fix: config files for dcgm exporter (#823)

* fix: dcgm exporter config * refactor: rename dirs for dcgm exporter
flashcatcloud · Mar 8, 2024 · 0205216 · 0205216
1 parent ca97095
commit 0205216
Show file tree

Hide file tree

Showing 25 changed files with 245 additions and 12 deletions.
diff --git a/agent/metrics_agent.go b/agent/metrics_agent.go
@@ -23,7 +23,7 @@ import (
 	_ "flashcat.cloud/categraf/inputs/conntrack"
 	_ "flashcat.cloud/categraf/inputs/consul"
 	_ "flashcat.cloud/categraf/inputs/cpu"
-	_ "flashcat.cloud/categraf/inputs/dcgm_exporter"
+	_ "flashcat.cloud/categraf/inputs/dcgm"
 	_ "flashcat.cloud/categraf/inputs/disk"
 	_ "flashcat.cloud/categraf/inputs/diskio"
 	_ "flashcat.cloud/categraf/inputs/dns_query"

diff --git a/conf/input.dcgm/1.x-compatibility-metrics.csv b/conf/input.dcgm/1.x-compatibility-metrics.csv
@@ -0,0 +1,66 @@
+# Format
+# If line starts with a '#' it is considered a comment
+# DCGM FIELD, Prometheus metric type, help message
+
+# Clocks
+dcgm_sm_clock,     gauge, SM clock frequency (in MHz).
+dcgm_memory_clock, gauge, Memory clock frequency (in MHz).
+
+# Temperature
+dcgm_memory_temp, gauge, Memory temperature (in C).
+dcgm_gpu_temp,    gauge, GPU temperature (in C).
+
+# Power
+dcgm_power_usage,              gauge, Power draw (in W).
+dcgm_total_energy_consumption, counter, Total energy consumption since boot (in mJ).
+
+# PCIe
+dcgm_pcie_tx_throughput,  counter, Total number of bytes transmitted through PCIe TX (in KB) via NVML.
+dcgm_pcie_rx_throughput,  counter, Total number of bytes received through PCIe RX (in KB) via NVML.
+dcgm_pcie_replay_counter, counter, Total number of PCIe retries.
+
+# Utilization (the sample period varies depending on the product)
+dcgm_gpu_utilization,      gauge, GPU utilization (in %).
+dcgm_mem_copy_utilization, gauge, Memory utilization (in %).
+dcgm_enc_utilization,      gauge, Encoder utilization (in %).
+dcgm_dec_utilization,      gauge, Decoder utilization (in %).
+
+# Errors and violations
+dcgm_xid_errors,            gauge, Value of the last XID error encountered.
+# dcgm_power_violation,       counter, Throttling duration due to power constraints (in us).
+# dcgm_thermal_violation,     counter, Throttling duration due to thermal constraints (in us).
+# dcgm_sync_boost_violation,  counter, Throttling duration due to sync-boost constraints (in us).
+# dcgm_board_limit_violation, counter, Throttling duration due to board limit constraints (in us).
+# dcgm_low_util_violation,    counter, Throttling duration due to low utilization (in us).
+# dcgm_reliability_violation, counter, Throttling duration due to reliability constraints (in us).
+
+# Memory usage
+dcgm_fb_free, gauge, Framebuffer memory free (in MiB).
+dcgm_fb_used, gauge, Framebuffer memory used (in MiB).
+
+# ECC
+# dcgm_ecc_sbe_volatile_total,  counter, Total number of single-bit volatile ECC errors.
+# dcgm_ecc_dbe_volatile_total,  counter, Total number of double-bit volatile ECC errors.
+# dcgm_ecc_sbe_aggregate_total, counter, Total number of single-bit persistent ECC errors.
