Hii Team

We’re using GoReplay v1.3 to mirror live traffic from our production environment in India to a newly deployed system in the US (both systems are identical in setup and configuration).

However, we consistently observe higher response times for replayed traffic on the US system only when using GoReplay. When sending the same traffic using multi-threaded Python scripts or load testing tools like JMeter, the response times are significantly lower.

Note: Response time is measured at the system level (from request arrival to response departure), so network latency is not a factor in this observation.

GoReplay Command Used:

sudo ./gor \
  --input-raw :8983 \
  --input-raw-bpf-filter "dst port 8983 and inbound" \
  --input-raw-allow-incomplete \
  --http-allow-url="/app/appPath" \
  --input-raw-buffer-size 10485760 \
  --output-http "http://<US-IP>:8983"

We’re capturing and replaying traffic to a specific application-level path exposed by our service.

We’ve also observed similar behaviour in another project targeting the same geo-location infrastructure; normal traffic behaves well, but GoReplay traffic shows degraded performance.

• Could GoReplay be replaying traffic unevenly or introducing delays internally??
• Are there flags or best practices for optimising replay to remote regions (e.g., high-latency links)??
• Is this a known behaviour in GoReplay v1.3??

Any insights would be appreciated. Thanks!!!