Hi All, I would like to thank all who answered and explain that sched is just waiting for the disk and should not be blame on. Especial thanks to Gary Paveza who payed my attention on high fpi which means an active reading from the file system. iosnoop from DtraceToolkit showed me a process which was reading a lot small files. Files were moved to CFS and the problem has been solved. Thanks again and best regards, es > > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Tue, 16 Mar 2010 18:08:11 +0100 > From: Eugenia Shuiskaya <redtigra@gmail.com> > Subject: heavy paging > To: sunmanagers@sunmanagers.org > Message-ID: > <5c0b76411003161008w7637b6f8r3dbbcb681d26e5f2@mail.gmail.com> > Content-Type: text/plain; charset="us-ascii" > > Hi All, > > We run into heavy I/O on our zfs disk and could not find the reason. > Solaris > 10 10/09. > > Here is zfs info: > > pool: rpool > state: ONLINE > status: The pool is formatted using an older on-disk format. The pool can > still be used, but some features are unavailable. > action: Upgrade the pool using 'zpool upgrade'. Once this is done, the > pool will no longer be accessible on older software versions. > scrub: none requested > config: > > NAME STATE READ WRITE CKSUM > rpool ONLINE 0 0 0 > mirror ONLINE 0 0 0 > c1t0d0s0 ONLINE 0 0 0 > c1t1d0s0 ONLINE 0 0 0 > > > dtrace script iotop shows: > > # iotop d sd0 Co ** > > UID PID PPID CMD DEVICE MAJ MIN D DISKTIME > > 0 0 0 sd0 32 0 60104 > > 0 0 0 sched sd0 32 0 W 4554988 > > > 2010 Mar 16 17:51:16, load: 3.25, disk_r: 0 KB, disk_w: 233648 KB > UID PID PPID CMD DEVICE MAJ MIN D DISKTIME > 0 0 0 sched sd0 32 0 W 4864154 > > > I run whoispaging.d ang got the result: > > > # ./whoispaging.d > Who's waiting for pagein (milliseconds): > Who's on cpu (milliseconds): > ScriptAgent 20 > Supervisor 24 > dtrace 28 > psSTAT64_10 33 > data 51 > dmh 65 > fsflush 178 > rdh 1999 > rlh 6515 > sched 217344 > > > I assumed that the virtual memory is low, but prstat shows I'm wrong: > #prstat -Z... > ZONEID NPROC SWAP RSS MEMORY TIME CPU ZONE > 0 188 40G 38G 59% 4:11:50 3.4% global > > # vmstat -p 3 > memory page executable anonymous > filesystem > swap free re mf fr de sr epi epo epf api apo apf fpi fpo > fpf > 68783448 43892160 235 226 22 0 3 0 0 0 0 0 0 1531 23 > 22 > 41939512 15828032 298 871 0 0 0 0 0 0 0 0 0 868 0 > 0 > 41902168 15797808 197 123 0 0 0 0 0 0 0 0 0 1312 0 > 0 > 41911104 15782904 320 3 0 0 0 0 0 0 0 0 0 2859 0 > 0 > 41910920 15759696 232 177 0 0 0 0 0 0 0 0 0 1007 0 > 0 > 41823336 15734752 212 6 0 0 0 0 0 0 0 0 0 2132 0 > 0 > 41779384 15712160 177 4 0 0 0 0 0 0 0 0 0 1697 0 > 0 > > A lot of free memory - and high fpi. It's paging a lot. > > And I cannot understand, how I could find the reason of such heavy paging. > Any advise, thought, opinion would be very appreciated. > > Thanks a lot and best regards, > Evgenia _______________________________________________ sunmanagers mailing list sunmanagers@sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagersReceived on Mon Mar 22 09:24:41 2010
This archive was generated by hypermail 2.1.8 : Thu Mar 03 2016 - 06:44:16 EST