Chronosphere
Send your logs, metrics, and traces to Chronosphere using the OpenTelemetry Protocol (OTLP).
Supported Telemetry Types
✓
✓
✓
Prerequisites
Before configuring the Chronosphere destination, ensure you have:
A Chronosphere account with tenant access
Your tenant name (e.g.,
mycompanyfrommycompany.chronosphere.io)A service account API token with write permissions for your telemetry types
For help creating a service account and API token, see the Chronosphere service accounts documentation.
Configuration
Basic Configuration
The minimum configuration requires your tenant name and API token.
Choose Telemetry Type
telemetrySelector
["Logs","Metrics","Traces"]
Select which telemetry types to export (Logs, Metrics, Traces).
Tenant Name
string
""
Your Chronosphere tenant subdomain (e.g., mycompany).
API Token
string
""
Service account API token with write permissions. Sensitive field.
Protocol
enum
http
OTLP protocol: http (recommended) or grpc. For gRPC, collector connects to <tenant>.chronosphere.io:443 over TLS.
Advanced Configuration
TLS Configuration
TLS is enabled by default and required for gRPC protocol.
Enable TLS
bool
true
Enable TLS for secure connections (required for gRPC).
Skip TLS Certificate Verification
bool
false
Skip certificate validation (not recommended in production).
TLS Certificate Authority File
string
""
Optional CA certificate file for server validation.
Mutual TLS
bool
false
Enable client TLS authentication.
TLS Client Certificate File
string
""
Client certificate file for mTLS.
TLS Client Private Key
string
""
Client private key file for mTLS.
HTTP Headers and Compression
Additional Headers
map
{}
Custom HTTP headers for requests (HTTP protocol only).
Compression
enum
zstd
Compression algorithm for HTTP OTLP: none, gzip, or zstd.
Batching Configuration
Chronosphere recommends specific batch settings to prevent payload size violations.
Batch Send Size
int
1000
Number of items to trigger a batch send.
Batch Max Size
int
2000
Maximum items per batch before splitting.
Batch Timeout
int
1
Seconds before sending incomplete batch.
Request Timeout
int
30
Seconds to wait for request completion.
Queuing and Retry
Enable Retry on Failure
bool
true
Retry failed requests with exponential backoff.
Initial Interval
int
5
Seconds to wait after first failure.
Max Interval
int
30
Maximum backoff interval between retries.
Max Elapsed Time
int
300
Maximum total retry time in seconds.
Enable Sending Queue
bool
true
Buffer telemetry in memory before sending.
Number of Consumers
int
10
Concurrent batch processors from queue.
Queue Size
int
5000
Maximum batches in memory queue.
Enable Persistent Queuing
bool
true
Buffer telemetry to disk during outages.
Persistent Queue Directory
string
${OIQ_OTEL_COLLECTOR_HOME}/storage
Directory for disk-based queue.
Synchronize Persistent Queue to Disk
bool
false
fsync queue writes (safer but slower).
Telemetry-Specific Settings
Metrics Validation Response Verbosity
enum
SHORT
Error detail level: SHORT, SUMMARY, or DETAILED.
Convert Resource Attributes to Telemetry Attributes
bool
false
Map resource attributes to telemetry attributes.
Examples
gRPC Configuration with mTLS
Use gRPC with mutual TLS for secure, high-performance connections in zero-trust or highly regulated environments. This configuration enables client certificate authentication.
High-Volume Scenario
For environments sending large volumes of telemetry, increase buffering and processing capacity by configuring larger queue sizes, more concurrent consumers, and persistent queuing for reliability.
Configuration Tips
Protocol Selection and Compression
HTTP (recommended): Broadly compatible, works through most firewalls and proxies. Use
zstdcompression for best balance.gRPC: Lower overhead and native binary protocol. Better for high-throughput scenarios but requires direct connectivity to port 443.
Compression: Use
zstdfor most cases; switch togzipornoneonly if CPU usage is a concern.
Queue Configuration for Reliability
Enable Persistent Queuing for mission-critical data paths to survive restarts and longer outages.
Queue Size: Start with 5000; increase to 10,000 for high-volume burst scenarios.
Number of Consumers: Use 10-20 for high-volume environments; adjust based on CPU availability.
Timeout and Batch Settings
Batch Size: Keep at 1000 items (Chronosphere recommendation) for optimal throughput.
Timeout: Use 30 seconds default; increase to 45-60 seconds for high-latency or intercontinental networks.
Batch Timeout: Set to 1 second for real-time responsiveness with reasonable batching efficiency.
Troubleshooting
Authentication and Connection Issues
Symptoms: "401 Unauthorized", "Invalid API Token", or connection timeouts.
Solutions:
Verify the API token is correct and has not expired in Chronosphere service accounts.
Confirm the token has write permissions for the telemetry types you're sending.
Test connectivity to
<tenant>.chronosphere.io:443:curl -I https://<tenant>.chronosphere.io/otlp(for HTTP) or usetelnet <tenant>.chronosphere.io 443(for gRPC).
Metrics Rejection and Validation
Symptoms: "Partial success" responses or metrics rejected with validation errors.
Solutions:
Set Metrics Validation Response Verbosity to
DETAILEDto see full error messages.Check for metrics exceeding Chronosphere limits: payload must be <16 MB compressed; metric items <10,000 per batch.
Review Chronosphere's incoming data dashboard for cardinality or formatting issues.
Data Not Arriving
Symptoms: No data appears in Chronosphere despite successful configuration.
Solutions:
Verify telemetry is being generated upstream (check source configuration logs).
Confirm the correct telemetry types are enabled in both BindPlane and your sources.
Check BindPlane and Chronosphere logs for validation errors or rate limiting.
Related Resources
Last updated
Was this helpful?