Hello, there is no such algorithm done internally. But you can get it via configuration: - use two set ids, first with the 3 nodes and then the second with the other 2 nodes - first to ds_select_*(setid1, 4) and ds_next_*() utill you consume all of them - if no answer yet and no more addresses with setid1, then call ds_select_*(setid2, 4) and ds_next_*() You can also look to see if weight based distribution fits your needs, so you can have the 3 good nodes with high weight and the other with something lower. But it won't try first the 3 good ones, it will distribute across all of them based on their weight. Cheers, Daniel On 11/14/13 2:54 AM, santosh satapathy wrote: -- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Kamailio Advanced Trainings - Berlin, Nov 25-28 - more details about Kamailio trainings at http://www.asipto.com -