Greg Ames wrote: > > Brian Pane wrote: > > One more thought: I graphed the CPU utilization from your vmstat > > output, and the spike is mostly sys time, not usr. So truss/strace > > data may be helpful--especially if truss on that platform can measure > > the time spent in each syscall. > > truss doesn't measure time on FreeBSD according to the man page. But > maybe there are a lot more syscalls in the failing case, so I'll try it.
~gregames/2.0.30.truss and ~gregames/2_0_28.truss are available on daedalus, if you're interested. The former was created by trussing a process while running log replay against the 2.0.30 server on port 8092; the latter was a process on the production server. There could be some differences in 2.0.30 due to my test environment: it listens on two ports (adds a poll() before the accept() ), and I was running thru a firewall which seems to introduce some network errors. Greg