DOCA Telemetry Service Guide

This guide provides instructions on how to use the DOCA Telemetry Service (DTS) container on top of hosts or NVIDIA® BlueField® DPU.

New Features

• Extended the OpenTelemetry exporter to support gRPC transport, as well as TLS/mTLS encryption across both HTTP and gRPC transports.

• Added support for the new SDK APIs (via IPC to DTS), specifically the Metrics API and the OTLP logs API.

• Enabled dpeserver support on RPM-based operating systems for BlueField deployments.

Bug Fixes

• BlueField inventory collection:

  • Resolved an issue that caused corrupted data points with identical timestamps (e.g., "data0", "data1") when inventory collection was enabled.

  • Corrected IPv4 data collection to reliably target the oob_net0 network interface.

• ppcc_eth provider: Implemented automatic detection for the libmlxreg_sdk.so library location on the host. This resolves pathing failures in Arm images caused by recent directory changes.

The DOCA Telemetry Service (DTS) delivers real-time system and workload metrics for BlueField platforms and supported host systems. It aggregates data from both built-in telemetry providers and external applications utilizing the DOCA Telemetry Exporter API library.

Supported Providers

DTS collects and aggregates metrics through the following supported providers:

Data Providers

• sysfs

  Note

  Enabled by default on BlueField platforms

• ethtool

• Advanced monitoring and bringup errors (amber)

• Port programmable congestion control (ppcc_eth)

• Diagnostic data

• ifconfig

• RDMA notifications

• Data Center GPU Manager (DCGM)

• NVIDIA System Management Interface (nvidia-smi)

• BlueField performance (bfperf)

• Traffic control (tc)

Aggregation Providers

• Fluent aggregator

• Prometheus aggregator

Data Export Options

DTS supports multiple mechanisms for exporting and storing collected telemetry data.

Export Mechanisms

• Prometheus endpoint (Pull)

• Fluent Bit (Push)

• OpenTelemetry (Push)

• Prometheus Remote Write (Push)

Local Storage Export

DTS can store collected telemetry as binary files in the /opt/mellanox/doca/services/telemetry/data directory.

• Default state: Data writing is disabled by default.

• Data retrieval: Use the clx_read binary (included in the collectx-clxapidev Linux package) to parse and read these binary files.

DTS is available as a built-in service on BlueField DPUs and as a container image on NVIDIA NGC for hosts.

Since DTS does not monitor all changes related to firmware, networking, firewall settings, or other system components, it is strongly recommended (and in some cases required) to restart DTS after any server configuration changes to ensure proper operation.

Built-in DOCA Service Image (BlueField)

DOCA Telemetry Service is included by default in the BlueField DPU image. Each DPU image ships with a fixed version of the service, providing out-of-the-box support for applications using the DOCA Telemetry Exporter library.

DOCA Telemetry Exporter Integration

Applications using the DOCA Telemetry Exporter API can leverage DTS Inter-Process Communication (IPC) to export collected telemetry data through DTS.

The export modules available in DTS are also integrated into the DOCA Telemetry Exporter. Both solutions support decoupling telemetry collection from export configuration and allow data to be exported either directly from the library (e.g., to Prometheus or OpenTelemetry) or indirectly via DTS using IPC.

DOCA Service on NGC

In addition to the built-in image, updated DTS versions are available on NVIDIA NGC. This is useful if you need to upgrade to a newer version than the one bundled with the BlueField OS. Refer to the container's NGC page for service-specific configuration and deployment steps.

DPU Deployment

On BlueField DPUs, DTS starts automatically during boot using the configuration defined in:

            

            
/etc/kubelet.d/doca_telemetry_standalone.yaml
        

• To disable auto-start, remove the YAML file from this path

• Changes to the YAML file are applied on system restart

DTS file location: /opt/mellanox/doca/services/telemetry/

• Container volume mounts:

  • config/

  • data/

  • ipc_sockets/

• Backup artifacts:

  • doca_telemetry_service_${version}_arm64.tar.gz – DTS container image

  • doca_telemetry_standalone.yaml – backup of the default boot YAML

Host Deployment

DTS supports deployment on x86_64 and aarch64 host architectures. All exporters and providers run from a single Docker container.

1. Initialize and configure host DTS with the desired DTS version:

   Copy

   Copied!

               
   
               
   export DTS_IMAGE=nvcr.io/nvidia/doca/doca_telemetry:<desired-DTS-host-version>
   docker run -v "/opt/mellanox/doca/services/telemetry/config:/config" --rm --name doca-telemetry-init -it $DTS_IMAGE /bin/bash -c "DTS_CONFIG_DIR=host /usr/bin/telemetry-init.sh"
           

   Note

   NGC does not support the latest tag. You must specify a full versioned tag (e.g., 1.23.4-doca3.2.0-host ).

   Copy

   Copied!

               
   
               
   export DTS_IMAGE=nvcr.io/nvidia/doca/doca_telemetry:1.23.4-doca3.2.0-host
           

   Info

   The purpose of telemetry-init.sh is to populate the config/ directory. If this directory already contains configuration files, the script is a no-op. When upgrading, mount an alternate directory to add new configuration files as needed.

2. Launch the DTS container:

   Copy

   Copied!

               
   
               
   docker run -d --net=host --uts=host --ipc=host \
              --privileged \
              --device=/dev/mst/ \
              --device=/dev/infiniband/ \
              --gpus all \
              -v "/opt/mellanox/doca/services/telemetry/config:/config" \
              -v "/opt/mellanox/doca/services/telemetry/ipc_sockets:/tmp/ipc_sockets" \
              -v "/opt/mellanox/doca/services/telemetry/data:/data" \
              -v "/opt/mellanox/doca/services/telemetry/log:/var/log" \
              -v "/usr/lib64/mft:/usr/lib64/mft" \
              --rm --name doca-telemetry -it $DTS_IMAGE /usr/bin/telemetry-run.sh
           

   Note

   If the --gpus all parameter causes the command to fail and DTS GPU providers are not needed, remove the parameter from the command.

   If GPU providers are required but the GPU driver is not available, install the nvidia-container-toolkit package and restart the container runtime.

   Note

   Required mounts per service:

   Service Component	Required Flags
   GPU providers (nvidia_smi, dcgm)	--gpus all
   amber, ppcc_eth	--device=/dev/mst/ -v "/usr/lib64/mft:/usr/lib64/mft"
   UCX/RDMA modes	--device=/dev/infiniband/

Deployment with Grafana Monitoring

For information on integrating DTS with Grafana dashboards, refer to the section "Deploying with Grafana Monitoring".

DTS configuration is located at /opt/mellanox/doca/services/telemetry/config.

This directory is initialized during service startup. It contains the main configuration file (dts_config.ini) and a folder for Fluent Bit configuration (fluent_bit_configs). The dts_config.ini file is used to control telemetry provider behavior, exporter settings, sampling rates, and data output. Fluent Bit integration is discussed separately in the Fluent Bit section.

Initialization Scripts (BlueField Deployment)

The DTS container’s initContainers section of the .yaml file executes two initialization scripts:

• /usr/bin/telemetry-init.sh – Generates default configuration files only if the config directory is empty

• /usr/bin/enable-fluent-forward.sh – Configures the Fluent Bit forwarding destination (host and port). This script only runs when both parameters are provided and will overwrite the fluent_bit_configs directory with a new .exp configuration file.

Enabling Fluent Bit Forwarding

To enable Fluent Bit forwarding, add the destination IP and port to the command line in the initContainers section of your deployment YAML:

            

            
command: ["/bin/bash", "-c", "/usr/bin/telemetry-init.sh && /usr/bin/enable-fluent-forward.sh -i=127.0.0.1 -p=24224"]
        

Generating Configuration

The config directory starts empty by default. The initialization process will generate default configuration files during first launch.

Resetting Configuration

To reset DTS to its default state:

1. Delete all files from /opt/mellanox/doca/services/telemetry/config/.

2. Restart the DTS container to regenerate default configuration files.

Sampling Interval

The sampling interval for most telemetry providers is controlled by the update parameter in dts_config.ini:

            

            
# Sampling interval for providers (milliseconds)
update=1000
        

            

            
# Timeout for forced page rotation in seconds
sync-time-limit=10000
        

Enabling Providers

To enable a telemetry provider, uncomment its enable-provider line in dts_config.ini. For example:

            

            
enable-provider=ethtool
        

Remote Collection (BlueField Only)

Some providers may not function correctly within a container and require remote execution. Enable this mode as follows:

1. Start the DOCA privileged executor (DPE):

   Copy

   Copied!

               
   
               
   systemctl start dpe
           

2. Prefix the provider name with grpc.. For example, the following line configures remote collection of the sysfs.fw_pages component:

   Copy

   Copied!

               
   
               
   enable-provider=grpc.sysfs.fw_pages
           

3. For provider-specific configuration, apply the same prefix. Building upon the previous example:

   Copy

   Copied!

               
   
               
   grpc.sysfs.fw_pages.mlx5_1=ignore 
           

Enabling Data Output

To enable writing collected data to disk, uncomment the following line in dts_config.ini:

            

            
output=/data
        

Enabling IPC with External Applications

To enable inter-process communication (IPC) between DTS and a non-containerized application, refer to the section "Using IPC with Non-container Application" in DOCA Telemetry Exporter.

Level Labels

DTS supports metadata static labels for telemetry exporters such as Prometheus and OpenTelemetry. Labels can be configured globally or per device, and are defined in /opt/mellanox/doca/services/telemetry/config/level_labels.ini.

Sample level_labels.ini

The following is an example level_labels.ini file, defining global and per-device labels:

            

            
# lines starting with hash are ignored
# 2 level labels are supported - global and device
# global labels will be added to all metrics
# device labels will be added to metrics matching the device description
 
[global_labels]
# global labels to be added to all metrics
# format:
# label_name=label_value
hostname=myhost
ip=127.127.127.127
 
[device_labels]
# device labels will be added to metrics related to the device
# format:
# device_index_label=device_label_name1|device_label_value1|device_label_name2|device_label_value2
# where device_index_label is either an RDMA device name or PCI address
mlx5_0=rail|0|network|compute
mlx5_1=rail|1|network|compute
mlx5_2=rail|2|network|storage
 
[device_mapping]
# optional: mapping of device names to network interfaces and mst devices
# this is done internally by DTS, but can be overridden here
# add clear_auto_detected_mapping=true to clear the auto-detected mapping
# format:
# device_name=mst_device_name|net_interface_names
# in case of multiple net interfaces, separate them with comma
 
[data_types_mapping]
# internal data types, don't change this section if you don't know what you are doing
# format:
# data_type|index|device_type_name
ethtool_event=device_name|netif
ppcc_eth=device_name|mst
ifconfig_event=device_name|netif
amber_event=device_name|mst
        

Enabling Level Labels

1. Confirm that level_labels.ini is properly configured.

2. Add the following line to dts_config.ini:

   Copy

   Copied!

