Mellanox WinOF VPI Documentation v5.50.53000
v6.5

Reported Driver Events

The driver records events in the system log of the Windows server event system which can be used to identify, diagnose, and predict sources of system problems.

To see the log of events, open System Event Viewer as follows:

  • Right click on My Computer, click Manage, and then click Event Viewer.

OR

  1. Click Start → Run and enter "eventvwr.exe".

  2. In Event Viewer, select the system log.

Severity

Message ID

Description

Error

0x0002

Failed to initialize < >.

0x0004

< > has been configured to use RSS while Windows' TCP RSS is disabled. This configuration prevents the initialization and enabling of the port. You need to either enable Windows' TCP RSS, or configure the adapter's port to disable RSS. For details, see the README file under the documentation folder.

0x000A

Failed to initialize the %2 Interface because it uses old firmware version (< >). You need to burn firmware version < > or higher, and to restart your computer. For details, see the README file under the documentation folder.

0x00022

There is an interface mismatch between ETH driver and the bus driver. The ETH driver interface version is < > while the bus driver interface version is < >. As a result the ETH driver has failed to start. This happened due to setup failures or partial update of the drivers. In order to resolve the issue, please reboot the computer. In case the problem had not been resolved, please reinstall the Driver Package.

0x0027

For port < >, the following address is considered as multicast address: < >. Please configure the registry value NetworkAddress with another address in the registry, then restart the driver.

0x0028

For port < >, OID_QOS_PARAMETERS was not called, or called with invalid parameters.

0x0033

Can't support VEA (Virtual Ethernet Adapter) feature.

Reason: Firmware doesn't support "Multiple MACs Per Port" feature.

Resolution: Please update your firmware.

0x0034

Can't support VEA (Virtual Ethernet Adapter) feature.

Reason: Firmware doesn't support DMS (Data Message Steering) feature.

Resolution: Please update your firmware.

0x00038

The registry enabled ignore FCS, but the firmware does not support it. Please update the firmware.

0x004A

The NDIS version of the driver (< >) is not supported by the OS (< >)

0x004B

DriverEntry failed with status < >.

0x0055

Failed to open Channel Adapter.

0x0056

Failed to enable NDK with status < >.

0x0057

Failed to disable NDK with status < >.

0x0058

CQ CPU mismatch, newly created CQ was connected to EQ (< >) which is affinitized to CPU (%4) instead of CPU (< >). (The issue is reported in Function < >)

Warning

0x0005

< > detected that Head of Queue feature is disabled. It is recommended to enable it in order to prevent the system from hanging.
For more details, please refer to WinOF User Manual.

0x0006

< > detected that Head-of-Queue life limit value (< >) does not correspond with the Self-Healing feature configuration - CheckForHangCQMaxNoProgress = < >, SHCheckForHangTimeInSeconds = < >
CheckForHangCQMaxNoProgress value is increased to < >
For more details, please refer to WinOF User Manual.

0x0008

< > reports a "Set packet Filter failed" with status #< >.(The issue is reported in Function < >).

0x000C

< > device couldn’t create the optimal number of MSI-X vectors. The Network interface will use only one CPU for interrupts, although RSS will still be in effect. This may slightly affect performance. For fully optimal performance, configure the number of MSI-X vectors to be at least < >.

0x000E

< > device detected that the link connected to port < > is down. This can occur if the physical link is disconnected or damaged, or if the other end-port is down.

0x000F

< > device has been configured not to use RSS. This configuration significantly affect the network performance.

0x0010

< > device reports an "Error event" on cqn #< >. The event type is: < >.(The issue is reported in Function < >).

0x0012

< > device reports a "Test operation error" on Port #< > with status #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0013

< > device reports a send=< > "CQE error" on cqn #<X> qpn #<Y> cqe_error->syndrome < >, cqe_error->vendor_error_syndrome < >, Opcode < >, Corrupted CQE < >. HCA NIC will be reset if resiliency is enabled. (The issue is reported in Function < >).

0x0014

< > device reports an "EQ stuck" on eqn #< > qpn #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0015

< > device reports a "TX cq stuck" on cqn #< > uncompleted send #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0016

< > device reports an "Rx ring stuck" on cqn #< > srqn #< > . Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0017

, > device reports a "Set information failed" on port #< > Oid #< > Status #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0018

< > device reports a "Rearm CQ Failed for NOP" on Rx ring, cqn #< > Error #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0019

< > device reports a "NOP Operation Post Send Failed" on cqn #< > with Status #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x001A

< > device reports a "Query operation error" on Port #< > with status #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x001B

< > device reports an "SQP Event" on qpn #< > the event type is #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x001C

< > device reports a "Customer reset request" on port #< > oid #< >. Therefore, the HCA NIC will be reset. (The issue is reported in Function < >).

