v0.4.0

What's new?

• Added retry mechanism
  • Enabled by default
  • Retries 3 times with 100 ms backoff
  • Configure using retryCount and retryBackoff options
• Allow completely silencing errors
  • By default, exceptions are still caught and logged to standard output
  • Use ignoreExceptions: true to silently ignore exceptions
  • Use throwExceptions: true to throw exceptions
• Make @logtail/node a regular dependency for @logtail/winston and @logtail/bunyan

Got questions?
Please let us know at hello@betterstack.com.
We're happy to help! 🙏

Full Changelog: v0.3.0...v0.4.1

v0.3.0

What's new

• Changed how we determine automatically added context system.main_file and runtime.file. You could see some change in these values depending on your setup. This was done to prevent the library from producing warnings in some cases.
• Extended support for log levels. Logtail Node and some frameworks now support sending custom log levels to Logtail.
• More useful logs are sent to Logtail when the log context is recursive or very deep. You can Prevent Logtail from producing warnings:

const logtail = new Logtail(token, {
    contextObjectCircularRefWarn: false,
    contextObjectMaxDepthWarn: false
});

• Minor fixes for Pino and .debug() loging method.

Full Changelog: v0.2.0...v0.3.0

v0.2.0

Changelog:

• Added logtail.flush() method that sends all currently batched logs to Logtail without waiting for the batching interval.
  • Just call logtail.flush() before the process exit or at the end of serverless functions.
  • Pino calls logtail.flush() automatically so you don't have to.

Want to see code examples? Check out Quick start guide ⇗

Got questions? Please let us know at hello@betterstack.com. We're happy to help!