Re: zpool - low speed write
Hi! I'm not sure if the problem was solved, so... On Thu, 5 Aug 2010 18:49:14 +0800 Alex V. Petrov wrote: > smartctl -a /dev/ad8 > Device Model: WDC WD10EADS-00M2B0 > Firmware Version: 01.00A01 > 193 Load_Cycle_Count0x0032 184 184 000Old_age Always > > - 49237 > smartctl -a /dev/ad10 > Device Model: WDC WD10EADS-00L5B1 > Firmware Version: 01.01A01 > 193 Load_Cycle_Count0x0032 200 200 000Old_age Always > > - 24 > smartctl -a /dev/ad12 > Device Model: WDC WD10EADS-00M2B0 > Firmware Version: 01.00A01 > 193 Load_Cycle_Count0x0032 200 200 000Old_age Always > > - 91 >From the above info I'd say that you may try to play with load cycles (set a bigger delay, etc.) with /dev/ad8. -- WBR, Boris Samorodov (bsam) Research Engineer, http://www.ipt.ru Telephone & Internet SP FreeBSD Committer, http://www.FreeBSD.org The Power To Serve ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
> Can you please remove use of the zpool entirely (e.g. zpool destroy > tank) and do a write test to each disk itself? E.g.: > > dd if=/dev/zero of=/dev/ad8 bs=64k count=100 > dd if=/dev/zero of=/dev/ad10 bs=64k count=100 > dd if=/dev/zero of=/dev/ad12 bs=64k count=100 > > Thanks. zpool destroy tank At one time 3 dd dd if=/dev/zero of=/dev/ad8 bs=64k count=100 100+0 records in 100+0 records out 6553600 bytes transferred in 605.173955 secs (108292830 bytes/sec) dd if=/dev/zero of=/dev/ad10 bs=64k count=100 100+0 records in 100+0 records out 6553600 bytes transferred in 759.946393 secs (86237662 bytes/sec) dd if=/dev/zero of=/dev/ad12 bs=64k count=100 100+0 records in 100+0 records out 6553600 bytes transferred in 605.139062 secs (108299074 bytes/sec) Disks ad4 ad6 ad8 ad10 ad12 da0 da1 KB/t 0,00 16,00 64,00 64,00 64,00 0,00 0,00 tps 0 1 1667 1329 1646 0 0 MB/s 0,00 0,02 104 83,04 103 0,00 0,00 %busy 0 0939493 0 0 ie this problem is not the controller - Alex V. Petrov ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
> > If only one of the dds shows bad throughput, then please: > > - Install ports/sysutils/smartmontools and run smartctl -a /dev/XXX, > where XXX is the disk which has bad throughput > - Try making a ZFS pool with all 3 disks, but then do "zpool offline > tank XXX" and then re-attempt the following dd: > dd if=/dev/zero of=/tank/test.zero bs=64k count=100 > And see what throughput looks like. > zpool destroy tank zpool create tank ad8 ad10 ad12 dd if=/dev/zero of=/tank/test.zero bs=3M count=1000 1000+0 records in 1000+0 records out 3145728000 bytes transferred in 83.219741 secs (37800262 bytes/sec) zpool offline tank ad10 cannot offline ad10: no valid replicas zpool destroy tank zpool create tank ad8 ad12 dd if=/dev/zero of=/tank/test.zero bs=3M count=1000 1000+0 records in 1000+0 records out 3145728000 bytes transferred in 126.873102 secs (24794286 bytes/sec) dd if=/dev/zero of=/tank/test.zero bs=64k count=100 100+0 records in 100+0 records out 6553600 bytes transferred in 1735.680273 secs (37758106 bytes/sec) zpool destroy tank zpool create tank ad8 dd if=/dev/zero of=/tank/test.zero bs=3M count=1000 1000+0 records in 1000+0 records out 3145728000 bytes transferred in 39.550739 secs (79536517 bytes/sec) dd if=/dev/zero of=/tank/test.zero bs=64k count=10 10+0 records in 10+0 records out 655360 bytes transferred in 90.810344 secs (72167990 bytes/sec) =-O - Alex V. Petrov ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