0x001D

< > device reports that the "QOS (ETS) capability is missing". The current firmware does not support the QOS (ETS) capability. Please burn the latest firmware and restart your machine. (The issue is reported in Function < >).

0x0001E

The firmware version burned on the < > device is not up-to-date. Hence, some of the NIC capabilities will either not function properly or affect the NIC performance. Please burn a newer firmware and restart the < > device. For additional information on firmware burning process, please refer to the User Manual document.

0x0001F

The firmware version that is burned on the < > device does not support Network Direct functionality. This may affect the File Transfer (SMB) performance. The current firmware version is < > while we recommend using firmware version 2.9.8350 or higher. Please burn a newer firmware and restart < > device. For more details about firmware burning process please refer to the User Manual document.

0x00020

The Flow Control on the Ethernet adapter < > (port < >) was disabled. Therefore, RoCE cannot function properly without Flow Control. To resolve this issue, please make sure that the Flow Control is enabled on both the hosts and switches in your network. For more details, please refer to the User Manual document.

0x00021

The firmware version burned on the < > device is not up-to-date. CQ to EQ mapping feature is missing Hence, RSS feature will not function properly and will affect the NIC performance. Please burn a newer firmware and restart the < > device. For additional information on firmware burning process please refer to the User Manual document.

0x00023

The Flow Control on the Ethernet adapter < > (port < >) was disabled. Therefore, RoCE v1.0/v2.0 cannot function properly without Flow Control. To resolve this issue, please make sure that the Flow Control is enabled on both the hosts and switches in your network. For more details, please refer to the User Manual document.

0x00024

There were not enough MSI vectors on Ethernet adapter< >. < > vectors were required. < > vectors were acquired.

0x00025

ECN was enabled for interface %2 but the adapter < > does not support it. ECN congestion control will not be enabled for this port. Please burn a newer firmware and restart the < > device. For more details, please refer to the User Manual document.

0x00026

The File Transfer (SMB) performance may be affected as Network Direct functionality is not supported in ConnectX-2 firmware version.

0x0029

< > failed to enable ECN congestion control on port < >. This port will continue its work without ECN congestion control. Please, verify ECN configuration and then restart the driver.

0x002A

< > failed to set bandwidth allocation on port < >. The device will continue working using default bandwidth allocation.

0x002B

, > device reports an "RQP Event" on qpn #< > the event type is #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x002C

< > device reports an "SRQ Event" on sqpn #< > the event type is #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function <>).

0x0002E

< > device detected that the state of the link connected to port <> is unknown. This can occur if the physical link state determination process is in progress. Once the link state is determined, link up or down events will be notified.

0x00036

The external switch loopback is enabled in Registry, but not supported on the Ethernet adapter < >. Please update FW in case you want to enable the feature.

0x0037

< > failed to set VXLAN UDP port on port < >. The device will continue working sing default (0x4789) VXLAN UDP port.

0x00039

The registry enabled Head-Of-Queue life time limit, but the firmware does not support it. Please update the firmware.

0x003B

Adapter < > detected that OID < > is stuck.

0x00040

Filter < >, Mac < >, QueueId < > is duplicated.

0x00042

Adapter < > detected issues with Processor groups on the system that may cause performance degradation. Disable processor groups for best results.

0x0045

Adapter < >, port < >: Ecn is supported only for RRoCE v2. ECN is disabled.

0x0046

Adapter < >, port < >: EcnPriorityEnable is a 8-bit field. The bits mean priorities of Roce data streams. At least one bit should be set to enable Ecn. ECN is disabled.

0x0047

Adapter < >, port < >: Ecn is enabled in Registry, but is not supported by FW. ECN is disabled.

0x0048

Adapter < >, port < >: Force eSWUCLoopback to go back to default value, since SRiOV is not enabled or not active.

0x0049

Adapter < >, port < >: Force eSWMCLoopback to go back to default value, since SRiOV is not enabled or not active.

0x004C

Mellanox ethernet counters are not supported for port number < >.

0x004D

Mellanox infiniband counters are not supported for port number < >.

0x004E

Dynamic configuration update initialization failure. Dynamic parameters won't updated in run-time

0x004F

Dynamic configuration: < > invalid value, refer to user manual for acceptable values.

