similar to: Pfilestat vs. prstat

Folks, Perplexing question about load average display with prstat -Z Solaris 10 OS U4 (08/07) We have 4 zones with very different processes and workloads.. The prstat -Z command issued within each of the zones, correctly displays the number of processes and lwps, but the load average value looks exactly the same on all non-global zones..I mean all 3 values (1,5,15 load averages) are the same

I''m looking for a suggestion on a good way to hunt down the source of high context switching and mutex contentions... Is dtrace the way to go now, or should I stick with something like lockstat? Russ This is a 5 second interval for mpstat: CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl 16 0 0 1115 1241 206 9095 912 2420 7393 0 12105 68 25

Hi world, I have a 10-disk RAID-Z2 system with 4 GB of DDR2 RAM and a 3 GHz Core 2 Duo. It''s exporting ~280 filesystems over NFS to about half a dozen machines. Under some loads (in particular, any attempts to rsync between another machine and this one over SSH), the machine''s load average sometimes goes insane (27+), and it appears to all be in kernel-land (as nothing in

Hi I have a process which reads a socket using recvfrom. After sometime, for no apparent reason, recvfrom exits with an EINTR causing the sender to barf too. Basically the socket on which these processes were communicating no longer exists. On writing a dtrace, I could determine that a signal was received at that socket causing the EINTR. I do not know of way to see who sent the signal (which

Hello, Using one of Brendan''s dtrace script prusage and prstat, I was able to find the process that is consuming high cpu. I also have a test case that can trigger this high cpu usage. Questions are a) how do I proceed from here? b) how do I figure out which thread is using this high cpu? c) how do I figure out which piece of code is causing this problem? Thanks This message posted from

Hi, I'm attempting to use the UPS network tools. I'm sure I got the configurations worked out as I get notified when the ups goes on battery and back online. I am having problems with upssched. I have the following in my upssched.conf: CMDSCRIPT /usr/local/ups/sbin/upssched-cmd AT ONBATT * EXECUTE onbatt The problem I have is that is seems like the script in CMDSCRIPT never gets

when check java process lock statistics, plockstat failed, please see below: # prstat -mLp 21162 PID USERNAME USR SYS TRP TFL DFL LCK SLP LAT VCX ICX SCL SIG PROCESS/LWPID 21162 7677 0.9 0.1 0.0 0.0 0.0 99 0.0 0.3 83 89 215 0 java/81 21162 7677 0.3 0.1 0.0 0.0 0.0 0.0 99 0.2 106 33 305 0 java/35 21162 7677 0.1 0.0 0.0 0.0 0.0 100 0.0 0.1 79 6 85 0 java/59

hello! First of all thank you for all your information you give us on this sites. It is a great help to learn for example dtrace. i''ve a little script in work, which shows the cpu utilization per process in percent. this output is refreshed every 2 seconds for example. now i want to extend this script with the output of the allocated bytes of memory per process. how could i display this

Chris Lattner wrote: > Check out http://nondot.org/sabre/LLVMNotes > > %Size = getelementptr %T* null, int 1 %SizeI = cast %T* %Size to uint How incredibly cunning. :-) Thanks for that. Next stupid question. I've put together a simple coroutine/fibre style threading system on top of the Linux setcontext/getcontext stuff, which surprisingly enough seems to work *almost*

Hello, I am initializing FDR mode and finalizing/flushing the buffers manually. XRay does not log calls from the main thread unless there is a function call after __xray_log_finalize(). This behavior is abnormal since one would expect the trace file to contain all function calls made up to the point when __xray_log_finalize() is called. To demonstrate this behavior, I have taken the test case

Check out http://nondot.org/sabre/LLVMNotes -Chris http://nondot.org/sabre http://llvm.org On Jul 27, 2007, at 12:00 PM, Sarah Thompson <thompson at email.arc.nasa.gov > wrote: > Hi folks, > > Assuming that I'm writing a pass and that for bizarre reasons I need > to > programmatically do the equivalent of a C/C++ sizeof on a Value (or a > Type, it doesn't

Hi folks, Assuming that I'm writing a pass and that for bizarre reasons I need to programmatically do the equivalent of a C/C++ sizeof on a Value (or a Type, it doesn't matter which really), yielding a result in bytes, what is the known-safe way to do this? I notice that doing something like struct thingy { ... some stuff ... }; ... printf("Size = %d",

I have a funny issue whereby when a application exits the proc:::exit probe doesn''t seem to be firing. A customer has written a dtrace script that outputs some info whenever proc:::exit fires but on occasion (random, no pattern in behaviour) dtrace exits without outputting the data. They are running dtrace via ''dtrace -s exit_trace.d -p <pid>'' Can anyone think

Hi all, I am trying to write a D script which would print ustack() for every program in the system receiving SIGSEGV. All the stacks printed in trap()/sigtoproc() context do not have meaningful symbols. The following solves the problem to some degree but I''d much rather have a self-contained D script. dtrace -w -n ''fbt:genunix:sigtoproc:entry/arg2 == 11/ {

hi, Why xapian don't support memory index/search ? I know there is a method can create memory datebase, like this: Xapian::WritableDatabase db(Xapian::InMemory::open()); *But, if i use these in multithread program, i need create many datebases!!* Xapian::WritableDatabase db1(Xapian::InMemory::open()); //used in thread1 Xapian::WritableDatabase db2(Xapian::InMemory::open()); //used in

I have a SunRay server that I am looking at to determine some sizing requirements in my department. The machine has 16G of ram and 10G of swap. Currently, I have about 4G of swap used. I am wondering if dtrace/mdb can be used to find out what lwp/processes have been swapped out? Any hints? This message posted from opensolaris.org

I am looking for Dtrace scripts that can be used to collect data during performance tests. I am especially interested in IO but CPU, memory, threads, etc are needed as well. Thanks, Dave

I've finally got it to work, kind of .. It should allow you to log in, but several settings don't work yet and there are probably many bugs. Anyway, I thought I'd show this already in case someone is interested of looking how it works :) One thing I'm still wondering a bit is if I should show process names as: 32602 pts/19 S 0:00 dovecot/config 32603 pts/19 S

Hi all, I have exchanged my Dell R610 in favor of a Sun Fire 4170 M2 which has 32 GB RAM installed. I am running Sol11Expr on this host and I use it to primarily serve Netatalk AFP shares. From day one, I have noticed that the amount of free RAM decereased and along with that decrease the overall performance of ZFS decreased as well. Now, since I am still quite a Solaris newbie, I seem to

On Apr 24, 2014, at 1:33 AM, Dmitry Vyukov <dvyukov at google.com> wrote: >> >> I can see that the behavior of our current instrumentation is going to be a >> problem for the kinds of applications that you’re looking at. If you can >> find a way to get the overhead down without losing accuracy > > What are your requirements for accuracy? > Current