AW: AW: [support] Delete tunnel with MCOA on HA

[Thomas Kempf]

Hello Romain,

I was wondering if the code on the website is the same as you sent? I checked out the code and found in src/ha.c the code some lines belower than in your previous mail. So the latest code on the website should work as well???

Thanks and regards,
Thomas


   1320             prefix_list_free(&bce->mob_net_prefixes);ope that it will
   1321             list_splice(&p.mob_net_prefixes, &bce->mob_net_prefixes);
   1322         }       status = IP6_MH_BAS_MCOA_BULK_FAILED;CULAR PURPOSE.
   1323         bcache_update_expire(bce);nse for more details.
   1324     }   }
   1325     /* bce is always valid here */sts->bid) { General Public License
   1326     bcache_release_entry(bce); for Linux; if not, write to the Free
   1327     if (!tsisset(lft)) has a BID = 0, this means no MCoA  Boston, MA
   1328         /* MCoA: We delete one specific entry (bid is set) 
   1329          * or all entries for this HoA (if bid is set to 0)
   1330          */  * New registratration is done later
   1331         bcache_delete(out.src, out.dst, bid);
   1332             MDBG("BID option but BCE has no BID\n");
   1333     if ((bu_flags & IP6_MH_BU_KEYM) &&ter new one.\n");
   1334         conf.pmgr.use_keymgm(out.dst, out.src))A_NO_BID);
   1335         ba_flags |= IP6_MH_BA_KEYM;_CONFLICT_BINDING;
   1336         }          
   1337     if (ba_



-----Ursprüngliche Nachricht-----
Von: support-bounces at l2tp.nautilus6.org [mailto:support-bounces at l2tp.nautilus6.org] Im Auftrag von Romain KUNTZ
Gesendet: Freitag, 26. September 2008 11:25
An: Support ML
Betreff: Re: AW: [support] Delete tunnel with MCOA on HA

Hello Thomas,

On 2008/09/25, at 12:31, Thomas Kempf wrote:
> The tunnel with CoA 2a01:348:109:3000: I'm switching off. At 12:13:30 
> there is the update for the CoA but the AP is already off.
> So it is the one with lifetime ==0. At 12:13:58 the tunnel is deleted. 
> I guess by the timer.

When receiving a BU with lifetime 0, the HA is supposed to delete the corresponding binding cache entry which triggers the deletion of the tunnel. So the described behaviour seems odd to me.

You may want to check why the entry is not deleted. In src/ha.c (line
1295) the following code is the one that is supposed to trigger the deletion of the binding cache. You could test what is the value of "lft" at that time:

1295     if (!tsisset(lft))
1296         /* MCoA: We delete one specific entry (bid is set)
1297          * or all entries for this HoA (if bid is set to 0)
1298          */
1299         bcache_delete(out.src, out.dst, bid);


Regards,
--
Romain KUNTZ
kuntz at lsiit.u-strasbg.fr
LSIIT - Networks and Protocols Team
http://clarinet.u-strasbg.fr/~kuntz/

_______________________________________________
Support mailing list
Support at ml.nautilus6.org
http://ml.nautilus6.org/mailman/listinfo/support