diff -uNr a/heartbeat/SAPHana b/heartbeat/SAPHana

--- a/heartbeat/SAPHana	2016-04-26 12:01:55.620889964 +0200

+++ b/heartbeat/SAPHana	2016-04-26 12:03:17.240897137 +0200

@@ -2,9 +2,9 @@

 #

 # SAPHana

 #

-# Description:	Manages two single SAP HANA Instance in System Replication 

+# Description:  Manages two single SAP HANA Instance in System Replication

 #               Planned: do also manage scale-up scenarios

-#               currently the SAPHana is dependent of the analysis of 

+#               currently the SAPHana is dependent of the analysis of

 #               SAPHanaTopology

 #               For supported scenarios please read the README file provided

 #               in the same software package (rpm)

@@ -16,16 +16,17 @@

 # Support:      linux@sap.com

 # License:      GNU General Public License (GPL)

 # Copyright:    (c) 2013,2014 SUSE Linux Products GmbH

+# Copyright:    (c) 2015      SUSE Linux GmbH

 #

-# An example usage: 

+# An example usage:

 #      See usage() function below for more details...

 #

 # OCF instance parameters:

-#	OCF_RESKEY_SID

-#	OCF_RESKEY_InstanceNumber       

-#	OCF_RESKEY_DIR_EXECUTABLE   (optional, well known directories will be searched by default)

-#	OCF_RESKEY_DIR_PROFILE      (optional, well known directories will be searched by default)

-#	OCF_RESKEY_INSTANCE_PROFILE (optional, well known directories will be searched by default)

+#   OCF_RESKEY_SID

+#   OCF_RESKEY_InstanceNumber

+#   OCF_RESKEY_DIR_EXECUTABLE   (optional, well known directories will be searched by default)

+#   OCF_RESKEY_DIR_PROFILE      (optional, well known directories will be searched by default)

+#   OCF_RESKEY_INSTANCE_PROFILE (optional, well known directories will be searched by default)

 #   OCF_RESKEY_PREFER_SITE_TAKEOVER (optional, default is no)

 #   OCF_RESKEY_DUPLICATE_PRIMARY_TIMEOUT (optional, time difference needed between two last-primary-tiemstampe (lpt))

 #   OCF_RESKEY_SAPHanaFilter    (optional, should only be set if been told by support or for debugging purposes)

@@ -71,7 +72,7 @@

         info )

         case "$shf" in

             all) skip=0

-            ;;          

+            ;;

             none )

                 skip=1

                 ;;

@@ -80,13 +81,13 @@

                 mtype=${mtype#fh}

                 echo "$shf"|  grep -iq ${mtype}; search=$?

                 if [ $search -eq 0 ]; then

-                     skip=0  

+                     skip=0

                 else

                     skip=1

                 fi

             ;;

         esac

-        ;;    

+        ;;

     esac

     if [ $skip -eq 0 ]; then

         ocf_log "$level" "$message"

@@ -103,8 +104,8 @@

     local rc=0

     methods=$(saphana_methods)

     methods=$(echo $methods | tr ' ' '|')

-  cat <<-!

-	usage: $0 ($methods)

+    cat <<-EOF

+    usage: $0 ($methods)

     $0 manages a SAP HANA Instance as an HA resource.

@@ -118,8 +119,17 @@

     The 'validate-all' operation reports whether the parameters are valid

     The 'methods'      operation reports on the methods $0 supports

-	!

-	return $rc

+EOF

+    return $rc

+}

+

+function backup_global_and_nameserver() {

+    super_ocf_log info "FLOW $FUNCNAME ($*)"

+    local rc=0

+    cp /hana/shared/LNX/global/hdb/custom/config/global.ini /hana/shared/LNX/global/hdb/custom/config/global.ini.$(date +"%s")

+    cp /hana/shared/LNX/global/hdb/custom/config/nameserver.ini /hana/shared/LNX/global/hdb/custom/config/nameserver.ini.$(date +"%s")

+    super_ocf_log info "FLOW $FUNCNAME rc=$rc"

+    return $rc

 }

 #

@@ -130,11 +140,12 @@

 function saphana_meta_data() {

     super_ocf_log info "FLOW $FUNCNAME ($*)"

     local rc=0

-	cat <

+#

+    cat <

 <resource-agent name="SAPHana">

-<version>0.149.7</version>

+<version>0.151.1</version>

 <shortdesc lang="en">Manages two SAP HANA instances in system replication (SR).</shortdesc>

 <longdesc lang="en">

@@ -157,7 +168,7 @@

 2. landscapeHostConfiguration

    The interface is used to monitor a HANA system. The python script is named landscapeHostConfiguration.py.

    landscapeHostConfiguration.py has some detailed output about HANA system status

-   and node roles. For our monitor the overall status is relevant. This overall 

+   and node roles. For our monitor the overall status is relevant. This overall

    status is reported by the returncode of the script:

    0: Internal Fatal, 1: ERROR, 2: WARNING, 3: INFO, 4: OK

    The SAPHana resource agent will interpret returncodes 0 as FATAL, 1 as not-running or ERROR and and returncodes 2+3+4 as RUNNING.

