cannot allocate new log, seq 5281, needs checkpointing

Oracle gurus !

We have DG/UX 5.4 and Oracle 6.0.33.

When we run large SQL jobs, I get the following message in 4_25434.trc :

  cannot allocate new log, seq 5281, needs checkpointing

Also the whole system hangs, if you kill the large job everything goes back to normal. I've tried to change the parameters in init.ora, with no real luck.
The following parameters are set at our site:

  log_archive_dest = /arch/IFS/arch.IFS.dbf
  log_archive_start = true
  log_allocation = 15000
  log_checkpoint_interval = 15000
  log_buffer = 65536

  three log files (15,7 Mbyte each)
  block OS size is 512

If I understand this correctly a value of 30000, checkpoint only happens at a log switch, so 15000 should work !, why lower the value since I don't want less performance. Long recover is OK.

Am I missing something ?

Thanks for any help, Jonas.

-- 

   /johe_at_pbutv.pb.se/