UDP Problems between HADB hosts ?
I try to create a management domain spanning two hosts as mentioned below.
hadbm createdomain --adminpasswordfile myfile.txt host1,host2
The management agents on these machines are already started with
/opt/SUNWhadb/4/bin/ma.
I see the following errors in ma.log file.
2006-05-17 14:56:01.590 SEVERE Thr{0011} DistributedRepository.enterDomain:
This node took too long with recovery. UDP Multicast problems?
After this, I enabled TCP logs on my hosts and I see UDP communication failures on host 1 and host2.
bash-2.05# ./tcpdump host host1 | grep udp
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on eri0, link-type EN10MB (Ethernet), capture size 68 bytes
15:54:23.425684 IP host1 > host2: ICMP host1 udp port 15225 unreachable, length 71
15:54:23.425694 IP host1 > host2: ICMP host1 udp port 15205 unreachable, length 71
The same failure is seen on host 2 as well for udp failures from host2 to host1.
Can someone please tell me where this UDP port configuration 15225 is made in the entire HADB setup ?
Thanks,
Srinivas
