unhappy with Zyte retry issues

If you’re unhappy with Zyte’s retry behavior, here are a few practical angles to look at:

1) Clarify what “retry issues” means

Common problems include:

• retries happening too many times
• retries not happening when they should
• duplicate requests or duplicate items
• delays caused by aggressive backoff
• retries masking real errors
• different behavior in Scrapy vs Zyte API vs Zyte Smart Proxy Manager

2) Check the retry layer involved

Zyte-related retries can come from multiple places:

• Scrapy RetryMiddleware
• Zyte API / Smart Proxy Manager
• your own custom retry logic
• downstream timeouts or network errors

If two layers are retrying, you can get unexpected behavior.

3) Adjust settings

Depending on your stack, you may want to review:

• retry count / max retry times
• HTTP status codes that trigger retries
• timeout values
• concurrency settings
• download delay / backoff
• proxy rotation behavior

4) Inspect logs

Look for:

• repeated status codes like 429, 403, 500, 502, 503
• timeout errors
• proxy errors
• “retrying” messages from Scrapy or Zyte
• whether retries are due to transient failures or site blocking

5) Prevent duplicate processing

If retries are causing duplicate items:

• make item processing idempotent
• deduplicate by URL, SKU, or external ID
• use request fingerprints carefully
• store retry state explicitly if needed

6) If you want to raise it with Zyte

A good support message should include:

• account/project name
• exact feature involved
• sample request IDs or logs
• what you expected vs what happened
• timestamps and target URLs
• whether this happens consistently or only on certain domains

If you want, I can help you draft:

1. a complaint/support email to Zyte, or
2. a troubleshooting checklist for your specific retry setup.