Sematext Agent for Docker collects Metrics, Events, and Logs from the Docker API for SPM Docker Monitoring & Logsene / Hosted ELK Log Management. You can do many things like these:
- ✔ Application Performance Monitoring
- ✔ Log Management
- ✔Container Monitoring
- ✔Metrics and Logs
- ❐ SPM’s over 40 integrations let you collect metrics and events across your whole stack.
- ❐Get started in minutes with out-of-the-box dashboards, alerts and anomaly detection rules
- ❐Slice and dice by a number of context-aware filters
- ❐Solr & SolrCloud, Elasticsearch, Spark, Storm, Kafka, Cassandra, HBase, MongoDB, Hadoop MapReduce, HDFS, YARN, Docker, CoreOS, Node.js, Express, Apache, Nginx (Plus), HAProxy, Redis, Memcached, MySQL, MariaDB, AWS EC2 ELB EBS RDS…
App & Database Monitoring
- ❐SPM goes beyond just collecting metrics and uncovers your slowest database transactions.
- ❐Find the slowest DB operation types
- ❐View full SQL statements
- ❐See end-to-end HTTP transaction context, through multiple tiers, including full call trace
- ❐Get top 10 operations by throughput, latency, or time consumed
- ❐Filter DB operations by type
- ❐App, Network & Infra Maps
- ❐SPM shows your whole infrastructure and how it’s all connected, in real-time.
- ❐View complete, self-organizing infrastructure & network maps
- ❐See throughput, latency, and error rates for all apps and servers
- ❐Get info about incoming and outgoing connections and receive/transmit rates
- ❐Multiple map views with zooming, panning, and filtering
Tracing & Profiling
SPM instruments your apps and captures your slowest transactions. The on-demand, low impact profiler finds bottlenecks in your code.
- See the end to end HTTP transactions
- Full HTTP call stack with all request parameters
- Annotate Java apps for custom method tracing
- Find slowest HTTP transaction components
- Customize tracing pointcuts
- Profile any Java app on demand
- Find methods using most CPU or wall-clock time
- Get top controllers by latency, throughput, or total time used
- See aggregate throughput, latency, errors/exception rates
Alerting & ChatOps
- SPM alerting integrates with ChatOps services and WebHooks.
- Alert on any metric and any filter combination
- Alerts based on: thresholds, anomaly detection, app heartbeat
- Get notified via email, ChatOps, or WebHooks
- Built-in integrations for PagerDuty, Slack, HipChat, and Nagios
- Works with VictorOps, OpsGenie, BigPanda, etc.
- Send Alerts events to any WebHook
- Pass custom HTTP parameters and headers to WebHooks
Correlate Metrics & Logs
Metrics are not enough! SPM correlates metrics with application and server logs, events, alerts, anomalies, and more. Really cool and empowering, and a huge time saver!
- Go from metric spikes to your apps’ and servers’ logs in seconds!
- Get to the root cause in minutes!
- Forget about ssh-ing to servers and grepping logs!
- See any metric any logs in a single unified same view!
- Find similar metric patterns with built-in metric correlation
APIs and Custom Metrics
- SPM is built by DevOps for DevOps. Custom Metrics? APIs? We’ve got ’em!
- Open HTTP APIs let you ship Custom Metrics from your apps
- Instrument your apps and send custom metrics
- Capture and send app, business, and other events
- Open-source metrics libraries for Java, Ruby, and Node.js
- Use APIs for alert automation and more
Sematext Tracing provides end to end visibility into your distributed applications so you can find bottlenecks quickly, resolve production issues faster and with less effort.
- End to end visibility from backend entry-point down to storage, and back up
- Support for fully distributed applications and distributed transaction tracing
- Bytecode instrumentation-based tracing for Java applications, including asynchronous calls to services
- Spot performance bottlenecks, identify hotspots, find the root cause of latency problems
- Support for OpenTracing API
- Integrated with Log Management, Infrastructure, and Real User Monitoring. Fewer siloed tools mean:
- Faster troubleshooting, shorter downtimes, faster MTTR
- Lower license, implementation, and integration costs
- Real-time service maps show the true application architecture
- Detect anomalies and receive alerts for latencies, error rates, or throughput
- The power of the ELK stack while relieving you of Elasticsearch management hassle, infrastructure, and high expert staff costs.
- ELK as a Service, Fully Managed
- Fully managed Elasticsearch in the Cloud. Elasticsearch API and fully integrated Kibana give you the power of the ELK stack while relieving you of Elasticsearch management hassle, infrastructure, and high expert staff costs.
- Ship data with your favorite data shipper – Logstash, Filebeat, rsyslog, Logagent, and many others
- Correlate logs with infrastructure and application metrics
- Extract business KPIs from logs to build rich reports and dashboards
- Automatic field and type detection with smart mappings and templates
Powerful Searching and Filtering
- If you know how to search with Google, you’ll know how to search your logs in Sematext Cloud.
- Use AND, OR, NOT operators – e.g. (error OR warn) NOT exception
- Group your AND, OR, NOT clauses – e.g. message:(exception OR error OR timeout) AND severity:(error OR warn)
- Don’t like Booleans? Use + and – to include and exclude – e.g. +message:error -message:timeout -host:db1.example.com)
- Use explicitly field references – e.g. message: timeout
- Need a phrase search? Use quotation marks – e.g. message:”fatal error”
- Syslog comes in many flavors of message formats, protocols, and daemons.
- There are copy-pastable configurations for all major Syslog daemons: traditional syslogd, rsyslog, syslog-ng and nxlog
- Send logs in traditional RFC-3164 message format or the newer RFC-5424
- Send syslog events as JSON over socket
- Traditional UDP syslog will work, of course, but you can also send data over TCP for more reliability. Even more, reliability is provided by RELP. If you need encryption, you can use TLS syslog.
Alerting on Logs
- You can alert on metrics, as well as logs.
- It’s simple: run a query, save it as the alert query, optionally select additional notification hooks like Slack, PagerDuty, etc.
- Use query filters to narrow down on logs you really want to alert on – e.g. severity: ERROR
- Use threshold-based alerts if you have specific thresholds and know what is alarming. Get notified when the number of logs hits the threshold.
- Use anomaly detection to be alerted on sudden abnormalities in your logs and let statistical algorithms find abnormal log counts.