Re: [SR-Users] [permissions] 'address' table and mask priorirty

2010-04-28 Thread Iñaki Baz Castillo
2010/4/28 Uriel Rozenbaum : > Iñaki, > > Maybe you can try with 2 masks that are not 32. > example: > > Entry 1: > - grp   = 1 > - ip_addr =  9.9.9.128 > - mask   =  25 > > Entry 2: > - grp   = 2 > - ip_addr =  9.9.9.0 > - mask   =  24 > > And then verify if the order or id matters. But as per Juha

Re: [SR-Users] [permissions] 'address' table and mask priorirty

2010-04-28 Thread Uriel Rozenbaum
Iñaki, Maybe you can try with 2 masks that are not 32. example: Entry 1: - grp = 1 - ip_addr = 9.9.9.128 - mask = 25 Entry 2: - grp = 2 - ip_addr = 9.9.9.0 - mask = 24 And then verify if the order or id matters. But as per Juha's comment, the result can be unpredictable. Maybe we an

[SR-Users] [permissions] 'address' table and mask priorirty

2010-04-28 Thread Juha Heinanen
Iñaki Baz Castillo writes: > However after checking the module code it seems that there are just > two cases: > - IP addresses (mask = 32). > - Subnets (mask != 32). > > So first the source address is always matched against he address hash, > and if it doesn't match then it is matched aga

[SR-Users] [permissions] 'address' table and mask priorirty

2010-04-28 Thread Iñaki Baz Castillo
Hi, some time ago I asked if allow_source_address_group() gives preference to the entries in 'address' table with lowest mask. I was replied that such case is not analyzed. But the fact is that in my tests it just works: Entry 1: - grp = 1 - ip_addr = 9.9.9.9 - mask = 32 Entry 2: - grp =