similar to: performance troubleshooting

HI, T5220, S10U4 + patches mdb -k > ::memstat While above is working (takes some time, ideally ::memstat -n 4 to use 4 threads could be useful) mpstat 1 shows: CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl 48 0 0 1922112 9 0 0 8 0 0 0 15254 6 94 0 0 So about 2mln xcalls per second. Let''s check with dtrace:

In crossbow, each mac_srs has a kernel thread called "mac_rx_srs_poll_ring" to poll the hardware and crossbow will wakeup this thread to poll packets from the hardware automatically. Does crossbow provide any method to disable the polling mechanism, for example disabling the this kernel thread? Thanks Zhihui -------------- next part -------------- An HTML attachment was scrubbed... URL:

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

I just had a quick play with gzip compression on a filesystem and the result was the machine grinding to a halt while copying some large (.wav) files to it from another filesystem in the same pool. The system became very unresponsive, taking several seconds to echo keystrokes. The box is a maxed out AMD QuadFX, so it should have plenty of grunt for this. Comments? Ian

Recently I found that the mac_srs_rx_poll_ring thread may never stop in kernel, please see the following mpstat, cpu 2 is in 100% kernel usage, but no syscalls and no interrupts. CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl 0 0 0 0 300 100 0 0 1 0 0 0 0 0 0 100 1 14 0 0 134 68 134 1

>From time to time a jboss process would end eating all, the available CPU and the load avg would skyrocket. Once the operators restarted jboss, the system''d be normal again (sometimes for weeks) until the next incident Since we moved the app from a v440 running Solaris 10 8/07 to a t2000 running Solaris 10 5/08 the problem started to happen more frequently (2-3 times a week). The

Hi. T2000 1.2GHz 8-core, 32GB RAM, S10U3, zil_disable=1. Command ''zpool export f3-2'' is hung for 30 minutes now and still is going. Nothing else is running on the server. I can see one CPU being 100% in SYS like: bash-3.00# mpstat 1 [...] CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl 0 0 0 67 220 110 20 0 0 0 0

So I have this dual 16-core Opteron Dell R715 with 128G of RAM attached to a SuperMicro disk enclosure with 45 2TB Toshiba SAS drives (via two LSI 9200 controllers and MPxIO) running OpenIndiana 151a4 and I''m occasionally seeing a storm of xcalls on one of the 32 VCPUs (>100000 xcalls a second). The machine is pretty much idle, only receiving a bunch of multicast video streams and

Hello ZFS, System was rebooted and after reboot server again System is snv_39, SPARC, T2000 bash-3.00# ptree 7 /lib/svc/bin/svc.startd -s 163 /sbin/sh /lib/svc/method/fs-local 254 /usr/sbin/zfs mount -a [...] bash-3.00# zfs list|wc -l 46 Using df I can see most file systems are already mounted. > ::ps!grep zfs R 254 163 7 7 0 0x4a004000

Does anyone has any ideas about this problem? 2008/1/15, ?? TaoJie <eulertao at gmail.com>: > > Hi all: > > I''m working on revealing system performance now. > My testing program is an infinite loop. Inside the loop, it will do some > mathematical opertaions and call function callee(), then go to the next > loop. > I install a alarm(30) in the program. It

Good morning, We have been noticing troubles browsing on a ZFS share, especially in the afternoon, and found our 8 cores going at 100% with over 100000 smtx running on each core on mpstat. We are running Solaris 5.11 with Samba 3.5.10, 48 GB of RAM and two 4 core Xeons. The fileserver is attached by domain mode to Windows 2003 R2 SP2 with Services for Unix installed and we only have around 80

Hi, I have a mysql instance which if I point more load towards it it suddenly gets 100% in SYS as shown below. It can work fine for an hour but eventually it gets to a jump from 5-15% of CPU utilization to 100% in SYS as show in mpstat output below: # prtdiag | head System Configuration: SUN MICROSYSTEMS SUN FIRE X4170 SERVER BIOS Configuration: American Megatrends Inc. 07060215

Hello, Sorry for the (very) long subject but I''ve pinpointed the problem to this exact situation. I know about the other threads related to hangs, but in my case there was no < zfs destroy > involved, nor any compression or deduplication. To make a long story short, when - a disk contains 2 partitions (p1=32GB, p2=1800 GB) and - p1 is used as part of a zfs mirror of rpool

Dtrace and Performance Teams, I have the following IO Performance Specific Questions (and I''m already savy with the lockstat and pre-dtrace utilities for performance analysis.. but in need of details regarding specifying IO bottlenecks @ the controller or IO bus..) : **Q.A*> Determining IO Saturation bottlenecks ( */.. beyond service times and kernel contention.. )/ I''m

I figured the following ZFS ''success story'' may interest some readers here. I was interested to see how much sequential read/write performance it would be possible to obtain from ZFS running on commodity hardware with modern features such as PCI-E busses, SATA disks, well-designed SATA controllers (AHCI, SiI3132/SiI3124). So I made this experiment of building a fileserver by

Dear Spencer and Prof. Fox, Thank you for your replies. I'll very appreciate, if you have any ideas concerning the problem described below. First, I'd like to describe the model in brief. In general I consider a model with three equations. First one is for annual GRP growth - in general it looks like: 1) GRP growth per capita = G(investment, migration, initial GRP per

Holding the vblk->lock across kick causes poor scalability in SMP guests. If one CPU is doing virtqueue kick and another CPU touches the vblk->lock it will have to spin until virtqueue kick completes. This patch reduces system% CPU utilization in SMP guests that are running multithreaded I/O-bound workloads. The improvements are small but show as iops and SMP are increased. Khoa Huynh

Holding the vblk->lock across kick causes poor scalability in SMP guests. If one CPU is doing virtqueue kick and another CPU touches the vblk->lock it will have to spin until virtqueue kick completes. This patch reduces system% CPU utilization in SMP guests that are running multithreaded I/O-bound workloads. The improvements are small but show as iops and SMP are increased. Khoa Huynh

Hi, Jens & Rusty This version is rebased against linux-next which resolves the conflict with Paolo Bonzini's 'virtio-blk: allow toggling host cache between writeback and writethrough' patch. Patch 1/3 and 2/3 applies on linus's master as well. Since Rusty will pick up patch 3/3 so the changes to block core (adding blk_bio_map_sg()) will have a user. Jens, could you please

Hi, Jens & Rusty This version is rebased against linux-next which resolves the conflict with Paolo Bonzini's 'virtio-blk: allow toggling host cache between writeback and writethrough' patch. Patch 1/3 and 2/3 applies on linus's master as well. Since Rusty will pick up patch 3/3 so the changes to block core (adding blk_bio_map_sg()) will have a user. Jens, could you please