Messages in this thread |  | | | From | "Jeffrey Noll" <> | | Subject | Q: Gathering I/O stats per device and NFS stats | | Date | Sun, 22 Mar 1998 12:16:04 -0500 |
| |
I'm in the process of reverse enginerring the hp-ux glanceplus program and am running into some problems with tracking down some statistics that are available at least from the hp-ux kernel. What i'm looking to gather in some way are the following: Disk I/O requests per device: ie: total read requests for sda1, sda2, hda1, etc. Disk transfers per device NFS read/write requests NFS transfers
Also, its trivial enough to figure out things such as %cpu time spent in User mode, and number of interrupts/cswitches, etc over that period of time, however is their anyway to calculate from this the %cpu time spent doing cswitches, servicing interrupts, syscalls, etc, like you apparently can do with hp-ux?
I've looked all over /proc and couldnt find a way and i've done some source searching but to no evail. Is there anyway that i can obtain at least some of these?
Thanks,
Jeff
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu
|  |