immnd going to headless twice when receive event NCSMDS_DOWN, NCSMDS_RED_DOWN
in the same time.
The fix is prevent go to headless many time.
---
src/imm/immnd/immnd_evt.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/src/imm/immnd/immnd_evt.c b/src/imm/immnd/immnd_evt.c
index 656ec9398..9257529e8 100644
--- a/src/imm/immnd/immnd_evt.c
+++ b/src/imm/immnd/immnd_evt.c
@@ -12230,6 +12230,7 @@ static uint32_t immnd_evt_proc_mds_evt(IMMND_CB *cb,
IMMND_EVT *evt)
/*TRACE_ENTER(); */
uint32_t rc = NCSCC_RC_SUCCESS;
bool is_headless = false;
+ bool already_headless = false;
IMMSV_MDS_INFO *mdsInfo = &evt->info.mds_info;
if ((mdsInfo->change == NCSMDS_DOWN) &&
@@ -12243,11 +12244,17 @@ static uint32_t immnd_evt_proc_mds_evt(IMMND_CB *cb,
IMMND_EVT *evt)
/* In multi partitioned clusters rejoin, IMMND may not realize
* headless due to see IMMDs from different partitions */
if (mdsInfo->svc_id == NCSMDS_SVC_ID_IMMD) {
+ if ((cb->immd_node_id == 0) && (cb->other_immd_id == 0)) {
+ TRACE_2("Node already headless");
+ already_headless = true;
+ }
switch (mdsInfo->change) {
case NCSMDS_DOWN:
- is_headless = true;
- cb->immd_node_id = 0;
- cb->other_immd_id = 0;
+ if (!already_headless) {
+ is_headless = true;
+ cb->immd_node_id = 0;
+ cb->other_immd_id = 0;
+ }
break;
case NCSMDS_RED_UP:
if ((mdsInfo->role == V_DEST_RL_STANDBY) &&
@@ -12286,7 +12293,8 @@ static uint32_t immnd_evt_proc_mds_evt(IMMND_CB *cb,
IMMND_EVT *evt)
TRACE_2("IMMD RED_DOWN EVENT %x role=%d ==> ACT:%x
SBY:%x",
mdsInfo->node_id, mdsInfo->role,
cb->immd_node_id, cb->other_immd_id);
- if ((cb->immd_node_id == 0) && (cb->other_immd_id ==
0)) {
+ if (!already_headless && (cb->immd_node_id == 0) &&
+ (cb->other_immd_id == 0)) {
LOG_WA("Both Active & Standby DOWN, going to
headless");
is_headless = true;
}
--
2.25.1
_______________________________________________
Opensaf-devel mailing list
Opensaf-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-devel