               
   
               
   level-labels-file=/config/level_labels.ini
           

Monitoring and Reloading

DTS continuously monitors level_labels.ini. When a change is detected:

• The current telemetry process exits automatically

• DTS restarts within 60 seconds and loads the updated labels

This ensures label changes take effect without requiring manual restarts.

Providers

DTS supports on-board data collection from sysf, ethtool, and tc providers. Fluent and Prometheus aggregator providers can collect the data from other applications.

Other providers are available based on different conditions (e.g., specific container mounts or host only such as amber, ppcc_eth, etc). Such providers are described with their dependencies in their corresponding sections.

Sysfs

The sysfs provider includes several components: ib_port, hw_port, devices, ib_mr_cache, eth, hwmon, and bf_ptm. The schema for sysfs counters defines the exact counters available for each component.

By default, the following components are disabled when the sysfs provider is enabled:

• ib_mr_cache

• rate

• bf_ptm

You can enable the provider and selectively disable components. For example, to disable eth:

            

            
enable-provider=sysfs
disable-provider=sysfs.eth
        

To filter which counters are collected, you may use a .cset (counterset) file. This file must not contain aliases. Place the .cset file in the config directory (or any accessible directory in the container), and specify the path in the configuration:

            

            
sysfs-cset-path=/config/path/to/my.cset
        

Counters List

• ib_port counters:

  Copy

  Copied!

              
  
              
  {hca_name}:{port_num}:ib_port_state
  {hca_name}:{port_num}:VL15_dropped
  {hca_name}:{port_num}:excessive_buffer_overrun_errors
  {hca_name}:{port_num}:link_downed
  {hca_name}:{port_num}:link_error_recovery
  {hca_name}:{port_num}:local_link_integrity_errors
  {hca_name}:{port_num}:multicast_rcv_packets
  {hca_name}:{port_num}:multicast_xmit_packets
  {hca_name}:{port_num}:port_rcv_constraint_errors
  {hca_name}:{port_num}:port_rcv_data
  {hca_name}:{port_num}:port_rcv_errors
  {hca_name}:{port_num}:port_rcv_packets
  {hca_name}:{port_num}:port_rcv_remote_physical_errors
  {hca_name}:{port_num}:port_rcv_switch_relay_errors
  {hca_name}:{port_num}:port_xmit_constraint_errors
  {hca_name}:{port_num}:port_xmit_data
  {hca_name}:{port_num}:port_xmit_discards
  {hca_name}:{port_num}:port_xmit_packets
  {hca_name}:{port_num}:port_xmit_wait
  {hca_name}:{port_num}:symbol_error
  {hca_name}:{port_num}:unicast_rcv_packets
  {hca_name}:{port_num}:unicast_xmit_packets
          

• ib_hw counters:

  Copy

  Copied!

              
  
              
  {hca_name}:{port_num}:hw_state
  {hca_name}:{port_num}:hw_duplicate_request
  {hca_name}:{port_num}:hw_implied_nak_seq_err
  {hca_name}:{port_num}:hw_lifespan
  {hca_name}:{port_num}:hw_local_ack_timeout_err
  {hca_name}:{port_num}:hw_out_of_buffer
  {hca_name}:{port_num}:hw_out_of_sequence
  {hca_name}:{port_num}:hw_packet_seq_err   
  {hca_name}:{port_num}:hw_req_cqe_error
  {hca_name}:{port_num}:hw_req_cqe_flush_error
  {hca_name}:{port_num}:hw_req_remote_access_errors   
  {hca_name}:{port_num}:hw_req_remote_invalid_request   
  {hca_name}:{port_num}:hw_resp_cqe_error
  {hca_name}:{port_num}:hw_resp_cqe_flush_error
  {hca_name}:{port_num}:hw_resp_local_length_error
  {hca_name}:{port_num}:hw_resp_remote_access_errors
  {hca_name}:{port_num}:hw_rnr_nak_retry_err   
  {hca_name}:{port_num}:hw_rx_atomic_requests   
  {hca_name}:{port_num}:hw_rx_dct_connect
  {hca_name}:{port_num}:hw_rx_icrc_encapsulated
  {hca_name}:{port_num}:hw_rx_read_requests   
  {hca_name}:{port_num}:hw_rx_write_requests
          

• ib_mr_cache counters:

  Copy

  Copied!

              
  
              
  {hca_name}:mr_cache:size_{n}:cur
  {hca_name}:mr_cache:size_{n}:limit
  {hca_name}:mr_cache:size_{n}:miss
  {hca_name}:mr_cache:size_{n}:size
          

  Note

  Where n ranges from 0 to 24.

• eth counters:

  Copy

  Copied!

              
  
              
  {hca_name}:{device_name}:eth_collisions
  {hca_name}:{device_name}:eth_multicast
  {hca_name}:{device_name}:eth_rx_bytes
  {hca_name}:{device_name}:eth_rx_compressed
  {hca_name}:{device_name}:eth_rx_crc_errors
  {hca_name}:{device_name}:eth_rx_dropped
  {hca_name}:{device_name}:eth_rx_errors
  {hca_name}:{device_name}:eth_rx_fifo_errors
  {hca_name}:{device_name}:eth_rx_frame_errors
  {hca_name}:{device_name}:eth_rx_length_errors
  {hca_name}:{device_name}:eth_rx_missed_errors
  {hca_name}:{device_name}:eth_rx_nohandler
  {hca_name}:{device_name}:eth_rx_over_errors
  {hca_name}:{device_name}:eth_rx_packets
  {hca_name}:{device_name}:eth_tx_aborted_errors
  {hca_name}:{device_name}:eth_tx_bytes
  {hca_name}:{device_name}:eth_tx_carrier_errors
  {hca_name}:{device_name}:eth_tx_compressed
  {hca_name}:{device_name}:eth_tx_dropped
  {hca_name}:{device_name}:eth_tx_errors
  {hca_name}:{device_name}:eth_tx_fifo_errors
  {hca_name}:{device_name}:eth_tx_heartbeat_errors
  {hca_name}:{device_name}:eth_tx_packets
  {hca_name}:{device_name}:eth_tx_window_errors
          

• devices counters

  Copy

  Copied!

              
  
              
  {hca_name}:current_link_width
  {hca_name}:current_link_speed
  {hca_name}:max_link_speed
  {hca_name}:max_link_width
          

• BlueField-2 hwmon counters:

  Collapse Source

  Copy

  Copied!

              
  
              
  {hwmon_name}:{l3cache}:CYCLES
  {hwmon_name}:{l3cache}:HITS_BANK0
  {hwmon_name}:{l3cache}:HITS_BANK1
  {hwmon_name}:{l3cache}:MISSES_BANK0
  {hwmon_name}:{l3cache}:MISSES_BANK1
  {hwmon_name}:{pcie}:IN_C_BYTE_CNT
  {hwmon_name}:{pcie}:IN_C_PKT_CNT
  {hwmon_name}:{pcie}:IN_NP_BYTE_CNT
  {hwmon_name}:{pcie}:IN_NP_PKT_CNT
  {hwmon_name}:{pcie}:IN_P_BYTE_CNT
  {hwmon_name}:{pcie}:IN_P_PKT_CNT
  {hwmon_name}:{pcie}:OUT_C_BYTE_CNT
  {hwmon_name}:{pcie}:OUT_C_PKT_CNT
  {hwmon_name}:{pcie}:OUT_NP_BYTE_CNT
  {hwmon_name}:{pcie}:OUT_NP_PKT_CNT
  {hwmon_name}:{pcie}:OUT_P_PKT_CNT
  {hwmon_name}:{tile}:MEMORY_READS
  {hwmon_name}:{tile}:MEMORY_WRITES
  {hwmon_name}:{tile}:MSS_NO_CREDIT
  {hwmon_name}:{tile}:VICTIM_WRITE
  {hwmon_name}:{tilenet}:CDN_DIAG_C_OUT_OF_CRED
  {hwmon_name}:{tilenet}:CDN_REQ
  {hwmon_name}:{tilenet}:DDN_REQ
  {hwmon_name}:{tilenet}:NDN_REQ
  {hwmon_name}:{trio}:TDMA_DATA_BEAT
  {hwmon_name}:{trio}:TDMA_PBUF_MAC_AF
  {hwmon_name}:{trio}:TDMA_RT_AF
  {hwmon_name}:{trio}:TPIO_DATA_BEAT
  {hwmon_name}:{triogen}:TX_DAT_AF
  {hwmon_name}:{triogen}:TX_DAT_AF
          

• BlueField-3 hwmon counters:

  Copy

  Copied!

              
  
              
  {hwmon_name}:{llt}:GDC_BANK0_RD_REQ
  {hwmon_name}:{llt}:GDC_BANK1_RD_REQ
  {hwmon_name}:{llt}:GDC_BANK0_WR_REQ
  {hwmon_name}:{llt}:GDC_BANK1_WR_REQ
  {hwmon_name}:{llt_miss}:GDC_MISS_MACHINE_RD_REQ
  {hwmon_name}:{llt_miss}:GDC_MISS_MACHINE_WR_REQ
  {hwmon_name}:{mss}:SKYLIB_DDN_TX_FLITS
  {hwmon_name}:{mss}:SKYLIB_DDN_RX_FLITS
          

• BlueField-3 bf_ptm counters:

  Copy

  Copied!

              
  
              
  bf:ptm:active_power_profile
  bf:ptm:atx_power_available
  bf:ptm:core_temp
  bf:ptm:ddr_temp
  bf:ptm:error_state
  bf:ptm:power_envelope
  bf:ptm:power_throttling_event_count
  bf:ptm:power_throttling_state
  bf:ptm:thermal_throttling_event_count
  bf:ptm:thermal_throttling_state
  bf:ptm:throttling_state
  bf:ptm:total_power
  bf:ptm:vr0_power
  bf:ptm:vr1_power
          

• rate counters – calculated counters showing the rate of raw counters of other components. Such counters are identified by a _rate suffix which correspond to the original raw counter name they track.