+# dcgm_ecc_dbe_aggregate_total, counter, Total number of double-bit persistent ECC errors.
+
+# Retired pages
+# dcgm_retired_pages_sbe,     counter, Total number of retired pages due to single-bit errors.
+# dcgm_retired_pages_dbe,     counter, Total number of retired pages due to double-bit errors.
+# dcgm_retired_pages_pending, counter, Total number of pages pending retirement.
+
+# NVLink
+# dcgm_nvlink_flit_crc_error_count_total, counter, Total number of NVLink flow-control CRC errors.
+# dcgm_nvlink_data_crc_error_count_total, counter, Total number of NVLink data CRC errors.
+# dcgm_nvlink_replay_error_count_total,   counter, Total number of NVLink retries.
+# dcgm_nvlink_recovery_error_count_total, counter, Total number of NVLink recovery errors.
+dcgm_nvlink_bandwidth_total,            counter, Total number of NVLink bandwidth counters for all lanes
+
+# Add DCP metrics
+dcgm_fi_prof_gr_engine_active,   gauge, Ratio of time the graphics engine is active (in %).
+# dcgm_fi_prof_sm_active,          gauge, The ratio of cycles an SM has at least 1 warp assigned (in %).
+# dcgm_fi_prof_sm_occupancy,       gauge, The ratio of number of warps resident on an SM (in %).
+dcgm_fi_prof_pipe_tensor_active, gauge, Ratio of cycles the tensor (HMMA) pipe is active (in %).
+dcgm_fi_prof_dram_active,        gauge, Ratio of cycles the device memory interface is active sending or receiving data (in %).
+dcgm_fi_prof_pcie_tx_bytes,      counter, The number of bytes of active pcie tx data including both header and payload.
+dcgm_fi_prof_pcie_rx_bytes,      counter, The number of bytes of active pcie rx data including both header and payload.
diff --git a/conf/input.dcgm/dcp-metrics-included.csv b/conf/input.dcgm/dcp-metrics-included.csv
@@ -0,0 +1,90 @@
+# Format
+# If line starts with a '#' it is considered a comment
+# DCGM FIELD, Prometheus metric type, help message
+
+# Clocks
+DCGM_FI_DEV_SM_CLOCK,  gauge, SM clock frequency (in MHz).
+DCGM_FI_DEV_MEM_CLOCK, gauge, Memory clock frequency (in MHz).
+
+# Temperature
+DCGM_FI_DEV_MEMORY_TEMP, gauge, Memory temperature (in C).
+DCGM_FI_DEV_GPU_TEMP,    gauge, GPU temperature (in C).
+
+# Power
+DCGM_FI_DEV_POWER_USAGE,              gauge, Power draw (in W).
+DCGM_FI_DEV_TOTAL_ENERGY_CONSUMPTION, counter, Total energy consumption since boot (in mJ).
+
+# PCIE
+# DCGM_FI_DEV_PCIE_TX_THROUGHPUT,  counter, Total number of bytes transmitted through PCIe TX (in KB) via NVML.
+# DCGM_FI_DEV_PCIE_RX_THROUGHPUT,  counter, Total number of bytes received through PCIe RX (in KB) via NVML.
+DCGM_FI_DEV_PCIE_REPLAY_COUNTER, counter, Total number of PCIe retries.
+
+# Utilization (the sample period varies depending on the product)
+DCGM_FI_DEV_GPU_UTIL,      gauge, GPU utilization (in %).
+DCGM_FI_DEV_MEM_COPY_UTIL, gauge, Memory utilization (in %).
+DCGM_FI_DEV_ENC_UTIL,      gauge, Encoder utilization (in %).
+DCGM_FI_DEV_DEC_UTIL ,     gauge, Decoder utilization (in %).
+
+# Errors and violations
+DCGM_FI_DEV_XID_ERRORS,            gauge,   Value of the last XID error encountered.
