Diagnose proxy errors at the right layer

Identify which request actually failed

A proxy workflow has a management API, a connection to the proxy and a request from the proxy to the destination. An error at one layer does not prove a failure at the others. Record whether the failure came from creating or reading the Lupin resource, establishing proxy authentication or receiving destination content. Keep credentials redacted. A successful resource creation is not proof that a particular client can connect, and a destination rejection does not mean the proxy user must be recreated. Layered diagnosis prevents unnecessary paid allocations while preserving the evidence needed for an operator to act.

Fix management validation and authority first

Lupin validates usernames, thread counts, sticky_days and nonzero manual top-up amounts. Creation and traffic changes require proxy_write, while reads require proxy_read. Human verification, billing readiness and a configured live provider remain additional gates; a bootstrap credential cannot allocate the resource. A validation error should be fixed against the contract rather than retried unchanged. proxy_upstream_error indicates a provider-side management call failed, not necessarily that existing proxy connections are unusable. Preserve the original Idempotency-Key when retrying the same allocation operation so an uncertain response does not duplicate traffic or users.

Inspect authentication and account constraints

For HTTP proxy connections, authentication-related responses can also represent exhausted traffic or thread capacity, not only a mistyped password. Read the sanitized provider reason and compare it to the configured endpoint and client protocol. Do not log a full credential-bearing URL to make the comparison easier. A protocol mismatch can fail before the destination is contacted. Thread settings describe concurrent connection capacity and do not guarantee request throughput. Reduce unnecessary concurrency when appropriate, but do not silently buy more traffic or create another user without establishing which constraint caused the failure.

curl --proxy "http://$PROXY_HOST:$PROXY_PORT" \
  --proxy-user "$PROXY_USER:$PROXY_PASSWORD" \
  --connect-timeout 10 --max-time 20 \
  --output /dev/null --write-out "%{http_code}\n" \
  https://example.com

Distinguish destination and availability failures

A target connection failure can mean the destination is unreachable from the proxy route, while an availability failure can mean no endpoint matches the requested targeting. A site or port can also be blocked under provider policy. These conditions should not be treated as permission to bypass access restrictions. Use a small approved diagnostic request and compare with the intended configuration. Keep target geography, rotation expectations and TLS behavior explicit. Do not disable certificate checks to hide an unrelated connection error, and do not interpret a reachable error page as successful retrieval of the required source.

Reconcile state before the next spend

Reading an active proxy user can synchronize traffic and trigger its configured refill, so a diagnostic read itself may relate to paid allocation. Follow the approved refill policy and account for possible stale data when synchronization fails. A proxy.pool.low event can signal a dry pool or a broader refill problem. Investigate the cause rather than assuming a purchase is the only remedy. Once evidence identifies a safe retry or configuration correction, record that reasoning and keep the request bounded. If the outcome remains uncertain, stop dependent work and give the operator the resource identifier and sanitized failure instead of a speculative success claim.

Sources & verification

Reviewed against the current Lupin route, provider adapter and shared input contracts, plus the linked official references. Examples were not run against live services.

• docs.dataimpulse.com · errors ↗Checked 2026-11-05
• curl.se · docs · manpage.html ↗Checked 2026-11-05

Prepared with AI assistance by Lupin editorial. How we research and maintain these guides.