@@ -168,14 +179,14 @@

    system replication takeover (sr_takeover) or to register a former primary to a newer one (sr_register).

 4. hdbsql / systemReplicationStatus

-   Interface is SQL query into HANA (system replication table).  The hdbsql query will be replaced by a python script 

+   Interface is SQL query into HANA (system replication table).  The hdbsql query will be replaced by a python script

    "systemReplicationStatus.py" in SAP HANA SPS8 or 9.

    As long as we need to use hdbsql you need to setup secure store users for linux user root to be able to

    access the SAP HANA database. You need to configure a secure store user key "SAPHANA${SID}SR" which can connect the SAP

-   HANA database: 

+   HANA database:

 5. saphostctrl

-   The interface saphostctrl uses the function ListInstances to figure out the virtual host name of the 

+   The interface saphostctrl uses the function ListInstances to figure out the virtual host name of the

    SAP HANA instance. This is the hostname used during the HANA installation.

 </longdesc>

@@ -207,7 +218,7 @@

     </parameter>

     <parameter name="DUPLICATE_PRIMARY_TIMEOUT" unique="0" required="0">

         <shortdesc lang="en">Time difference needed between to primary time stamps, if a dual-primary situation occurs</shortdesc>

-        <longdesc lang="en">Time difference needed between to primary time stamps, 

+        <longdesc lang="en">Time difference needed between to primary time stamps,

         if a dual-primary situation occurs. If the time difference is

         less than the time gap, then the cluster hold one or both instances in a "WAITING" status. This is to give an admin

         a chance to react on a failover. A failed former primary will be registered after the time difference is passed. After

@@ -231,12 +242,8 @@

         <content type="string" default="" />

     </parameter>

     <parameter name="SAPHanaFilter" unique="0" required="0">

-        <shortdesc lang="en">Define SAPHana resource agent messages to be printed</shortdesc>

-        <longdesc lang="en">Define SAPHana resource agent messages to be printed.

-        This parameter should only be set if requested by support. The default is sufficient for normal operation.

-        Values: ra-act-lpa-dec-flow

-        You could specify any combination of the above values like "ra-act-flow"

-        </longdesc>

+        <shortdesc lang="en">OUTDATED PARAMETER</shortdesc>

+        <longdesc lang="en">OUTDATED PARAMETER</longdesc>

         <content type="string" default="" />

     </parameter>

 </parameters>

@@ -271,7 +278,7 @@

     for m in start stop status monitor promote demote notify validate-all methods meta-data usage; do

         echo "$m"

     done

-	return $rc

+    return $rc

 }

 #

@@ -298,7 +305,7 @@

     local remoteNode=""

     local rc=1

     for cl in ${otherNodes[@]}; do

-        vHost=$(get_hana_attribute $cl  ${ATTR_NAME_HANA_VHOST[@]})

+        vHost=$(get_hana_attribute $cl  ${ATTR_NAME_HANA_VHOST[@]} "$cl")

         if [ "$vHost" = "$remoteHost" ]; then # we found the correct node

             remoteNode=$cl

             rc=0

@@ -347,9 +354,10 @@

 }

 #

-# function: get_hana_attribute 

+# function: get_hana_attribute

 # params:   NODE ATTR [STORE]

 # globals:  -

+# output:   attribute value

 #

 function get_hana_attribute()

 {

@@ -358,14 +366,20 @@

     local attr_node=$1

     local attr_name=$2

     local attr_store=${3:-reboot} # DONE: PRIO5 get this (optional) from parameter

-    local attr_default=${5:-}

+    local attr_default=${4:-}

+    local dstr

     local attr_val=""

-    attr_val=$(crm_attribute -N ${attr_node} -G -n "$attr_name" -l $attr_store -q -d "$attr_default"); rc=$?

-    if [ $debug_attributes -eq 1 ]; then

-        dstr=$(date)

-        echo "$dstr: SAPHana: crm_attribute -N ${attr_node} -G -n \"$attr_name\" -l $attr_store -q --> $attr_val" >> /var/log/fhATTRIBUTE

-    fi

-    echo "$attr_val"

+    dstr=$(date)

+    case "$attr_store" in

+        reboot | forever )

+            echo "$dstr: SAPHana: crm_attribute -N ${attr_node} -G -n \"$attr_name\" -l $attr_store -q" >> /var/log/fhATTRIBUTE

+            crm_attribute -N ${attr_node} -G -n "$attr_name" -l $attr_store -q -d "$attr_default" 2>>/var/log/fhATTRIBUTE; rc=$?