+# DCGM_FI_DEV_POWER_VIOLATION,       counter, Throttling duration due to power constraints (in us).
+# DCGM_FI_DEV_THERMAL_VIOLATION,     counter, Throttling duration due to thermal constraints (in us).
+# DCGM_FI_DEV_SYNC_BOOST_VIOLATION,  counter, Throttling duration due to sync-boost constraints (in us).
+# DCGM_FI_DEV_BOARD_LIMIT_VIOLATION, counter, Throttling duration due to board limit constraints (in us).
+# DCGM_FI_DEV_LOW_UTIL_VIOLATION,    counter, Throttling duration due to low utilization (in us).
+# DCGM_FI_DEV_RELIABILITY_VIOLATION, counter, Throttling duration due to reliability constraints (in us).
+
+# Memory usage
+DCGM_FI_DEV_FB_FREE, gauge, Framebuffer memory free (in MiB).
+DCGM_FI_DEV_FB_USED, gauge, Framebuffer memory used (in MiB).
+
+# ECC
+# DCGM_FI_DEV_ECC_SBE_VOL_TOTAL, counter, Total number of single-bit volatile ECC errors.
+# DCGM_FI_DEV_ECC_DBE_VOL_TOTAL, counter, Total number of double-bit volatile ECC errors.
+# DCGM_FI_DEV_ECC_SBE_AGG_TOTAL, counter, Total number of single-bit persistent ECC errors.
+# DCGM_FI_DEV_ECC_DBE_AGG_TOTAL, counter, Total number of double-bit persistent ECC errors.
+
+# Retired pages
+# DCGM_FI_DEV_RETIRED_SBE,     counter, Total number of retired pages due to single-bit errors.
+# DCGM_FI_DEV_RETIRED_DBE,     counter, Total number of retired pages due to double-bit errors.
+# DCGM_FI_DEV_RETIRED_PENDING, counter, Total number of pages pending retirement.
+
+# NVLink
+# DCGM_FI_DEV_NVLINK_CRC_FLIT_ERROR_COUNT_TOTAL, counter, Total number of NVLink flow-control CRC errors.
+# DCGM_FI_DEV_NVLINK_CRC_DATA_ERROR_COUNT_TOTAL, counter, Total number of NVLink data CRC errors.
+# DCGM_FI_DEV_NVLINK_REPLAY_ERROR_COUNT_TOTAL,   counter, Total number of NVLink retries.
+# DCGM_FI_DEV_NVLINK_RECOVERY_ERROR_COUNT_TOTAL, counter, Total number of NVLink recovery errors.
+DCGM_FI_DEV_NVLINK_BANDWIDTH_TOTAL,            counter, Total number of NVLink bandwidth counters for all lanes.
+# DCGM_FI_DEV_NVLINK_BANDWIDTH_L0,               counter, The number of bytes of active NVLink rx or tx data including both header and payload.
+
+# VGPU License status
+DCGM_FI_DEV_VGPU_LICENSE_STATUS, gauge, vGPU License status
+
+# Remapped rows
+DCGM_FI_DEV_UNCORRECTABLE_REMAPPED_ROWS, counter, Number of remapped rows for uncorrectable errors
+DCGM_FI_DEV_CORRECTABLE_REMAPPED_ROWS,   counter, Number of remapped rows for correctable errors
+DCGM_FI_DEV_ROW_REMAP_FAILURE,           gauge,   Whether remapping of rows has failed
+
+# Static configuration information. These appear as labels on the other metrics
+DCGM_FI_DRIVER_VERSION,        label, Driver Version
+# DCGM_FI_NVML_VERSION,          label, NVML Version
+# DCGM_FI_DEV_BRAND,             label, Device Brand
+# DCGM_FI_DEV_SERIAL,            label, Device Serial Number
+# DCGM_FI_DEV_OEM_INFOROM_VER,   label, OEM inforom version
+# DCGM_FI_DEV_ECC_INFOROM_VER,   label, ECC inforom version
+# DCGM_FI_DEV_POWER_INFOROM_VER, label, Power management object inforom version
+# DCGM_FI_DEV_INFOROM_IMAGE_VER, label, Inforom image version
