Big checkpoints

Hi,
I've go a small db of about 2Gb in size and only 40 sessions. The server is NT with 3CPU, 2.5Gb
memory, and RAID5 for data, raid 1 for logs and raid1 for arch logs. The version of Oracle is 8.1.6.

The problem I'm getting is that during normal system operation, a log switch checkpoint will start,
and then does not appear to complete for 30 mins! During this time there seems to be no undue disk
activity and the system is not under any real stress. Is there a reason for this. SOme params are:

db_block_lru_latches=1
dbwr_io_slaves=2
db_block_size = 4096
log_buffer = 512000
db_block_buffers = 102400

There are 5 redo groups each with 1 member of 10M (although I'm adding a new member per group ASAP).

Any ideas why the check point would take so long? It's driving me mad.

I've set log_checkpoint_time_out to 1800 but this doesn't seem to cause a checkpoint (how does this
work?) I've even tried to set db_file_max_dirty but I'm not sure this is doing anything (whats the
recommendation for this).

Thanks in anticipation

• Posted via NewsOne.Net: Free (anonymous) Usenet News via the Web ----- http://newsone.net/ -- Free reading and anonymous posting to 60,000+ groups NewsOne.Net prohibits users from posting spam. If this or other posts made through NewsOne.Net violate posting guidelines, email abuse_at_newsone.net