Re: [B.A.T.M.A.N.] [RFC] batman-adv: prevent TT request storms by not sending inconsistent TT TLVLs

[Sven Eckelmann <sven@narfation.org>]

[-- Attachment #1: Type: text/plain, Size: 1475 bytes --]

On Mittwoch, 9. Mai 2018 20:20:58 CEST Sven Eckelmann wrote:
> On Mittwoch, 9. Mai 2018 19:21:32 CEST Sven Eckelmann wrote:
> [...]
> > [...]
> > > diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
> > > index 0225616d..def8d109 100644
> > > --- a/net/batman-adv/translation-table.c
> > > +++ b/net/batman-adv/translation-table.c
> > > @@ -855,7 +855,8 @@ batadv_tt_prepare_tvlv_global_data(struct batadv_orig_node *orig_node,
> > 
> > What about batadv_tt_prepare_tvlv_local_data?
> 
> Let me rephrase that: Are we sure that this is from intermediate nodes or is 
> there still the possibility that it is from the actual originator? Regarding 
> the latter: Should batadv_tt_prepare_tvlv_local_data be modified or will you 
> wait for Leonardo to debug why it happens in the first place?

Wait. When it is a problem of an intermediate node that doesn't have (for 
whatever reason) not a valid copy of the entries, shouldn't the intermediate
node just ignore the request and not answer it? Otherwise the next OGM with 
the data from batadv_tt_prepare_tvlv_local_data from the actual node will 
just trigger the next request by the receiver.

And the next thing is that your commit message is talking about OGMs all the 
time. But the OGM TVLV are not filled by the 
batadv_tt_prepare_tvlv_global_data but by the 
batadv_tt_prepare_tvlv_local_data. I am unable to find out what you actually
doing here and why you do it.

Kind regards,
	Sven

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 833 bytes --]