+# DCGM_FI_DEV_VBIOS_VERSION,     label, VBIOS version of the device
+
+# DCP metrics
+DCGM_FI_PROF_GR_ENGINE_ACTIVE,   gauge, Ratio of time the graphics engine is active (in %).
+# DCGM_FI_PROF_SM_ACTIVE,          gauge, The ratio of cycles an SM has at least 1 warp assigned (in %).
+# DCGM_FI_PROF_SM_OCCUPANCY,       gauge, The ratio of number of warps resident on an SM (in %).
+DCGM_FI_PROF_PIPE_TENSOR_ACTIVE, gauge, Ratio of cycles the tensor (HMMA) pipe is active (in %).
+DCGM_FI_PROF_DRAM_ACTIVE,        gauge, Ratio of cycles the device memory interface is active sending or receiving data (in %).
+# DCGM_FI_PROF_PIPE_FP64_ACTIVE,   gauge, Ratio of cycles the fp64 pipes are active (in %).
+# DCGM_FI_PROF_PIPE_FP32_ACTIVE,   gauge, Ratio of cycles the fp32 pipes are active (in %).
+# DCGM_FI_PROF_PIPE_FP16_ACTIVE,   gauge, Ratio of cycles the fp16 pipes are active (in %).
+DCGM_FI_PROF_PCIE_TX_BYTES,      gauge, The rate of data transmitted over the PCIe bus - including both protocol headers and data payloads - in bytes per second.
+DCGM_FI_PROF_PCIE_RX_BYTES,      gauge, The rate of data received over the PCIe bus - including both protocol headers and data payloads - in bytes per second.
+
diff --git a/conf/input.dcgm/default-counters.csv b/conf/input.dcgm/default-counters.csv
@@ -0,0 +1,77 @@
+# Format
+# If line starts with a '#' it is considered a comment
+# DCGM FIELD, Prometheus metric type, help message
+
+# Clocks
+DCGM_FI_DEV_SM_CLOCK,  gauge, SM clock frequency (in MHz).
+DCGM_FI_DEV_MEM_CLOCK, gauge, Memory clock frequency (in MHz).
+# DCGM_EXP_CLOCK_EVENTS_COUNT, gauge, Count of clock events within the user-specified time window (see clock-events-count-window-size param).
+
+# Temperature
+DCGM_FI_DEV_MEMORY_TEMP, gauge, Memory temperature (in C).
+DCGM_FI_DEV_GPU_TEMP,    gauge, GPU temperature (in C).
+
+# Power
+DCGM_FI_DEV_POWER_USAGE,              gauge, Power draw (in W).
+DCGM_FI_DEV_TOTAL_ENERGY_CONSUMPTION, counter, Total energy consumption since boot (in mJ).
+
+# PCIE
+DCGM_FI_DEV_PCIE_TX_THROUGHPUT,  counter, Total number of bytes transmitted through PCIe TX (in KB) via NVML.
+DCGM_FI_DEV_PCIE_RX_THROUGHPUT,  counter, Total number of bytes received through PCIe RX (in KB) via NVML.
+DCGM_FI_DEV_PCIE_REPLAY_COUNTER, counter, Total number of PCIe retries.
+
+# Utilization (the sample period varies depending on the product)
+DCGM_FI_DEV_GPU_UTIL,      gauge, GPU utilization (in %).
+DCGM_FI_DEV_MEM_COPY_UTIL, gauge, Memory utilization (in %).
+DCGM_FI_DEV_ENC_UTIL,      gauge, Encoder utilization (in %).
+DCGM_FI_DEV_DEC_UTIL ,     gauge, Decoder utilization (in %).
+
+# Errors and violations
+DCGM_FI_DEV_XID_ERRORS,              gauge,   Value of the last XID error encountered.
+# DCGM_FI_DEV_POWER_VIOLATION,       counter, Throttling duration due to power constraints (in us).
+# DCGM_FI_DEV_THERMAL_VIOLATION,     counter, Throttling duration due to thermal constraints (in us).
+# DCGM_FI_DEV_SYNC_BOOST_VIOLATION,  counter, Throttling duration due to sync-boost constraints (in us).