0x0050

Dynamic configuration updated: < > changed from < > to < >.

0x0051

The registry enabled VMQ or SR-IOV, but those parameters are only supported on host machines and not VFs. Please remove those settings.

0x0052

< > provided by the user is an invalid RSS processor, auto-switching to the closest RSS processor.

0x0059

CFI filtering is requested but not supported by FW, device will ignore CFI flag in received packets steering.

0x0060

The value of NumaNodeId is configured out of range. Thus, its value is reset to default - undefined (ANY_NODE).

0x0061

The priorities enabled on the PFC was %3, and changed to < > by miniport regestry key PFC_IGNORE_PAUSE_PER_PRIO (%5).

0x0062

< > device reports an VMQ "Rx ring stuck" on cqn #< > srqn #< > QueueID #< >. (The issue is reported in Function < >).

Severity

Message ID

Description

Error

0x0004

< > has started in non-operational mode.

0x0007

MAP_FA command failed with error < >. The adapter card is non-functional. Most likely a FW problem. Please burn the last FW and restart the mlx4_bus driver.

0x0008

RUN_FW command failed with error < >. The adapter card is non-functional. Most likely a FW problem. Please burn the last FW and restart the mlx4_bus driver.

0x0009

QUERY_FW command failed with error < >. The adapter card is non-functional. Most likely a FW problem. Please burn the last FW and restart the mlx4_bus driver.

0x000b

QUERY_DEV_CAP command failed with error < >. The adapter card is non-functional. Most likely a FW problem. Please burn the last FW and restart the mlx4_bus driver.

0x000c

QUERY_ADAPTER command failed with error < >. The adapter card is non-functional. Most likely a FW problem. Please burn the last FW and restart the mlx4_bus driver.

0x000d

Too few QPs were requested (requested < >, reserved for FW < >). The adapter card is non-functional. Please increase the Registry LogNumQp parameter under HKLM\System\CurrentControlSet\Services\mlx4_bus\Parameters.

0x000f

Port module change event is ignored as the PCI driver does not support interface re-enumeration. Therefore, the driver could not be restarted. For changes to take effect, please restart the mlx4_bus driver.

0x0010

Are you using SRIOV-enabled firmware?

0x0011

Failed to move location string < >, status < >.

0x0012

WdfDeviceAllocAndQueryProperty failed, status < >.

0x0016

Only same port types supported on this HCA. Please go to the Port Protocol UI, and change the port types to be either ETH or IB.

0x0017

Your port type configuration (eth,ib) is not supported. If you have connected the ports to Ethernet and InfiniBand switches please switch the ports, connecting the Infiniband switch to port 1.

0x0018

Problem - The port was configured to use auto sensing for deciding the port type. Port < > failed to detect the port type automatically.
Impact - As a result the port is being started as an IB port (unless the port is ETH only). This may cause a connection problem if the other side is an ETH port.
Reason and suggestion to fix - This problem may happen since the computer is connected back to back or the cable is unplugged.
To solve this issue connect the port to a switch or define the port type explicitly (IB or ETH) instead of auto.

0x001f

Error, Allocating memory for device driver failed (memory size > <). Either close any running applications, or reboot your computer or add additional memory.

0x0020

Error, Allocating memory for device driver failed (memory size < >). The miniport driver cannot start. Either close any running applications, or reboot your computer or add additional memory.

0x0022

Problem - On port < > device capabilities indicate IB only, which is not supported on a multi protocol machine.

0x0024

Driver startup failed due to system call WdfDeviceAddQueryInterface function failure. Error=< >.

0x0028

Driver startup failed due to failure in creation of the child device, IB or ETH. Error=< >.

0x0029

Creation of < > device failed. Error=< >.

0x0030

< > FW command failed. op < >, status < >, errno < >, token < >, in_modifier < >, op_modifier < >, in_param < >.

0x0038

Illegal ETH-IB port configuration.
Port 1 is configured as Ethernet without RoCE/RRoCE enabled.
Port 2 is configured as IB.
The driver will fail. If you insist on this configuration, configure it manually from Device Manager.

0x0047

Driver startup failed because < >. (status < >)

0x0049

Driver startup failed because < > could not be initialized.

0x004a

Driver startup failed because the pci device with vendor id < >/device id < > could not be found.

0x004b

Driver startup failed because < > bytes of memory could not be allocated for < >.
Either close any running applications, or reboot your computer or add additional memory.

0x004c

Driver startup failed because the driver could not take ownership of the device after < >3 tries.

