Hello all,
I noticed that with a UDP convergence layer, I cannot send bundles more than ~1300B in size. Does IBR-DTN support fragmentation and reassembly? Would really appreciate any help on this.
Thanks, Harsha
Hi Harsha,
at the moment IBR DTN does not support fragmentation of bundles. Also the DTN UDP convergence layer has no fragmentation capabilities. However, normally you should be able to get bigger sizes because IP supports fragmentation. Could it be that you use IPv6? Then there is no fragmentation on IP level, and the usable payload size is limited by the MTU (the minimum MTU for IPv6 is 1280 bytes, so if your IPv6 setup is configured with this MTU, this might be the limit you are hitting)
MfG
Sebastian
Am 10.11.2011 um 19:57 schrieb Harsha Chenji:
Hello all,
I noticed that with a UDP convergence layer, I cannot send bundles more than ~1300B in size. Does IBR-DTN support fragmentation and reassembly? Would really appreciate any help on this.
Thanks, Harsha -- !! This message is brought to you via the `ibr-dtn' mailing list. !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust !! your settings, send a mail message to ibr-dtn-request@ibr.cs.tu-bs.de !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/ibr-dtn.
Hello,
No it was IPv4 for sure. I was able to send huge packets on TCP/ibr-dtn but not over UDP. This was on version 0.6.4 though...has it been fixed in 0.6.5?
Thanks again, Harsha
On Fri, Nov 11, 2011 at 6:58 AM, Sebastian Schildt schildt@ibr.cs.tu-bs.de wrote:
Hi Harsha,
at the moment IBR DTN does not support fragmentation of bundles. Also the DTN UDP convergence layer has no fragmentation capabilities. However, normally you should be able to get bigger sizes because IP supports fragmentation. Could it be that you use IPv6? Then there is no fragmentation on IP level, and the usable payload size is limited by the MTU (the minimum MTU for IPv6 is 1280 bytes, so if your IPv6 setup is configured with this MTU, this might be the limit you are hitting)
MfG
Sebastian
Am 10.11.2011 um 19:57 schrieb Harsha Chenji:
Hello all,
I noticed that with a UDP convergence layer, I cannot send bundles more than ~1300B in size. Does IBR-DTN support fragmentation and reassembly? Would really appreciate any help on this.
Thanks, Harsha -- !! This message is brought to you via the `ibr-dtn' mailing list. !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust !! your settings, send a mail message to ibr-dtn-request@ibr.cs.tu-bs.de !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/ibr-dtn.
-- !! This message is brought to you via the `ibr-dtn' mailing list. !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust !! your settings, send a mail message to ibr-dtn-request@ibr.cs.tu-bs.de !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/ibr-dtn.
Hi.
The TCP convergence layer split up bundles during the transfer into chunks. The UDP convergence layer just puts one bundle into one UDP frame as specified in http://tools.ietf.org/html/draft-irtf-dtnrg-udp-clayer-00. That means the bundle size is limited by the MTU.
By default the MTU for a UDP convergence layer is hard-coded and limited to 1280 bytes. To change this you could add a value in the Main.cpp. Change this line...
UDPConvergenceLayer *udpcl = new UDPConvergenceLayer( net.interface, net.port );
like this...
UDPConvergenceLayer *udpcl = new UDPConvergenceLayer( net.interface, net.port, 65535 );
Whether 0.6.5 or 0.6.4, both versions are equal with this behavior. In the next version the MTU of convergence layers will be configurable in the config file.
Best regards, Johannes
Am 15.11.2011 00:25, schrieb Harsha Chenji:
Hello,
No it was IPv4 for sure. I was able to send huge packets on TCP/ibr-dtn but not over UDP. This was on version 0.6.4 though...has it been fixed in 0.6.5?
Thanks again, Harsha
participants (3)
-
Harsha Chenji -
Johannes Morgenroth -
Sebastian Schildt