+# DCGM_FI_DEV_BOARD_LIMIT_VIOLATION, counter, Throttling duration due to board limit constraints (in us).
+# DCGM_FI_DEV_LOW_UTIL_VIOLATION,    counter, Throttling duration due to low utilization (in us).
+# DCGM_FI_DEV_RELIABILITY_VIOLATION, counter, Throttling duration due to reliability constraints (in us).
+# DCGM_EXP_XID_ERRORS_COUNT,         gauge,   Count of XID Errors within user-specified time window (see xid-count-window-size param).
+# Memory usage
+DCGM_FI_DEV_FB_FREE, gauge, Frame buffer memory free (in MB).
+DCGM_FI_DEV_FB_USED, gauge, Frame buffer memory used (in MB).
+
+# ECC
+# DCGM_FI_DEV_ECC_SBE_VOL_TOTAL, counter, Total number of single-bit volatile ECC errors.
+# DCGM_FI_DEV_ECC_DBE_VOL_TOTAL, counter, Total number of double-bit volatile ECC errors.
+# DCGM_FI_DEV_ECC_SBE_AGG_TOTAL, counter, Total number of single-bit persistent ECC errors.
+# DCGM_FI_DEV_ECC_DBE_AGG_TOTAL, counter, Total number of double-bit persistent ECC errors.
+
+# Retired pages
+# DCGM_FI_DEV_RETIRED_SBE,     counter, Total number of retired pages due to single-bit errors.
+# DCGM_FI_DEV_RETIRED_DBE,     counter, Total number of retired pages due to double-bit errors.
+# DCGM_FI_DEV_RETIRED_PENDING, counter, Total number of pages pending retirement.
+
+# NVLink
+# DCGM_FI_DEV_NVLINK_CRC_FLIT_ERROR_COUNT_TOTAL, counter, Total number of NVLink flow-control CRC errors.
+# DCGM_FI_DEV_NVLINK_CRC_DATA_ERROR_COUNT_TOTAL, counter, Total number of NVLink data CRC errors.
+# DCGM_FI_DEV_NVLINK_REPLAY_ERROR_COUNT_TOTAL,   counter, Total number of NVLink retries.
+# DCGM_FI_DEV_NVLINK_RECOVERY_ERROR_COUNT_TOTAL, counter, Total number of NVLink recovery errors.
+DCGM_FI_DEV_NVLINK_BANDWIDTH_TOTAL,            counter, Total number of NVLink bandwidth counters for all lanes
+
+# VGPU License status
+DCGM_FI_DEV_VGPU_LICENSE_STATUS, gauge, vGPU License status
+
+# Remapped rows
+DCGM_FI_DEV_UNCORRECTABLE_REMAPPED_ROWS, counter, Number of remapped rows for uncorrectable errors
+DCGM_FI_DEV_CORRECTABLE_REMAPPED_ROWS,   counter, Number of remapped rows for correctable errors
+DCGM_FI_DEV_ROW_REMAP_FAILURE,           gauge,   Whether remapping of rows has failed
+
+# Static configuration information. These appear as labels on the other metrics
+DCGM_FI_DRIVER_VERSION,        label, Driver Version
+# DCGM_FI_NVML_VERSION,          label, NVML Version
+# DCGM_FI_DEV_BRAND,             label, Device Brand
+# DCGM_FI_DEV_SERIAL,            label, Device Serial Number
+# DCGM_FI_DEV_OEM_INFOROM_VER,   label, OEM inforom version
+# DCGM_FI_DEV_ECC_INFOROM_VER,   label, ECC inforom version
+# DCGM_FI_DEV_POWER_INFOROM_VER, label, Power management object inforom version
+# DCGM_FI_DEV_INFOROM_IMAGE_VER, label, Inforom image version
+# DCGM_FI_DEV_VBIOS_VERSION,     label, VBIOS version of the device
diff --git a/conf/input.dcgm/exporter.toml b/conf/input.dcgm/exporter.toml
@@ -1,6 +1,6 @@
 #[[instances]]
 # path to the file, that contains the DCGM fields to collect
-# collectors = "/etc/dcgm-exporter/default-counters.csv"
+# collectors = "conf/input.dcgm/default-counters.csv"
 
