NIC failover

Similar Questions

• The NIC failover does not work in LBT mode

  Dear community,
  

  
  

  We met a strange behavior in our network environment while testing our new Cisco 4500 switches x 10g. If you have a TCP Session (example JJ | netcat) data transfer between two virtual machines on different host esx failover does not work on the VM receiver if a physical connection fails.

  The test configuration is as follows:

  |    | - 4 - switch 1 vmnic - vmnic 4-|    |

  VM1 | ESX1 |                  | VSS |                    | ESX2 |-VM2

  |    | - 5 - switch 2 vmnic - vmnic 5-|    |

  More details:

  • The switches are configured as a cluster VSS. New version of the software
  • ESX1/2 are running the latest version 5.5
  • The two vmnic is active on a DV-switch in LBT mode
  • Exchanges are configured with "Port based on the physical NIC load."
  • VM1 and VM2 are invited linux, NICS are vmxnet3

  Steps to reproduce:

  1. Beginning listening mode on VM2 (netcat-l Pei 12345 |) JJ = / dev/null)
  2. Start the transfer on VM1 (dd if = / dev/zero bs = 1 M | netcat VM2 12345)
  3. Check which link is used (in my example vmnic5 on ESX2)
  4. Unplug the ESX2 vmnic5

  What you observe:

  • DD-traffic dies immediately
  • If you kill the dd on machines (it does not end by itself) and start all over again, it works immediately
  • If you ping VM1 VM2 or another system, it works
  • If you plugged vmnic5 again once the virtual machine is switched to new vmnic5-> if you didn't kill dd will continue

  Now to test the reverse:

  1. Beginning listening mode on VM2 (netcat-l Pei 12345 |) JJ = / dev/null)
  2. Start the transfer on VM1 (dd if = / dev/zero bs = 1 M | netcat VM2 12345)
  3. Check which link is used (in my example vmnic4 on ESX1)
  4. Disconnect vmnic1 ESX1

  What you observe:

  • Failover works like a charm
  • Failback works like a charm

  Conclusion:

  • Only the virtual machine at the front desk has a problem
  • Only an already established TCP session behaves badly, work of new sessions very well (as does ping, no losses)
  • If you use LACP place LBT on uplink the problem disappeared
  • Same problem if you connect two links to Switch1
  • Same problem if you use a physical box with linux and LACP as system
  • I did the same test with two c3750x (no cluster VSS) and has not encountered the problem

  Any ideas? It's really a good one.

  Thank you!

  Sascha

  
  

  
  

  Update:

  To make it clear, it doesn't have the transmission of UDP.

  Post edited by: sewert1

  I did some further testing and also involved cisco. There is no problem with the switches and network in general.

  Turns out he works with Windows clients, but it does not work with Linux. But it works with Linux with adapters e1000 and e1000e but not with vmxnet3.

  So I did some tests and learned that the vmxnet3 kernel module must be at least version 1.3. Version 1.1 (SLES 11 default) or 1.2 (default SLES12) does not work.

  (./vmware-install.pl--clobber-noyau-modules = vmxnet3)

• Virt Center 2.5 shows no nic failover status

  Hello

  We bring a new ESX3.5 U3 DR/HA with Sun X 2200 cluster as nodes and experiencing several network problems.

  When testing failover nic that the Virtual Center displays correctly that nic is not available but does not show standby nic is active even if the old standby nic seems to be an OK traffic. Has anyone else seen elsewhere?

  Thank you

  Tom D

  Welcome to the forums,

  This is a known issue, only way to see what is happening is via esxtop, you would see the mb of the jump from the first to the second network adapter.

  Duncan

  Blogs: http://www.yellow-bricks.com

  If you find this information useful, please give points to "correct" or "useful".

• The team of NIC failover problem

  Hi all

  We have a problem with the failover of vSwitches.