0x004d

Driver startup failed because of an unsupported feature: < >.

0x004e

Driver startup failed because slave is not allowed by customer or disallowed because IB is not supported by Flex10 slaves. (max_allowed_slaves < >, cur < >)

0x004f

Driver startup failed because insufficient Event Queues (EQs) are available. (< > are required, < > are available)

0x0050

Driver startup failed because port < > could not < >.

0x0053

SR-IOV cannot be enabled as an old firmware (< >) that does not supports SR-IOV is burned on the HCA. Please upgrade your firmware. For more details please refer to product user manual.

0x005C

Failed to add extension to Eth hardware_id, status < >.

0x005E

__check_mtt_before_free: MTT is already free, n_mtt < >, p_mtt < >, val < >.

0x005F

__check_mtt_before_free: want release < >: < >, expected < >: < >, p_mtt < >.

0x0060

__check_mtt_before_write: MTT is already used, n_mtt < >, p_mtt < >, val < >.

0x0063

Failed to reserve QP Range for VF < >, Consider increasing LogNumQP. For more information, please refer to the Troubleshooting section in the User Manual.

0x0064

SR-IOV cannot be enabled due to an error in the PCI_VIRTUALIZATION_INTERFACE. Possible reason, the machine does not support SR-IOV.

0x0070

file < > was created due to fw fatal error.

0x0071

file < > was created due to command timeout.

0x0072

file < > was created due to EQ error.

0x0073

file < > was created due to TXCQ error.

0x0083

Failed to install < > performance counters. < >.

0x0084

Failed to remove < > performance counters. < >.

0x0085

Failed to set Eth port type for port < > with roce_mode < >, error < >.

0x0089

Fast FW Load: < >

0x008a

Self Healing - Ignores error that was reported by sensors< > (0x< >) as a result of reaching the maximum number of Self-Healing resets (< >). Please clear the counters of the Self-Healing feature.
For more details, please refer to WinOF User Manual.

0x008b

Self Healing - Failed to activate the resiliency flow as a result of a SW reset failure, error=< >.
The error was reported by the sensors < > (0x< >).

0x008c

Restart < > as a result of error that was reported by sensors < > (0x< >)
Self healing state:
Restarts count: < >

0x008d

Stopped < > activity as a result of an error that was reported by sensors < > (0x <>).

0x0093

Can't get unloaded. < >3 applications are still active.

0x0095

Restart < > as a result of error that was reported by sensors < > (0x< >)
Self healing state:
Restarts count: < >
Max restarts count: < >

0x0096

Stopped < > activity as a result of exceeding the maximum amount of allowed restarts (< >).

0x0097

Failed to initialize self-healing mechanism as a result of error < >.

0x009a

Restart < > as a result of error that was reported by sensors < > (0x< >)
Self healing state:
Restarts count: < >
Restart time: < >
Max restarts count in time interval < > seconds: < >

0x009b

Stopped < > activity as a result of exceeding the maximum amount of allowed restarts (< >) in time interval of < > seconds.

0x0101

Failed to create full dump me now.
Dump me now root directory: < >
Failure: < >
Status: < >

0x0107

file < > was created due to dump-me-now request.

0x0109

Self-Healing second tier policy was activated on Virtual Function (VF) #< > with sensors < > (0x< >), and Bus driver restart is needed on the VF. Please restart the Bus driver.

0x010a

Failed to create full dump me now.
Status: < >

0x010b

QP was moved from error state to hibernation state.
QP number: < >
Is SQ: < >
Is RQ: < >

0x010c

A virtual function's QP was moved from error state to hibernation state.
Port number: < >
Function id: < >
QP number: < >
Is SQ: < >
Is RQ: < >

0x010e

Command from Virtual Function (VF) #< .: was blocked as it failed to wait for the previous command to complete.

0x0113

< > QPs were not released !!

Warning

0x0005

Self Healing - Bus device mode (recovery from Bus errors) was requested, but it is not supported, as the < >. The feature starts in Miniport mode (recovery only from Miniport errors).

0x000a

Found PF (non-primary physical function) at < >.

0x000e

Failed to generate an event (0x< >) for slave < > as eq 0x< > is full.

0x0013

< > failed on < > with status < >.

0x0014

WdfDeviceOpenRegistryKey failed on opening SW (=driver) key for mlx4_bus with status < >.

0x0015

Port type registry value for device < > contains invalid value (PortType = < >). Default value will be set.

