For the complete documentation index, see llms.txt. This page is also available as Markdown.

Pebble

Overview

Pebble is a key-value storage engine that provides efficient data buffering for the Bindplane OTel Collector. It offers good performance with minimal configuration and is the recommended default for most production deployments.

Under the hood, Pebble Persistent Queue Storage is the Bindplane pebble extension.

Configuration

The following parameters control Pebble storage behavior:

Parameter
Type
Default
Description

Persistent Queue Directory

string

${OIQ_OTEL_COLLECTOR_HOME}/storage

Directory where Pebble stores data files

Synchronize Persistent Queue to Disk

bool

true

Whether to sync writes immediately to disk for durability

Cache Size

int

8 MB

Size in bytes of the block cache. When 0, uses Pebble's default cache behavior of 8 MB. Larger values improve read performance at the cost of memory usage.

When to Use Pebble

Use Pebble for:

  • Single collector deployments

  • Standard gateway deployments

  • Edge collectors and agent deployments

  • When configuration simplicity is preferred

Monitoring

Monitor these indicators after configuring Pebble:

Disk Usage

  • Growth is normal and proportional to data volume

  • Pebble handles cleanup automatically

  • Ensure sufficient disk space remains available

Processing Latency

  • Should be consistent without significant spikes

  • If degrading, check if at capacity for current data volume

Memory Usage

  • Should remain stable and predictable

  • If growing unexpectedly, adjust Cache Size if configured

Troubleshooting

Performance degradation:

  • Check if data volume exceeds current capacity

  • Monitor CPU and disk I/O utilization

High disk usage:

  • Normal with high-throughput deployments

  • Verify adequate disk space remains available

Next Steps

  1. If switching to or from Pebble, follow the Switching Queues guide to avoid data loss.

  2. Monitor per-collector throughput, memory usage, and disk space.

For help choosing your persistent queue storage, see the Persistent Queue guide.

Last updated

Was this helpful?