В сообщении от 5 августа 2010 14:19:59 вы написали: > > Can you please remove use of the zpool entirely (e.g. zpool destroy > tank) and do a write test to each disk itself? E.g.: > > dd if=/dev/zero of=/dev/ad8 bs=64k count=100 > dd if=/dev/zero of=/dev/ad10 bs=64k count=100 > dd if=/dev/zero of=/dev/ad12 bs=64k count=100 > > I don't recommend using large block sizes (e.g. bs=1M, bs=3M). dd if=/dev/zero of=/dev/ad8 bs=64k count=100 100+0 records in 100+0 records out 6553600 bytes transferred in 604.849406 secs (108350937 bytes/sec) dd if=/dev/zero of=/dev/ad10 bs=64k count=100 100+0 records in 100+0 records out 6553600 bytes transferred in 757.755459 secs (86487005 bytes/sec) dd if=/dev/zero of=/dev/ad12 bs=64k count=100 100+0 records in 100+0 records out 6553600 bytes transferred in 604.857282 secs (108349526 bytes/sec) > If all of the above dds show good/decent throughput, then there's > something strange going on with ZFS. If this is the case, I would > recommend filing a PR and posting to freebsd-fs about the problem, > pointing folks to this thread. > > If all of the dds show bad throughput, then could you please do the > following: > > - Provide vmstat -i output > - Install ports/sysutils/smartmontools and run smartctl -a /dev/ad8, > smartctl -a /dev/ad10, and smartctl -a /dev/ad12 > > If only one of the dds shows bad throughput, then please: > > - Install ports/sysutils/smartmontools and run smartctl -a /dev/XXX, > where XXX is the disk which has bad throughput > - Try making a ZFS pool with all 3 disks, but then do "zpool offline > tank XXX" and then re-attempt the following dd: > dd if=/dev/zero of=/tank/test.zero bs=64k count=100 > And see what throughput looks like. > > Thanks. - Alex V. Petrov ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
В сообщении от 8 августа 2010 22:00:57 автор Vladislav V. Prodan написал: > 05.08.2010 6:43, Alex V. Petrov wrote: > > Intel® ICH10 > > motherboard Gigabyte GA-EP43-DS3 (rev. 1.0) P43 / Socket 775 > > CPU: Intel(R) Core(TM)2 Quad CPU Q8200 @ 2.33GHz (2335.41-MHz > > K8-class CPU) > > Please, show output: > atacontrol mode ada2 > atacontrol mode ada3 > atacontrol mode ada4 > > Install /usr/ports/sysutils/smartmontools and show output: > smartctl -x /dev/ada2 > smartctl -x /dev/ada3 > smartctl -x /dev/ada4 > > And run the Western Digital Data Lifeguard Diagnostic and check all the > HDD(screws :)) . > ___ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org" System without ahci now. atacontrol mode ad8 current mode = UDMA100 SATA 3Gb/s atacontrol mode ad10 current mode = UDMA100 SATA 3Gb/s atacontrol mode ad12 current mode = UDMA100 SATA 3Gb/s smartctl -x /dev/ad8 smartctl 5.39.1 2010-01-28 r3054 [FreeBSD 8.1-STABLE amd64] (local build) Copyright (C) 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net === START OF INFORMATION SECTION === Model Family: Western Digital Caviar Green family Device Model: WDC WD10EADS-00M2B0 Serial Number:WD-WCAV51709425 Firmware Version: 01.00A01 User Capacity:1 000 204 886 016 bytes Device is:In smartctl database [for details use: -P show] ATA Version is: 8 ATA Standard is: Exact ATA specification draft version not indicated Local Time is:Sun Aug 8 23:35:53 2010 KRAST SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x84) Offline data collection activity was suspended by an interrupting command from host. Auto Offline Data Collection: Enabled. Self-test execution status: ( 25) The self-test routine was aborted by the host. Total time to complete Offline data collection: (20400) seconds. Offline data collection capabilities:(0x7b) SMART execute Offline immediate. Auto Offline data collection on/off support. Suspend Offline collection upon new command. Offline surface scan supported. Self-test supported. Conveyance Self-test supported. Selective Self-test supported. SMART capabilities:(0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability:(0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time:( 