Networking :: Write A New Protocol On Transport Layer
Jul 23, 2010
I want to write a new protocol on transport layer. These are the features that new protocol should include:
-Real time
-Concurrent transfer
-Multipath and Multihoming
the reason why i should write this protocol is in order to transport encoded video stream over 3G link. When i investigated TCP UDP or SCTP source codes, i couldnot understand them at all. What ı want is to help me about steps of the protocol writing or guideline about it. How should i start, is there any book or source about it (i couldnot find over my internet research).
once again banging the head against a brick wall, entered the following commands in to client and server, but get the above message on both, is there something wrong with the script, or thw computers? mount 192.168.1.68:/export/shared /mnt/nfs_on_debian
im trying to send pages of 4096 bytes from kernel layer of server to kernel layer of client over a network. previously i tried the foll. code , for data less than a 100 bytes it worked fine , but for something larger than that the computer hangs......(even the dmesg's wont say why) i also wanted to know how we could use the 'sendpage' function to solve this problem.
I am trying to find out how I can estimate the IP Layer bitrate I would get from an ethernet link? Say at 100 Mbits/s or 200 Mbits/s (powerline)? Is there a formula that I could use to calculate that?
I'm looking for a way (kernel patches, configuration, etc) to bond multiple network interfaces together but for limited purposes. Here's the setup. Machines A, B, C, and D each have 4 NICs, each of which are on separate unmanaged switches. The connections are made in a corresponding way. e.g. eth0 of each machine are connected via switch 0, eth1 are connected via switch 1, etc. There are also other machines which have only one NIC and are connected to switch 0 only. All NICs for A/B/C/D and the switches are gigabit speed. The remaining machines have a low traffic level. Machines A/B/C/D need the extended bandwidth. And this bandwidth need usually involves only one connection at a time.
E.g. machine A transferring files to machine C with no other traffic going on. The speed need is to cut the transfer times from several hours to few hours (such as 8 hours to 2 hours). Transfers of up to a few terabytes at a time are involved. IEEE 802.1AX won't accomplish this. It requires special support from a single switch that all connection go to (raising costs and reducing reliability). Also, from technical details of 802.1AX, it appears that a decision process is made for which traffic goes over which physical link based on destination information. It's unclear what impact this will have, but it looks like at least a single TCP connection cannot use all physical links.
And possibly all traffic from host A to host B is limited to a physical link (not any better than a round robin of crossover cables). What I am looking for is something that works entirely on an end-to-end basis within a LAN. If it works at the link layer, that could be OK as long as it doesn't have the limitations of 802.1AX. Working at the IP layer would be OK, too (as I can already envision the logic of how to make that work). This might be an experimental patch to the Linux kernel if anyone has tried it. I have not dug into kernel source to see what might be in there, yet, but will eventually do that if there isn't a patch already available.
Our ISP provides us with Layer 2 hardware (Modem?/Switch? (Hatteras)) for our leased line and internet access. I have been told that I should set up a VLAN capable router to separate the internet traffic from the internal traffic. I found that linux is capable of VLAN routing. Nice!
I have setup opensuse 10.3 put in two NICs and did the following vconfig add eth0 10 vconfig add eth0 20 ifconfig public.ip.add.ress netmask 255.255.255.252 eth0.10 up ifconfig 192.168.0.1 netmask 255.255.255.0 eth0.20 up
Plugged this NIC into the HATTERAS hardware (with a straight cable), and thought that this way I should be able to ping the public gateway or any ip out on the internet. My ISP is telling me that I should create a VLAN trunk to be able to 'use the internet', but as I understand in linux if I create any number of VLANs on one NIC they are already trunked. I also got the info, that the traffic is tagged, and I can separate the traffic reading the tags. I already read that some NICs are not able to VLAN because they are not able to handle the increased packet size.
Also that the MTU setting is important (dono' the exact value though, only that its important). I thought that a linux machine can act as a router and firewall in such a case, because proprietary switches/routers use linux as embedded os. This is my first meeting with VLAN so if this whole post does not make any sense or you think that I just need to RTFM more then tell me! I also have some (3) Dlink 3226 Layer2 switches around, but I think it would be waste to use a 24 port switch on this subject.
I was wondering how to install the Link Layer Topology on CentOS 5. I have installed the lld2d daemon from a Debian how to after compiling it. It seems to be running but it is not talking to the Windows 7 network mapping.
We run redundant switches that two nic's on each server connect to. We also run bonding on our servers. Because we have two switches, we can't run lacp or anything. If a switch goes into a crashed state where it doesn't pass traffic but still provides link, bonding thinks the interface is still up and thus will still send traffic through it. Does anybody know a better way to configure the fail over of the interface? This would be a similar situation to somebody using a media converter.
I was trying to install a new protocol under ns2-34 when I execute make command, I got this error :
In file included from ./config.h:60, from ./routing/address.h:38, from ant/ant.cc:4: /usr/local/include/tclcl.h:42: fatal error: tcl.h: Aucun fichier ou dossier de ce type compilation terminated. make: *** [ant/ant.o] Erreur 1
I can't find this tcl.h !! I don't know how to add this file !
We are running in to a issue where we don't seem to get our bridging to work as we expected. Meaning passing all traffic from Point A to Point B. As explained below:
[Code]...
