[ECOS] TCP/IP stack work collapsing

[ECOS] TCP/IP stack work collapsing

[Marek]

Hi
I  use stack TCP/IP on MPC860;
I use eCos Configuration Tool 1.3.1.2 (Oct 20, 2000);
I have new packages (23-02-2001);
I run server aplication using TCP/IP stack;
It works correct if my client (or clients) send below 30 connnecting per sec
(client TCP);
My server stops if it receives above 40 connecting per sec and I have to
reset everything;
My server opens a socket in response to requests and closes it (the socket);

When server starts working cyg_kmem_print_stats function shows the following
information:
            Network stack mbuf stats:
               mbufs 0, clusters 1, free clusters 1
               Failed to get 0 times
               Waited to get 0 times
               Drained queues to get 0 times
            Misc mpool: total   65520, free   62032, max free block 61056
            Mbufs pool: total   65408, free   65280, blocksize  128
            Clust pool: total  131072, free  126976, blocksize 2048

after 15 mins of working
            Network stack mbuf stats:
               mbufs 0, clusters 9, free clusters 9
               Failed to get 0 times
               Waited to get 0 times
               Drained queues to get 0 times
            Misc mpool: total   65520, free   61808, max free block 60832
            Mbufs pool: total   65408, free   65280, blocksize  128
            Clust pool: total  131072, free  110592, blocksize 2048

When server stops its working cyg_kmem_print_stats function shows the
following information:
            Network stack mbuf stats:
               mbufs 118, clusters 63, free clusters 0
               Failed to get 0 times
               Waited to get 0 times
               Drained queues to get 0 times
            Misc mpool: total   65520, free   31184, max free block 10032
            Mbufs pool: total   65408, free   51328, blocksize  128
            Clust pool: total  131072, free            0, blocksize 2048

after work collapsing (about 50 mins after stopping) cyg_kmem_print_stats
function shows the following information:
            Network stack mbuf stats:
               mbufs 629, clusters 63, free clusters 1
               Failed to get 0 times
               Waited to get 0 times
               Drained queues to get 0 times
            Misc mpool: total   65520, free   31184, max free block 10032
            Mbufs pool: total   65408, free       128, blocksize  128
            Clust pool: total  131072, free            0, blocksize 2048

after work collapsing (about 2 h after stopping) cyg_kmem_print_stats
function shows the following information:
            Network stack mbuf stats:
               mbufs 996, clusters 63, free clusters 1
               Failed to get 0 times
               Waited to get 0 times
               Drained queues to get 0 times
            Misc mpool: total   65520, free   31184, max free block 10032
            Mbufs pool: total   65408, free       128, blocksize  128
            Clust pool: total  131072, free            0, blocksize 2048

I dont know what is wrong?
Why server's working collapses?
What I should to do now?

Marek ( pekalski@rts.com.pl )