Greetings, I have been looking at AWR Reports on a two-node RAC cluster running RHEL4, 32-bit and Oracle 10.2.0.3 SE and comparing the PHYSICAL_MEMORY_BYTES value with the same information on another identical cluster and finding a big discrepancy in the values which I don't understand. I've been looking around and so far have not been able to come up with an explanation for what I am seeing. Let's take one cluster at a time. Both machines in the cluster have 4GB physical memory - myucd1:/local/d02/oracle->cat /proc/meminfo MemTotal: 4090376 kB MemFree: 530076 kB Buffers: 205836 kB Cached: 2389484 kB SwapCached: 0 kB Active: 2832112 kB Inactive: 386320 kB HighTotal: 3211100 kB HighFree: 64000 kB LowTotal: 879276 kB LowFree: 466076 kB SwapTotal: 2096376 kB SwapFree: 2096376 kB Dirty: 332 kB Writeback: 0 kB Mapped: 1820464 kB Slab: 94728 kB CommitLimit: 4141564 kB Committed_AS: 5474452 kB PageTables: 225628 kB VmallocTotal: 106488 kB VmallocUsed: 7388 kB VmallocChunk: 98664 kB HugePages_Total: 0 HugePages_Free: 0 Hugepagesize: 2048 kB Yet selecting from v$osstat shows only 916644 bytes - SQL> select * from v$osstat; STAT_NAME VALUE OSSTAT_ID -------------------------------- ---------- ---------- NUM_CPUS 4 0 IDLE_TIME 591089966 1 BUSY_TIME 30269232 2 USER_TIME 12758301 3 SYS_TIME 17435075 4 IOWAIT_TIME 12346622 5 NICE_TIME 301 6 RSRC_MGR_CPU_WAIT_TIME 0 14 LOAD .709960938 15 NUM_CPU_SOCKETS 2 17 PHYSICAL_MEMORY_BYTES 916644 1008 As I understand it PHYSICAL_MEMORY_BYTES should indicate the actual physical memory available so I don't understand why the discrepancy. I have not gathered system statistics and don't know if that would have an impact. It also doesn't make sense to me that being a node in a cluster would affect this but I don't know. Any help gratefully appreciated. Thanks. Bill Wagman Univ. of California at Davis IET Campus Data Center wjwagman@xxxxxxxxxxx (530) 754-6208 -- //www.freelists.org/webpage/oracle-l