Packet analysis lab 8 of 12

Application performance: server delay against network delay

Three slow applications cross the same 100 ms link, each slow for a different reason. You measure the network's round trip, then show that one page waits on the server, one waits on dozens of round trips, and one is held back by a small receive window, using http.time, iRTT, the I/O graph and the tcptrace graph.

Run this lab

It needs Lite: its 3 nodes are set to 3 vCPU and 2,560 MB in total. That fits the free demo too.

In your workspace: Shared, then Packet analysis with Wireshark and tcpdump guided labs, then 08 - Application performance - server delay against network delay

Try it free for 30 minutesRun it on Lite from £3 a day

The free demo needs no card and no email. Every lab plan includes every guided lab.

Level
Advanced
Time
About 70 minutes
Nodes need
3 vCPU, 2,560 MB
Smallest plan
Lite

The scenario

The Kendal shop says three things are "slow": the daily sales report page, the online catalogue on the kiosk, and the stock-sync tool that downloads the stock file. They all use the head-office application server SRV across a link with about 100 ms of round-trip time (WAN). There's no packet loss on this link, so "the network is slow" can only mean the round trip. Your job is to show, with packets, what each one is waiting for.

DeviceInterfaceAddressRole
WSeth010.90.0.10/24your workstation; runs curl, catalogue and stock-sync
WANeth0 ↔ eth1bridge, no addressadds 50 ms in each direction
SRVeth010.90.0.20/24the shop's web application

Your workstation, WS. Open WS's console (a remote desktop, root / eve), start Wireshark from Applications > Internet > Wireshark, and open a terminal. Write answers in /root/answers.txt (pluma /root/answers.txt &).

What you’ll practise

  • Separate server response time from network round-trip time in a capture
  • Show how round-trip time multiplies in chatty request and response applications
  • Explain window-limited throughput (window divided by round-trip time) and spot it in a stream graph
  • Use http.time, tcp.analysis.initial_rtt, tcp.analysis.ack_rtt and bytes in flight as columns and filters
  • Turn the evidence into a clear statement of who must fix what

The tasks

  1. Recreate the link
  2. Capture the three complaints
  3. Server delay or network delay?
  4. Death by round trips
  5. The window that holds back stock-sync

The Lab Tasks pane has the scenario, these tasks, the checks to run, hints and the full solution.

Topology and devices

The network as it opens in EVE-NG, drawn from the lab’s own canvas, and the image each node runs.

Topology of Application performance - server delay against network delay3 nodes, 2 links. Every link is listed as text below the diagram.WSWANSRVeth0eth0eth1eth0
3 nodes, with the interface names each link uses. On a narrow screen, scroll the diagram sideways.
Nodes in Application performance: server delay against network delay, with the image each runs and the vCPU and RAM it’s set to
NodeDeviceImage in Add NodeSet to
WSEVE-NG GUI server (Docker)eve-gui-server:latest1 vCPU, 1,024 MB
WANAlpine WAN emulatorclbalpinewanem-alpine-wanem-3-20-31 vCPU, 512 MB
SRVEVE-NG GUI server (Docker)eve-gui-server:latest1 vCPU, 1,024 MB

How to open this lab

  1. Sign in to your workspace at lab.cloudlabbox.com.

  2. In File Manager, open Shared, then the folder Packet analysis with Wireshark and tcpdump guided labs.

  3. Open 08 - Application performance - server delay against network delay and start its nodes, then open Lab Tasks for the scenario and the steps.

The lab is read-only. You can start the nodes and configure the devices, but changes to the topology can’t be saved. To change it, rebuild it in your own folder.

More Packet analysis labs

All Packet analysis labs

Related labs in other courses

CloudLabBox is independent of the Wireshark Foundation, CompTIA and Cisco. The lab follows exam topics and isn’t vendor training material; vendor names identify the exams and the images in the workspace.