 # Enable kubernetes mapping metrics to kubernetes pods
 # kubernetes=false
@@ -11,15 +11,15 @@
 # Use old 1.x namespace
 # use-old-namespace = false
 
-# cpu-devices = "f"
+  cpu-devices = "f"
 
 # gpu devices
-# devices = "f"
+  devices = "f"
 
-# switch-devices = "f"
+  switch-devices = "f"
 
 # ConfigMap <NAMESPACE>:<NAME> for metric data
-# configmap-data = "none"
+  configmap-data = "none"
 
 # Connect to remote hostengine at <HOST>:<PORT>
 # remote-hostengine-info = "localhost:5555"
@@ -28,4 +28,4 @@
 # fake-gpus = false
 
 # Replaces every blank space in the GPU model name with a dash, ensuring a continuous, space-free identifier.
-# replace-blanks-in-model-name = false
+# replace-blanks-in-model-name = false
diff --git a/...er/dcgmexporter/clock_events_collector.go → ...gm/dcgmexporter/clock_events_collector.go b/...er/dcgmexporter/clock_events_collector.go → ...gm/dcgmexporter/clock_events_collector.go
diff --git a/inputs/dcgm_exporter/dcgmexporter/config.go → inputs/dcgm/dcgmexporter/config.go b/inputs/dcgm_exporter/dcgmexporter/config.go → inputs/dcgm/dcgmexporter/config.go
diff --git a/inputs/dcgm_exporter/dcgmexporter/const.go → inputs/dcgm/dcgmexporter/const.go b/inputs/dcgm_exporter/dcgmexporter/const.go → inputs/dcgm/dcgmexporter/const.go
diff --git a/inputs/dcgm_exporter/dcgmexporter/dcgm.go → inputs/dcgm/dcgmexporter/dcgm.go b/inputs/dcgm_exporter/dcgmexporter/dcgm.go → inputs/dcgm/dcgmexporter/dcgm.go
diff --git a/...cgm_exporter/dcgmexporter/expcollector.go → inputs/dcgm/dcgmexporter/expcollector.go b/...cgm_exporter/dcgmexporter/expcollector.go → inputs/dcgm/dcgmexporter/expcollector.go
diff --git a/...exporter/dcgmexporter/exporter_metrics.go → inputs/dcgm/dcgmexporter/exporter_metrics.go b/...exporter/dcgmexporter/exporter_metrics.go → inputs/dcgm/dcgmexporter/exporter_metrics.go
diff --git a/...xporter/field_entity_group_system_info.go → ...xporter/field_entity_group_system_info.go b/...xporter/field_entity_group_system_info.go → ...xporter/field_entity_group_system_info.go
diff --git a/...gm_exporter/dcgmexporter/gpu_collector.go → inputs/dcgm/dcgmexporter/gpu_collector.go b/...gm_exporter/dcgmexporter/gpu_collector.go → inputs/dcgm/dcgmexporter/gpu_collector.go
diff --git a/.../dcgm_exporter/dcgmexporter/kubernetes.go → inputs/dcgm/dcgmexporter/kubernetes.go b/.../dcgm_exporter/dcgmexporter/kubernetes.go → inputs/dcgm/dcgmexporter/kubernetes.go
diff --git a/inputs/dcgm_exporter/dcgmexporter/parser.go → inputs/dcgm/dcgmexporter/parser.go b/inputs/dcgm_exporter/dcgmexporter/parser.go → inputs/dcgm/dcgmexporter/parser.go
diff --git a/...ts/dcgm_exporter/dcgmexporter/pipeline.go → inputs/dcgm/dcgmexporter/pipeline.go b/...ts/dcgm_exporter/dcgmexporter/pipeline.go → inputs/dcgm/dcgmexporter/pipeline.go
diff --git a/...ts/dcgm_exporter/dcgmexporter/registry.go → inputs/dcgm/dcgmexporter/registry.go b/...ts/dcgm_exporter/dcgmexporter/registry.go → inputs/dcgm/dcgmexporter/registry.go
diff --git a/inputs/dcgm_exporter/dcgmexporter/server.go → inputs/dcgm/dcgmexporter/server.go b/inputs/dcgm_exporter/dcgmexporter/server.go → inputs/dcgm/dcgmexporter/server.go
diff --git a/...dcgm_exporter/dcgmexporter/system_info.go → inputs/dcgm/dcgmexporter/system_info.go b/...dcgm_exporter/dcgmexporter/system_info.go → inputs/dcgm/dcgmexporter/system_info.go
diff --git a/inputs/dcgm_exporter/dcgmexporter/types.go → inputs/dcgm/dcgmexporter/types.go b/inputs/dcgm_exporter/dcgmexporter/types.go → inputs/dcgm/dcgmexporter/types.go
diff --git a/inputs/dcgm_exporter/dcgmexporter/utils.go → inputs/dcgm/dcgmexporter/utils.go b/inputs/dcgm_exporter/dcgmexporter/utils.go → inputs/dcgm/dcgmexporter/utils.go
diff --git a/...gm_exporter/dcgmexporter/xid_collector.go → inputs/dcgm/dcgmexporter/xid_collector.go b/...gm_exporter/dcgmexporter/xid_collector.go → inputs/dcgm/dcgmexporter/xid_collector.go
diff --git a/inputs/dcgm_exporter/exporter.go → inputs/dcgm/exporter.go b/inputs/dcgm_exporter/exporter.go → inputs/dcgm/exporter.go
@@ -1,6 +1,6 @@
 //go:build dcgm
 
