Hot answers tagged nic
No all 127.0.0.1 traffic never hits the physical network, it gets processed by a loop back adapter in the kernel.
Yeah it was ethtool I was looking for but specifically this will flash the link light for two minutes: ethtool -p eth0 120
Yes, it can be done in Windows: Go to the Control Panel > Network Connections Right click on the Local Area Connection (or whichever network connection you want to add the 2nd IP Address) and click Properties Click on Internet Protocol (TCP/IP) in the connection box and click properties Enter the first IP address in the properties box Click Advanced Click ...
Is ethtool what you're looking for ethtool --identify eth0
You don't mention a particular OS but for most all that happens is that the data travels down the stack until it gets to IP at which point it's pretty much sent back. That's a massive oversimplification but means that the entire process is usually CPU bound so its performance is therefore directly linked to CPU speed plus stack efficiency. In practical terms ...
(2a). Load balancing. (3). Separation of traffic (i.e. you could have a combo web/database server, same network, put all web traffic on one NIC, db traffic on the other, makes it easier to calculate loads for traffic types). This also makes it easier to split the two later on, nobody has to change connection strings.
Yes, you can have multiple; they are not one to one. A single network card can do this in a couple of ways; it can answer for multiple IP addresses with a single MAC address (assigning multiple addresses to a NIC in most operating systems will do this), or with multiple MAC addresses (virtualization platforms behave in this way).
The main principle behind interrupt moderation is to generate less than one interrupt per received frame (or one interrupt per transmit frame completion), reducing the OS overhead encountered when servicing interrupts. The BCM5709 controller supports a couple of methods in hardware for coalescing interrupts, including: Generate an interrupt after ...
What should be: The PCIe spec states that all slots start at 1x/v1.0 and negotiate how many lanes they can use and what clock speed. It shouldn't matter which supports more lanes/clock, some slots are designed to take larger cards and smaller cards fit in larger slots. Whatever the highest spec both sides can communicate at (both the number of lanes and the ...
A separate network for backups. A separate network for console access (these are typically now integrated on the system board for mainstream servers). Increased throughput (multiple GB networks) bonded together.
Ok so after posting this question last night night I continued to do some research the only real solution I came across seems to have taken care of the problem. Disabling TSO, GSO and GRO using ethtool: ethtool -K eth0 gso off gro off tso off According to a post found here: http://ehc.ac/p/e1000/bugs/378/ From what I understand this will or can cause a ...
Bonding is not applicable for this case. For your solution: You will find some useful notes at Increasing bandwidth with multiple NICs. For Bonding: You can start at the Wikipedia Link Aggregation and related Channel bonding pages for initial reading on bonding.
You're not going to want to do this with direct manipulation of the registry. You're much better off using the netsh command to make these kinds of changes. Assuming the NIC is named "Local Area Connection", you can do netsh interface ip "Local Area Connection" x.x.x.x y.y.y.y where x.x.x.x is the IP address and y.y.y.y is the subnet mask.
I've found this article amazingly useful: JLS2009: Generic receive offload. It gives a great overview of how GRO works. Some adapters might do it, but the associated drivers have to be aware of it as well. Also, drivers themselves can do this in software. As this happens before entering the Kernel TCP/IP stack, by the time the kernel-space TCP/IP stack is ...
They probably mean your MAC address. (The hexidecimal string under the Physical Address column of getmac. - Either D4-BE-D9-1C-4F-C7, 24-77-03-3E-B2-9C, 7C-E9-D3-FE-40-41, or 24-77-03-3E-B2-9D, depending on which is your NIC/the Network interface this license will be run against. Most likely D4-BE-D9-1C-4F-C7.) Certainty will require asking whomever it is ...
Sure, you simply have to enable vlan tagging on the switch and the network adapter and setup both sides with all the vlans you want the computer to see. The details about how to configure vlan trunking vary depending on what OS, and what you have to do for a specific nic or switch. Keep in mind that this may be a bit of a security issue. Suppose this is ...
Wait - you're asking for server grade NICs but want to buy a 350 bucks switch?! I don't get that ... Usually "server grade" 48 port GigE switches go for somewhere around 3000-5000 USD list price. Maybe you want to look out for switch side things like stacking for cross-stack LACP. Regarding the NIC, things like: proper DMA interface and good drivers that ...
It's bollocks. If you take a strike on a network cable, it'll go right through the card and into the PCI bus, which will make your system just as dead as if you'd had the NIC built-in.
For that I am planning to use an old Pentium 3 machine. Since you are using an older machine, I suspect I would go with whatever is inexpensive. Does your computer have a 64 bit PCI-X slot? Most of the 4 port PCI interfaces I have seen prefer one? Since you are already choosing to use an older machine, I am not sure you will really see much difference ...
With Server 2008 Service Pack 2 (not R2), or Vista SP2 and MS hotfix KB975808 there is a solution, although a bit clumsy. You would remove all the addresses you DON'T want as a source, then re-add each one at the command line using Netsh int ipv4 add address <Interface Name> <ip address> skipassource=true The hotfix enables the "Skip As ...
Edit /etc/network/interface and remove all but one of the gateway statements.
You may be interested in the Loopback Fast Path feature we added in Windows 8 and Windows Server 2012. I've written a short blog about it here http://blogs.technet.com/b/wincat/archive/2012/12/05/fast-tcp-loopback-performance-and-low-latency-with-windows-server-2012-tcp-loopback-fast-path.aspx It also provides an illustration of the path taken by the ...
Not only is this possible, it is very common. A network card may have multiple IP addresses. This is a concept called multihoming. There are other variants of multihoming as well.
A "default" gateway is just one that is used when there isn't a more specific route defined. You don't want to have a default route on two different interfaces (unless you're doing it for redundancy). What you want is to have a default route on your "main" interface that most of the traffic uses and then you want to create a persistent static route for each ...
VMware is a company, not a product. I'll assume that you're referring to vSphere, based on the link you posted in your question. Yes it is possible to use only one pNIC. In this case you would have a Virtual Machine port group and a VMkernel port on the same vSwitch.
If your NICs have the same MAC addresses, you should stop fiddling around with workarounds and return them to the manufacturer as defective. Get proper replacements and continue on as normal. Incur the downtime once instead of the recurring issues that will pop up from continuing to hack together a "solution."
Yes, they're completely hot-swappable. Think of them the same way as cables (note the link lights). Same for SFP+ direct-attach cables. It wouldn't make sense for them not to be hot-swappable. Heck, most network switches don't even have power switches :)
"Bonding" (teaming, trunking, or a host of other terms) NICs is used to increase bandwidth into a switch, or for redundancy for LAN connections. Redundancy and/or bandwidth expansion for WAN connections (like, to the Internet) is accomplished with routing protocols (typically BGP). Bonding the NICs together on a server computer isn't going to accomplish ...
Yes. Not an answer to your question, but something I didn't believe until I had seen it with my own eyes, it is sometimes possible to take a hacksaw to a PCIe x16 card and cut the connector down to fit into a PCIe x4 slot. "Why on earth would you want to do that?" is the obvious question. Well we inherited half a dozen Poweredge 440s that we wanted to use ...
For a single server, what you're essentially asking about in maximizing network throughput is not so much load balancing, but teaming. Load Balancing is typically done by a "front-end" device that doesn't do a whole lot of processing, other than to keep track of traffic, and send any new incoming "requests" to the currently least-used application server. ...
Only top voted, non community-wiki answers of a minimum length are eligible