+            ;;

+        props )

+            echo "$dstr: SAPHana: crm_attribute -G -n \"$attr_name\" -t crm_config -q" >> /var/log/fhATTRIBUTE

+            crm_attribute -G -n "$attr_name" -t crm_config -q -d "$attr_default" 2>>/var/log/fhATTRIBUTE; rc=$?

+            ;;

+    esac

     super_ocf_log info "FLOW $FUNCNAME rc=$rc"

     return $rc

 }

@@ -388,11 +402,17 @@

     attr_old=$(get_hana_attribute $attr_node $attr_name $attr_store $attr_default); get_rc=$?

     if [ "$attr_old" != "$attr_value" ]; then

         super_ocf_log debug "DBG: SET attribute $attr_name for node ${attr_node} to ${attr_value} former ($attr_old) get_rc=$get_rc "

-        crm_attribute -N $attr_node -v $attr_value -n "$attr_name" -l $attr_store; rc=$?

-        if [ $debug_attributes -eq 1 ]; then

-            dstr=$(date)

-            echo "$dstr: SAPHana: crm_attribute -N $attr_node -v $attr_value -n \"$attr_name\" -l $attr_store" >> /var/log/fhATTRIBUTE

-        fi

+        dstr=$(date)

+        case "$attr_store" in

+            reboot | forever )

+                echo "$dstr: SAPHana: crm_attribute -N $attr_node -v $attr_value -n \"$attr_name\" -l $attr_store" >> /var/log/fhATTRIBUTE

+                crm_attribute -N $attr_node -v $attr_value -n "$attr_name" -l $attr_store 2>>/var/log/fhATTRIBUTE; rc=$?

+                ;;

+            props )

+                echo "$dstr: SAPHana: crm_attribute -v $attr_value -n \"$attr_name\" -t crm_config -s SAPHanaSR" >> /var/log/fhATTRIBUTE

+                crm_attribute -v $attr_value -n "$attr_name" -t crm_config  -s SAPHanaSR 2>>/var/log/fhATTRIBUTE; rc=$?

+                ;;

+        esac

     else

         super_ocf_log debug "DBG: LET attribute $attr_name for node ${attr_node} still be ${attr_value}"

         rc=0

@@ -408,7 +428,8 @@

 #

 function assert() {

     super_ocf_log info "FLOW $FUNCNAME ($*)"

-    local err_msg=$1 local default_rc=$OCF_NOT_RUNNING

+    local err_msg=$1

+    local default_rc=$OCF_NOT_RUNNING

     # DONE: Check, if we need to destinguish between probe and others

     if ocf_is_probe; then

         default_exit=$OCF_NOT_RUNNING

@@ -435,7 +456,7 @@

     local score=0

     if [ -n "$1" ]; then

         score=$1

-    fi 

+    fi

     # DONE: PRIO2: Only adjust master if value is really different (try to check that)

     oldscore=$(${HA_SBIN_DIR}/crm_master -G -q -l reboot)

     if [ "$oldscore" != "$score" ]; then

@@ -452,7 +473,7 @@

 #

 # function: scoring_crm_master - score instance due to role ans sync match (table SCORING_TABLE_PREFERRED_SITE_TAKEOVER)

 # params:   NODE_ROLES NODE_SYNC_STATUS

-# globals:  SCORING_TABLE_PREFERRED_SITE_TAKEOVER[@], 

+# globals:  SCORING_TABLE_PREFERRED_SITE_TAKEOVER[@],

 #

 scoring_crm_master()

 {

@@ -467,7 +488,7 @@

             if grep "$rolePatt" <<< "$roles"; then

                if grep "$syncPatt" <<< "$sync"; then

                   skip=1

-                  myScore=$score 

+                  myScore=$score

                fi

             fi

         fi

@@ -496,7 +517,7 @@

 # function: saphana_init - initialize variables for the resource agent

 # params:   InstanceName

 # globals:  OCF_*(r), SID(w), sid(rw), sidadm(w), InstanceName(w), InstanceNr(w), SAPVIRHOST(w), PreferSiteTakeover(w), 

-# globals:  sr_name(w), remoteHost(w), otherNodes(w) 

+# globals:  sr_name(w), remoteHost(w), otherNodes(w), rem_SR_name(w)