0x0018

Problem - The port was configured to use auto sensing for deciding the port type. Port < >3 failed to detect the port type automatically.
Impact - As a result the port is being started as an IB port (unless the port is ETH only). This may cause a connection problem if the other side is an ETH port.
Reason and suggestion to fix - This problem may happen since the computer is connected back to back or the cable is unplugged.
To solve this issue connect the port to a switch or define the port type explicitly (IB or ETH) instead of auto.

0x0019

Problem - The port was configured to use auto sensing for deciding the port type. Port 2 failed to detect the port type automatically.
Impact - Since the first port is configured to be ETH the second port is started as an ETH port. This may cause a connection problem if the other side is an IB port.
Reason and suggestion to fix - This problem may happen since the computer is connected back to back or the cable is unplugged.
To solve this issue connect the port to a switch or define the port type explicitly (IB or ETH) instead of auto.

0x001a

Single port optimization feature was requested, but it is disabled as the < >.

0x001b

The driver will be restarted as a result of the Port module change event.

0x001d

Problem - FW sense command could not be run on port < >. We recommend upgrading your FW image. For further details, please refer to the README file in the documentation folder.

0x001e

Problem - The port was configured to use auto sensing for deciding the port type. Port 1 failed to detect the port type automatically.
Impact - Since the second port is configured to be IB the first port is started as an IB port. This may cause a connection problem if the other side is an ETH port.
Reason and suggestion to fix - This problem may happen since the computer is connected back to back or the cable is unplugged.
To solve this issue connect the port to a switch or define the port type explicitly (IB or ETH) instead of auto.

0x0021

Warning - IB configuration on Multi Protocol is prohibited. Forcing port < > to be ETH.

0x0023

Warning - RoCE configuration on Multi Protocol is prohibited. Forcing RoCE off.

0x0025

The BUS driver started working in Legacy mode, which may affect network performance. This can be caused by a resource limitation.
(CPUs: < >, HW EQs: < >, SW EQs: , >, Eternert EQs: < >).

0x0026

The number of allocated MSI-X vectors is less than required. As a result, multiple EQs will share the same MSI-X vector. This may decrease network performance.
The number of requested MSI-X vectors is: < > while the number of allocated MSI-X vectors is: < >.

0x0027

The requested port type for port %3 is unsupported on this HCA. The driver was configured to use the supported type.

0x0031

SL change is unsupported on this HCA. QoS is partly operational.

0x0032

Too many IPs in-use for RRoCE.
< >: RRoCE supports only < > IPs per port.
Please Reduce Number of IPs to use the new IPs.

0x0036

SRIOV cannot be enabled. Running in single-function mode.

0x003f

RRoCE is not supported for ConnectX®-2 device; as a result, RRoCE is disabled and the NIC starts in RoCE mode.
NOTE: If your environment contains a mix of different NIC types, you need to make sure that the whole environment is configured to use RoCE; otherwise the traffic between the different NICs will not work.

0x0040

< > mode was requested, but it is not supported. The NIC starts in < > mode.
NOTE: If your environment contains mix of different NIC types, you need to make sure that the whole environment is configured to use the same RoCE mode, otherwise the traffic between the different NICs does not work.

0x0051

The link on port < > is down. Bad cable was detected. Please replace the cable to continue working.

0x0052

The link on port < > is down. Unsupported cable was detected. Please replace the cable to continue working.

0x0054

SR-IOV cannot be enabled. The device does not support SR-IOV in InfiniBand mode and cannot be used when one of the ports is configured to use InfiniBand. In order to resolve this issue please set the port type of the HCA to Ethernet. For more details please refer to product user manual.

0x0058

Delay drop timeout occurred on port < .. Drop mode entered; packets may now be dropped.

0x0059

SR-IOV cannot be enabled. the device does not support SR-IOV in InfiniBand mode, it cannot be used when one of the ports is auto-sensed to be connected to InfiniBand switch. To resolve this issue please set the port type of the HCA to Ethernet or connect the port to an Ethernet switch. For further details please refer to product user manual.

0x005A

bus: EQ 0x< > was stuck. We inserted DPC for polling it which was expected to solve the problem.

0x005B

The link on port < > is down. Bad/unreadable EEPROM module was detected.

0x005D

Cmd timeout. < > eqn < >, cons_index < >, type < >, subtype < >, owner < >, token < >, status < >, out_param < >.

0x0067

Port # < > is configured to Ethernet. Since Ethernet is not supported in this device, it will automatically be configured to IB instead. Check PortType registry key.

