What can I help you with?
NVIDIA UFM Telemetry Documentation v1.19.0

Bug Fixes History

Ref. #

Description

3988610

Description: Increasing memory usage

Keywords: Memory, Usage

Discovered in Release: 1.17

4013278

Description: Prometheus remote write exporter is missing counter-level labels

Keywords: Prometheus, Remote, Write

Discovered in Release: 1.17

4013275

Description: OTLP exporter to handle counter-level labels

Keywords: OTLP Exporter

Discovered in Release: 1.17

3988901

Description: Missing Port info about link_speed

Keywords: link_speed_enabled

Discovered in Release: 1.17

3923318

Description: Link partner info is not collected when the counter_setoption is used

Keywords: link_partner

Discovered in Release: 1.17

3920844

Description: Wrong metric type is reported for BER in prometheus format

Keywords: Prometheus type, Gauge/Counter

Discovered in Release: 1.17

3898362

Description: [Multi-rate] is unable to collect down ports data without collecting cable data

Keywords: Down, Ports

Discovered in Release: 1.17

3961429

Description: CollectX client (clxcli) and bringup env_vars separator changed

Keywords: env_vars, Separator

Discovered in Release: 1.17

3863854

Description: Warning is reported once running run_bringup.sh

Keywords: Warning, Dependency

Discovered in Release: 1.17

3854242

Description: prometheus_configs/cset/ib_labels_lookup.xcset file is missing

Keywords: ib_labels_lookup.xcset

Discovered in Release: 1.17

3728036

Description: Extended counter sets with the labeled output in CSV format

Keywords: CSV, Labels

Discovered in Release: 1.17

3733919

Description: PEMI SNR does not return values – all 0

Keywords: PEMI, SNR empty values

Discovered in Release: 1.16

3828214

Description: PEMI SNR - only 4 lanes are shown in telemetry instead of 8

Keywords: PEMI, SNR, 4 lanes

Discovered in Release: 1.16

3841532

Description: Data from different sources show different timestamp

Keywords: Timestamp

Discovered in Release: 1.16

3807154

Description: Cannot set counter dt_num_imported_recent_ports, no context yet

Keywords: log error, dt

Discovered in Release: 1.16

3768320

Description: Missing RX Power from optical transceivers

Keywords: RX Power, Optical Transceivers

Discovered in Release: 1.15.3

3639657

Description: telemetry_initialize.sh ignores flags if they do not exist on the base configuration

Keywords: telemetry_initialize.sh, flag

Discovered in Release: 1.15.3

3689874

Description: Requesting specific fields in xcset leads to wrong Prometheus format output.

Keywords: prometheus

Discovered in Release: 1.15.6

3682405

Description: generate_basic_results_csvfails with 'ascii' codec due to inability to encode character ‘\xa0’

Keywords: ascii, generate_basic_results_csv

Discovered in Release: 1.15.6

3681267

Description: pack_data crashes Bringup CLI if directory does not exist

Keywords: pack_data, Bringup CLI

Discovered in Release: 1.15.6

3716902

Description: RxPowerType reported as empty

Keywords: RxPowerType, empty

Discovered in Release: 1.15.6

3734856

Description: HTTP server stops data transfer after 30 seconds

Keywords: HTTP server, stop, 30 seconds

Discovered in Release: 1.15.5

3745754

Description: reset_counters lacks option to reset disabled cables

Keywords: reset_counters, disabled cables

Discovered in Release: 1.15.6

3751954

Description: Running with CC configuration crashes after multiple iterations

Keywords: CC configuration, Crash

Discovered in Release: 1.15.0

3667820

NormalizedXmitData 100% - miscalculated

Keyword: NormalizedXmitData 100%, miscalculated field

Discovered in release: 1.14

3597364

Description: Multi-rate usage causes a crash

Keywords: Coredump, Multi-rate

Discovered in release: 1.14

3648187

Description: Hundreds of 'could not retrieve data for' warning messages

Keywords: Warn, Warning

Discovered in release: 1.14

3590777

Description: After upgrading UFM new telemetry data is not being collected and presented in UI Telemetry tab.

Keywords: Telemetry, Coredump

Discovered in release: 1.14

3442081

Description: The gen_metadata script fails and cannot generate metadata file.