 # globals:  ATTR_NAME_HANA_SYNC_STATUS(w), ATTR_NAME_HANA_CLONE_STATE(w)

 # globals:  DIR_EXECUTABLE(w), SAPSTARTSRV(w), SAPCONTROL(w), DIR_PROFILE(w), SAPSTARTPROFILE(w), LD_LIBRARY_PATH(w), PATH(w)

 # globals:  LPA_DIRECTORY(w), SIDInstanceName(w), remoteNode(w), hdbSrQueryTimeout(w)

@@ -506,6 +527,8 @@

     super_ocf_log info "FLOW $FUNCNAME ($*)"

     local rc=$OCF_SUCCESS

     local vName

+    local clN

+    # local site

     # two parameter models (for transition only)

     # OLD: InstanceName

     # NEW: SID InstanceNumber

@@ -528,11 +551,10 @@

        #

        # if saphostctrl does not know the answer, try to fallback to attribute provided by SAPHanaTopology

        #

-       vName=$(get_hana_attribute ${NODENAME} ${ATTR_NAME_HANA_VHOST[@]});

+       vName=$(get_hana_attribute ${NODENAME} ${ATTR_NAME_HANA_VHOST[@]} "$NODENAME");

     fi

     SAPVIRHOST=${vName}

     PreferSiteTakeover="$OCF_RESKEY_PREFER_SITE_TAKEOVER"

-    SAPHanaFilter="${OCF_RESKEY_SAPHanaFilter:-ra-act-dec-lpa}"

     AUTOMATED_REGISTER="${OCF_RESKEY_AUTOMATED_REGISTER:-false}"

     LPA_DIRECTORY=/var/lib/SAPHanaRA

     LPA_ATTR=("lpa_${sid}_lpt" "forever")

@@ -591,6 +613,8 @@

        *openais* ) otherNodes=($(crm_node -l | awk '$3 == "member" { if ($2 != me) { print $2 }}' me=${NODENAME}));;

        *cman*    ) otherNodes=($(crm_node -l | awk '{for (i=1; i<=NF; i++) { if ($i != me) { print $i }}}' me=${NODENAME}));;

     esac

+    #

+    #

     remoteHost=$(get_hana_attribute ${NODENAME} ${ATTR_NAME_HANA_REMOTEHOST[@]});

     if [ -z "$remoteHost" ]; then

@@ -611,9 +635,13 @@

     #  ATTR_NAME_HANA_SITE

     sr_name=$(get_hana_attribute ${NODENAME} ${ATTR_NAME_HANA_SITE[@]});

     sr_mode=$(get_hana_attribute "${NODENAME}" ${ATTR_NAME_HANA_SRMODE[@]})

+

     if [ -z "$sr_mode" ]; then

         sr_mode="sync"

     fi

+    if [ -n "$remoteNode" ]; then

+        rem_SR_name=$(get_hana_attribute ${remoteNode} ${ATTR_NAME_HANA_SITE[@]});

+    fi

     super_ocf_log debug "DBG: sr_name=$sr_name, remoteHost=$remoteHost, remoteNode=$remoteNode, sr_mode=$sr_mode"

     # optional OCF parameters, we try to guess which directories are correct

     if  [ -z "$OCF_RESKEY_DIR_EXECUTABLE" ]

@@ -706,7 +734,7 @@

         then

             runninginst=$(echo "$output" | grep '^0 : ' | cut -d' ' -f3)

             if [ "$runninginst" != "$InstanceName" ]

-            then 

+            then

                 super_ocf_log warn "ACT: sapstartsrv is running for instance $runninginst, that service will be killed"

                 restart=1

             else

@@ -784,38 +812,113 @@

     node_full_status=$(su - ${sidadm} -c "hdbnsutil -sr_state" 2>/dev/null )

     node_status=$(echo "$node_full_status" | awk '$1=="mode:" {print $2}')

     super_ocf_log debug "DBG: check_for_primary: node_status=$node_status"

+    # TODO: PRIO2: Maybe we need to use a fallback interface when hdbnsitil does not answer properly -> lookup in config files?

+    #              This might also solve some problems when we could not figure-out the ilocal or remote site name

     for i in  1 2 3 4 5 6 7 8 9; do

        case "$node_status" in

-           primary ) 

-                  super_ocf_log info "FLOW $FUNCNAME rc=HANA_STATE_PRIMARY"

-                  return $HANA_STATE_PRIMARY;;

+           primary )

+               super_ocf_log info "FLOW: $FUNCNAME rc=HANA_STATE_PRIMARY"

+               return $HANA_STATE_PRIMARY;;

            syncmem | sync | async )

-                  super_ocf_log info "FLOW $FUNCNAME rc=HANA_STATE_SECONDARY"

-                  return $HANA_STATE_SECONDARY;;