0x0068

Although Port #< > is configured as IB, the software sets it to Ethernet only.
This occurs when the Ethernet link is already up due to one of the following: BMC is enabled or the firmware version installed in not up-to-date.

0x0076

VF allowed TX ether types feature is not supported by FW. Please burn the last FW and restart the driver.

0x0077

VF allowed TX ether types feature - < > ether types requested but only < > ether types are supported by FW. Only the first < > ether types will be set.

0x0079

Virtual Function (VF) #< > issued an invalid or out-of-sequence device-command (channel=< >, fragment=< >, cmd=< >, status=< >) - command blocked.

0x007a

SR-IOV cannot be enabled as it is unsupported in the machine. Please check first the BIOS configuration and enable SR-IOV if disabled, and afterwards verify the firmware used supports SR-IOV.

0x007b

FW doesn't support Multi Protocol. Setting port < > to IB.

0x007f

Adjusting of QP1 for ECN Burst Control failed for port < > with error < >.

0x0080

RDMA is disabled as a part of the healing policy.
For more details, please refer to the Self-Healing section in the WinOF User Manual.

0x0081

Illegal ETH-IB port configuration.
Port 1 is configured as Ethernet without RoCE/RRoCE enabled.
Port 2 is configured as IB.
To resolve the conflict, the driver enables RoCE on Port 1.

0x0082

Illegal ETH-IB port configuration.
Port 1 is configured as Ethernet without RoCE/RRoCE enabled.
Port 2 is configured as AUTO.
To resolve the conflict, the driver enables RoCE on Port 2.

0x0090

Port type registry value for device < > could not be modified to value (PortType = < >). Previous value will be set.

0x0091

Manual VF per port allocation (p1=< >, p2=< .) was auto-truncated to (p1=< >, p2=< >) to fit the max supported by the adapter (< >)

0x0092

Problem - RDMA in VF: Temporary this facility is allowed only for one VF in VM. RDMA is disabled for this VF.

0x0094

A VM (or the host) attempted to set one or more gids on a vf (< >) which were already in use by another vf.

0x0098

MAP_ICM_AUX < > CMPT < > EQC < > MTT < > DMPT < > QPC < > AUXC < > ALTC < > RDMARC < > CQC < > SRQC < > MCG < >.

0x0099

< > will be unloaded while < > applications are still active.

0x009c

Lost interrupt was detected, inserting DPC to process EQE.
EQE found on EQ index: < >
Number of ETH EQs: < >
Last consumer index: 0x< >

0x009d

SriovPortMode registry key is configured to use port < > but the connected port is < >.
In order to configure the Virtual Functions (VFs) correctly, please connect the correct port or update the value of SriovPortMode registry key.

0x0100

dump was created at folder (< >) due to dump-me-now request.
Dump-me-now dumps are placed by default in folder SystemRoot\temp\Mlx4_Dump_Me_Now tor folder that was set by registry keyword HKLM\SYSTEM\CurrentControlSet\Services\mlx4_bus\Parameters\DumpMeNowDirectory

0x0102

Device dynamic Registry update initialization failure. Dynamic Registry parameters won't updated in run-time.

0x0103

Device dynamic Registry configuration: < > invalid value, refer to user manual for acceptable values.

0x0105

< > has detected that the NIC resiliency feature is not supported by this firmware. It is recommended to update the firmware in order to prevent the system from hanging. For more details, please refer to the WinOF User Manual.

0x0106

< > has detected that the NIC resiliency feature is disabled. It is recommended to enable it in order to prevent the system from hanging. For more details, please refer to the WinOF User Manual.

0x0108

Device has been reset and is now dysfunctional. To activate it, restart the Bus driver.

0x010d

ExtraVFsQuotas registry key value (< >) is greater than the ExtraVFsQuotasScale registry key value (< >). The ExtraVFsQuotas value will be set to < >. For more details, please refer to the Mellanox WinOF User Manual.

0x0010f

Virtual Function (VF) #< > issued an invalid or out-of-sequence device-command (channel=< >, fragment=< >, cmd=< >, status=< >) - command blocked.
To prevent the guest VM from flooding the Event Log, this event will not be logged again from this VF until the host driver is restarted.

0x0111

< > registry key value (0x< >) is insufficient for physical function (PF) and < > virtual functions (VFs). The initialization of the PF or the VFs may fail as the configured limit of the resources is insufficient.