  Basically the problem is when I unplug the cable of the two physical natachasery associated with my vSwitch there is a loss of contents of a sachet.  The content of a packet loss creates no real problems but the biggest problem is when you plug the network cable from back in.  That's where we lose 10 packs, what causes host ESXi due relinquish in VCenter since I did the test on the uplink system and causing all virtual machines that are running under it restart on another Member of the same Cluster ESXi host.

  After a lot of reading these forums and Ken vSwitch debate blog http://kensvirtualreality.wordpress.com/2009/04/05/the-great-vswitch-debate%E2%80%93part-3/

  Using the default route based on the original virtual port code should work just fine without the need of any special Cisco switch configuration.

  We run ESXi 4.1 HP DL380 / DL360 servers.  This is the configuration for VMware and switch cisco so any help is greatly appreciated.

  VMware VSwitch and PortGroup configs are attached as images.

  Cisco config which only the two switchports connected to the specific ESXi is attached as a text file.

  What us has monitored him looking at the physical cisco switch when attempting to connect the network to switch the switchport cisco cable flashes orange so that network on ESXi reports linking as down and as soon as the switchport goes green ESXi comes back online.

  Any ideas or recommendations how to solve this are greatly appreciated.

  ... should work just fine without the need of any special Cisco switch configuration...

  There is an important parameter, you must configure. It is the spanning tree portfast

  See http://kb.vmware.com/kb/1003804 for more details

  André

• dvSwitch - Failover VMNIC and network control?

  Hello.

  We move to a new data center and have decided to move away from the Nexus 1000v for Vmware vDS. Therefore, it is still a bit new to us.

  We have 2 x 10 Gbit NIC. One for the traffic of the virtual machine and the other for vMotion, and they're tipping (standby) to eachother.

  We are concerned that in the case of a NIC failover that vMotion traffic will affect the VM traffic too. What is the best practice to avoid that?

  We considered on vMotion portgroup vDS assigning up to max 7Gbit traffic shaping traffic. This means that we will lose 3 Gbps when there is no failover... Advice or better ideas?

  Info:

  ESXi/vCenter 5.0 Update 3

  Best regards
  Pete

  Use the actions of NIOC instead of limits on groups of ports for difficult traffic shaping bandwidth management. In this way, you don't waste bandwidth and you can even have two active uplinks VM traffic and vMotion. The NIOC will ensure that traffic is prioritized based on the actions that you specify.

• Override the port NIC teaming with powercli group?

  Hi all

  Any chance you could lend a hand?

  I have a powercli script that goes out to all of my ESX 4.0 host and add a new port group to vswitch1 with a new VLAN ID. That works well, but I also need to override the NIC teaming on this port group, to set an active NIC and the other to be the backup. (we put NIC failover port groups not the vswitches).

  I see ways to change NIC vswitches grouping settings, but am yet to find a way to change the settings of the NIC collection for groups of ports themselves with powercli?

  Can someone shine a light?

  Thank you

  Try something like:

  Get-VirtualPortGroup-name '' | Get-NicTeamingPolicy | Game-NicTeamingPolicy - MakeNicActive "vmnic1" - MakeNicStandby "vmnic0".

  I hope this helps!

• Card NIC for FT assignments: where is Ko 1011966?

  I'm on a quest to find KB 1011966 as shown in VMware vSphere 4 Fault Tolerance: Architecture and Performance white paper (see excerpt below).  Anyone know where I can find it?  I searched vmware.com without success.

  Anyone can join me in this quest?

  *

  2.6 NIC assignments for logging traffic

  *

  FT generates two types of network traffic:

  • Migration traffic to create the secondary virtual machine

  • FT logging traffic

  Migration traffic happens on the NETWORK card that is designated for VMotion and it cause bandwidth utilization of the network boost for a short period.

  Cards separate and dedicated network is recommended for FT logging, and VMotion traffic, especially when several virtual FT

  resident on the same host machine. Sharing the same NETWORK card for FT logging times and VMotion can affect the performance of virtual FT

  machines whenever a secondary antibody is created for another pair FT or a VMotion operation for any other reason.