-package dcgm_exporter
+package dcgm
 
 import (
 	"bytes"
@@ -15,7 +15,7 @@ import (
 
 	"flashcat.cloud/categraf/config"
 	"flashcat.cloud/categraf/inputs"
-	"flashcat.cloud/categraf/inputs/dcgm_exporter/dcgmexporter"
+	"flashcat.cloud/categraf/inputs/dcgm/dcgmexporter"
 	"flashcat.cloud/categraf/parser/prometheus"
 	"flashcat.cloud/categraf/types"
 )
@@ -176,7 +176,7 @@ func (ins *Instance) Init() error {
 	groups, err = dcgm.GetSupportedMetricGroups(0)
 	if err != nil {
 		cfg.CollectDCP = false
-		log.Printf("Not collecting DCP metrics: ", err)
+		log.Println("Not collecting DCP metrics: ", err)
 	} else {
 		log.Println("Collecting DCP Metrics")
 		cfg.MetricGroups = groups
@@ -349,7 +349,7 @@ func (ins *Instance) Gather(slist *types.SampleList) {
 		log.Println("E! dcgm exporter collects error:", err)
 		return
 	}
-	parser := prometheus.NewParser(inputName, labels, http.Header{}, false,
+	parser := prometheus.NewParser("", labels, http.Header{}, false,
 		nil, nil)
 	err = parser.Parse(buf.Bytes(), slist)
 	if err != nil {

diff --git a/inputs/dcgm/exporter_none.go b/inputs/dcgm/exporter_none.go
@@ -0,0 +1 @@
+package dcgm
diff --git a/inputs/dcgm_exporter/exporter_none.go b/inputs/dcgm_exporter/exporter_none.go