Another checkpoint situation

Here's a situation hopefully someone else has seen: HPUX 10.01 OnLine 7.13.UC1 I was monitoring a specific checkpoint; we've had somewhat of a problem with the checkpoint durations. In this instance, only had an average of 50pgs per LRU queue (8 queues) dirty, but the ckpoint took 18 seconds. Seems like it should have taken a shorter amount of time. Checked the disks; not too busy (BTW, another situation I've discovered is that a checkpoint may take longer if that particular disk is busy. Had a report that hammered a table, found the checkpoint hanging on that chunk. But I digress . . ) Here's sample output from the onstat -F. I thought that the chunk number would show in the event of a chunk write. This output appears to be mistaken. I've never noticed the "-1" or the hex number. address flusher state data c93be444 0 C -1 = 0xffffffff c93be878 1 I 0 = 0X0 c93becac 2 I 0 = 0X0 c93bf0e0 3 I 0 = 0X0 c93bf514 4 I 0 = 0X0 c93bf948 5 I 0 = 0X0 c93bfd7c 6 I 0 = 0X0 c93c01b0 7 I 0 = 0X0 states: Exit Idle Chunk Lru Any ideas? Just for the record, we're in the process of a major upgrade on our test system. I'll be tuning our production system (the one reported above) after the upgrade is completed. thanks in advance, John Carlson Informix DBA WH Smith, Inc.