Does anyone know if the shadow replicates created during a cdr sync repl
will be automatically deleted or will I have to manually remove them?  If
they are removed automatically, when are they removed?  I still see them
hanging around on the target servers that were sync'd.

This is what I've got and what I've done:

IDS 10.00.UC5 - Linux RHEL3 - 1 primary ER node, 3 read-only leaf ER nodes

cdr def ser -c oninit01 -I g_oninit01
cdr def ser -c oninit02 -I -S g_oninit01 -L g_oninit02
cdr def ser -c oninit03 -I -S g_oninit01 -L g_oninit03
cdr def ser -c oninit04 -I -S g_oninit01 -L g_oninit04

cdr def repl orders -c g_oninit01 -i -I -C "ignore" -f y -n y -M g_oninit01
"P stores_demo@g_oninit01:informix.orders" "select * from orders" "R
stores_demo@g_oninit02:informix.orders" "select * from orders" "R
stores_demo@g_oninit03:informix.orders" "select * from orders" "R
stores_demo@g_oninit04:informix.orders" "select * from orders"

# insert some rows into orders on oninit01 (master)

cdr start repl orders

cdr sync repl -c g_oninit01 -a -m g_oninit01 -r orders

After the sync completes, the leaf nodes are sync'd with the master node but
the shadow replicates remain on the leaf nodes.

cdr lis repl -c g_oninit02 (same for g_oninit03 and 04)

CURRENTLY DEFINED REPLICATES
-------------------------------
REPLICATE:        orders
STATE:            Active ON:g_oninit02
CONFLICT:         Ignore
FREQUENCY:        immediate
QUEUE SIZE:       0
PARTICIPANT:      stores_demo:informix.orders
OPTIONS:          transaction,floatieee,fullrow
REPLID:           65539 / 0x10003
REPLMODE:         READ-ONLY ON:g_oninit02
APPLY-AS:         INFORMIX ON:g_oninit02
REPLTYPE:         Master

REPLICATE:        sync_65539_1166596919
STATE:             ON:g_oninit02
CONFLICT:         Always Apply
FREQUENCY:        immediate
QUEUE SIZE:       0
PARTICIPANT:      :.
OPTIONS:          row
REPLID:           65542 / 0x10006
REPLMODE:         OTHER () ON:g_oninit02
APPLY-AS:         OWNER ON:g_oninit02
REPLTYPE:         Shadow
PARENT REPLICATE: orders

And the following appears in the master node online.log, anything to be
worried about?

15:24:10  CDR GC: operation sparse repl lookup failed (error 0).

Thanks,

Andrew