jwboyer at gmail.com
Fri Jun 27 09:18:22 PDT 2008
On Fri, 2008-06-27 at 09:50 -0600, Matthew Wilcox wrote:
> On Fri, Jun 27, 2008 at 10:23:33AM -0500, James Bottomley wrote:
> > DTrace is more a piece of sun marketing coolaid which they use to beat
> > us up at every opportunity.
> Sure. And it's working, even amongst open source people who're supposed
> to be immune to marketing ;-)
> > We actually have a reasonably functional equivalent piece of technology
> > called systemtap.
> > When I go around end users, I find people in two camps: The ones who've
> > drunk the sun coolaid and won't take anything on linux that isn't a
> > fully replicated dtrace (sort of like windows people who demand the
> > availability of outlook on linux) and people who are migrating to Linux
> > and trying to use systemtap for tracing. These latter seem to have a
> > number of genuine concerns including latency, the time it takes to
> > actually go from command executing to functional trace, the inability to
> > trace user programs (dtrace can) and concerns about the amount of
> > perturbation the probes actually place inside the kernel.
> The ability to trace userspace programs is the key one for the postgres
> people. They get reports saying how long queries took to execute, which
> tables were accessed, how long locks (postgres locks, that is, not
> kernel ones) were held, cache hits, cache misses, etc. It's _really_
> useful for Postgres developers and admins. And they only get it on
> Solaris and MacOS, not Linux.
utrace (and uprobes) are supposed to enable that functionality in Linux
from my limited understanding.
More information about the Ksummit-2008-discuss