Skip to main content
Glama

Convex MCP server

Official
by get-convex
consts.rs2.58 kB
use std::time::Duration; // LogManager pub const MANAGER_INITIAL_BACKOFF: Duration = Duration::from_secs(1); pub const MANAGER_MAX_BACKOFF: Duration = Duration::from_secs(30); pub const SINK_STARTUP_TIMEOUT: Duration = Duration::from_secs(15); // MockSink #[cfg(any(test, feature = "testing"))] pub const MOCK_SINK_EVENTS_BUFFER_SIZE: usize = 5; #[cfg(any(test, feature = "testing"))] pub const MOCK_SINK_INITIAL_BACKOFF: Duration = Duration::from_secs(1); #[cfg(any(test, feature = "testing"))] pub const MOCK_SINK_MAX_BACKOFF: Duration = Duration::from_secs(10); // LocalSink pub const LOCAL_SINK_EVENTS_BUFFER_SIZE: usize = 50; pub const LOCAL_SINK_INITIAL_BACKOFF: Duration = Duration::from_secs(1); pub const LOCAL_SINK_MAX_BACKOFF: Duration = Duration::from_secs(10); // Datadog pub const DD_SINK_EVENTS_BUFFER_SIZE: usize = 8; pub const DD_SINK_INITIAL_BACKOFF: Duration = Duration::from_millis(500); pub const DD_SINK_MAX_BACKOFF: Duration = Duration::from_secs(60); pub const DD_SINK_MAX_REQUEST_ATTEMPTS: usize = 6; /// We currently limit logs to `MAX_LOG_LINE_LENGTH` (4kb) and Datadog has a 200 /// log and 1MB limit per batch. Thus, this batch size makes the payload size /// 800kb so we still have 200kb legroom per batch for system fields. /// https://docs.datadoghq.com/agent/logs/log_transport/?tab=https pub const DD_SINK_MAX_LOGS_PER_BATCH: usize = 200; // Axiom pub const AXIOM_SINK_EVENTS_BUFFER_SIZE: usize = 8; pub const AXIOM_SINK_INITIAL_BACKOFF: Duration = Duration::from_millis(500); pub const AXIOM_SINK_MAX_BACKOFF: Duration = Duration::from_secs(60); pub const AXIOM_SINK_MAX_REQUEST_ATTEMPTS: usize = 6; /// This is Axiom's hard limit: https://axiom.co/docs/send-data/ingest#limits /// In practice, this is impossible to hit in one batch since the LogManager /// aggregation recv buffer size (LOG_MANAGER_EVENT_RECV_BUFFER_SIZE) is /// controlled by a knob which, by default, is much less than this. pub const AXIOM_SINK_MAX_LOGS_PER_BATCH: usize = 10000; // Webhook pub const WEBHOOK_SINK_EVENTS_BUFFER_SIZE: usize = 8; pub const WEBHOOK_SINK_INITIAL_BACKOFF: Duration = Duration::from_secs(1); pub const WEBHOOK_SINK_MAX_BACKOFF: Duration = Duration::from_secs(60); pub const WEBHOOK_SINK_MAX_REQUEST_ATTEMPTS: usize = 6; pub const WEBHOOK_SINK_MAX_LOGS_PER_BATCH: usize = 128; // Sentry pub const SENTRY_SINK_EVENTS_BUFFER_SIZE: usize = 8; pub const SENTRY_SINK_INITIAL_BACKOFF: Duration = Duration::from_millis(500); pub const SENTRY_SINK_MAX_BACKOFF: Duration = Duration::from_secs(60); pub const SENTRY_SINK_MAX_LOGS_PER_BATCH: usize = 100;

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/get-convex/convex-backend'

If you have feedback or need assistance with the MCP directory API, please join our Discord server