <https://clojurians.slack.com/archives/C06ALA6EEUA/p1713805333272469> Previously: Attempting to run AOT'd code using Telemere would result in errors like: "Attempting to call unbound fn: #'taoensso.telemere.handlers.open-telemetry/handler:open-telemetry-logger" The approach I was using of conditionally requiring namespaces and then aliasing vars seems to be inherently fragile under AOT, and was leading to the remote source var being unbound. With this commit I've now switched to a simpler approach - where we conditionally require namespaces *without* the need for any aliasing.
5 KiB
See below for config by topic-
Signal filtering
A signal will be provided to a handler iff ALL of the following are true:
- Signal creation is allowed by compile-time filter config
- Signal creation is allowed by runtime filter config
- Signal handling is allowed by handler filter config
- Signal middleware does not suppress the signal (return nil)
- Handler middleware does not suppress the signal (return nil)
For 1-3, filtering may depend on (in order): Sample rate → namespace → kind → id → level → when form/fn → rate limit
- See
help:signal-filtersfor info on signal creation filters. - See
add-handler!for info on signal handler filters.
Signal handlers
See section 4-Handlers.
Interop
clojure.tools.logging
clojure.tools.logging can use Telemere as its logging implementation.
To do this:
- Ensure that you have the
clojure.tools.loggingdependency, and - Call
tools-logging->telemere!, or set the relevant system config as described in its docstring.
Verify successful intake with check-intakes:
(check-intakes) ; =>
{:tools-logging {:sending->telemere? true, :telemere-receiving? true}}
Java logging
SLF4J can use Telemere as its logging backend.
To do this, ensure that you have the following dependencies:
[org.slf4j/slf4j-api "x.y.z"] ; >= 2.0.0
[com.taoensso/slf4j-telemere "x.y.z"]
When com.taoensso/slf4j-telemere is on your classpath AND no other SLF4J backends are, SLF4J will direct all its logging calls to Telemere.
Verify successful intake with check-intakes:
(check-intakes) ; =>
{:slf4j {:sending->telemere? true, :telemere-receiving? true}}
For other (non-SLF4J) logging like Log4j, java.util.logging (JUL), and Apache Commons Logging (JCL), use an appropriate SLF4J bridge and the normal SLF4J config as above.
In this case logging will be forwarded:
- From Log4j/JUL/JCL/etc. to SLF4J, and
- From SLF4J to Telemere
System streams
The JVM's System/out and/or System/err streams can be set to flush to Telemere signals.
To do this, call streams->telemere!.
Note that Clojure's *out*, *err* are not necessarily automatically affected.
Verify successful intake with check-intakes:
(check-intakes) ; =>
{:system/out {:sending->telemere? true, :telemere-receiving? true}
:system/err {:sending->telemere? true, :telemere-receiving? true}}
OpenTelemetry
Telemere can send signals as LogRecords to OpenTelemetry.
To do this:
- Ensure that you have the OpenTelemetry Java dependency.
- Use
handler:open-telemetry-loggerto create an appropriately configured handler, and register it withadd-handler!.
Tufte
Tufte is a simple performance monitoring library for Clojure/Script by the author of Telemere.
Telemere can easily incorporate Tufte performance data in its signals, just like any other data:
(let [[_ perf-data] (tufte/profiled <opts> <form>)]
(t/log! "Performance data" {:perf-data perf-data}))
Telemere and Tufte work great together:
- Their functionality is complementary.
- The upcoming Tufte v4 will share the same core as Telemere and offer an identical API for managing filters and handlers.
Truss
Truss is an assertions micro-library for Clojure/Script by the author of Telemere.
Telemere can easily incorporate Truss assertion failure information in its signals, just like any other (error) data.
The catch->error! signal creator can be particularly convenient for this:
(t/catch->error! <form-with-truss-assertion/s>)