On 8/02/2013 10:00 AM, Greg Troxel wrote:
Carl Brewer <carl%bl.echidna.id.au@localhost> writes:# iostat tty cd0 wd0 wd1 CPU tin tout KB/t t/s MB/s KB/t t/s MB/s KB/t t/s MB/s us ni sy in id 0 67 0.000 0 0.000 63.90 25 1.542 52.36 31 1.600 0 0 4 25 72You can also use 'systat vmstat', which shows the same info, but I prefer it; with a window with more than 24 lines, it can show more disks. I would try to test the drives with dd, separate from raidframe. But, a 1TB drive, assuming 80 MB/s, should take about 3.6h, so someth9ing is off. Your iostat output shows 1.5 MB/s, which is not healthy. Also, you might check your SATA cables.
I've removed & reseated the SATA cables, no difference. systat vmstat shows :
Disks: wd0 wd1 raid0 1024 fmin seeks 1365 ftarg xfers 25 27 2 itarg bytes 1602K 1621K 24K 1506 wired %busy 95.7 98.3 pdfre top also shows :load averages: 0.15, 0.11, 0.06; up 0+00:07:38 10:38:25
24 processes: 21 sleeping, 3 on CPUCPU0 states: 0.0% user, 0.0% nice, 0.0% system, 100% interrupt, 0.0% idle CPU1 states: 0.0% user, 0.0% nice, 21.4% system, 0.0% interrupt, 78.6% idle CPU2 states: 0.0% user, 0.0% nice, 4.5% system, 0.0% interrupt, 95.5% idle CPU3 states: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle
Memory: 46M Act, 6024K Wired, 8440K Exec, 29M File, 3738M Free Swap: 4096M Total, 4096M FreeSomething's driving CPU0 to 100% interrupts. I'm guessing that's the raid reconstruct.
I think the drives are running at close to 100% "busy". I guess I should dd wd0 and see how long it takes? wd1 seems to be fine, it did a newfs in ~15 mins from memory on the raid0 array