-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Titus Brown wrote: | That's definitely the medium-term goal. Err, one question -- do you | think this is a software problem, or a hardware problem? i.e. will we | need Genuine XO Hardware (tm) to reproduce?
Determining which layer of the networking stack contains a bug is one of the most challenging problems we have had with recent network debugging. One lead developer recently proposed running all our tests twice, over the wireless network and also using RJ-45 USB dongles, in order to help classify these problems. Running tests on other hardware, or in virtual machines, would also be worthwhile. It should be noted that there are already two active OLPC networking testbeds: http://wiki.laptop.org/go/Collaboration_Network_Testbed http://wiki.laptop.org/go/Mesh_OLPC_Testbed If your project this summer requires XO's, I suspect OLPC will be able to provide you with several. However, the main network testbed has over 100 XOs, which is necessary for testing scalability bugs (of which there are many). As that page suggests, essentially no automation has yet been achieved on the network testbeds. If you develop a way to perform automated testing on the network testbeds, OLPC will happily adopt it. - --Ben -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFH6xo5UJT6e6HFtqQRAvFaAJ90pTS/exEJtiPvk7grn2kXBU0nJQCeIIoR vzQYa0NDnpNteg4CLiSy2BI= =Qo12 -----END PGP SIGNATURE----- _______________________________________________ Testing mailing list [email protected] http://lists.laptop.org/listinfo/testing
