Documentation
LiveSPInstallation Operating GuideDownload PDF
Monitoring : Data Collection : Check 4 – Netflow data collection works fine
Check 4 – Netflow data collection works fine
Process
API endpoint: /api/v1/collect/status//NETFLOW
CUSTOMER_ID: identifier of the customer tenant from provisioning
Expected: status value should equal to HEALTHY
Severity: High
Services involved in the Netflow collection chain are:
livesp_flower-balancer
livesp_flower-node
livesp_flower-master
livesp_esus-collector
livesp_esus-farcaster*
Support action
1. Are all these services up and running?
Source: See Services
If yes: Fix this first
2. Was there a recent degradation and when did it start?
Source: ELK dashboard Netflow Analysis (last 4, 12 and 48 hours)
Expected observation on graphs:
Good stability of the number of realms (total, and per Farcaster node)
Relative stability of the number of lines in bridge
No alert regarding CPU load, memory load and disk space
Capture snapshots, summarize observations
3. Is Netflow bridge collection working fine for the customer?
Source: Run command checkFlowsCollectedFor
Expected outputs: JSON element with a lineCount property greater than 0
{
  'clientId':'2314459',
  'date':'2020-03-28-01-00',
  'lineCount':10178,
  'hostname':'flowcollector2',
  'ipaddress':'10.154.129.222',
  'bridgePath':'/data/files/bridge/flows/traffic/5minutes/data-2020-03-28-01-00-00'
}
If no (lineCount = 0): The source of the problem is on Netflow packet collection side
Select 1 or 2 CPEs with viewpoints for this customer (LiveSP inventory screen)
Any error or warn in flower-node logs filtered on this CPE IP address?
Enable debug mode for this CPE: any data after 10 minutes?
If no to all above: capture Netflow UDP packets with tcpdump on this CPE
If yes (lineCount > 0): The source of the problem is on Memdex collection side.
Any error in esus-collector logs?
Example
Request URL: https://mylivesp.com/api/v1/collect/status/7/NETFLOW
Response JSON content:
{
  "id": "netflow",
  "name": "NETFLOW",
  "type": "TECHNOLOGY",
  "status": "HEALTHY",
  "message": "Collect is healthy for technology NETFLOW"
}