2) minutes. Extended self-test routine recommended polling time:( 235) minutes. Conveyance self-test routine recommended polling time:( 5) minutes. SCT capabilities: (0x303f) SCT Status supported. SCT Feature Control supported. SCT Data Table supported. SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 200 200 051Pre-fail Always - 0 3 Spin_Up_Time0x0027 122 109 021Pre-fail Always - 6875 4 Start_Stop_Count0x0032 100 100 000Old_age Always - 72 5 Reallocated_Sector_Ct 0x0033 200 200 140Pre-fail Always - 0 7 Seek_Error_Rate 0x002e 200 200 000Old_age Always - 0 9 Power_On_Hours 0x0032 093 093 000Old_age Always - 5636 10 Spin_Retry_Count0x0032 100 253 000Old_age Always - 0 11 Calibration_Retry_Count 0x0032 100 253 000Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000Old_age Always - 41 192 Power-Off_Retract_Count 0x0032 200 200 000Old_age Always - 35 193 Load_Cycle_Count0x0032 184 184 000Old_age Always - 49237 194 Temperature_Celsius 0x0022 108 104 000Old_age Always - 39 196 Reallocated_Event_Count 0x0032 200 20
Re: zpool - low speed write
05.08.2010 6:43, Alex V. Petrov wrote: > Intel® ICH10 > motherboard Gigabyte GA-EP43-DS3 (rev. 1.0) P43 / Socket 775 > CPU: Intel(R) Core(TM)2 Quad CPU Q8200 @ 2.33GHz (2335.41-MHz K8-class > CPU) > Please, show output: atacontrol mode ada2 atacontrol mode ada3 atacontrol mode ada4 Install /usr/ports/sysutils/smartmontools and show output: smartctl -x /dev/ada2 smartctl -x /dev/ada3 smartctl -x /dev/ada4 And run the Western Digital Data Lifeguard Diagnostic and check all the HDD(screws :)) . ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
08.08.2010 8:57, Alex V. Petrov wrote: > I think these settings are relevant only for i386, but not for amd64 These settings are just for amd64! They guarantee maximum performance ZFS. On my system amd64 + ZFS speed read-write into ZFS on 5-20% different from the physical read-write speed on single discs. ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
On Sun, Aug 08, 2010 at 01:57:09PM +0800, Alex V. Petrov wrote: > В сообщении от 8 августа 2010 12:13:29 автор Vladislav V. Prodan написал: > > 08.08.2010 6:07, Alex V. Petrov wrote: > > > sysctl -a | grep vm.kmem > > > vm.kmem_size_max: 329853485875 > > > > > > sysctl -a | grep vfs.zfs.arc > > > vfs.zfs.arc_meta_used: 13797096 > > > vfs.zfs.arc_max: 858721280 > > > > > > > > > sysctl -a | grep vm.kvm > > > vm.kvm_free: 547673337856 > > > vm.kvm_size: 549755809792 > > > > Please, insert into /boot/loader.conf only this options: > > > > vm.kmem_size="999M" > > vm.kmem_size_max="999M" > > vfs.zfs.arc_max="160M" > > > > And after reboot, please run dd... > > I think these settings are relevant only for i386, but not for amd64 They're relevant to both, I can assure you. -- | Jeremy Chadwick j...@parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB | ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
В сообщении от 8 августа 2010 12:13:29 автор Vladislav V. Prodan написал: > 08.08.2010 6:07, Alex V. Petrov wrote: > > sysctl -a | grep vm.kmem > > vm.kmem_size_max: 329853485875 > > > > sysctl -a | grep vfs.zfs.arc > > vfs.zfs.arc_meta_used: 13797096 > > vfs.zfs.arc_max: 858721280 > > > > > > sysctl -a | grep vm.kvm > > vm.kvm_free: 547673337856 > > vm.kvm_size: 549755809792 > > Please, insert into /boot/loader.conf only this options: > > vm.kmem_size="999M" > vm.kmem_size_max="999M" > vfs.zfs.arc_max="160M" > > And after reboot, please run dd... I think these settings are relevant only for i386, but not for amd64 dd if=/dev/zero of=/tank/test.zero bs=3M count=1000 1000+0 records in 1000+0 records out 3145728000 bytes transferred in 425.627495 secs (7390801 bytes/sec) zpool