Port Counters

The following parameters are located in /sys/class/infiniband/mlx5_0/ports/1/counters.

Hardware Counters

The hardware counters, found under /sys/class/infiniband/mlx5_0/ports/1/hw_counters/, are counted per function and exposed on the function. Some counters are not counted per function. These counters are commented with a relevant comment.

Debug Status Counters

The following parameters are located in /sys/class/net/<interface>/debug.

            

            
# cat /sys/class/net/eth2/debug/lro_timeout
Actual timeout: 32
Supported timeout: 8 16 32 1024
        

            

            
$ cat /sys/class/net/ethXX/debug/link_down_reason
monitor_opcode: 0x0
status_message: The port is Active.
        

Power Thermal Counters

The bf_ptm component collects BlueField-3 power thermal counters using remote collection. It is disabled by default and can be enabled as follows:

1. Load kernel module mlxbf-ptm:

   Copy

   Copied!

               
   
               
   modprobe -v mlxbf-ptm
           

2. Enable component using remote collection:

   Copy

   Copied!

               
   
               
   enable-provider=grpc.sysfs.bf_ptm
           

   Note

   DPE server should be active before changing the dts_config.ini file. See section "Remote Collection" for details.

Ethtool Counters

Ethtool counters deliver a comprehensive, per-device list of hardware and software statistics generated by the Ethtool utility. Administrators rely on these counters to monitor traffic flow, measure hardware offload efficiency, and identify network or device errors.

Counter Groups

The system categorizes counters based on their origin location within the datapath:

• Ring: Software ring counters.

• Software port: An aggregation of all software ring counters for a given port.

• vPort: Traffic and drop counters related to internal steering or buffer availability.

  • Scope: Includes standard Ethernet traffic (including raw) and RDMA/RoCE traffic.

  • Troubleshooting: Errors in this group typically indicate issues internal to the BlueField DPU.

• Physical port: Counters for the physical port connecting the BlueField DPU to the external network.

  • Standards Tracking: Includes IEEE 802.3, RFC 2863, RFC 2819, RFC 3635, flow control, and Forward Error Correction (FEC) counters.

  • Troubleshooting: Errors here generally indicate physical link issues (e.g., defective cables or transceivers) or external network switch issues.

    Note

    The system does not expose physical port counters to virtual machines.

• Priority port: A specialized subset of physical port counters, broken down by priority level per port.

Counter Types

Within each group, counters fall into three functional classifications:

• Traffic informative: Tracks general traffic volume. Use these counters for load estimation and high-level debugging.

• Traffic acceleration [A]: Tracks traffic processed specifically by NVIDIA hardware offloads or drivers.

  Note

  Acceleration tracking adds an additional layer to the informative set. The system counts the identical packet in both the informative and acceleration counters. Acceleration counters always feature the [A] designation.

• Error counters: Tracks dropped packets or malformed frames. Increments in these counters serve as primary indicators of network or device problems.

Acceleration Mechanism Counters

Specific hardware offload mechanisms utilize dedicated counters to verify operational efficiency:

• TCP segmentation offload (TSO): Tracks outbound throughput improvements where the kernel aggregates multiple packets into a single large buffer, which the BlueField hardware subsequently splits into individual packets.

• Large receive offload (LRO): Tracks inbound throughput improvements where the hardware aggregates multiple incoming packets from a single stream into a single buffer.

• CQE compress: Tracks the compression of Completion Queue Events (CQE). This mechanism spares PCIe bandwidth and improves overall performance.

• CHECKSUM: Tracks TCP checksum calculations performed by the BlueField hardware. Status indicators include:

  • CHECKSUM_UNNECESSARY: No checksum required.

  • CHECKSUM_NONE: No acceleration applied.

  • CHECKSUM_COMPLETE: The device successfully verified the checksum for the entire packet.

  • CHECKSUM_PARTIAL: The device provided partial checksum validation.

    Info

    Consult the Linux skbuff.h header file for detailed definitions of these states

Configuration Parameters

Configure which interfaces to monitor and exactly what data to collect using the parameters below in the DTS configuration file.

Interface Selection

Define the network interfaces to monitor using a comma-separated list. By default, the system uses a "contains" matching method (any interface name containing the specified string matches).

            

            
ethtool-interfaces=enp1s0f0,enp1s0f1,enp1s0f2
        

To force a strict, exact match for the interface names, configure the following parameter:

            

            
ethtool-interfaces-match-method=exact
        

Fieldset Customization (Filtering)

To customize exactly which fields the service collects, point the configuration to a specific fieldset file (.fset). Ensure this file resides in the DTS configuration folder.

            

            
ethtool-fieldset-file=/config/ethtool.fset
        

Example .fset filtering for Receive (Rx) and Transmit (Tx) pause control counters:

            

            
[ethtool_event_*]
rx_pause_ctrl_phy
tx_pause_ctrl_phy
        

Ring and Software Port Counters

The counters listed in the subsequent table supply detailed information regarding the volume of traffic accelerated by the BlueField DPU. Keep in mind that these counters tally the accelerated traffic in addition to the standard informative counters (accelerated traffic counts twice).

The counter names apply to both ring and software port counters depending on the presence of an index:

• Ring counters: Include the specific [i] index number (e.g., rx0_packets for ring 0).

• Software port counters: Aggregate the rings and omit the index entirely (e.g., rx_packets for the entire software port).

vPort Counters

These counters reside on the eSwitch port connected to the vNIC. They are critical for monitoring traffic steering between the DPU and the host or virtual machines.

Physical Port Counters

These counters monitor the external interface connecting the adapter to the network. They include standardized metrics (IEEE 802.3, RFCs) and hardware-specific indicators for Flow Control and FEC.

Priority Port Counters

These counters are a specialized subset of physical port metrics, indexed by priority [p].

Device Counters

Device counters monitor the health and performance of the PCIe subsystem. High error rates here typically indicate physical hardware issues or bus-level bottlenecks rather than network problems.

Full List of Counters

            

