Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

added more detailed logs around ES communication failure #992

Merged
merged 2 commits into from
Dec 28, 2021
Merged

added more detailed logs around ES communication failure #992

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
cb9bfa7
added more detailed logs around ES communication failure
lykkin Dec 13, 2021
1e55584
pinned clarifying comment to a version tag
lykkin Dec 28, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 12 additions & 0 deletions cmd/fleet/error.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,7 @@ import (
"encoding/json"
"net/http"
"os"
"strings"

"github.com/elastic/fleet-server/v7/internal/pkg/dl"
"github.com/elastic/fleet-server/v7/internal/pkg/limit"
Expand Down Expand Up @@ -144,6 +145,17 @@ func NewErrorResp(err error) errResp {
}
}

// Check if we have encountered a connectivity error
// Predicate taken from https://github.com/golang/go/blob/go1.17.5/src/net/dial_test.go#L798
if strings.Contains(err.Error(), "connection refused") {
return errResp{
http.StatusServiceUnavailable,
"ServiceUnavailable",
"Fleet server unable to communicate with Elasticsearch",
zerolog.InfoLevel,
}
}

// Default
return errResp{
StatusCode: http.StatusBadRequest,
Expand Down
13 changes: 13 additions & 0 deletions internal/pkg/coordinator/monitor.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -128,11 +128,17 @@ func (m *monitorT) Run(ctx context.Context) (err error) {
// Start timer loop to ensure leadership
lT := time.NewTimer(m.checkInterval)
defer lT.Stop()

// Keep track of errored statuses
erroredOnLastRequest := false
numFailedRequests := 0
for {
select {
case hits := <-s.Output():
err = m.handlePolicies(ctx, hits)
if err != nil {
erroredOnLastRequest = true
numFailedRequests++
m.log.Warn().Err(err).Msgf("Encountered an error while policy leadership changes; continuing to retry.")
}
case <-mT.C:
Expand All @@ -141,13 +147,20 @@ func (m *monitorT) Run(ctx context.Context) (err error) {
case <-lT.C:
err = m.ensureLeadership(ctx)
if err != nil {
erroredOnLastRequest = true
numFailedRequests++
m.log.Warn().Err(err).Msgf("Encountered an error while checking/assigning policy leaders; continuing to retry.")
}
lT.Reset(m.checkInterval)
case <-ctx.Done():
m.releaseLeadership()
return ctx.Err()
}
if err == nil && erroredOnLastRequest {
erroredOnLastRequest = false
m.log.Info().Msgf("Policy leader monitor successfully recovered after %d attempts", numFailedRequests)
numFailedRequests = 0
}
}
}

Expand Down