-           none )     # have seen that mode on second side BEFEORE we registered it as replica

-                  super_ocf_log info "FLOW $FUNCNAME rc=HANA_STATE_STANDALONE"

-                  return $HANA_STATE_STANDALONE;;

+               super_ocf_log info "FLOW: $FUNCNAME rc=HANA_STATE_SECONDARY"

+               return $HANA_STATE_SECONDARY;;

+           none ) # have seen that mode on second side BEFEORE we registered it as replica

+               super_ocf_log info "FLOW: $FUNCNAME rc=HANA_STATE_STANDALONE"

+               return $HANA_STATE_STANDALONE;;

            * )

-	          super_ocf_log err "ACT: check_for_primary:  we didn't expect node_status to be: <$node_status>"

-                  dump=$( echo $node_status | hexdump -C );

-	          super_ocf_log err "ACT: check_for_primary:  we didn't expect node_status to be: DUMP <$dump>"

-                  node_full_status=$(su - ${sidadm} -c "hdbnsutil -sr_state" 2>/dev/null )

-                  node_status=$(echo "$node_full_status" | awk '$1=="mode:" {print $2}')

-                  super_ocf_log debug "DEC: check_for_primary: loop=$i: node_status=$node_status"

-                  # TODO: PRIO1: Maybe we need to keep the old value for P/S/N, if hdbnsutil just crashes

+               super_ocf_log err "ACT: check_for_primary:  we didn't expect node_status to be: <$node_status>"

+               dump=$( echo $node_status | hexdump -C );

+               super_ocf_log err "ACT: check_for_primary:  we didn't expect node_status to be: DUMP <$dump>"

+               node_full_status=$(su - ${sidadm} -c "hdbnsutil -sr_state" 2>/dev/null )

+               node_status=$(echo "$node_full_status" | awk '$1=="mode:" {print $2}')

+               super_ocf_log debug "DEC: check_for_primary: loop=$i: node_status=$node_status"

+               # TODO: PRIO1: Maybe we need to keep the old value for P/S/N, if hdbnsutil just crashes

        esac;

     done

     super_ocf_log info "FLOW $FUNCNAME rc=$rc"

     return $rc

 }

+# function: analyze_hana_sync_statusSRS

+# params:   -

+# globals:  DIR_EXECUTABLE(r), FULL_SR_STATUS(w), remoteNode

+#

+# systemReplicationStatus.py return-codes:

+#    NoHSR        = 10

+#    Error        = 11

+#    Unkown       = 12

+#    Initializing = 13

+#    Syncing      = 14

+#    Active       = 15

+function analyze_hana_sync_statusSRS()

+{

+    super_ocf_log info "FLOW $FUNCNAME ($*)"

+    local rc=-1 srRc=0 all_nodes_other_side="" n="" siteParam=""

+    if [ -n "$rem_SR_name" ]; then

+       siteParam="--site=$rem_SR_name"

+    fi

+    FULL_SR_STATUS=$(su - $sidadm -c "python $DIR_EXECUTABLE/python_support/systemReplicationStatus.py $siteParam" 2>/dev/null); srRc=$?

+    super_ocf_log info "DEC $FUNCNAME systemReplicationStatus.py (to site '$rem_SR_name')-> $srRc"

+    super_ocf_log info "FLOW $FUNCNAME systemReplicationStatus.py (to site '$rem_SR_name')-> $srRc"

+    #

+    # TODO: PRIO2: Here we might also need to filter additional sites (if multi tier should be supported)

+    #              And is the check for return code capable for chains?

+    #

+    if [ $srRc -eq 15 ]; then

+       # Fix for a HANA BUG, where a non-working SR resulted in RC 15:

+       if grep -q "ACTIVE" <<< "$FULL_SR_STATUS"; then

+          super_ocf_log info "FLOW $FUNCNAME SOK"

+          set_hana_attribute "$remoteNode" "SOK" ${ATTR_NAME_HANA_SYNC_STATUS[@]}

+          super_ocf_log info "ACT site=$sr_name, seting SOK for secondary (1)"

+          lpa_set_lpt 30 "$remoteNode"

+          rc=0;

+       else

+          # ok we should be careful and set secondary to SFAIL

+          super_ocf_log info "FLOW $FUNCNAME SFAIL"

+          set_hana_attribute "$remoteNode" "SFAIL" ${ATTR_NAME_HANA_SYNC_STATUS[@]}

+          super_ocf_log info "ACT site=$sr_name, seting SFAIL for secondary (6) - srRc=$srRc lss=$lss No ACTIVES found in cmd output"

+          # TODO: PRIO1 - P004: need to check LSS again to avoid dying primary to block (SFAIL) secondary

+          lpa_set_lpt 10 "$remoteNode"

+       fi

+    elif [ $srRc -le 11 ]; then # 11 and 10

+       # if systemReplicationStatus is ERROR and landscapeHostConfiguration is down than do NOT set SFAIL

+       get_hana_landscape_status; lss=$?

