Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Mailing Lists -> Oracle-L -> Transportable tablespaces
I'm rebuilding a test database in 8.1.6 using transportable tablespaces
to copy the data. All is going well except I was just importing a
partitioned table (17 partitions) and the imp was sitting on that
particular table for 2+ hours. The only lock I saw was against
sys.tab$, I'm sure there were more but that was the only one I caught
near the end of the import. I have a few waits for the session listed
below but surely nothing so outstanding for it to take hours to imp.
Any thoughts on what caused a partitioned table to import for so long versus a regular table? It is a beast of a table, 455 columns, and I saw some procedure executing (I lost the name of it) which appeared to be walking through each column per-partition. Anyone experienced this type of behavior before when transporting tablespaces and if so any thoughts on speeding it up? I was going to take this approach to re-building the production database but this table has 56 partitions in production and that could mean several more hours just to import this one table.
EVENT TOTAL TOTAL TIME WAITS TIMEOUTS WAITED ----------------------------- ----- -------- ------ file identify 40 0 0 file open 84 0 7 refresh controlfile command 1 0 8 control file sequential read 197 0 15 single-task message 1 0 17 log file sync 42 0 19 latch free 8 8 27 control file parallel write 95 0 48 rdbms ipc reply 18 0 84 db file sequential read 225 0 377
Thanks - Brian