Folks, I have been going through some code samples on kstat. In the simplest case, you open the kstat chain (kstat_open()), walk down the chain, and read values (kstat_read).
Is the data being collected during kstat_open() or during kstat_read()? If it is getting collected during kstat_open, it must be collecting ALL the counters whether or not the user needs them. This must be performance-expensive. If it is getting collected during kstat_read, it is collecting only one counter at a time, making it performance-inefficient. I don't need all the data. Just some counters on CPU and disk. Is there a way to inform kstat upfront on the names of counters that need to be collected? Another related question. I need to continuously collect data on a 30 second interval. What is the best way to achieve this? Thank you in advance for your help. Regards, Peter -- This message posted from opensolaris.org _______________________________________________ opensolaris-code mailing list opensolaris-code@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