            
# ethtool -S eth5
 
NIC statistics:
rx_packets: 10
rx_bytes: 3420
tx_packets: 18
tx_bytes: 1296
tx_tso_packets: 0
tx_tso_bytes: 0
tx_tso_inner_packets: 0
tx_tso_inner_bytes: 0
tx_added_vlan_packets: 0
tx_nop: 0
rx_lro_packets: 0
rx_lro_bytes: 0
rx_ecn_mark: 0
rx_removed_vlan_packets: 0
rx_csum_unnecessary: 0
rx_csum_none: 0
rx_csum_complete: 10
rx_csum_unnecessary_inner: 0
rx_xdp_drop: 0
rx_xdp_redirect: 0
rx_xdp_tx_xmit: 0
rx_xdp_tx_full: 0
rx_xdp_tx_err: 0
rx_xdp_tx_cqe: 0
tx_csum_none: 18
tx_csum_partial: 0
tx_csum_partial_inner: 0
tx_queue_stopped: 0
tx_queue_dropped: 0
tx_xmit_more: 0
tx_recover: 0
tx_cqes: 18
tx_queue_wake: 0
tx_udp_seg_rem: 0
tx_cqe_err: 0
tx_xdp_xmit: 0
tx_xdp_full: 0
tx_xdp_err: 0
tx_xdp_cqes: 0
rx_wqe_err: 0
rx_mpwqe_filler_cqes: 0
rx_mpwqe_filler_strides: 0
rx_buff_alloc_err: 0
rx_cqe_compress_blks: 0
rx_cqe_compress_pkts: 0
rx_page_reuse: 0
rx_cache_reuse: 0
rx_cache_full: 0
rx_cache_empty: 2688
rx_cache_busy: 0
rx_cache_waive: 0
rx_congst_umr: 0
rx_arfs_err: 0
ch_events: 75
ch_poll: 75
ch_arm: 75
ch_aff_change: 0
ch_eq_rearm: 0
rx_out_of_buffer: 0
rx_if_down_packets: 15
rx_steer_missed_packets: 0
rx_vport_unicast_packets: 0
rx_vport_unicast_bytes: 0
tx_vport_unicast_packets: 0
tx_vport_unicast_bytes: 0
rx_vport_multicast_packets: 2
rx_vport_multicast_bytes: 172
tx_vport_multicast_packets: 12
tx_vport_multicast_bytes: 936
rx_vport_broadcast_packets: 37
rx_vport_broadcast_bytes: 9270
tx_vport_broadcast_packets: 6
tx_vport_broadcast_bytes: 360
rx_vport_rdma_unicast_packets: 0
rx_vport_rdma_unicast_bytes: 0
tx_vport_rdma_unicast_packets: 0
tx_vport_rdma_unicast_bytes: 0
rx_vport_rdma_multicast_packets: 0
rx_vport_rdma_multicast_bytes: 0
tx_vport_rdma_multicast_packets: 0
tx_vport_rdma_multicast_bytes: 0
tx_packets_phy: 0
rx_packets_phy: 0
rx_crc_errors_phy: 0
tx_bytes_phy: 0
rx_bytes_phy: 0
tx_multicast_phy: 0
tx_broadcast_phy: 0
rx_multicast_phy: 0
rx_broadcast_phy: 0
rx_in_range_len_errors_phy: 0
rx_out_of_range_len_phy: 0
rx_oversize_pkts_phy: 0
rx_symbol_err_phy: 0
tx_mac_control_phy: 0
rx_mac_control_phy: 0
rx_unsupported_op_phy: 0
rx_pause_ctrl_phy: 0
tx_pause_ctrl_phy: 0
rx_discards_phy: 0
tx_discards_phy: 0
tx_errors_phy: 0
rx_undersize_pkts_phy: 0
rx_fragments_phy: 0
rx_jabbers_phy: 0
rx_64_bytes_phy: 0
rx_65_to_127_bytes_phy: 0
rx_128_to_255_bytes_phy: 0
rx_256_to_511_bytes_phy: 0
rx_512_to_1023_bytes_phy: 0
rx_1024_to_1518_bytes_phy: 0
rx_1519_to_2047_bytes_phy: 0
rx_2048_to_4095_bytes_phy: 0
rx_4096_to_8191_bytes_phy: 0
rx_8192_to_10239_bytes_phy: 0
link_down_events_phy: 0
rx_prio0_bytes: 0
rx_prio0_packets: 0
tx_prio0_bytes: 0
tx_prio0_packets: 0
rx_prio1_bytes: 0
rx_prio1_packets: 0
tx_prio1_bytes: 0
tx_prio1_packets: 0
rx_prio2_bytes: 0
rx_prio2_packets: 0
tx_prio2_bytes: 0
tx_prio2_packets: 0
rx_prio3_bytes: 0
rx_prio3_packets: 0
tx_prio3_bytes: 0
tx_prio3_packets: 0
rx_prio4_bytes: 0
rx_prio4_packets: 0
tx_prio4_bytes: 0
tx_prio4_packets: 0
rx_prio5_bytes: 0
rx_prio5_packets: 0
tx_prio5_bytes: 0
tx_prio5_packets: 0
rx_prio6_bytes: 0
rx_prio6_packets: 0
tx_prio6_bytes: 0
tx_prio6_packets: 0
rx_prio7_bytes: 0
rx_prio7_packets: 0
tx_prio7_bytes: 0
tx_prio7_packets: 0
module_unplug: 0
module_bus_stuck: 0
module_high_temp: 0
module_bad_shorted: 0
ch0_events: 9
ch0_poll: 9
ch0_arm: 9
ch0_aff_change: 0
ch0_eq_rearm: 0
ch1_events: 23
ch1_poll: 23
ch1_arm: 23
ch1_aff_change: 0
ch1_eq_rearm: 0
ch2_events: 8
ch2_poll: 8
ch2_arm: 8
ch2_aff_change: 0
ch2_eq_rearm: 0
ch3_events: 19
ch3_poll: 19
ch3_arm: 19
ch3_aff_change: 0
ch3_eq_rearm: 0
ch4_events: 8
ch4_poll: 8
ch4_arm: 8
ch4_aff_change: 0
ch4_eq_rearm: 0
ch5_events: 8
ch5_poll: 8
ch5_arm: 8
ch5_aff_change: 0
ch5_eq_rearm: 0
rx0_packets: 0
rx0_bytes: 0
rx0_csum_complete: 0
rx0_csum_unnecessary: 0
rx0_csum_unnecessary_inner: 0
rx0_csum_none: 0
rx0_xdp_drop: 0
rx0_xdp_redirect: 0
rx0_lro_packets: 0
rx0_lro_bytes: 0
rx0_ecn_mark: 0
rx0_removed_vlan_packets: 0
rx0_wqe_err: 0
rx0_mpwqe_filler_cqes: 0
rx0_mpwqe_filler_strides: 0
rx0_buff_alloc_err: 0
rx0_cqe_compress_blks: 0
rx0_cqe_compress_pkts: 0
rx0_page_reuse: 0
rx0_cache_reuse: 0
rx0_cache_full: 0
rx0_cache_empty: 448
rx0_cache_busy: 0
rx0_cache_waive: 0
rx0_congst_umr: 0
rx0_arfs_err: 0
rx0_xdp_tx_xmit: 0
rx0_xdp_tx_full: 0
rx0_xdp_tx_err: 0
rx0_xdp_tx_cqes: 0
rx1_packets: 10
rx1_bytes: 3420
rx1_csum_complete: 10
rx1_csum_unnecessary: 0
rx1_csum_unnecessary_inner: 0
rx1_csum_none: 0
rx1_xdp_drop: 0
rx1_xdp_redirect: 0
rx1_lro_packets: 0
rx1_lro_bytes: 0
rx1_ecn_mark: 0
rx1_removed_vlan_packets: 0
rx1_wqe_err: 0
rx1_mpwqe_filler_cqes: 0
rx1_mpwqe_filler_strides: 0
rx1_buff_alloc_err: 0
rx1_cqe_compress_blks: 0
rx1_cqe_compress_pkts: 0
rx1_page_reuse: 0
rx1_cache_reuse: 0
rx1_cache_full: 0
rx1_cache_empty: 448
rx1_cache_busy: 0
rx1_cache_waive: 0
rx1_congst_umr: 0
rx1_arfs_err: 0
rx1_xdp_tx_xmit: 0
rx1_xdp_tx_full: 0
rx1_xdp_tx_err: 0
rx1_xdp_tx_cqes: 0
rx2_packets: 0
rx2_bytes: 0
rx2_csum_complete: 0
rx2_csum_unnecessary: 0
rx2_csum_unnecessary_inner: 0
rx2_csum_none: 0
rx2_xdp_drop: 0
rx2_xdp_redirect: 0
rx2_lro_packets: 0
rx2_lro_bytes: 0
rx2_ecn_mark: 0
rx2_removed_vlan_packets: 0
rx2_wqe_err: 0
rx2_mpwqe_filler_cqes: 0
rx2_mpwqe_filler_strides: 0
rx2_buff_alloc_err: 0
rx2_cqe_compress_blks: 0
rx2_cqe_compress_pkts: 0
rx2_page_reuse: 0
rx2_cache_reuse: 0
rx2_cache_full: 0
rx2_cache_empty: 448
rx2_cache_busy: 0
rx2_cache_waive: 0
rx2_congst_umr: 0
rx2_arfs_err: 0
rx2_xdp_tx_xmit: 0
rx2_xdp_tx_full: 0
rx2_xdp_tx_err: 0
rx2_xdp_tx_cqes: 0
...
tx0_packets: 1
tx0_bytes: 60
tx0_tso_packets: 0
tx0_tso_bytes: 0
tx0_tso_inner_packets: 0
tx0_tso_inner_bytes: 0
tx0_csum_partial: 0
tx0_csum_partial_inner: 0
tx0_added_vlan_packets: 0
tx0_nop: 0
tx0_csum_none: 1
tx0_stopped: 0
tx0_dropped: 0
tx0_xmit_more: 0
tx0_recover: 0
tx0_cqes: 1
tx0_wake: 0
tx0_cqe_err: 0
tx1_packets: 5
tx1_bytes: 300
tx1_tso_packets: 0
tx1_tso_bytes: 0
tx1_tso_inner_packets: 0
tx1_tso_inner_bytes: 0
tx1_csum_partial: 0
tx1_csum_partial_inner: 0
tx1_added_vlan_packets: 0
tx1_nop: 0
tx1_csum_none: 5
tx1_stopped: 0
tx1_dropped: 0
tx1_xmit_more: 0
tx1_recover: 0
tx1_cqes: 5
tx1_wake: 0
tx1_cqe_err: 0
tx2_packets: 0
tx2_bytes: 0
tx2_tso_packets: 0
tx2_tso_bytes: 0
tx2_tso_inner_packets: 0
tx2_tso_inner_bytes: 0
tx2_csum_partial: 0
tx2_csum_partial_inner: 0
tx2_added_vlan_packets: 0
tx2_nop: 0
tx2_csum_none: 0
tx2_stopped: 0
tx2_dropped: 0
tx2_xmit_more: 0
tx2_recover: 0
tx2_cqes: 0
tx2_wake: 0
tx2_cqe_err: 0
...
        

Traffic Control Info

The following TC objects are supported and reported regarding the ingress filters:

• Filters

  • flower

• Actions

  • mirred

  • tunnel_key

The info is provided as one of the following events:

• Basic filter event

• Flower/IPv4 filter event

• Flower/IPv6 filter event

• Basic action event

• Mirred action event

• Tunnel_key/IPv4 action event

• Tunnel_key/IPv6 action event

General notes:

• Actions always belong to a filter, so action events share the filter event's ID via the event_id data member

• Basic filter event only contains textual kind (so users can see which real life objects' support they are lacking)

• Basic action event only contains textual kind and some basic common statistics if available

Amber Provider

Amber data for both InfiniBand and Ethernet MST devices in amBER format.

The following configuration parameters are optional:

            

            
amber_devices=mt41692_pciconf0,mt41692_pciconf0.1   # Default:all, or a comma-separated list of devices under /dev/mst
amber_update_interval_sec=30                        # Sample rate for collection amber counters
        

PPCC_ETH Provider

Programmable congestion control counters are based on algorithms defined by the user, although default algorithms are also available.

Counters are collected per MST device and algorithm parameters.

The counter list depends on the installed MFT version.

The provider supports automatic detection of available Ethernet MST devices for collection. To override this behavior, specify a comma-separated list of device names:

            

            
ppcc_eth_devices=mt41692_pciconf0,mt41692_pciconf0.1
        

When correctly configured, the ppcc_eth provider operates in read-only mode and does not write to device registers. The following parameter enforces this behavior:

            

            
ppcc_read_only=1
        

The following algorithm parameters use default values and typically do not require changes:

            

            
ppcc_algo_slot=1
ppcc_algo_param_index=0
ppcc_local_port=1
ppcc_pnat=0
ppcc_lp_msb=0
        

Fluent Aggregator

fluent_aggr listens on a port for Fluent Bit Forward protocol input connections. Received data can be streamed via a Fluent Bit exporter.

The default port is 42442. This can be changed by updating the following option:

            

            
fluent-aggr-port=42442
        

Prometheus Aggregator

prometheus_aggr polls data from a list of Prometheus endpoints.

Each endpoint is listed in the following format:

            

            
prometheus_aggr_endpoint.{N}={host_name},{host_port_url},{poll_inteval_msec}
        

Where N starts from 0.

Aggregated data can be exported via a Prometheus Aggr Exporter endpoint.

Network Interfaces

ifconfig collects network interface data. To enable, set:

            

            
enable-provider=ifconfig
        

If the Prometheus endpoint is enabled and autodetection of fset indexes is disabled, add the following configuration to cache every collected network interface and arrange the index according to their names:

            

            
prometheus-fset-indexes=name
        

Metrices are collected for each network interface as follows:

            

            
name
rx_packets
tx_packets
rx_bytes
tx_bytes
rx_errors
tx_errors
rx_dropped
tx_dropped
multicast
collisions
rx_length_errors
rx_over_errors
rx_crc_errors
rx_frame_errors
rx_fifo_errors
rx_missed_errors
tx_aborted_errors
tx_carrier_errors
tx_fifo_errors
tx_heartbeat_errors
tx_window_errors
rx_compressed
tx_compressed
rx_nohandler
        

NVIDIA System Management Interface

The nvidia-smi provider collects GPU and GPU process information provided by the NVIDIA system management interface.

