DATA PATH FAILOVER METHOD FOR SR-IOV CAPABLE ETHERNET CONTROLLER

• US 20160342437A1
• Filed: 05/21/2015
• Published: 11/24/2016
• Est. Priority Date: 05/21/2015
• Status: Active Grant

• Current Assignee: Dell Products LP (Dell Technologies Inc.)
• Original Assignee: Dell Products LP (Dell Technologies Inc.)
• Inventors: Khemani, Lucky Pratap, Rabeela, Janorious
• Granted Patent: US 9,858,102 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 11/1464 for networked environments
• CPC Class Codes: G06F 11/1658 Data re-synchronization of a redundant component, or initial sync of replacement, additional or spare unit
• CPC Class Codes: G06F 11/2005 using redundant communication controllers
• CPC Class Codes: G06F 11/3006 where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems multiprogramming arrangements G06F9/46; allocation of resources G06F9/50
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 2009/45579 I/O management, e.g. providing access to device drivers or storage
• CPC Class Codes: G06F 2201/805 Real-time
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/815 Virtual middleware or OS functionality using virtual machines to implement generic software techniques for error detection or fault masking G06F11/1484
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: H04L 41/0654 using network fault recovery ring fault isolation or reconfiguration in loop networks without recovery actions by a network management system H04L12/437
• CPC Class Codes: H04L 41/0816 the condition being an adaptation, e.g. in response to network events
• CPC Class Codes: H04L 41/0897 by horizontal or vertical scaling of resources, or by migrating entities, e.g. virtual resources or entities
• CPC Class Codes: H04L 43/0811 by checking connectivity

Abstract

A method and an information handling system (IHS) provide data path failover for a single root input/output virtualization network interface controller (SR-IOV NIC) card. According to one aspect, an IOV failover module (FM) enables a physical function (PF) driver in a service virtual machine (VM) to access physical resources and configure virtual functions (VFs) within the SR-IOV NIC card. The IOVFM monitors data paths established between VMs and VFs. If a data path fails, the IOVFM initiates a failover by migrating a VF coupled to a failed data path to an alternate VF. However, if a threshold number of consecutive failed migration attempts occurs and the SR-IOV NIC card has an additional and available port(s), the IOVFM migrates the VF to an alternate VF of the available port. The IOVFM enables the VM to access and configure the alternate VF of the available port via a failover data path.

Claims

1. An information handling system comprising:

an input/output (I/O) virtualization network interface controller (IOV NIC);

a processor;

a memory system communicatively coupled to the processor and having a host operating system (OS) having an IOVM module having a physical function (PF) driver for managing a physical function of the IOV NIC;

at least one virtual machine (VM) having a respective virtual function (VF) device driver and direct access to a corresponding VF device of the IOV NIC;

a virtual machine manager (VMM) having an extended support capability for input/output virtual machine (IOVM) functionality; and

an IOVM module having a physical function (PF) driver for managing a physical function of the IOV NIC; and

an IOV failover module executed by said processor to;

establish one or more network paths between at least one VM and a VF, respectively;

determine a health status of network paths between the at least one VM and respective VF devices to identify whether any network path is a compromised network path having a failed health status; and

in response to determining that at least one network path coupled between a VM and a VF device has a failed health status, migrate each one of the VF devices coupled to a compromised network path to a respective alternate VF device identified using a pre-configured mapping, wherein said migrating a VF device to an alternate VF device is performed using the IOV NIC without performing VMM switching.

2. The IHS of claim 1, wherein the processor-executed IOV failover module:

in response to migrating each one of the VF devices coupled to a compromised network path to a respective alternate VF device, determine a health status of network paths between the at least one VM and respective alternate VF devices; and

in response to determining that at least one network path coupled between a VM and an alternate VF device has a failed health status, migrate each one of the alternate VF devices coupled to a compromised network path to a respective next alternate VF device identified using the pre-configured mapping.

3. The IHS of claim 2, wherein the processor-executed IOV failover module:

retrieves a value of a threshold number of consecutive failed attempts caused by compromised network paths that is permitted for migrating a VF device to a next alternate VF device using a same IOV NIC port;

determines whether at least the threshold number of alternate network paths for coupling a VM to respective alternate VF devices has been reached; and

in response to the threshold number not having been reached, repeatedly performs said sequence of functions comprising (i) said determining a health status of a network path from a respective VM to an alternate VF device in response to said migrating and (ii) said migrating each one of the VF devices coupled to said compromised network path to said respective alternate VF device.

4. The IHS of claim 3, wherein the processor-executed IOV failover module:

in response to determining that a number of migrations of a VM to respective alternate VF devices has reached the threshold number;

determine a number of ports with which the IOV NIC is configured;

in response to determining that the IOV NIC is configured as a multi-port IOV NIC;

determine a health status of a network path corresponding to an alternate VF device from a most recent migration; and

in response to determining that the network path corresponding to the alternate VF device associated with the most recent migration has a failed health status, migrate each one of the alternate VF devices coupled to a compromised network path to a respective alternate VF device on a next port of the IOV NIC, identified using the pre-configured mapping.

