Nerea Toledo Gandarias wrote: Hi,
Hi, Miika Komu escribió:Varjonen Samu wrote: Hi,We have one guy who is working on a rewrite of the code on the experiences we have had with the mobility parts of the code. This problem is on the list also :)the current mobility code has been bug fixed multiple times and now it is time to reorganize the inflated code. Baris is also working on to support fault-tolerance better in the current mobility code. Fault tolerance means that daemon will detect problems in connectivity with ICMPv6 messages to HITs and will try to recover automatically. The daemon will send UPDATE messages through all possible address pairs to find even a single working locator pair. A similar procedure will be applied to base exchange (I1 packets) as well. I would expect incrementally mergeable results throughout this Spring. Stay tuned!any notice about those results concerning UPDATE message exchange??
Baris has fixed some related bugs (679, 695, 801). I solved some problems (370, 676 and 864). They have been merged already.
Baris is working on the following bugs this summer: 181 lazy vs. active handover 183 handling of UPDATE retransmissions 370 adding and deleting of security associations 391 delayed handovers 392 input packet queue 395 Learning peer addresses from update packets 396 shotgun-style mobility 434 esp_info and spi numbers 435 mobility code clean-up656 Some ideas on the revised code for mobility especially the address selection parts PSEUDO CODE
I'll merge the changes when they don't break any existing functionality.I would expect more progress towards the end of the summer. Would you like to subscribe to the bugs in bugzilla to get quicker notifications?