This provider is supported only on hosts with installed GPUs. All GPU cards supported by nvidia-smi are supported by this provider.

The counter list is GPU dependent. Additionally, per-process information is collected for the first 20 (by default) nvidia_smi_max_processes processes.

Counters can be either collected as string data "as is" in nvidia-smi or converted to numbers when nvsmi_with_numeric_fields is set.

To enable nvidia-smi provider and change parameters, set:

            

            
enable-provider=nvidia_smi
 
# Optional parameters:
#nvidia_smi_max_processes=20
#nvsmi_with_numeric_fields=1
        

NVIDIA Data Center GPU Manager

The dcgm provider collects GPU information provided by the NVIDIA data center GPU manager (DCGM) API.

This provider is supported only on hosts with installed GPUs, and requires running the nv-hostengine service (refer to DCGM documentation for details).

DCGM counters are split into several groups by context:

• GPU – basic GPU information (always)

• COMMON – common fields that can be collected from all devices

• PROF – profiling fields

• ECC – ECC errors

• NVLINK / NVSWITCH / VGPU – fields depending on the device type

To enable DCGM provider and counter groups, set:

            

            
enable-provider=dcgm
 
dcgm_events_enable_common_fields=1
#dcgm_events_enable_prof_fields=0
#dcgm_events_enable_ecc_fields=0
#dcgm_events_enable_nvlink_fields=0
#dcgm_events_enable_nvswitch_fields=0
#dcgm_events_enable_vgpu_fields=0
        

BlueField Performance

The bfperf provider collects calculated performance counters of BlueField Arm cores. It requires the executable bfperf_pmc, which is integrated in the DOCA BFB bundle of BlueField-3, as well as an active DPE.

To enable BlueField performance provider, set:

            

            
enable-provider=bfperf
        

Diagnostic Data

Diagnostic data is comprised of two providers which gather diagnostic data counters from network interface cards (NICs). These providers support the same counters (as defined in a YAML file), but they differ in usage and collection frequency:

• Low frequency provider is defined in dts_config.ini and is controlled by DTS collection loop.

• High frequency provider is expected to run ad-hoc remotely (via REST API) or from dts_ad_hoc_runner_config.ini , and operates in a distinct flow for a limited duration.

Both providers get the counter set from a YAML file - see below.

Prerequisites:

• Firmware version 28.43.1000 onwards for ConnectX-7 or 32.43.1000 onwards for BlueField-3.

• fwctl driver should be installed and loaded (see instructions in NVIDIA MLNX_OFED Documentation v24.07-0.6.1.0 ), which is packaged in doca-ofed profile. Ensure that fwctl and mlx5_fwctl drivers are loaded with:

  Copy

  Copied!

              
  
              
  lsmod | grep fwctl
          

Diagnostic Data Low Frequency

To enable the diagnostic data low frequency provider, set:

            

            
enable-provider=diagnostic_data_low_freq
        

Configure the counter set YAML file (all default files are available in the mounted folder host /opt/mellanox/doca/services/telemetry/config/diagnostic_data_configs).

If there is no such folder, you must update the config folder by running telemetry-init.sh. See "DTS Deployment" section for information.

            

            
diagnostic-data-yml-file=/config/diagnostic_data_configs/all-single-port.yml
        

To configure the diagnostic data timestamp collection type, set the following:

            

            
diagnostic-data-timestamp-collection-type=<method>
        

Where <method> can be one of the following:

• no_counters – Do not collect timestamp counters. This is the default method.

• start_and_end – Collect sample start and end timestamps

• per_counter – Collect every counter collection timestamp

To configure the clock firmware should use when collecting time stamps, set the following:

            

            
diagnostic-data-timestamp-source=<clock>
        

Where <clock> can be one of the following:

• RTC – Real-time clock (default clock used)

• RFC – Free-running clock

Diagnostic Data High Frequency

The Diagnostic Data High Frequency provider supports higher sampling frequencies with sub-millisecond resolution, enabling detailed and precise telemetry collection. Due to the large volume of collected data, this provider is designed to run ad-hoc for limited periods, unlike standard DTS providers configured via the DTS configuration file at /opt/mellanox/doca/services/telemetry/config/dts_config.ini.

While the standard DTS flow functions as an endless collect-export loop, High Frequency Telemetry (HFT) operates as an external flow triggered by an HFT HTTP API or by HFT configuration file.

Providers Compatibility

Both low and high frequency providers can run concurrently. The low frequency provider samples at the DTS standard frequency (defined in dts_config.ini), and the high frequency provider samples counters based on the HFT configuration file ( dts_ad_hoc_runner_config.ini ).

To allow both providers to run concurrently, verify that the counters, the timestamp collection type, and the timestamp collection source are identical. Otherwise, when the high frequency provider starts sampling, the low frequency provider hangs until the end of the HFT session.

HFT HTTP API

The HFT HTTP API allows to start and stop remotely high frequency sampling sessions at scale on devices running DTS, it also allows to get and delete the data as needed.

The API is disabled by default. To enable it, add the following parameter to DTS configuration file /opt/mellanox/doca/services/telemetry/config/dts_config.ini.

            

            
enable-http-api=true
        

The default port is 9117. To modify it, set the http-port option with the desired port number.

The general expected API usage is:

1. Start an HFT ad hoc collection, wait for it to stop or manually stop it. By default, the collected telemetry is stored locally to binary files in the filesystem.

2. Get the binary data files. The binary data can be read using the /opt/mellanox/collectx/bin/clx_read app, packaged in collectx-clxapidev, a DOCA dependency package.

3. Delete the collection from DTS host. In any case, DTS runs a configurable retention policy for the collected data files:

   Copy

   Copied!

               
   
               
   # by size - set the max size limit of the collection files in bytes. Default is 1GB
   ad-hoc-runner-max-size-bytes=1073741824
   # by time - set the max age of collection files in seconds. Default is 1 week
   ad-hoc-runner-max-age-seconds=604800
   # set the minimum time for collection files to exist. Default is 10 minutes
   ad-hoc-runner-min-age-seconds=600
           

Summary

Query parameters are key-value pairs appended to the URL using ? and separated by &. Example: /URL?key1=value1&key2=value2.

Request body is a JSON object containing key-value pairs passed in the body of the request. Example: {"key1": "value1", "key2": "value2"}.

Response body is the JSON-formatted reply returned by the server. The Success Response section of each endpoint specifies the expected response format.

Status code:

• 200 indicates a successful request

• Other status codes indicate errors

POST /ad-hoc-collection/start

Start ad hoc collection.

Request Body

The following table describes supported parameters in the request body:

Success Response

            

            
{
  "status": "success",
  "message": "Collection created",
  "collection_id": 2
}
        

The collection_id parameter is generated automatically by this API and used in the GET and DELETE requests.

POST /ad-hoc-collection/stop

Stop active collection.

Success Response

            

            
{
  "status": "success",
  "message": "Collection stopped successfully"
}
        

POST /ad-hoc-collection/hft-counterset

Posts an HFT counterset that can later be used by the POST /ad-hoc-collection/start endpoint via the hft-counterset parameter.

Two methods are supported for posting a counterset to this endpoint:

• By providing a JSON payload

• By uploading a YAML file (using the format used by the diagnostic data high-frequency provider)

YAML File Request

The request is expected to be a multipart/form-data containing a valid counterset YAML file.

            

            
curl -X POST localhost:9117/ad-hoc-collection/hft-counterset?filename=my_cx7_hft_cset -F "file=@my_cx7_hft_cset.yml"
        

JSON Request

JSON of key-value strings pairs, keys are the counter names and the value is a string representation of the hexadecimal data ID of that counter.

For example, the following JSON contains two counters

            

            
curl -X POST localhost:9117/ad-hoc-collection/hft-counterset?filename=my_cx7_hft_cset -d '{"port_rx_bytes_0": "0x1020000100000000", "port_rx_packets_0": "0x1020000300000000"}'
        

Query Parameters

Success Response

            

            
{
  "status": "success",
  "message": "HFT counterset saved successfully",
  "filename": "my_cset_filename",
  "overriden": false
}
        

overriden is true if there was already a counterset with this name and it was replaced.

GET /ad-hoc-collection

This request is feasible for collections that were made with file-write: true and were not removed from the system (by DELETE /ad-hoc-collection or the retention policy).

Query Parameters

Success Response

Content of tar.gz file. when extracted, a collection_data folder should appear in the working directory. Example with bash:

            

            
$ curl -s localhost:9117/ad-hoc-collection?collection_id=2 -o response.tar.gz
$ tar xf response.tar.gz
$ tree collection_data
collection_data/
├── 2025
│   └── 0401
│       └── c-237-169-100-103
│           ├── diagnostic_data_1743529761189538.bin
│           ├── diagnostic_data_1743529761802460.bin
│           ├── diagnostic_data_1743529762402948.bin
│           ├── diagnostic_data_1743529763003488.bin
│           ├── diagnostic_data_1743529763604078.bin
│           ├── diagnostic_data_1743529764204606.bin
│           ├── diagnostic_data_1743529764805022.bin
│           ├── diagnostic_data_1743529765405535.bin
│           └── diagnostic_data_1743529766005931.bin
└── schema
    └── schema_c4d0317ee6cecbab995bfbcc237c76b2.json
        

Example with delta=true:

            

            
$ curl -s -H "X-Client-ID: my_app" "localhost:9117/ad-hoc-collection?collection_id=3&delta=true" -o response1.tar.gz
$ curl -s -H "X-Client-ID: my_app" "localhost:9117/ad-hoc-collection?collection_id=3&delta=true" -o response2.tar.gz
$ curl -s -H "X-Client-ID: my_app" "localhost:9117/ad-hoc-collection?collection_id=3&delta=true" -o response3.tar.gz
$ tar xf response1.tar.gz
$ tar xf response2.tar.gz
$ tar xf response3.tar.gz
$ tree collection_data/
collection_data/
├── 2025
│   └── 0709
│       └── swx-ray02
│           ├── diagnostic_data_1752050796207691.bin
│           ├── diagnostic_data_1752050796512800.bin
│           ├── diagnostic_data_1752050796949445.bin
│           ├── diagnostic_data_1752050797235709.bin
│           ├── diagnostic_data_1752050797524730.bin
│           └── diagnostic_data_1752050797826312.bin
└── schema
    └── schema_300a8aceac1d63c968ae837f366c9db4.json
        

GET /ad-hoc-collection/list

List the collections available on the file system, to be gathered by GET /ad-hoc-collection.