+       if [ $lss -lt 2 ]; then

+          # keep everithing like it was

+          rc=2

+       else

+          # ok we should be careful and set secondary to SFAIL

+          super_ocf_log info "FLOW $FUNCNAME SFAIL"

+          set_hana_attribute "$remoteNode" "SFAIL" ${ATTR_NAME_HANA_SYNC_STATUS[@]}

+          super_ocf_log info "ACT site=$sr_name, seting SFAIL for secondary (5) - srRc=$srRc lss=$lss"

+          # TODO: PRIO1 - P004: need to check LSS again to avoid dying primary to block (SFAIL) secondary

+          lpa_set_lpt 10 "$remoteNode"

+          rc=1

+       fi

+    else

+       super_ocf_log info "FLOW $FUNCNAME SFAIL"

+       set_hana_attribute "$remoteNode" "SFAIL" ${ATTR_NAME_HANA_SYNC_STATUS[@]}

+       super_ocf_log info "ACT site=$sr_name, seting SFAIL for secondary (2) - srRc=$srRc"

+       # TODO: PRIO1 - P004: need to check LSS again to avoid dying primary to block (SFAIL) secondary

+       lpa_set_lpt 10 "$remoteNode"

+       rc=1;

+    fi

+    super_ocf_log info "FLOW $FUNCNAME PRIM+LPA"

+    super_ocf_log info "DBG PRIM"

+    super_ocf_log info "FLOW $FUNCNAME rc=$rc"

+    return $rc

+}

+

 #

-# function: analyze_hana_sync_status - query and check hana system replication status

+####

+#### OLD HDBSQL STUFF FOR SPS6,7,8 AND SCALE-UP ONLY

+####

+# function: analyze_hana_sync_statusSQL - query and check hana system replication status

 # params:   -

 # globals:  DIR_EXECUTABLE(r), remoteHost(r)

 # get the HANA sync status

-# 

-function analyze_hana_sync_status()

+#

+function analyze_hana_sync_statusSQL()

 {

     super_ocf_log info "FLOW $FUNCNAME ($*)"

     local -a clusterNodes=()

@@ -863,35 +966,9 @@

             # TODO PRIO1: REMOVE remoteNode dependency - set SFAIL

             set_hana_attribute "$remoteNode" "SFAIL" ${ATTR_NAME_HANA_SYNC_STATUS[@]}

         fi

-        # first get a list of all secondary hosts, than a list of all secondary hosts, if the is ANY failure at this site

-        #    TODO: PRIO9: for first we assume there is only ONE secondary site (like ROT)

-        #    TODO: PRIO3: should we loop over all cluster nodes fetching their roles-attribute? To minimize sql-queries?

-        #

-        all_secondary_hosts=$(timeout $hdbSrQueryTimeout hdbsql -a -x -U $secUser $query_secondaries ); sqlrc=$?

-        all_secondary_hosts=$(echo $all_secondary_hosts | dequote);

-        if [ "$sqlrc" -eq 0 ]; then

-            all_broken_secondary_hosts=$(timeout $hdbSrQueryTimeout hdbsql -a -x -U $secUser $query_failed_secondaries); sqlrc=$?

-            all_broken_secondary_hosts=$(echo $all_broken_secondary_hosts | dequote);

-            if  [ "$sqlrc" -eq 0 ]; then                 

-                 if [ -n "$all_broken_secondary_hosts" ]; then

-                     #

-                     # we have a broken secondary site - set all hosts to "SFAIL"

-                     #

-                     # Note: since HANA hostname can be different from nodename we need to check all vhost attributes

-                     for n in $all_broken_secondary_hosts; do

-                         for cl in ${otherNodes[@]}; do

-                             vHost=$(get_hana_attribute $cl  ${ATTR_NAME_HANA_VHOST[@]})

-                             if [ "$vHost" = "$n" ]; then # we found the correct node

-                                set_hana_attribute $cl "SFAIL" ${ATTR_NAME_HANA_SYNC_STATUS[@]}

-                             fi

-                         done  

-                     done

-                 fi

-            fi

-        fi                  

     else

         case "$sqlrc" in

-            19 ) 

+            19 )

