ºÝºÝߣ

ºÝºÝߣShare a Scribd company logo

Monitoring @haptik

•
•
Vishal Uderani
Vishal Uderani

The document summarizes the monitoring infrastructure of haptik.co, which is hosted on AWS across multiple availability zones in the Singapore region. It monitors instances using New Relic to detect capacity and health issues, application performance using APM to track transactions and queries, and AWS resources like RDS databases and ELB load balancers using Cloudwatch metrics. Over 15 metrics are tracked for each RDS instance and over 10 for each ELB.

1 of 8
Download to read offline
Monitoring @haptik.co
A little bit about the infrastructure • Hosted in AWS • Production in Singapore region with multiple AZ • 8 instances and growing steadily
And now the gory details
Instance Monitoring • Newrelic based monitoring • Detects capacity issues so corrective action can be taken • Alerts for server health issues: • CPU utilization, memory utilization, • disk I/O utilization and disk capacity
APM(Application monitoring) • Used to follow the performance of critical transactions in applications • Drill down view of specific code segments and SQL queries • Flag critical transactions to spot response times , call counts or errors
Amazon Cloudwatch • collect and track metrics • set alarms for AWS resources • used for ELB and RDS 
RDS Monitoring • Over 15 metrics tracked per db-instance • Metrics include : BinLogDiskUsage,CPU Util • DbConnections,FreeStorage/Memory • NetworkRx/Tx,ReadIOPS/Latency
ELB(Load Balancer) Monitoring • Over 10 metrics tracked per ELB  • Metrics include: BackendConnErr, • HTTP_CODE_2xx/3xx/4xx/5xx • HealthyHostCount,Latency, •

More Related Content

Monitoring @haptik

  • 2. A little bit about the infrastructure • Hosted in AWS • Production in Singapore region with multiple AZ • 8 instances and growing steadily
  • 3. And now the gory details
  • 4. Instance Monitoring • Newrelic based monitoring • Detects capacity issues so corrective action can be taken • Alerts for server health issues: • CPU utilization, memory utilization, • disk I/O utilization and disk capacity
  • 5. APM(Application monitoring) • Used to follow the performance of critical transactions in applications • Drill down view of specific code segments and SQL queries • Flag critical transactions to spot response times , call counts or errors
  • 6. Amazon Cloudwatch • collect and track metrics • set alarms for AWS resources • used for ELB and RDS 
  • 7. RDS Monitoring • Over 15 metrics tracked per db-instance • Metrics include : BinLogDiskUsage,CPU Util • DbConnections,FreeStorage/Memory • NetworkRx/Tx,ReadIOPS/Latency
  • 8. ELB(Load Balancer) Monitoring • Over 10 metrics tracked per ELB  • Metrics include: BackendConnErr, • HTTP_CODE_2xx/3xx/4xx/5xx • HealthyHostCount,Latency, •