Success Response

            

            
{
  "status": "success",
  "num_collections": 3,
  "collection_ids": [1,2,3]
}
        

DELETE /ad-hoc-collection

Delete data of an ad-hoc collection. Note that in addition to this API there is a configurable retention policy.

Query Parameters

Success Response

            

            
{
  "status": "success",
  "message": "Collection removed"
}
        

HFT Configuration File

The configuration file is mounted to DTS container, the host path is /opt/mellanox/doca/services/telemetry/config/clx_ad_hoc_runner_config.ini

This configuration file specifies the HFT session's timing parameters, provider settings, and export configurations. HFT allows data export to endpoints or protocols distinct from those used in the standard DTS collection loop. The DTS configuration references the HFT configuration file as an "ad hoc runner," enabling DTS to monitor its status via the following directive:

            

            
ad-hoc-runner-file=/config/clx_ad_hoc_runner_config.ini
        

The HFT configuration file serves as both the trigger and definition for HFT sessions. Modifying the file stops the current session and applies the new configuration for the next session. Deleting the file stops any pending HFT sessions entirely.

HFT File Parameters

This table provides the details of the HFT parameters. Refer to section "HFT Configuration File Example" for more helpful tips.

HFT Configuration File Example

            

            
## DTS configuration file for ad-hoc high frequency collection
## When modified, the file is parsed and applied.
## Note that the folders path is the container path, not the host path.
 
## Each section defines a collection. A file may have several sections, each one defines a high frequency collection.
## Section names must be unique and will be used as collection name by clx.
[collection-session]
 
### Time between samples in microseconds
sample-time-us=100000
 
### Start time of high frequency collection. Can be in the format HH:MM:SS or HH:MM or as epoch timestamp in microseconds
### Note - in container, the time is in UTC
start-time=18:00:00
 
### End time of high frequency collection. Can be in the format HH:MM:SS or HH:MM or as epoch timestamp in microseconds
### Note - in container, the time is in UTC
end-time=18:01:00
### Alternatively, you can set the number of iterations. This and start_time field will determine the end time
#num-iterations=300
 
### Data provider to use
provider=diagnostic_data_high_freq
 
### Write data to file system. Could potentially fill up the disk
file-write=false
 
### Root directory to store the data
# Ignored if file-write is set to false
data-root=/data
 
### Enable busy wait between iterations, for a more accurate sample time (default is false)
#busy-wait-sampling=true
 
### Set prometheus endpoint to enable http endpoint
#prometheus-endpoint=http://0.0.0.0:9112
 
### Set fluentbit config dir to enable fluentbit export
#fluentbit-config-dir=/config/fluent_bit_configs
 
### Set open telemetry receiver to enable open telemetry export
#open-telemetry-receiver=http://0.0.0.0:9502/v1/metrics
 
### Set remote write receiver to enable remote write export
#remote-write-receiver=http://0.0.0.0:9090/api/v1/write
 
### Provider specific parameters. Format is 'provider.$KEY=$VALUE'.
### The options below are specific to the diagnostic data high frequency provider
 
# Number of samples to collect on each iteration
provider.diagnostic-data-num-samples=1000
 
# The time period (in nanoseconds) between samples
provider.diagnostic-data-sample-period-nsec=100000
 
# The YAML file with the configuration for the diagnostic data provider
provider.diagnostic-data-yml-file=/config/diagnostic_data_configs/all-single-port.yml
 
# Diagnostic Data timestamp collection type. Options are ['no_counters', 'start_and_end', 'per_counter']. default: 'no_counters'
#provider.diagnostic-data-timestamp-collection-type=start_and_end
 
# Diagnostic Data timestamp source. Options are ['RTC', 'FRC']. default: 'RTC'
#provider.diagnostic-data-timestamp-source=FRC
        

Diagnostic Data YAML File

For compatibility with other related tools, the counter set is defined in YAML format.

There are 4 YAML files within a DTS container (one per permutation of BlueField-3 and ConnectX-7 with dual or single ports). The path to the YAMLs folder is /opt/mellanox/doca/services/telemetry/config/diagnostic_data_configs which is mounted to /config/diagnostic_data_configs.

By default, YAML files include a counter set that is not device-specific. This implies that the same counter set is utilized across all devices by default.

It is possible to assign a specific device within a YAML file; however, this requires maintaining a separate copy of the YAML file for each device. To manage multiple devices, use the diagnostic-data-yml-dir option to specify a directory for YAML files, where each .yml/.yaml file is utilized. This folder should be available to the container under /opt/mellanox/doca/services/telemetry/config .

The following list describes the expected entries in the YAML file:

• counters – sequence of counters to collect

  • id – counter data ID

  • desc – counter description (optional)

  • unit – name of unit to collect from (optional)

  • name – name of counter to use (optional). If not specified, the generated name is based on the counter description. Otherwise, it is based on the data ID.

• device – name of the mlx devices to collect, comma separated (optional). If not used, the provider requires a single file containing a list of counters, which it then applies to all available devices on the host.

YAML File Example

The following is the content of the all-dual-port.yml file provided by DTS:

            

            
counters:
  - id: 0x1020000100000000
    name: port_rx_bytes_0
  - id: 0x1020000100000001
    name: port_rx_bytes_1
  - id: 0x1020000300000000
    name: port_rx_packets_0
  - id: 0x1020000300000001
    name: port_rx_packets_1
  - id: 0x1140000100000000
    name: port_tx_bytes_0
  - id: 0x1140000100000001
    name: port_tx_bytes_1
  - id: 0x1140000300000000
    name: port_tx_packets_0
  - id: 0x1140000300000001
    name: port_tx_packets_1
  - id: 0x1100000100000000
    name: port_tx_transport_cnp_sent_packets_0
  - id: 0x1100000100000001
    name: port_tx_transport_cnp_sent_packets_1
  - id: 0x1080000500000000
    name: port_rx_transport_cnp_handled_packets_0
  - id: 0x1080000500000001
    name: port_rx_transport_cnp_handled_packets_1
  - id: 0x1080000400000000
    name: port_rx_transport_ecn_packets_0
  - id: 0x1080000400000001
    name: port_rx_transport_ecn_packets_1
  - id: 0x1160000b00000000
    name: pcie_link_latency_total_read_ns
  - id: 0x1160000c00000000
    name: pcie_link_latency_total_read_packets
  - id: 0x1160000d00000000
    name: pcie_link_latency_max_read_ns
  - id: 0x1160000e00000000
    name: pcie_link_latency_min_read_ns
  - id: 0x1180000100000000
    name: global_icmc_request
  - id: 0x1180000200000000
    name: global_icmc_hit
  - id: 0x1180000300000000
    name: global_icmc_miss
        

RDMA Notifications

The RDMA Notifications events provider collects notifications from the RDMA firmware and converts them into DTS events for telemetry purposes.

To enable the RDMA notifications provider, add the following line to the DTS configuration file (/opt/mellanox/doca/services/telemetry/config/dts_config.ini):

            

            
enable-provider=rdma_notifications
        

Prerequisites

• RDMA device(s) configured to operate in Ethernet protocol

• OFED installed

• ConnectX-7 or BlueField-3 and later

• Firmware version x.44.0820 or newer

Options

Events List

There are two types of events provided by the RDMA notifications:

• Notification Event – A direct translation of an RDMA notification into one of four possible event types, based on the notification content

• Counter Event – A periodic aggregation event that reports the total number of events collected for each syndrome

Notification Events

Additional fields are provided for convenience:

• guid – A globally unique identifier for the device

• idx – The event index, determined by the provider option rdma-notifications-num-event-indexes

• device_name – The name of the RDMA device, based on the devices listed in rdma-notifications-hca

These fields are not specified in the NVIDIA Adapters PRM but are included in the event data to enhance usability.

cqe_with_error_responder_ipv4

            

            
telemetry_counter
msn
syndrome
qp_type
vport_id
source_qpn
destination_qpn
psn
timestamp
dest_ipv4
node_guid
idx
device_name
        

cqe_with_error_responder_ipv6

            

            
telemetry_counter
msn
syndrome
qp_type
vport_id
source_qpn
destination_qpn
psn
timestamp
dest_ipv6
node_guid
idx
device_name
        

cqe_with_error_requestor_ipv4

            

            
telemetry_counter
msn
syndrome
opcode
qp_type
vport_id
source_qpn
destination_qpn
psn
timestamp
dest_ipv4
node_guid
idx
device_name
        

cqe_with_error_requestor_ipv6

            

            
telemetry_counter
msn
syndrome
opcode
qp_type
vport_id
source_qpn
destination_qpn
psn
timestamp
dest_ipv6
node_guid
idx
device_name
        

Counter Events

Describing the number of events per syndrome, for each device.

cqe_with_error_syndrome_counter

            

            
syndrome
device_name
node_guid
CQEwE_events
        

vNIC Provider

To enable the virtual NIC counters provider:

            

            
enable-provider=vnic
        

Operation and retrieval:

• Counters are automatically provided for all valid Ethernet devices listed in the devlink dev output.

• The service retrieves these metrics by executing:

  Copy

  Copied!

              
  
              
  devlink health diagnose <device_name> reporter vnic
          

• The exact set of available counters is dynamic and depends on the specific driver version currently in use.

The following is an example of a counter list:

            

            
total_error_queues
send_queue_priority_update_flow
comp_eq_overrun
async_eq_overrun
cq_overrun
invalid_command
quota_exceeded_command
generated_pkt_steering_fail
handled_pkt_steering_fail
        

Adaptive Retx Provider

The doca_telemetry_adp_retx provider exposes real-time statistics for the Adaptive Retransmission (ADP-RETX) algorithm configured on supported NVIDIA BlueField DPUs and ConnectX network adapters.

Adaptive Retransmission optimizes RDMA performance by dynamically adjusting retransmit timers based on network conditions. The provider collects this telemetry in a histogram format, organizing retransmit counts into specific time-range bins. This allows administrators to visualize the distribution of network latency and recovery efficiency.