                 # return codes 19: license error -> set SFAIL!

                 # DONE: PRIO1: We should NOT set SFAIL, if HDB is exactly broken now

                 #       When HDB breaks during monitor this could prevent a prositive remote failover

@@ -901,7 +978,7 @@

                 done

                 ;;

         esac

-    fi  

+    fi

     return $rc

 }

@@ -932,10 +1009,18 @@

     local remoteInstance="";

     remoteInstance=$InstanceNr

     if ocf_is_true ${AUTOMATED_REGISTER}; then

+       #

+       #

+       #

+       #

+       #

        super_ocf_log info "ACT: REGISTER: hdbnsutil -sr_register --remoteHost=$remoteHost --remoteInstance=$remoteInstance --mode=$sr_mode --name=$sr_name"

+       #

+       #

        su - $sidadm -c "hdbnsutil -sr_register --remoteHost=$remoteHost --remoteInstance=$remoteInstance --mode=$sr_mode --name=$sr_name"; rc=$?

+       # backup_global_and_nameserver

     else

-       super_ocf_log info "ACT: IGNORE REGISTER because AUTOMATED_REGISTER is set to FALSE"

+       super_ocf_log info "ACT: SAPHANA DROP REGISTER because AUTOMATED_REGISTER is set to FALSE"

        rc=1

     fi

     super_ocf_log info "FLOW $FUNCNAME rc=$rc"

@@ -945,7 +1030,7 @@

 #

 # function: saphana_status - pure status check

 # params:   -

-# globals:  SIDInstanceName, OCF_*, 

+# globals:  SIDInstanceName, OCF_*,

 function saphana_status() {

     local binDeam="hdb.sap${SIDInstanceName}" rc=0

     binDeam=${binDeam:0:15}   # Process name is limited to the first 15 characters

@@ -956,13 +1041,13 @@

 #

 # function: saphana_start - start a hana instance

 # params:   -

-# globals:  OCF_*, SAPCONTROL, InstanceNr, SID, InstanceName, 

+# globals:  OCF_*, SAPCONTROL, InstanceNr, SID, InstanceName,

 #

 function saphana_start() {

     super_ocf_log info "FLOW $FUNCNAME ($*)"

     local rc=$OCF_NOT_RUNNING

     local output=""

-    local loopcount=0  

+    local loopcount=0

     check_sapstartsrv

     rc=$?

     #

@@ -1000,11 +1085,11 @@

 # saphana_stop: Stop the SAP instance

 #

 function saphana_stop() {

-  super_ocf_log info "FLOW $FUNCNAME ($*)"

-  local output=""

-  local rc=0

-  check_sapstartsrv; rc=$?

-  if [ $rc -eq $OCF_SUCCESS ]; then

+    super_ocf_log info "FLOW $FUNCNAME ($*)"

+    local output=""

+    local rc=0

+    check_sapstartsrv; rc=$?

+    if [ $rc -eq $OCF_SUCCESS ]; then

     output=$($SAPCONTROL -nr $InstanceNr -function Stop)

     rc=$?

     super_ocf_log info "ACT: Stopping SAP Instance $SID-$InstanceName: $output"

@@ -1032,7 +1117,7 @@

 # function: saphana_validate - validation of (some) variables/parameters

 # params:   -

 # globals:  OCF_*(r), SID(r), InstanceName(r), InstanceNr(r), SAPVIRHOST(r)

-# saphana_validate: Check the symantic of the input parameters 

+# saphana_validate: Check the symantic of the input parameters

 #

 function saphana_validate() {

   super_ocf_log info "FLOW $FUNCNAME ($*)"

@@ -1060,12 +1145,12 @@

 #

 # function: saphana_start_primary - handle startup of PRIMARY in M/S

 # params:

-# globals: OCF_*(r), NODENAME, ATTR_NAME_*, HANA_STATE_*, 

+# globals: OCF_*(r), NODENAME, ATTR_NAME_*, HANA_STATE_*,

 #

 function saphana_start_primary()

 {

     super_ocf_log info "FLOW $FUNCNAME ($*)"

-	local primary_status sync_attr score_master rc=$OCF_NOT_RUNNING

+    local primary_status sync_attr score_master rc=$OCF_NOT_RUNNING

     local lss sqlrc;

     local rc=0

     local lpa_dec=4

@@ -1074,7 +1159,7 @@

     # we will be a master (PRIMARY) so checking, if the is an OTHER master

     #

     super_ocf_log debug "DBG: saphana_primary - check_for_primary reports HANA_STATE_PRIMARY"

-    # 

+    #

     lpa_init_lpt $HANA_STATE_PRIMARY

     lpa_check_lpt_status; lpa_dec=$?

     get_hana_landscape_status; lss=$?

@@ -1139,7 +1224,7 @@

                 1 ) # landcape says we are down, lets start and adjust scores and return code

                     super_ocf_log info "LPA: landcape: DOWN, LPA: start ==> start instance"

                     saphana_start

-                    rc=$?                    