iostat -v 10 10 capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank 759G 1,98T 0 43 31,7K 4,37M ad12 230G 701G 0 15 10,8K 1,55M ad8244G 684G 0 15 11,6K 1,55M ad10 286G 642G 0 13 9,33K 1,27M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank 759G 1,98T 0 80 0 3,26M ad12 230G 701G 0 27 0 1,24M ad8244G 684G 0 27 0 1,08M ad10 286G 642G 0 25 0 962K -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank 759G 1,98T 0100 6,39K 3,31M ad12 230G 701G 0 34 0 1,23M ad8244G 684G 0 33 0 1,13M ad10 286G 642G 0 33 6,39K 970K -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank 760G 1,98T 0132 0 7,21M ad12 230G 701G 0 41 0 2,23M ad8244G 684G 0 47 0 2,79M ad10 286G 642G 0 43 0 2,20M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank 760G 1,98T 0110 6,39K 5,19M ad12 230G 701G 0 39 0 2,12M ad8244G 684G 0 36 6,39K 1,67M ad10 286G 642G 0 33 0 1,40M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank 760G 1,98T 0100 6,39K 4,81M ad12 230G 701G 0 33 0 1,79M ad8244G 684G 0 34 0 1,65M ad10 286G 642G 0 32 6,39K 1,37M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank 760G 1,98T 0117 0 6,21M ad12 230G 701G 0 41 0 2,23M ad8244G 684G 0 39 0 2,14M ad10 286G 642G 0 37 0 1,84M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank 760G 1,98T 0113 0 5,87M ad12 230G 701G 0 36 0 1,77M ad8244G 684G 0 40 0 2,19M ad10 286G 642G 0 37 0 1,90M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank 760G 1,98T 0102 12,8K 5,19M ad12 230G 701G 0 37 6,39K 2,02M ad8244G 684G 0 34 0 1,84M ad10 286G 642G 0 30 6,39K 1,32M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank 760G 1,98T 0124 0 6,32M ad12 230G 701G 0 43 0 2,31M ad8244G 684G 0 41 0 2,16M ad10 286G 642G 0 39 0 1,85M -
Re: zpool - low speed write
08.08.2010 6:07, Alex V. Petrov wrote: > sysctl -a | grep vm.kmem > vm.kmem_size_max: 329853485875 > > sysctl -a | grep vfs.zfs.arc > vfs.zfs.arc_meta_used: 13797096 > vfs.zfs.arc_max: 858721280 > > sysctl -a | grep vm.kvm > vm.kvm_free: 547673337856 > vm.kvm_size: 549755809792 > > Please, insert into /boot/loader.conf only this options: vm.kmem_size="999M" vm.kmem_size_max="999M" vfs.zfs.arc_max="160M" And after reboot, please run dd... ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
В сообщении от 8 августа 2010 07:01:02 автор Vladislav V. Prodan написал: > 04.08.2010 15:08, Alex V. Petrov пишет: > > Hi All! > > > > $ dd if=/dev/random of=/tank/test bs=3M count=1000 > > 1000+0 records in > > 1000+0 records out > > 3145728000 bytes transferred in 298.153293 secs (10550707 bytes/sec) > > > > > > Any ideas? > > Please, show output: > vmstat -z > netstat -m > sysctl -a | grep vm.kmem > sysctl -a | grep vfs.zfs.arc > sysctl -a | grep kern.maxvnodes > sysctl -a | grep vm.kvm vmstat -z ITEM SIZE LIMIT USED FREE REQUESTS FAILURES UMA Kegs: 208,0, 191, 13, 191,0 UMA Zones:320,0, 191,1, 191,0 UMA Slabs:568,0, 5146, 1301, 24288196,0 UMA RCntSlabs:568,0, 1631, 1015, 329163,0 UMA Hash: 256,0, 79, 11, 82,0 16 Bucket:152,0, 16, 109, 167,0 32 Bucket:280,0, 30, 82, 274,2 64 Bucket:536,0, 49, 63, 793, 89 128 Bucket: 1048,0, 514, 95, 187117, 123359 VM OBJECT:216,0,34974, 213300, 413596412,0 MAP: 232,0,7, 25,7,0 KMAP ENTRY: 120, 150474, 183, 7040, 66175818,0 MAP ENTRY:120,0,24333,10759, 1025909468,0 DP fakepg:120,0, 216, 187, 544,0 SG fakepg:120,0, 667, 1689, 6212,0 mt_zone: 2056,0, 295, 54, 295,0 16:16,0, 3916,13724, 325519968,0 32:32,0, 3598, 1654, 65243323,0 64:64,0,16671,29529, 905000121,0 128: 128,0,12097,30011, 473239749,0 256: 256,0, 3300, 4650, 793867992,0 512: 512,0, 5211, 7690, 54770551,0 1024:1024,0, 239, 697, 5912498,0 2048:2048,0, 1335, 721, 35819870,0 4096:4096,0, 715, 617, 7671562,0 Files: 80,0, 2258, 1162, 