To enable the Adaptive Retx events provider, add the following line to your configuration:

            

            
enable-provider=apt_retx
        

Data Outputs

DTS can send the collected data to the following outputs:

• Data writer (saves binary data to disk)

• Prometheus endpoint (keeps the most recent data to be pulled)

• Fluent Bit (push-model streaming)

• Open Telemetry exporter (push-model streaming)

• Prometheus Remote write exporter (push-model streaming)

Data Writer

The data writer is disabled by default to save space on BlueField. Steps for activating data write during debug can be found under section Enabling Data Output.

The schema folder contains JSON-formatted metadata files which allow reading the binary files containing the actual data. The binary files are written according to the naming convention shown in the following example (apt install tree):

            

            
tree /opt/mellanox/doca/services/telemetry/data/
/opt/mellanox/doca/services/telemetry/data/
├── {year}
│   └── {mmdd}
│        └── {hash}
│             ├── {source_id}
│             │   └── {source_tag}{timestamp}.bin
│             └── {another_source_id}
│                  └── {another_source_tag}{timestamp}.bin
└── schema
    └── schema_{MD5_digest}.json
        

New binary files appears when the service starts or when binary file age/size restriction is reached. If no schema or no data folders are present, refer to the Troubleshooting section.

Reading the binary data can be done from within the DTS container using the following command:

            

            
crictl exec -it <Container ID> /opt/mellanox/collectx/bin/clx_read -s /data/schema /data/path/to/datafile.bin
        

Example output:

            

            
{
    "timestamp": 1634815738799728,
    "event_number": 0,
    "iter_num": 0,
    "string_number": 0,
    "example_string": "example_str_1"
}
{
    "timestamp": 1634815738799768,
    "event_number": 1,
    "iter_num": 0,
    "string_number": 1,
    "example_string": "example_str_2"
}
…
        

Prometheus Exporter

The Prometheus endpoint maintains a buffer of the most recent telemetry data for pull-based collection by a Prometheus server. This mechanism is enabled by default, allowing external monitoring systems to scrape BlueField metrics at defined intervals.

Data Retrieval and Verification

To verify that the Prometheus exporter is active and providing data, execute the following command on the BlueField DPU:

            

            
curl -s http://0.0.0.0:9100/metrics
        

This command returns each counter in the format counter_name {list of meta fields} counter_value timestamp.

The endpoint also supports alternative output formats:

• JSON: curl -s http://0.0.0.0:9100/json/metrics

• CSV: curl -s http://0.0.0.0:9100/csv/metrics

Schema Metadata

DTS provides schema metadata to help users understand the structure and origin of specific counters.

• Retrieve all schema IDs:

  Copy

  Copied!

              
  
              
  curl -s http://0.0.0.0:9100/management/schema
          

• Fetch a specific schema:

  Copy

  Copied!

              
  
              
  curl -s http://0.0.0.0:9100/management/schema?schema_id=xxx
          

• View all available schemas:

  Copy

  Copied!

              
  
              
  curl -s http://0.0.0.0:9100/management/schema?schema_id=all
          

Configuration Details

The Prometheus exporter is managed via dts_config.ini. By default, the HTTP endpoint listens on port 9100.

Standard vs. Secure Configuration

To disable the exporter, comment out the prometheus line. To enable HTTPS, provide the certificate and key paths:

            

            
# For standard HTTP
prometheus=http://0.0.0.0:9100
 
# For Secure HTTPS (Requires certs in DTS config volume)
prometheus=https://0.0.0.0:9100
prometheus-cert=/config/certs/server.crt
prometheus-key=/config/certs/server.key
        

Filtering and Indexing

• Name filtering: Ignore specific counters by providing a comma-separated list to prometheus-ignore-names.

• Tag filtering: Ignore data sources by tag. By default, streaming metrics (like FI_metrics) are disabled as the pull-based exporter is not suitable for high-frequency streaming.

  Copy

  Copied!

              
  
              
  prometheus-ignore-tags=FI_metrics
          

• Custom indexing: Users can override default indexing to preserve multiple records with different index values. These index fields are automatically converted into Prometheus labels.

  Copy

  Copied!

              
  
              
  # Manual override for fset indexes (Autodetected in version 1.22+)
  prometheus-fset-indexes=idx1,idx2
          

Prometheus Aggregator Exporter

The Prometheus Aggregator Exporter creates a specialized endpoint that maintains the latest aggregated data from multiple sources. It labels data according to its origin (endpoint name), allowing for centralized viewing of distributed telemetry.

The following configuration example enables aggregation from two distinct endpoints with different sampling intervals (1000ms and 3000ms) and exports the combined data on port 33333.

            

            
# Enable the aggregator provider
enable-provider=prometheus_aggr
 
# Define aggregation endpoints: [Name],[URL],[Interval_ms]
prometheus_aggr_endpoint.0=AAA,http://0.0.0.0:9000,1000
prometheus_aggr_endpoint.1=BBB,http://0.0.0.0:9002/metrics,3000
 
# Configure the aggregated exporter endpoint
prometheus-aggr-exporter-host=0.0.0.0
prometheus-aggr-exporter-port=33333
        

Fluent Bit

Fluent Bit allows streaming to multiple destinations. Destinations are configured in .exp files that are documented in-place and can be found under:

            

            
/opt/mellanox/doca/services/telemetry/config/fluent_bit_configs
        

Fluent Bit allows exporting data via "Forward" protocol which connects to the Fluent Bit/FluentD instance on customer side.

Export can be enabled manually:

1. Uncomment the line with fluent_bit_configs=… in dts_config.ini.

2. Set enable=1 in required .exp files for the desired plugins.

3. Additional configurations can be set according to instructions in the .exp file if needed.

4. Restart the DTS.

5. Set up receiving instance of Fluent Bit/FluentD if needed.

6. See the data on the receiving side.

Export file destinations are set by configuring .exp files or creating new ones. It is recommended to start by going over documented example files. Documented examples exist for the following supported plugins:

• forward

• file

• stdout

• kafka

• es (elastic search)

• influx

Export File Configuration Details

Each export destination has the following fields:

• name – configuration name

• plugin_name – Fluent Bit plugin name

• enable – 1 or 0 values to enable/disable this destination

• host – the host for Fluent Bit plugin

• port – port for Fluent Bit plugin

• msgpack_data_layout – the msgpacked data format. Default is flb_std. The other option is custom. See section Msgpack Data Layout for details.

• plugin_key=val – key-value pairs of Fluent Bit plugin parameter (optional)

• counterset/fieldset – file paths (optional). See details in section Cset/Fset Filtering.

• source_tag=source_tag1,source_tag2 – comma-separated list of data page source tags for filtering. The rest tags are filtered out during export. Event tags are event provider names. All counters can be enabled/disabled only simultaneously with a counters keyword.

Msgpack Data Layout

Data layout can be configured using .exp files by setting msgpack_data_layout=layout. There are two available layouts: Standard and Custom.

The standard flb_std data layout is an array of 2 fields:

• timestamp double value

• a plain dictionary (key-value pairs)

The standard layout is appropriate for all Fluent Bit plugins. For example:

            

            
[timestamp_val, {"timestamp"->ts_val, type=>"counters/events", "source"=>"source_val", "key_1"=>val_1, "key_2"=>val_2,...}]
        

The custom data layout is a dictionary of meta-fields and counter fields. Values are placed into a separate plain dictionary. Custom data format can be dumped with stdout_raw output plugin of Fluent-Bit installed or can be forwarded with forward output plugin.

Counters example:

            

            
{"timestamp"=>timestamp_val, "type"=>"counters", "source"=>"source_val", "values"=> {"key_1"=>val_1, "key_2"=>val_2,...}}
        

Events example:

            

            
{"timestamp"=>timestamp_val, "type"=>"events", "type_name"=>"type_name_val", "source"=>" source_val", "values"=>{"key_1"=>val_1, "key_2"=>val_2,...}}
        

Cset/Fset Filtering

Each export file can optionally use one cset and one fset file to filter DTS counters and events data.

• cset contains tokens per line to filter data with "type"="counters".

• fset contains several blocks started with the header line [event_type_name] and tokens under that header. An Fset file is used to filter data with "type"="events".

  Note

  Event type names could be prefixed to apply the same tokens to all fitting types. For example, to filter all ethtool events, use [ethtool_event_*].

If several tokens must be matched simultaneously, use <tok1>+<tok2>+<tok3>. Exclusive tokens are available as well. For example, the line <tok1>+<tok2>-<tok3>-<tok4> filters names that match both tok1 and tok2 and do not match tok3 or tok4.

The following are the details of writing cset files:

            

            
# Put tokens on separate lines
# Tokens are the actual name 'fragments' to be matched
# port$ # match names ending with token "port"
# ^port # match names starting with token "port"
# ^port$ # include name that is exact token "port
# port+xmit # match names that contain both tokens "port" and "xmit"
# port-support # match names that contain the token "port" and do not match the "-" token "support"
#
# Tip: To disable counter export put a single token line that fits nothing
        

The following are the details of writing fset files:

            

            
# Put your events here
# Usage:
#
# [type_name_1]
# tokens
# [type_name_2]
# tokens
# [type_name_3]
# tokens
# ...
# Tokens are the actual name 'fragments' to be matched
# port$ # match names ending with token "port"
# ^port # match names starting with token "port"
# ^port$ # include name that is exact token "port
# port+xmit # match names that contain both tokens "port" and "xmit"
# port-support # match names that contain the token "port" and do not match the "-" token "support"
 
# The next example will export all the "tc" events and all events with type prefix "ethtool_" "ethtool" are filtered with token "port":
# [tc]
#
# [ethtool_*]
# packet
 
# To know which event type names are available check export and find field "type_name"=>"ethtool_event_p0"
# ...
# Corner cases:
# 1. Empty fset file will export all events.
# 2. Tokens written above/without [event_type] will be ignored.
# 3. If cannot open fset file, warning will be printed, all event types will be exported.
        

NetFlow Exporter

NetFlow exporter must be used when data is collected as NetFlow packets from the telemetry client applications. In this case, DOCA Telemetry Exporter NetFlow API sends NetFlow data packages to DTS via IPC. DTS uses NetFlow exporter to send data to the NetFlow collector (3rd party service).

To enable NetFlow exporter, set netflow-collector-ip and netflow-collector-port in dts_config.ini. netflow-collector-ip could be set either to IP or an address.

For additional information, refer to the dts_config.ini file.