+                    rc=$?

                     LPTloc=$(date '+%s')

                     lpa_set_lpt $LPTloc

                     ;;

@@ -1152,7 +1237,7 @@

                     # DONE: PRIO3: check if this reaction is correct - tell cluster about failed start

                     super_ocf_log info "LPA: landcape: UP, LPA: register ==> take down"

                     set_crm_master -inf

-                    rc=$OCF_NOT_RUNNING 

+                    rc=$OCF_NOT_RUNNING

                     ;;

                 1 ) # lets try to register

                     # DONE: PRIO2: Like Action in start_secondary

@@ -1160,7 +1245,7 @@

                     super_ocf_log info "DEC: AN OTHER HANA IS AVAILABLE ==> LETS REGISTER"

                     set_crm_master  0

                     if wait_for_primary_master 1; then

-                        register_hana_secondary 

+                        register_hana_secondary

                         check_for_primary; primary_status=$?

                         if [ $primary_status -eq $HANA_STATE_SECONDARY ]; then

                             super_ocf_log info "ACT: Register successful"

@@ -1169,11 +1254,11 @@

                             set_crm_master  0

                             saphana_start_secondary

                             rc=$?

-                            lpa_set_lpt  30

+                            lpa_set_lpt  10

                         else

                             super_ocf_log err "ACT: Register failed"

                             rc=$OCF_NOT_RUNNING

-                        fi                                         

+                        fi

                     else

                         # lets check next monitor, if we can register

                         rc=$OCF_SUCCESS

@@ -1185,6 +1270,9 @@

             case "$lss" in

                 2 | 3 | 4 ) # as we ARE up we just keep it up

                     # TODO: PRIO3: I now change from "just keep it up to take that down"

+# TODO: PRIO1 differ lpt_advice!!

+# 2 => DOWN

+# 3 => KEEP

                     # TODO: PRIO3: OCF_SUCCESS, OCF_NOT_RUNNING or OCF_ERR_xxxx ?

                     set_crm_master -9000

                     #scoring_crm_master "$my_role" "$my_sync"

@@ -1193,7 +1281,7 @@

                 1 ) # we are down, so we should wait --> followup in next monitor

                     super_ocf_log info "LPA: landcape: DOWN, LPA: wait ==> keep waiting"

                     # TODO: PRIO3: Check, if WAITING is correct here

-                    set_hana_attribute ${NODENAME} "WAITING" ${ATTR_NAME_HANA_CLONE_STATE[@]}

+                    set_hana_attribute ${NODENAME} "WAITING4LPA" ${ATTR_NAME_HANA_CLONE_STATE[@]}

                     set_crm_master -9000

                     rc=$OCF_SUCCESS

                     ;;

@@ -1202,7 +1290,7 @@

         fail )   # process a lpa FAIL

             super_ocf_log info "LPA: LPA reports FAIL"

             set_crm_master -inf

-            rc=$OCF_NOT_RUNNING 

+            rc=$OCF_NOT_RUNNING

             ;;

     esac

     super_ocf_log info "FLOW $FUNCNAME rc=$rc"

@@ -1278,12 +1366,12 @@

 #

 # function: saphana_start_secondary - handle startup of PRIMARY in M/S

 # params:

-# globals: OCF_*(r), NODENAME, ATTR_NAME_*, 

+# globals: OCF_*(r), NODENAME, ATTR_NAME_*,

 #

 function saphana_start_secondary()

 {

     super_ocf_log info "FLOW $FUNCNAME ($*)"

-	local primary_status sync_attr score_master rc=$OCF_NOT_RUNNING

+    local primary_status sync_attr score_master rc=$OCF_NOT_RUNNING

     local sqlrc;

     set_crm_master 0

     #

@@ -1291,9 +1379,9 @@

     #

     lpa_push_lpt 10

     lpa_set_lpt  10

-    # 

+    #

     ####### LPA - end

-    # 

+    #

     #

     # we would be slave (secondary)

     # we first need to check, if there are Master Nodes, because the Scecondary only starts

@@ -1311,16 +1399,16 @@

                # It seams the stating secondary could not start because of stopping primary

                #    so this is a WAITING situation

                super_ocf_log info "ACT: PRIMARY seams to be down now ==> WAITING"

-               set_hana_attribute ${NODENAME} "WAITING" ${ATTR_NAME_HANA_CLONE_STATE[@]}

+               set_hana_attribute ${NODENAME} "WAITING4PRIM" ${ATTR_NAME_HANA_CLONE_STATE[@]}

                set_crm_master -INFINITY

                rc=$OCF_SUCCESS