Ack. Thanks, Ramesh.
On 9/10/2014 9:48 AM, [email protected] wrote: > Suppose if the entry retrieved form subscription table is of | 0 > osaf/libs/core/mds/mds_c_sndrcv.c | 8 +++++++- > 2 files changed, 7 insertions(+), 1 deletions(-) > > > While sending Mcast message , only ONE fist entry of to_svc_id entry is > used/picked form subscription table and its architecture is used to do send > Full encode or Falt encode . > > In case of mixed architecture cluster, there is possibility of to_svc_id > entry can be randomly > different architecture or same architecture. Suppose if the entry retrieved > form subscription table is of > type the Mcast message is getting sent as Flat encode > because of that the different architecture nodes not able to receive the > Mcast message as they expect Full encode message. > > Now Mcast message is always sent as Full encode (DESTINATION_OFF_NODE ). > > diff --git a/osaf/libs/core/mds/mds_c_sndrcv.c > b/osaf/libs/core/mds/mds_c_sndrcv.c > --- a/osaf/libs/core/mds/mds_c_sndrcv.c > +++ b/osaf/libs/core/mds/mds_c_sndrcv.c > @@ -3855,7 +3855,13 @@ static uint32_t mcm_pvt_process_svc_bcas > break; > } > > - mcm_query_for_node_dest_on_archword(info_result->key.adest, > &to, info_result->rem_svc_arch_word); > + /* If Mcast allwasy send full encode */ > + if ((svc_cb->subtn_info->prev_ver_sub_count == 0) && > (strcmp(tipc_or_tcp, "TIPC") == 0) > + && (to_msg.data.info.len < > MDS_DIRECT_BUF_MAXSIZE)) { > + to = DESTINATION_OFF_NODE; > + } else { > + > mcm_query_for_node_dest_on_archword(info_result->key.adest, &to, > info_result->rem_svc_arch_word); > + } > > if (to == DESTINATION_SAME_PROCESS) { > if (to_msg.msg_type != MSG_NCSCONTEXT) { ------------------------------------------------------------------------------ Want excitement? Manually upgrade your production database. When you want reliability, choose Perforce Perforce version control. Predictably reliable. http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk _______________________________________________ Opensaf-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/opensaf-devel
