Mcrouter (pronounced mc router) is a memcached protocol router for scaling memcached deployments. It's a component of the cache infrastructure at Facebook and Instagram and is a tool in the Memcached stack.

Why use a Telegraf plugin for Mcrouter?

The Mcrouter Input Plugin gathers statistics data from a mcrouter instance to help you ensure the uptime and performance of your infrastructure.

How to use the Mcrouter Telegraf Plugin

You can configure the Mcrouter Telegraf Plugin to gather metrics from one or many mcrouter servers. You can also set a timeout for the metric collection.

Key Mcrouter metrics to use for monitoring

Some of the important Mcrouter metrics that you should proactively monitor include:

  • uptime
  • num_servers
  • num_servers_new
  • num_servers_up
  • num_servers_down
  • num_servers_closed
  • num_clients
  • num_suspect_servers
  • destination_batches_sum
  • destination_requests_sum
  • outstanding_route_get_reqs_queued
  • outstanding_route_update_reqs_queued
  • outstanding_route_get_avg_queue_size
  • outstanding_route_update_avg_queue_size
  • outstanding_route_get_avg_wait_time_sec
  • outstanding_route_update_avg_wait_time_sec
  • retrans_closed_connections
  • destination_pending_reqs
  • destination_inflight_reqs
  • destination_batch_size
  • asynclog_requests
  • proxy_reqs_processing
  • proxy_reqs_waiting
  • client_queue_notify_period
  • rusage_system
  • rusage_user
  • ps_num_minor_faults
  • ps_num_major_faults
  • ps_user_time_sec
  • ps_system_time_sec
  • ps_vsize
  • ps_rss
  • fibers_allocated
  • fibers_pool_size
  • fibers_stack_high_watermark
  • successful_client_connections
  • duration_us
  • destination_max_pending_reqs
  • destination_max_inflight_reqs
  • retrans_per_kbyte_max
  • cmd_get_count
  • cmd_delete_out
  • cmd_lease_get
  • cmd_set
  • cmd_get_out_all
  • cmd_get_out
  • cmd_lease_set_count
  • cmd_other_out_all
  • cmd_lease_get_out
  • cmd_set_count
  • cmd_lease_set_out
  • cmd_delete_count
  • cmd_other
  • cmd_delete
  • cmd_get
  • cmd_lease_set
  • cmd_set_out
  • cmd_lease_get_count
  • cmd_other_out
  • cmd_lease_get_out_all
  • cmd_set_out_all
  • cmd_other_count
  • cmd_delete_out_all
  • cmd_lease_set_out_all

For more information, please check out the documentation.

Project URL   Documentation

Related resources

Monitoring Network Interface Template

Monitoring network interface performance is important in identifying the root cause of performance degradation and network bottlenecks.

Network Performance Monitoring

This solution blueprint describes how you can maintain a responsive and performant network no matter how distributed your resources are.

Cisco NX-OS and InfluxData for Pervasive Network Visibility

This webinar shows how the Cisco Nexus data center provides telemetry monitoring built on Cisco NX-OS, Telegraf and InfluxDB.

Scroll to Top