search for: procsystim

Opening folders containing 10K files totaling 20GB is slow (takes 1.5 - 2.0 minutes) in FreeBSD 9.2 with Samba 4.1.11 on a system with 32GB RAM and a newer Xeon processor. Server is a standalone server that I'm using for testing. I am using nfsv4 acls. I used the procsystime dtrace script from here: http://www.brendangregg.com/DTrace/procsystime with the following syntax: procsystime -n smbd Top offenders were as follows: fcntl 64334950 extattr_list_file 88368438 stat 305050713 getdirentries 480578501...

...e resource controls shows no problems. # ./errinfo -cn oracle Sampling... Hit Ctrl-C to end. ^C EXEC SYSCALL ERR COUNT DESC oracle ioctl 12 1 Not enough core oracle semsys 11 4 Resource temporarily unavailable # ./procsystime -can oracle Hit Ctrl-C to stop sampling... ^C Elapsed Times for processes oracle, SYSCALL TIME (ns) sysconfig 4400 pset 7700 lwp_park 34600 times 39400 kill 4620...

...ference may be down to hardware specification but if that is the case, I wish to fully understand which aspect of the hardware (CPU, memory, disk etc.) is limiting the performance of the T5120. In order to try to explain this time difference, I the test on Solaris using the dtrace toolkit. Running procsystime gives the following results (I''ve condensed them to keep the email short): Elapsed Times: ? ? ? ? SYSCALL ? ? ? ? ?TIME (ns) ? ? ?... ? ? ? ? ? ?stat ? ? ? ? ? 18498184 ? ? ? ? fstat64 ? ? ? ? ? 18720808 ? ? ? ? ? ?open ? ? ? ? ? 21688773 ? ? ? ? ?open64 ? ? ? ? ? 41127563 ? ? ? ? ?sta...

I see Intel has released a new tool. Oh, it requires some patches to the kernel to record latency times. Good thing people don''t mind patching their kernels, eh? So who can write the equivalent latencytop.d the fastest? ;-) http://www.latencytop.org/ -- cburgess at qnx.com