  VMware vSwitch Networking lets you send traffic as redundant for VMotion and FT to separate network cards using the links

  NIC failover. See KB article 1011966 for more information.

  The KB is now here. http://KB.VMware.com/kb/1011966

  Rick Blythe

  Social media specialist

  VMware Inc.

  http://Twitter.com/vmwarecares

• Direct NFS/CIFS connection with 1.4

  I see how 1.4 device ports can help with direct connect multi-channel iSCSI. Each tissue Gets a VLAN / subnet, iSCSI has multiple targets via every tissue, etc.

  But I do not see how device ports helps direct connect NFS/CIFS. For example,.

  

  In this diagram, the active hosts on the fabric B traffic must pass via the LAN to the North to reach the NAS. Even if each host and unit have its NIC failover configured to prefer the fabric and make pre-emptive, you'll still have failures such as NIC NAS and IOM, which will cause traffic NAS some/all to go through the LAN to the North. So, you have one of two situations:
  (1) the LAN peut process NAS traffic. If so, why not plug the NAS in the local network in the first place?
  (2) the LAN cannot handle the traffic, in which case you have not built tolerance of failure real and, worse still, a UCS problem can affect the LAN.

  Am I missing something here? How the ports of the device with a switch mode, shown below?

  

  Yes - you have a valid point on the traffic model following which breaks down etc.

  Just to back up a moment...

  Version 1.3 and below has 2 types of ports Ethernet - uplink port and the server.

  Version 1.4 has 4 port Ethernet - Uplink, server, device, monitoring port types

  In point 1.3, to directly connect a NAS to YEW meant that you move the mode switch to the FI, and then set the port to connect to the NAS server as an uplink port. What this has done is that the connection to the NAS port was an a trunk at the end END port allowing all the VLANS (IE no filtering VLANS), no QoS etc. settings.

  So if you want to end host mode AND not love you warnings above, connected the NAS to a switch upstream and not UCS and this option is still the spirit you today.

  That the port of the unit gives you now, is that the load VLAN the SIN belongs to can be filtered, QoS settings possible and it just works in end host Mode (most deployments are based on it).

  Above is the raison d'etre for the port of the device and the port type is neeeded even if it operates today in switch mode.

  Now, the question comes back, way mode switch or EHM material (which is what is the question).

  In EHM, you're right, the East-West traffic between NAS blades can use the network upstream.

  You can design effectively by specifying the id of fabric (A or B as primary) or set the affinity of the fabric if using a switching software but guaranteed location total (without using the network upstream) can be made as you say properly according to 'what' fails.

  If all uplinks in failure, so this should switch, but if a link between IOM and the IF fails, then the servers pinned to this link will begin using the external network.

  So yes, the network must be designed to keep the flow and that if scenarios in mind. It cannot be assumed not to hit above all East-West traffic.

  The long-term solution is to have data between the FI to EHM links or they are even vPC and that's why the two links to the NAS from the FI will be active/active.

  Port device in switch mode can be used, but it depends also on which links are blocked etc. PLEASE ensure that.

  The topology you mentioned did... but then you should also keep in mind on failurea etc happening is to say that the ISL between YEW should always be forwarding for this VLAN.

  Thank you

  -Matt

• Network status reports "an incorrect configuration detected.

  Network status reports "an incorrect configuration detected.

  Ensure that the physical switch and the ports used by Virtual SAN are active and enabled multicast. Enabling multicast can be made in one of the two tracks on your physical switches:

  • Disabling IGMP surveillance.
  • Configure IGMP snooping for selective traffic

  Also, validate the configuration of the virtual switch for correct uplink VLAN, NIC failover configuration of team strategy and the virtual SAN traffic service is enabled on the VMkernel interfaces.

  Virtual SAN requires a VMkernel network with active virtual SAN traffic interface. All members of the cluster need to communicate on the same network segment of L2 with multicast enabled, and all members of the cluster should be able to ping each other. Having failed to meet this requirement will prevent virtual SAN to be configured successfully, since it prevents the hosts communicate.

