> On Sat, Jan 30, 2010 at 3:59 AM, Joel Fernandes <agnel.joel@xxxxxxxxx> wrote: >> My question is, Is there a better way (better than logging) of >> tracking operations internal to the kernel continuously over long >> period of time in a way that doesn't affect latency (very >> drastically)? > > Maybe you also need to check Linux Trace Toolkit Next Generation > (ltt-ng). It's actually a tracing toolkit that delivers low latency > tracing (according to the its website). Once it uses relayfs, but not > sure what it uses today Thanks. will check it out, it seems its faster than systemtap and supports very large datasets. :) nice > > Anyway, do you need to track certain events and records i.e timestamp > for quite long time e.g 1 week or more? That's really massive IMO. You > really a fast logging and fast data transfer between kernel and user > space and very fast data writer too. Actually I will be dumping a few data structures over a long period of time. I just did that with ftrace and it didn't have much of an overhead. But yeah, the corruptions I've seen occur at over a 5 day period of regular use so its quite difficult. Thanks! Joel -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
