Communication Channels

Issue reports - FIL Slack: #bifrost-community

Team Members: @Cameron Wood @Nikolai Spasov

Team responsibilities


prd = prod

stg = staging

Area Footprint DRI / Co-DRI Gateways 142x nodes (prd)
4x node (stg)
Rhea / Bifrost-gw 14x nodes (prd)
1x node (stg)
Bootstrap nodes 6x nodes (prd) 7x nodes (prd)
JS Preload nodes 4x nodes (prd)
1x node (stg)
IPFS Check 1x node (prd)


Service Notes DRI / Co-DRI
Equinix Primary consumers of the Equinix account, and current maintainers of account access there. Will be transitioning away from this, as Equinix haven’t been willing to negotiate on pricing with us. 😒
@Cameron Wood
Gateway Metrics Metrics stack and dashboards for the Gateways. Currently, this consists of VictoriaMetrics, Elasticsearch, Kafka, Logstash, Loki, and Grafana. Running on an AWS EKS cluster, and leveraging blob storage from S3. @Nikolai Spasov
@Cameron Wood
Badbits Pipeline
i.e. abuse process This consists of the website at and a processing pipeline running on AWS. It uses a mixture of SES, SQS, S3, and a small service written in Go, that runs on our EKS cluster there. We run that pipeline in cooperation with the Legal team at PL, whom vet the requests received at [email protected] and forward them to the pipeline for processing.
@Cameron Wood

πŸ” Data access

As maintainers of the IPFS gateway infrastructure we collect and store various types of data generated by the systems - Kubo performance metrics and HTTP access logs. These can be accessed via the ProtocolLabs Grafana or programmatically via the respective APIs. For API access please contact member of the Bifrost team.

Alternatively, for data exploration of the HTTP access logs, stored in Elasticsearch, we maintain a Kibana interface - the login credentials are stored in 1Password Bifrost Elasticsearch/Kibana (read-only)


<aside> πŸ’‘ The data fields correspond to the variable names used by nginx. Reference:



πŸ“Š Gateway Report dashboard