0x0112

EXT_QP_MAX_RETRY_LIMIT/EXT_QP_MAX_RETRY_PERIOD registry keys were requested by user but FW does not support this feature. Please upgrade your firmware to support it.
For more details, please refer to WinOF User Manual.

0x0114

Opened RoCE QP on priority < > which is not set to lossless (enabled priority mask on PFC is < >).
For more details, please refer to WinOF User Manual.

0x0115

Port < > - PKey table is not configured correctly for supporting VFs:.
index0 = < >.
index1 = < >.
index2 = < >.
For more details, please refer to WinOF User Manual.

0x0117

Opened RoCE QP with Qos not configured, traffic is not lossless.

0x0118

Opened RoCE QP with Global Pause not configured, traffic is not lossless.

0x0119

Registry key DumpMeNowTotalCount must be greater than registry key DumpMeNowPreservedCount, setting new values: [DumpMeNowTotalCount: - DumpMeNowPreservedCount: < >].

0x011A

Internal memory error event was reported by FW.
Memory error information=< >

0x011B

Fatal warning event was reported by FW with unknown subtype < >.

0x011C

Thermal warning event was reported by FW.
Current temperature=< >
Warning threshold=< >

0x011E

< > Firmware version x.xx.xxxx is below the minimum FW version recommended for this driver.
Minimum recommended Firmware version for this driver: x.xx.yyyy.
It is recommended to upgrade the FW, for more details, please refer to WinOF User Manual.

Severity

Message ID

Description

Error

0x0002

Failed to initialize < >.

0x0004

< > has been configured to use RSS while Windows' TCP RSS is disabled. This configuration prevents the initialization and enabling of the port. You need to either enable Windows' TCP RSS, or configure the adapter's port to disable RSS.

0x000A

Failed to initialize the < > device because it uses old firmware version (< >). You need to burn firmware version < > or higher, and to restart your computer.

0x00021

There is an interface mismatch between IPoIB driver and the bus driver. The IPoIB driver interface version is < > while the bus driver interface version is < >. As a result the IPoIB driver has failed to start.
This happened due to setup failures or partial update of the drivers. In order to resolve the issue, please reboot the computer. In case the problem had not been resolved, please reinstall the Mellanox Package.

0x00022

Device < > is configured to work on a team and VMQ. This configuration is not allowed. For more information on teaming support for ipoib please refer to the User Manual document.

0x0028

For port < >, the following address is considered as multicast address: < >.
Please configure the registry value NetworkAddress with another address in the registry, then restart the driver.

0x00032

< > detected a null port GUID for port #< >. A Virtual Function device may have a null port GUID if there is no OpenSM instance on its network. Please make sure the network has an active OpenSM and restart the driver.

0x0035

According to the configuration under the "Jumbo Packets" advanced property, the MTU configured for device < > is < >. The effective MTU is the supplied value + 4 bytes (for the IPoIB header). This configuration exceeds the MTU reported by OpenSM, which is < >. This inconsistency may result in communication failures. Please change the MTU of IPoIB or OpenSM, and restart the driver.

0x0037

The NDIS version of the driver (< >) is not supported by the OS (< >)

0x0038

DriverEntry failed with status #< >.

0x0055

Failed to open Channel Adapter.

0x0056

Failed to enable NDK with status < >.

0x0057

Failed to disable NDK with status < >.

Warning

0x000C

< > device couldn’t create the optimal number of MSI-X vectors. The Network interface will use only one CPU for interrupts, although RSS will still be in effect. This may slightly affect performance. For fully optimal performance, configure the number of MSI-X vectors to be at least < >.

0x000E

< > device detected that the link connected to port < >is down. This can occur if the physical link is disconnected or damaged, or if the other end-port is down.

0x000F

< > device has been configured not to use RSS. This configuration significantly affect the network performance.

0x0010

< > device reports an "Error event" on cqn #< >. The event type is:< >. Therefore, the HCA Nic will be reset.(The issue is reported in Function < >).

0x0012

< > device reports a "Test operation error" on Port #< > with status #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0013

< > device reports a send=< > "CQE error" on cqn #< > qpn #< > cqe_error->syndrome < >, cqe_error->vendor_error_syndrome < >, Opcode < >, Corrupted CQE < >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0014

< > device reports an "EQ stuck" on eqn #< > cq #%< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0015

< > device reports a "TX ring stuck" on cqn #< > qpn #< > uncompleted send #<>. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0016

< > device reports an "Rx ring stuck" on cqn #< > srqn #< > . Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0017

< > device reports a "Set information failed" on port #< > Oid #< > Status #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0018

< > device reports a "Rearm CQ Failed for NOP" on Rx ring, cqn #< > Error #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0019

< > device reports a "NOP Operation Post Send Failed" on cqn #< > with Status #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x001A

< > device reports a "Query operation error" on Port #< > with status #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < .).