5. The IHS of claim 4, wherein the processor-executed IOV failover module:

in response to migrating each one of the alternate VF devices coupled to a compromised network path to a respective alternate VF device on a next port of the IOV NIC, repeatedly performs a sequence of functions comprising (i) determining a health status of a network path from a respective VM to an alternate VF device in response to migrating an alternate VF device to a next alternate VF device and (ii) migrating each one of the VF devices coupled to a compromised network path to a respective next alternate VF device, according to a threshold number for consecutive failed attempts and a number of available ports of the IOV NIC.

6. The IHS of claim 4, wherein the processor-executed IOV failover module:

in response to determining that the IOV NIC is configured as a SR-IOV NIC having a single port, provides one or more of;

updates to one or more health status flags; and

a notification about network path or port status failures.

7. The IHS of claim 1, wherein the IOV failover module:

pre-configures alternate VF devices and additional PFs using the one or more VF device drivers.

8. The IHS of claim 1, wherein the IOV failover module:

in response to determining that at least one network path coupled between a VM and a newly provided alternate VF device has an operational health status, utilizing the IOVM module to enable the VM to efficiently configure the newly provided alternate VF device.

9. The IHS of claim 1, wherein the memory system further comprises:

a health monitoring module (HMM);

wherein the IOV failover module receives, from the HMM, health status information for at least one of (a) network paths used to connect a VM to a VF device and (b) PFs of the IOV NIC.

10. The IHS of claim 1, wherein the IOV failover module:

allocates to failed network path recovery mechanisms a selected number of redundant, alternate VF devices from among an identified total number of VF devices that can be configured within the IOV NIC.

11. In an information handling system (IHS), a method comprising:

establishing one or more network paths between at least one VM and a VF, respectively;

determining a health status of network paths between the at least one VM and respective VF devices to identify whether any network path is a compromised network path having a failed health status; and

in response to determining that at least one network path coupled between a VM and a VF device has a failed health status, migrating each one of the VF devices coupled to a compromised network path to a respective alternate VF device identified using a pre-configured mapping, wherein said migrating a VF device to an alternate VF device is performed using the IOV NIC without performing VMM switching.

12. The method of claim 11, further comprising:

in response to migrating each one of the VF devices coupled to a compromised network path to a respective alternate VF device, determining a health status of network paths between the at least one VM and respective alternate VF devices; and

in response to determining that at least one network path coupled between a VM and an alternate VF device has a failed health status, migrating each one of the alternate VF devices coupled to a compromised network path to a respective next alternate VF device identified using the pre-configured mapping.

13. The method of claim 12, further comprising:

retrieving a value of a threshold number of consecutive failed attempts caused by compromised network paths that is permitted for migrating a VF device to a next alternate VF device using a same IOV NIC port;

determining whether at least the threshold number of alternate network paths for coupling a VM to respective alternate VF devices has been reached; and

in response to the threshold number not having been reached, repeatedly performing said sequence of functions comprising (i) said determining a health status of a network path from a respective VM to an alternate VF device in response to said migrating and (ii) said migrating each one of the VF devices coupled to said compromised network path to said respective alternate VF device.

14. The method of claim 13, further comprising:

in response to determining that a number of migrations of a VM to respective alternate VF devices has reached the threshold number;

determining a number of ports with which the IOV NIC is configured;

in response to determining that the IOV NIC is configured as a multi-port IOV NIC;

determining a health status of a network path corresponding to an alternate VF device from a most recent migration; and

in response to determining that the network path corresponding to the alternate VF device associated with the most recent migration has a failed health status, migrating each one of the alternate VF devices coupled to a compromised network path to a respective alternate VF device on a next port of the IOV NIC, identified using the pre-configured mapping.

15. The method of claim 14, further comprising:

in response to migrating each one of the alternate VF devices coupled to a network path with a failed health status to a respective alternate VF device on a next port of the IOV NIC, repeatedly performing a sequence of functions comprising (i) determining a health status of a network path from a respective VM to an alternate VF device in response to migrating an alternate VF device to a next alternate VF device and (ii) migrating each one of the VF devices coupled to a compromised network path to a respective next alternate VF device, according to a threshold number for consecutive failed attempts and a number of available ports of the IOV NIC.

16. The method of claim 14, further comprising:

in response to determining that the IOV NIC is configured as a SR-IOV NIC having a single port, provides one or more of;

updates to one or more health status flags; and

a notification about network path or port status failures.

17. The method of claim 11, further comprising:

pre-configuring alternate VFs and additional PFs via the one or more VF drivers.

18. The method of claim 11, further comprising:

in response to determining that at least one network path coupled between a VM and a newly provided alternate VF has an operational health status, utilizing the IOVM module to enable the VM to efficiently configure the newly provided alternate VF.

19. The method of claim 11, further comprising:

receiving health status information for at least one of (a) network paths connecting a VM to a VF and (b) PFs of the IOV NIC.

20. The method of claim 11, further comprising:

allocating to failed network path recovery mechanisms a selected number of redundant, alternate VFs from among an identified total number of VFs that can be configured/simulated within the IOV NIC.