[#307] add package trace, integrate it with logging, and adopt it throughout zrepl

package trace:

- introduce the concept of tasks and spans, tracked as linked list within ctx
    - see package-level docs for an overview of the concepts
    - **main feature 1**: unique stack of task and span IDs
        - makes it easy to follow a series of log entries in concurrent code
    - **main feature 2**: ability to produce a chrome://tracing-compatible trace file
        - either via an env variable or a `zrepl pprof` subcommand
        - this is not a CPU profile, we already have go pprof for that
        - but it is very useful to visually inspect where the
          replication / snapshotter / pruner spends its time
          ( fixes #307 )

usage in package daemon/logging:

- goal: every log entry should have a trace field with the ID stack from package trace

- make `logging.GetLogger(ctx, Subsys)` the authoritative `logger.Logger` factory function
    - the context carries a linked list of injected fields which
      `logging.GetLogger` adds to the logger it returns
    - `logging.GetLogger` also uses package `trace` to get the
      task-and-span-stack and injects it into the returned logger's fields
This commit is contained in:
Christian Schwarz
2020-04-11 15:49:41 +02:00
parent bcb5965617
commit 10a14a8c50
75 changed files with 1934 additions and 462 deletions
+14 -5
View File
@@ -6,6 +6,7 @@ import (
"github.com/pkg/errors"
"github.com/prometheus/client_golang/prometheus"
"github.com/zrepl/zrepl/daemon/logging/trace"
"github.com/zrepl/zrepl/config"
"github.com/zrepl/zrepl/daemon/filters"
@@ -164,12 +165,14 @@ func (j *PassiveSide) SenderConfig() *endpoint.SenderConfig {
func (*PassiveSide) RegisterMetrics(registerer prometheus.Registerer) {}
func (j *PassiveSide) Run(ctx context.Context) {
ctx, endTask := trace.WithTaskAndSpan(ctx, "passive-side-job", j.Name())
defer endTask()
log := GetLogger(ctx)
defer log.Info("job exiting")
ctx = logging.WithSubsystemLoggers(ctx, log)
{
ctx, cancel := context.WithCancel(ctx) // shadowing
ctx, endTask := trace.WithTask(ctx, "periodic") // shadowing
defer endTask()
ctx, cancel := context.WithCancel(ctx)
defer cancel()
go j.mode.RunPeriodic(ctx)
}
@@ -179,8 +182,14 @@ func (j *PassiveSide) Run(ctx context.Context) {
panic(fmt.Sprintf("implementation error: j.mode.Handler() returned nil: %#v", j))
}
ctxInterceptor := func(handlerCtx context.Context) context.Context {
return logging.WithSubsystemLoggers(handlerCtx, log)
ctxInterceptor := func(handlerCtx context.Context, info rpc.HandlerContextInterceptorData, handler func(ctx context.Context)) {
// the handlerCtx is clean => need to inherit logging and tracing config from job context
handlerCtx = logging.WithInherit(handlerCtx, ctx)
handlerCtx = trace.WithInherit(handlerCtx, ctx)
handlerCtx, endTask := trace.WithTaskAndSpan(handlerCtx, "handler", fmt.Sprintf("job=%q client=%q method=%q", j.Name(), info.ClientIdentity(), info.FullMethod()))
defer endTask()
handler(handlerCtx)
}
rpcLoggers := rpc.GetLoggersOrPanic(ctx) // WithSubsystemLoggers above