0x001B

< > device reports an "SQP Event" on qpn #< > the event type is #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x001C

< > device reports a "Customer reset request" on port #< > oid #< >. Therefore, the HCA NIC will be reset. (The issue is reported in Function < >).

0x001D

< > device failed to handle SM PNP Event due to insufficient resources.(The issue is reported in Function < >).

0x001E

< > device reports a "Modify QP error" on qpn #< > Status #< .. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x0001F

The firmware version burned on the < > device is not up-to-date. Hence, some of the NIC capabilities will either not function properly or affect the NIC performance. Please burn a newer firmware and restart the < > device. For additional information on firmware burning process please refer to the User Manual document.

0x00020

The firmware version that is burned on the < > device does not support Network Direct functionality. This may affect the File Transfer (SMB) performance. The current firmware version is < > while we recommend using firmware version 2.9.8350 or higher. Please burn a newer firmware and restart the < > device. For more details about firmware burning process please refer to the User Manual document.

0x00022

The firmware version burned on the < > device is not up-to-date. CQ to EQ mapping feature is missing Hence, RSS feature will not function properly and will affect the NIC performance. Please burn a newer firmware and restart the < > device. For additional information on firmware burning process, please refer to the User Manual document.

0x00023

There were not enough MSI vectors on ipoib adapter < >. < > vectors were required. < > vectors were acquired.

0x00026

The File Transfer (SMB) performance may be affected as Network Direct functionality is not supported in ConnectX-2 firmware version.

0x00027

IPoIB driver detected a possible incompatibility with Windows 8 inbox IPv6 support due to which there is no IPv6 connectivity. Please, consider to upgrade all the hosts to a new driver version or use Windows 8 compatibility mode.

0x002B

< > device reports an "RQP Event" on qpn #< > the event type is #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x002C

< > device reports an "SRQ Event" on sqpn #< > the event type is #< >. Therefore, the HCA Nic will be reset. (The issue is reported in Function < >).

0x00033

Network Direct is not operational on device < > port #< >. This may affect the File Transfer (SMB) performance. Please, restart the driver to fix this issue.

0x00036

Adapter < > detected issues with Processor groups on the system that may cause performance degradation. Disable processor groups for best results.

0x0039

< > Mellanox Ethernet counters are not supported for port number < >.

0x003A

< > Mellanox Infiniband counters are not supported for port number < >.

0x003B

Adapter < > detected that OID < > is stuck.

0x003C

< > Sent packets with unrecognized Ethernet protocol. This message appear only once.

0x0003D

Filter < >, Mac < >, QueueId < > is duplicated.

0x004E

Dynamic configuration update initialization failure. Dynamic parameters won't updated in run-time

0x004F

Dynamic configuration: < > invalid value, refer to user manual for acceptable values.

0x0051

The registry enabled VMQ or SR-IOV, but those parameters are only supported on host machines and not VFs. Please remove those settings.

0x0052

< > provided by the user is an invalid RSS processor, auto-switching to the closest RSS processor.

Severity

Message ID

Description

Error

0x0004

One of the HCAs failed to start since the same GUID is burned on multiple HCAs. This issue probably occurred due to FW burning mistake. Currently the system has detected < > duplicated GUIDs
To resolve the issue, please reburn one of the HCAs with another GUID.
For information on how to change the GUID please consult the HCA vendor support.

0x0005

Bus failed to initialize the Network interface, this issue occurred due to inappropriate card revision.
card revision=< >, please use newer driver that supports this card revision.

0x0007

GUID of port < > is equal to zero, which is incorrect. The possible reason is absence of opensm.
To resolve the issue, please run opensm before starting this card.

Warning

0x0006

Some of the ports on the HCA with GUID < > may not be functional since the same GUID is burned on multiple ports. This issue probably occurred due to FW burning mistake.
To resolve the issue, please reburn one of the ports with another GUID.
For information on how to change the GUID please consult the HCA vendor support.

Severity

Message ID

Description

Error

0x0001

ND is in invalid state as a result of a mismatch between the ndfltr.sys driver version and mlx4_bus.sys driver version. In order to fix this issue, please close all applications that use ND, and restart the bus driver.

© Copyright 2023, NVIDIA. Last updated on May 23, 2023.