Skip to content

fix(feed): Improve NVD feed download resilience #3343

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

Merged
dcaravel merged 1 commit into from
Jun 30, 2026
+56 −17
Merged

fix(feed): Improve NVD feed download resilience #3343

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
2 changes: 1 addition & 1 deletion pkg/vulnloader/nvdloader/loader_api.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,7 @@ import (
const urlFmt = `https://services.nvd.nist.gov/rest/json/cves/2.0?noRejected&startIndex=%d`

var client = http.Client{
Timeout: 5 * time.Minute,
Timeout: 6 * time.Minute,
Transport: proxy.RoundTripper(),
}

Expand Down
71 changes: 55 additions & 16 deletions pkg/vulnloader/nvdloader/loader_feed.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,7 @@ import (
"compress/gzip"
"encoding/json"
"fmt"
"io"
"os"
"path/filepath"
"time"
Expand Down Expand Up @@ -48,26 +49,27 @@ func (l *feedLoader) DownloadFeedsToPath(outputDir string) error {

func (l *feedLoader) downloadFeedForYear(enrichments map[string]*FileFormatWrapper, outputDir string, year int) error {
url := fmt.Sprintf("https://nvd.nist.gov/feeds/json/cve/2.0/nvdcve-2.0-%d.json.gz", year)
resp, err := client.Get(url)
if err != nil {
return errors.Wrapf(err, "failed to download feed for year %d", year)
}
defer utils.IgnoreError(resp.Body.Close)

// Un-gzip it.
gr, err := gzip.NewReader(resp.Body)
if err != nil {
return errors.Wrapf(err, "couldn't read resp body for year %d", year)
}

apiFeed := new(apischema.CVEAPIJSON20)
if err := json.NewDecoder(gr).Decode(apiFeed); err != nil {
return errors.Wrapf(err, "decoding feed response")
const maxRetries = 5
backoff := 10 * time.Second
var apiFeed *apischema.CVEAPIJSON20
for attempt := 1; ; attempt++ {
var err error
apiFeed, err = fetchFeed(url, year)
if err == nil {
break
}
if attempt >= maxRetries {
return errors.Wrapf(err, "failed to download feed for year %d after %d attempts", year, attempt)
}
log.Warnf("Feed year %d: attempt %d failed: %v; retrying in %s", year, attempt, err, backoff)
time.Sleep(backoff)
backoff *= 2
Comment on lines +56 to +67

@coderabbitai coderabbitai Bot Jun 30, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🩺 Stability & Availability | 🟠 Major | ⚡ Quick win

Stop retrying permanent HTTP client errors.

fetchFeed turns every non-200 into the same generic error, so downloadFeedForYear also retries 400/401/403/404 responses. With a 6-minute client timeout plus 10→80s backoff, one permanent 4xx can stall a single year for 30+ minutes before failing. Please classify errors so only transport failures and retryable statuses (for example 429/5xx) loop.

Also applies to: 101-103

🤖 Prompt for AI Agents 
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@pkg/vulnloader/nvdloader/loader_feed.go` around lines 56 - 67,
downloadFeedForYear is retrying all failures from fetchFeed, including permanent
4xx responses, so classify errors in fetchFeed and only retry transport errors
and retryable HTTP statuses like 429/5xx. Update the retry loop in
downloadFeedForYear to inspect the error type/status before sleeping, and
preserve the existing maxRetries/backoff behavior for transient failures while
failing fast on non-retryable client errors.

Source: Path instructions

}

cveItems, err := toJSON10(apiFeed.Vulnerabilities)
if err != nil {
return fmt.Errorf("failed to convert feed vulns to JSON: %w", err)
return fmt.Errorf("failed to convert feed vulns to JSON for year %d: %w", year, err)
}

enrichCVEItems(&cveItems, enrichments)
Expand All @@ -76,8 +78,45 @@ func (l *feedLoader) downloadFeedForYear(enrichments map[string]*FileFormatWrapp
CVEItems: cveItems,
}
if err := writeFile(filepath.Join(outputDir, fmt.Sprintf("%d.json", year)), feed); err != nil {
return errors.Wrapf(err, "writing to file")
return errors.Wrapf(err, "writing to file for year %d", year)
}

log.Infof("Feed year %d: completed with %d vulnerabilities", year, len(cveItems))
return nil
}

func fetchFeed(url string, year int) (*apischema.CVEAPIJSON20, error) {
log.Infof("Downloading NVD feed for year %d from %s", year, url)

start := time.Now()
resp, err := client.Get(url)
if err != nil {
return nil, errors.Wrapf(err, "HTTP request failed (elapsed: %s)", time.Since(start))
}
defer utils.IgnoreError(resp.Body.Close)

log.Infof("Feed year %d: HTTP %d, Content-Length: %d, Proto: %s (connect took %s)",
year, resp.StatusCode, resp.ContentLength, resp.Proto, time.Since(start))

if resp.StatusCode != 200 {
return nil, fmt.Errorf("unexpected status code %d", resp.StatusCode)
}

gr, err := gzip.NewReader(resp.Body)
if err != nil {
return nil, errors.Wrapf(err, "creating gzip reader")
}

body, err := io.ReadAll(gr)
if err != nil {
return nil, errors.Wrapf(err, "reading feed body (read %d bytes, elapsed: %s)", len(body), time.Since(start))
}
log.Infof("Feed year %d: read %d decompressed bytes (elapsed: %s)", year, len(body), time.Since(start))

apiFeed := new(apischema.CVEAPIJSON20)
if err := json.Unmarshal(body, apiFeed); err != nil {
return nil, errors.Wrapf(err, "decoding feed JSON (%d bytes)", len(body))
}

return apiFeed, nil
}
Loading