• Map-daughter of Dual IBM 1 GB

  Hola a todos,

  Hace tiempo con antigua HCL me salia than estar IBM tarjeta no estaba soportada, pero ahora is not the encuentro y no tr TR compatible o are there works correctamente.

  Alguien tiene alguna referecia of operation ok?

  Gracias! :-D

  Hola Jose, as tal buenos dias.

  Te Le esta tarjeta what are normal in all USO las facilities of VMware that material he sober s IBM, there sober X 3650, 3550, 3850, etc..

  Esta tarjeta add back bocas has the card than viene integrada.

  In reality no're una tarjeta adicional, proporciona fisica unit only has the Red en integrada Plaça tarjeta, sea o aren't una tarjeta en TR misma sino provee unit has the card on board.

  Te dejo a link a UN documento donde extraje el aqui abajo side.

  http://www.Redbooks.IBM.com/abstracts/tips0805.html?open#network

  Network cards

  x 3650 M3 supports up to four built-in Gigabit Ethernet ports. Two ports are standard, and two ports can be added by installing the card Dual-port Gigabit Ethernet daughter, 46M 1076. This card uses a connector on the motherboard and does not consume a PCI expansion slot.

  Integrated network adapters have the following features:

  • Chip Broadcom BCM5709
  • Support TCP Offload Engine (TOE)
  • Wake on LAN support
  • 802 1 q VLAN tagging support
  • Grouping of NICs (failover and load balancing)

  Por otra parte of promotion as a cuento you he probado todos los entornos ESX 4.0, 4.1, 4.1 U1, U2 4.1 y todas sus versiones including 5.0 ESXi ESXi y el ve como Broadcom NetXtreme II BCM5709 1000Base-T operating system

• Enable and configure the card NETWORK FT with KS.cfg

  All,

  Did someone familiar with orders that I would need to put KS.cfg to configure and activate FT?  I've included the part of my KS.cfg here, which will install and configure VMotion.  Background information, I have set up with 3 uplinks vSwitch0, all defined for the delivery by port id w/VST virtual.  Each group of ports (Service Console, FT, VMotion) has a primary and standby two uplinks.  The primary it's Eve on the other and vice versa.  My another vSwitch is configured with 4 uplinks and routes based on IP hash w/VST.

  Anyone have any ideas or suggestions to get this to work?

  -

  ---

  #create exchanges

  / usr/sbin/esxcfg-vswitch - a "VMotion" vSwitch0

  / usr/sbin/esxcfg-vswitch - a vSwitch0 "FT".

  #create id vlan for the port groups

  / usr/sbin/esxcfg-vswitch Pei "VMotion" - v 117 vSwitch0

  / usr/sbin/esxcfg-vswitch Pei "FT" - v 117 vSwitch0

  NIC #add to the vmnic0 vswitches is already added to vSwitch0

  / usr/sbin/esxcfg-vswitch-L vmnic1 vSwitch0

  / usr/sbin/esxcfg-vswitch-L vmnic10 vSwitch0

  #configure vmotion

  / usr/sbin/esxcfg-Vmknic - a "255.255.255.0"-i "10.35.17.253" - n "VMotion".

  / usr/sbin/esxcfg-road - a default value 10.35.17.1

  FT #configure

  / usr/sbin/esxcfg-Vmknic - a "255.255.255.0"-i "10.35.17.254" - n "FT".

  #waiting 60 seconds to restart the service

  Service mgmt-vmware restart

  sleep 60

  #enable vmotion

  cmd-vim-VMware vmotion/hostsvc/vnic_set vmk0

  #enable ft

  #

  #change nic failover command

  VMware-vim-cmd hostsvc/net/portgroup_set-nicorderpolicy-active = vmnic0 - nicorderpolicy-watch = vmnic1, vmnic10 vSwitch0 'Service Console'

  VMware-vim-cmd hostsvc/net/portgroup_set-nicorderpolicy-active = vmnic1 - nicorderpolicy-watch = vmnic0, vmnic10 vSwitch0 "VMotion".