OpenTelemetry Exporter

DTS) can stream telemetry data to an OpenTelemetry (OTel) receiver using the OpenTelemetry Protocol (OTLP). This exporter is disabled by default and supports both HTTP and gRPC transports with robust security options.

Transport and Security

• Protocols: Supports OTLP over HTTP (JSON encoded) and OTLP over gRPC (Protobuf).

• Security: Supports both insecure connections and mutual TLS (mTLS) using client certificates, keys, and CA bundles.

• Compatibility: Optimized for the OTLP Specification 1.3.0.

Configuration Examples (dts_config.ini)

Set the open-telemetry-receiver parameter to define your destination.

OTLP over HTTP

            

            
# Insecure HTTP (JSON)
open-telemetry-receiver=http://0.0.0.0:9502/v1/metrics
 
# Secure HTTP with TLS
open-telemetry-receiver=https://0.0.0.0:9504/v1/metrics
open-telemetry-cert-file=certs/client.pem
open-telemetry-key-file=certs/client.key
open-telemetry-ca-file=certs/ca.pem
        

This example expects the running Open Telemetry receiver to be bound to port 9502 on the host running the DTS instance. Section "Open Telemetry Metrics Receiver Sample Configuration" provides an example of the Open Telemetry receiver configuration.

OTLP over gRPC

            

            
# Insecure gRPC
open-telemetry-transport=grpc
open-telemetry-receiver=0.0.0.0:9501
 
# Secure gRPC with TLS
open-telemetry-transport=grpc
open-telemetry-receiver=0.0.0.0:9503
open-telemetry-cert-file=certs/client.pem
open-telemetry-key-file=certs/client.key
open-telemetry-ca-file=certs/ca.pem
        

Advanced Streaming Controls

You can fine-tune how OTel handles your data attributes and throughput:

• Filtering: Apply counter-set or field-set to the stream to limit data volume (see Counterset and Fieldset section for more details).

  Copy

  Copied!

              
  
              
  open-telemetry-counter-set=slow_counters
  open-telemetry-field-set=slow_counters
          

• Attribute mapping: By default, DTS uses space-efficient hierarchical attributes. For compatibility with backends like Prometheus DB, enable per-data-point attributes.

  Copy

  Copied!

              
  
              
  open-telemetry-with-data-point-attributes=true
          

• Performance tuning: Adjust the bulk size for combining data bursts (default is 100).

  Copy

  Copied!

              
  
              
  open-telemetry-bulk-size=20
          

• Serialization: To use the official Protobuf serializer instead of the default JSON encoder:

  Copy

  Copied!

              
  
              
  open-telemetry-fast-serializer=false
  open-telemetry-binary-protobuf=true
          

• Aggregation temporality: To ensure all "Sum" metrics (like packet counters) are treated as Cumulative values (which is the standard for high-reliability monitoring) configure this parameter:

  Copy

  Copied!

              
  
              
  # Ensures Sum metrics are marked as Cumulative (Industry Standard)
  open-telemetry-sum-is-always-cumulative=off
          

OpenTelemetry Receiver: Sample Collector Configuration

To receive data from DTS, configure your OpenTelemetry Collector (collector-config.yaml) as follows:

            

            
receivers:
  otlp/insecure:
    protocols:
      grpc:
        endpoint: 0.0.0.0:9501
      http:
        endpoint: 0.0.0.0:9502
  # Secure endpoints with TLS
  otlp/secure:
    protocols:
      grpc:
        endpoint: 0.0.0.0:9503
        tls:
          cert_file: /etc/otel/tls/server.crt
          key_file: /etc/otel/tls/server.key
      http:
        endpoint: 0.0.0.0:9504
        tls:
          cert_file: /etc/otel/tls/server.crt
          key_file: /etc/otel/tls/server.key
 
exporters:
  debug:
    verbosity: detailed
 
service:
  pipelines:
    traces:
      receivers: [otlp/insecure, otlp/secure]
      exporters: [debug]
    metrics:
      receivers: [otlp/insecure, otlp/secure]
      exporters: [debug]
    logs:
      receivers: [otlp/insecure, otlp/secure]
      exporters: [debug]
        

Run the Collector via Docker

            

            
docker run --rm --name "otel_collector" \
    -p 9501:9501 -p 9502:9502 -p 9503:9503 -p 9504:9504 \
    -v $(pwd)/collector-config.yaml:/etc/otelcol/config.yaml \
    -v $(pwd)/certs:/etc/otel/tls \
    otel/opentelemetry-collector
        

Prometheus Remote Write Exporter

DTS can stream the telemetry data towards an external Prometheus service with an enabled metrics receiver. If enabled, the DTS data exporter acts as a Prometheus remote write protocol "Sender", streaming telemetry data towards a Prometheus server using remote write protocol.

To enable telemetry data streaming:

            

            
remote-write-receiver=http://0.0.0.0:9090/api/v1/write
        

This example assumes the running Prometheus remote write receiver to be bound to port 9090 on the host running the DTS instance.

For example, a Prometheus remote write receiver could be run using the following command:

            

            
docker run -p 9090:9090 prom/prometheus --config.file=/etc/prometheus/prometheus.yml \
    --storage.tsdb.path=/prometheus --web.console.libraries=/usr/share/prometheus/console_libraries \
    --web.console.templates=/usr/share/prometheus/consoles --web.enable-remote-write-receiver
        

The Prometheus remote write exporter can be configured to apply counter-set or field-set to the data stream, similar to the Open Telemetry exporter:

            

            
remote-write-counter-set=slow_counters
remote-write-field-set=slow_counters
        

The Prometheus remote write exporter sends data in bulk, combining bursts when possible. By default, the size of the bulk is up to 100 data points. If necessary, the size of the bulk can be altered using the bulk-size configuration parameter. For example:

            

            
remote-write-bulk-size=20
        

Load Balancer Exporter

The load balancer exporter is an exporter which distributes or copies data between multiple instances of another (secondary) exporter.

For example, the number of instances may be set as follows depending on the use case:

When loadbalancer-INDEX-KEY and loadbalancer-exporter-config-var are not used, all instances are created equal, and their configuration is done according to other available configuration. If none of the instance configuration variables exist, that instance is created with other available configuration.

For example, the following is a section of the dts_config.ini configuration file which provides a sample of enabling load balancer to run 2 OTLP exporter instances:

            

            
loadbalancer-exporter-name=open_telemetry
loadbalancer-mode=replication
loadbalancer-0-open-telemetry-receiver=http://10.141.160.210:9502/v1/metrics
loadbalancer-1-open-telemetry-receiver=http://10.141.160.211:9502/v1/metrics
        

The default behavior is to export the entire collected data. To restrict the export interval of different instances (multirate export), add the required interval in seconds:

            

            
loadbalancer-0-open-telemetry-export-interval=30
loadbalancer-1-open-telemetry-export-interval=60
        

Distributed Real Time Analysis Exporter

The Real Time Analysis (RTA) Exporter is a powerful tool designed to export real-time analytics data from DTS. It provides seamless integration with various data sources and destinations, allowing you to efficiently collect, process, and export analytics information in real-time. One of its key features is the ability to run an HTTP server at a preconfigured port, allowing for remote management of Lua scripts used in data processing.

It allows to get relevant insights based on telemetry data without the overhead of aggregation and central analysis. Potential use cases can be running link health tests (by analyzing L1 data) before executing an AI job (saving cluster time by avoiding starting a job on a bad link), or any other custom logic that is able to run on-demand \ on-going and uses telemetry data as input.

To enable this exporter, the lua scripts folder path (in DTS container filesystem) should be specified in DTS config file

            

            
lua-script-dir=/config/rta/lua_scripts
        

The default DTS configuration folder contains several Lua scripts that can be used as an example. The section below describes the API

            

            
rta-enable-http-api=true
rta-http-port=1812
        

Additional Configuration Parameters

Lua Script Requirements

• If the script implements the isEnabled() function, it must return true for the script to be enabled. If this function is not implemented, the script will be disabled by default.

• Scripts must provide a set of metric names to be filtered for processing by implementing the getMetricNames() function.

• If a script is disabled, doesn't implement getMetricNames(), or returns an empty set of metric names, it will not be loaded.

Lua Scripts Workflow

The RTA Exporter plugin processes data in two ways:

1. 1. Periodic:

      • At the end of each data page, the plugin calls the processDataPageCompleted() Lua function.

      • This function allows the Lua script to process the collected data.

   2. On demand via http server - see <link>

Lua Script Examples

Lua scripts for RTA Exporter are available in /opt/mellanox/doca/services/telemetry/config/rta/lua_scripts. The following are examples for analyzing sysfs provider counters:

• sysfs_normalized_cbw.lua – Calculates the percentage of theoretical maximum bandwidth lost due to congestion on InfiniBand ports

• sysfs_check_link_state.lua – Monitors six key InfiniBand link metrics to verify that links are properly configured and operational

Lua Scripts API

There are 2 directions of function calls:

• DTS calls Lua script functions

• Lua script calls DTS functions

Lua Script Functions

2 functions are required - isEnabled and getMetricNames

isEnabled()

Determines if the script should be loaded.

            

            
function isEnabled()
    return true -- or false
end
        

getMetricNames()

Provides a set of metric names to be filtered for processing.

            

            
function getMetricNames()
    return {"metric1", "metric2", "metric3"}
end
        

processDataPageCompleted()

Processes collected data at the end of each data page.

            

            
function processDataPageCompleted()
    -- Implementation for processing data
end
        

getMetricsCapacity()

Defines the maximum size of the buffer that stores the most recent metric values.

            

            
function getMetricsCapacity()
    return 20 -- Example: Set the buffer size to 20 entries
end
        

processChangedMetricsOnly()

This function instructs the host application (CLX) to provide only metrics that have changed since the last time the script accessed them.

            

            
function processChangedMetricsOnly()
    return true -- Receive only updated metrics data
end
        

• Purpose: Optimizes data processing by instructing the host application to provide only metrics with new values since the script last accessed them. This can significantly improve performance and reduce unnecessary data transfer.

• Return value:

  • true: The host application will provide only metrics that have changed since the last access by the script.

  • false: The host application will provide all metrics, regardless of whether they've changed.

processDataMinPeriodMs()

Returns the minimum delay in milliseconds between calls to processDataPageCompleted(). This function helps reduce the frequency of calls if data pages are completed too quickly.