This is the original connection. Now we are adding our Linux IP impairment box for IP tables in between Application Server A and Router X so the figure would look as below: Application Server A --> IP Impairment box --> Router X --> router Y --> Application Server B
The issue we are having is When a packets are sent from Server A to Server B they are going through for PING, TCP, UDP, but for BGP the connection is established at a TCP level (SYN, SYN-ACK, ACK), but the BGP OPEN packet does not get through and our BGP relation ship does not come up between Application Server A and B. Application Server A and B are a routers also but there are applications running on those router. BGP is underlying protocol used for Application Server A to be talking with Application Server B and getting routes for each other.
Even with the iptables empty (all rules default ACCEPT) we still can't seem to get these BGP OPEN's through. We see both sides of the application servers sending BGP OPEN's, but the Linux bridge never passes those packets to the other side of the bridge. Tracking the BGP packets, we see the packets at iptables raw PREROUTING and filter FORWARD tables but the packet doesn't get out the other interface.
i'm new in linux world i would like to know how can i add the rtp protocol to my iptables rule for Netfilter firewall,but without installing the asterisk server
I have linux running from a cd rom when I need to start linux. I have the dhcp program on my computer and I need to save it to the cd rom with the linux system on it. The dhcp program is in my /users/bennett/downloads/dualserverinstallationv6.83 folder. So how do I get ths save to my linux system on the cd rom so I can use it in Voiphopper.
I have a small home/office network, which currently consists of a very old mac running osx10.4, a windoze 7 laptop and an ubuntu 10.04/windoze vista laptop, along with a homebrew setup on a wii which understands smb. I'm in the process of setting up an additional ubuntu PC which will, among other things act as a file and web server, mainly for web development and for sharing files, music and videos.
Up until now I've just run everything on SMB, and I will have to set samba up on the new server, but I would prefer to use a different protocol for the ubuntu and mac systems, as SMB doesn't recognise/create file permissions properly and it takes a while sometimes to notice new files in folders it's recently looked in.
Normal Operation: Transmission daemon is running on a dedicated server. Connecting to the transmission web interface from a remote machine: Entering the server IP followed by the transmission port number (i.e. URL... in the web browser -> a window pops up asking for the credentials -> entering my credentials -> transmission web interface opens in the web browser.Problem Description: Can't open the transmission web interface from a remote machine behind squid proxy: Read Error.The system returned: (107) Transport endpoint is not connected
Bypassing squid proxy enables me to open the transmission web interface. Also from a WinXP machine the transmission web interface is accessible.Besides the squid error (above), SELinux alert appears too: Can someone help me with this?I'm not sure if the problem is in squid or SELinux configuration.
This message appears very often in teh output of very different programs, for example Firefox/Swiftfox, Thunderbird, Mplayer In case of Mozilla apps it sometimes freezes KDE interface too (can't switch windows or panel does not pop-up) for a few seconds. IPV6 is not compiled into kernel and module is not built. It is disabled on the eth0 interface too.
is it possible in the standard remote desktop viewer in Fedora 14 to use RDP instead of VNC protocol? I know that Ubuntu provides this feature. I already installed tsclient and rdesktop but that didn't help.Before you say I have to use VNC: I know... I am just curious how I can set up RDP to connect to Windows based machines.
I am testing MINE protocol now. I got the basic concept of MINE protocol, but i dont know how to configure MINE protocol by creating a tunnel between two systems.
I did this configuration with linux boxes for GRE protocol.
creating a tunnel for MINE protocol and give me configuration details.?
I configure IPtable on RadHat as firewall and i want to allow for IP Phone using SIP protocol.I already allow port 5060 for IP Phone using SIP Protocol and I can call out.The problem is:
1. I can call out and in but when they pick up my call they hear what i'm saying but i can't hear they are saying.
But after a while i`ll see this message if i run ssh with -vvv debug2: channel 4: chan_shutdown_write: shutdown() failed for fd 9: Transport endpoint is not connected
After this message i can do nothing anymore i have to restart the ssh session on the server side i used: openssh-server 1:6.7p1-5 + Fail2Ban 0.8.13-1 (jessie) on the client side: openssh-client 1:6.9p1-2+b1 (stretch).I cannot find anything on the internet.
I have a production box I am trying to get CA's XCOM transport utility running on. I have been able to successfully transmitt a file but on a receive I get the following error.
"XCOMU0505E Received a signal from TCP/IP"
and in the debug I get the following:
E - catotxpi.c(1623): TxpiReceive: Receive. E - catotxpi.c(1624): Entering TxpiReceive E - catotxpi.c(2835): GetConnect: 0x095006BC E - catotxpi.c(1684): Receiving header. MaxLength: 32784 E - catotcp.c(704): No timeout; select omitted for socket 5.
[Code].....
I do have a ticket open with CA and they are doing there best to assist me, but this is not a supported version, they only support RedHat distros. They want to know what the equivilant RedHat kernel is to my CentOS install. I guess my biggest question is: Anyone out there running XCOM on CentOS? I searched and searched and can't find anyone else running XCOM like this. It's hard to believe I'm the only one out there trying to do this.
i'm having an error when running a python script that uses the python-twisted-web components... i get the impression that this program was developed and tested in a debian environment. heres the readout:
Quote:
Traceback (most recent call last): File "/usr/lib/python2.7/site-packages/twisted/internet/base.py", line 1165, in run self.mainLoop()
Trying to install 10.10 netbook edition on my MSI U230 netbook from a USB drive. Keep getting the error "hyper transport sync flood error occurred on last boot" Press F1 to Resume. F1 causes just a reboot and the same thing happens. Anyone seen this error? It happens with both the netbook and desktop version.