  VMware-vim-cmd hostsvc/net/portgroup_set-nicorderpolicy-active = vmnic10 - nicorderpolicy-watch = vmnic0, vmnic1 vSwitch0 "FT".

  Post edited by: paetecsfb

  Hello

  Friday I finally managed to find the key that allows to FT.

  VMware-vim-cmd hostsvc/advopt/update FT. Vmknic string vmk1

  You can find more details on my blog http://vmware-for-starters.blogspot.com/ or the blog of Mike Laverick for the powershell version http://www.rtfm-ed.co.uk/

  See you soon,.

  David

• UCS C200 and NIC Teaming/FailOver

  As UCS C200 with management of a 10/100 and 10/100/two interface 1000 interfaces and PCIe with 4NIC

  I want to install CUCM 8.5 is the NIC Teaming/failover supported by UCS C200 and how to set up the NIC Managment interface with failover?

  Thank you.

  Hello

  As you install apps from the CPU on the server of the UCS, speech application team listed their recommendations here.

  http://docwiki.Cisco.com/wiki/QoS_Design_Considerations_for_Virtual_UC_with_UCS#Guidelines_for_Physical_LAN_Links.2C_Trunking_and_Traffic_Sizing

  You can create a NIC teaming in the ESXi via vSphere client software for traffic from from / destined for virtual machines.

  For C200, we have only one management port and if you use the management for MMIC traffic port, switch option is not available.

  However, if you choose to use host for MMIC traffic NETWORK ports, you can set up CIMC NIC mode for 'LOM shared' which provides the NIC teaming options.

  http://www.Cisco.com/en/us/docs/unified_computing/UCS/c/SW/GUI/config/Guide/1.4.1/b_Cisco_UCS_C-Series_GUI_Configuration_Guide_141_chapter_01000.html#concept_AC4EC4E9FA3F4536A26BAD49734F23D0

  HTH

  Padma

• NIC teaming on ESX 5.1 and NGC failover

  Hi all

  A quick question. I know I read it before, but I can't find the reference again once. What is the recommended setting between allow vNic activate failover on cisco UCS and strategy setting on ESX 5.1 of the network?

  I know that I had read somewhere when the grouping of network cards is configured on ESX failover authorized UCS, ESX goes haywire if a path is lost the UCS.

  Our installation program running on UCS 2.04 Manager has with a mixture of blades M2 and M3 with M81kr and 1240/1280 cards respectively.

  Never use a fabric with a hypervisor vswitch failover. Let the failover of the vSwitch handle network failure mechanism.

  For ESX once you have two NICs attached to the vSwitch or a DVS it will automatically load balance and failover. You don't want you combine this with the UCS fabric failover.

  Louis

• NIC team Active standby failover - how to see which physical adapter is actively used at a specific time by a VM or VMKernel port

  Hello, simple question.

  I have a vSwitch with two portgroup VMKernel. The switch is connected with two network adapters. The failover policy is set at the level of Portgroup. PG1 has active nic1 and nic 2 in standby mode. PG2 nic 1 standby and active NIC 2. I want to know via cli or gui, command which nic is actually used for a portgroup at a specific time. Is this possible?

  Thank you!
  Francesco

  You can see the active uplink for each vNIC in the network view (press 'n') with esxtop (r):

  

• Order of NIC used real failover

  Hello

  Is it possible to know through client web or client of windows on which nic packets are transferred when using 2 NICs with the explicit order of failover and failback defined on no. It seems that the GUI does not reflect the fact that ESXi is passed to the standby NIC at startup or in other cases and because that failback is set to NO the transmission is not back to the active map.

  Concerning

  You can take the port of the virtual machine to the relationship of uplink programmatically or esxtop to view the rising current of the virtual machine (called team-Teddy in network view)