Keywords: Metadata, Labels

Discovered in release: 1.12

3331553

Description: [installation]: no check for supervisord as a prerequisite.

Keywords: run_bringup, supervisord

Discovered in release: 1.12

3546901

Description: Invalid Cable Power data retrieved

Keywords: Cable power, Power infinity

Discovered in release: 1.13

3481178

Description: Fixed UFM-Telemetry data refresh delay.

Keywords: Data Delay; Refresh;

Discovered in release: 1.13.5

3477852

Description: Calculated values are set to (-1) when unavailable.

Keywords: Unavailable Data;

Discovered in release: 1.13.5

3481011

Description: Fixed cable voltage values presentation in microvolts (uV) instead of millivolts (mV).

Keywords: Cable; Voltage; Unit of measurement;

Discovered in release: 1.13.5

3438270

Description: Added support to starting UFM telemetry using the bringup tool with multi_port_sm configured.

Keywords: Multiple HCAs; HCA list;

Discovered in release: 1.13.5

3488824

Description: Fixed MADs statistics clearance between iterations.

Keywords: MAD Statistics;

Discovered in release: 1.13.5

3477561

Description: Fixed faulty Telemetry results in segmentation.

Keywords: segfault;

Discovered in release: 1.13.5

3461058

Description: UFM Telemetry log rotation applications are not used

Keywords: Log, rotation applications

Discovered in release: 1.13.0

3397908

Description: UFM Telemetry fails to start due to supervisord

Keywords: supervisord, start failure

Discovered in release: 1.12.0

3440964

Description: _MEI folders created in tmp are not removed

Keywords: _MEI, temp folders

Discovered in release: 1.12.0

3214566

Description: Dynamic Fabric unexpected header warning

Keywords: Log warning, Dynamic Fabric

Discovered in release: 1.10

3292572

Description: [systemd service]: executable path is not absolute

Keywords: systemd

Discovered in release: 1.11

3305919

Description: Timestamp error in CSV serialize of fset

Keywords: timestamp

Discovered in release: 1.11

3306476

Description: HTTP endpoint: fset "[CableInfo]Temperature" is not rendered

Keywords: HTTP endpoint, Cable temperature

Discovered in release: 1.11

3327188

Description: Missing amBER configuration for fluentbit

Keywords: amber, Container configuration

Discovered in release: 1.11

3327193

Descrition: CSV output contains redundant comma at EOL

Keywords: csv

Discovered in release: 1.11

3332112

Description: Failed to execute script 'gen_metadata'

Keyword: get_metadata

Discovered in release:1.11

3229888

Description: Error on launch_ibdiagnet.log

Keywords: Log Error

Discovered in release: 1.10

3225679

Description: Adding eff_ber counter to low_freq.cset file

Keywords: cset, eff_ber

Discovered in release: 1.10

N/A

Description: The Temperature fset does not contain labels

Keywords: Temperature, labels

Discovered in release: 1.9

N/A

Description: Cannot rename fields in fset

Keywords: fset, events

Discovered in release: 1.9

N/A

Description: The node_guid field in fsets is missing leading zeros

Keywords: fset, events

Discovered in release: 1.9

3051843

Description: Can't start telemetry without setting arg_12= to empty

Keywords: startup failure

Discovered in release: 1.9

3076045

Description: gen_metadata doesn't use a rules file when provided

Keywords: metadata, labels

Discovered in release: 1.9

2943459

Description: Custom labels lost after UFM restart

Keywords: UFM restart, labels

Discovered in release: 1.8

2921452

Description: raw_ber from UFM Telemetry doesn't match output of mlxlink -e -c -m

Keywords: BER

Discovered in release: 1.8

2923525

Description: Enabling BER collection on an NDR fabric causes segmentation fault of ibdiagnet

Keywords: BER, ibdiagnet, segmentation fault

Discovered in release: 1.8

3037715

Description: ConnectX-7 infiniband_CBW, Normalized_CBW are always zero

Keywords: InfiniBand

Discovered in release: 1.8

3018638

Description: Normalized TX data returned 0% for ConnectX-7

Keywords: TX data

Discovered in release: 1.8

© Copyright 2024, NVIDIA. Last updated on Nov 7, 2024.