323524959,0 TURNSTILE:136,0, 1673, 67, 1677,0 umtx pi: 96,0,0,0,0,0 MAC labels:40,0,0,0,0,0 PROC:1120,0, 187, 824, 6336408,0 THREAD: 984,0, 1236, 436,20593,0 SLEEPQUEUE:80,0, 1673, 67, 1677,0 VMSPACE: 392,0, 159, 831, 6334753,0 cpuset:72,0,2, 98,2,0 audit_record: 952,0,0,0,0,0 mbuf_packet: 256,0, 371, 525, 348284916,0 mbuf: 256,0, 231, 1888, 2247652340,0 mbuf_cluster:2048,25600, 896, 804, 188084,0 mbuf_jumbo_page: 4096,12800, 27, 754, 593845769,0 mbuf_jumbo_9k: 9216, 6400,0,0,0,0 mbuf_jumbo_16k: 16384, 3200,0,0,0,0 mbuf_ext_refcnt:4,0,0, 672, 774185,0 g_bio:232,0, 20, 1644, 54750899,0 ttyinq: 160,0, 330, 126, 1275,0 ttyoutq: 256,0, 168, 87, 632,0 ata_request: 320,0, 10, 782, 21908538,0 ata_composite:336,0,0,0,0,0 nv_stack_t: 12288,0, 10, 10, 22,0 VNODE:472,0,32144,63544, 25299060,0 VNODEPOLL:112,0, 63, 135, 65,0 S VFS Cache: 108,0,23801, 111928, 38399522,0 L VFS Cache: 328,0, 4752, 1368, 2652656,0 NAMEI: 1024,0,0, 764, 1639822943,0 NFSMOUNT: 616,0,0,0,0,0 NFSNODE: 656,0,0,0,0,
Re: zpool - low speed write
04.08.2010 15:08, Alex V. Petrov пишет: > Hi All! > > $ dd if=/dev/random of=/tank/test bs=3M count=1000 > 1000+0 records in > 1000+0 records out > 3145728000 bytes transferred in 298.153293 secs (10550707 bytes/sec) > > > Any ideas? Please, show output: vmstat -z netstat -m sysctl -a | grep vm.kmem sysctl -a | grep vfs.zfs.arc sysctl -a | grep kern.maxvnodes sysctl -a | grep vm.kvm ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
On Sat, Aug 07, 2010 at 01:51:21PM +0200, Ivan Voras wrote: > On 5.8.2010 6:47, Alex V. Petrov wrote: > > > camcontrol identify ada2 > > pass2: ATA-8 SATA 2.x device > > Aren't those 4k sector drives? no, 4k drives have 'R' in type (ie. EARS and AARS models) > To verify this hypotesis though, you will have to destroy the zpool, use > gnop to create a virtual 4k sector drive for each physical drive and try > testing everything again, using these new virtual drives. > > Unfortunately, if this is the case, it will be troublesome to find a > production solution just yet. I have an idea but no time to try it. Buki pgpGeiBg7dh4Q.pgp Description: PGP signature
Re: zpool - low speed write
On Sat, Aug 7, 2010 at 4:51 AM, Ivan Voras wrote: > On 5.8.2010 6:47, Alex V. Petrov wrote: > >> camcontrol identify ada2 >> pass2: ATA-8 SATA 2.x device > > Aren't those 4k sector drives? EADS drives use regular 512-byte sectors AFAIK. It's EA*R*S models that use 4K sectors. --Artem ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
On 5.8.2010 6:47, Alex V. Petrov wrote: > camcontrol identify ada2 > pass2: ATA-8 SATA 2.x device Aren't those 4k sector drives? To verify this hypotesis though, you will have to destroy the zpool, use gnop to create a virtual 4k sector drive for each physical drive and try testing everything again, using these new virtual drives. Unfortunately, if this is the case, it will be troublesome to find a production solution just yet. I have an idea but no time to try it. ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
On Wed, Aug 4, 2010 at 9:47 PM, Alex V. Petrov wrote: ... >> > vfs.zfs.cache_flush_disable=1 >> > vfs.zfs.zil_disable=1 >> >> I question both of these settings, especially the latter. Please remove >> them both and re-test your write performance. > > I removed all settings of zfs. > Now it default. > ZFS would throttle writes if it thinks that not enough memory is available. Did you by any chance tinker with VM parameters, too? Could you post output of following commands? sysctl vm |grep kmem sysctl vfs.zfs sysctl kstat.zfs (before and after after you do some of your write speed tests) --Artem ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
> > Can you please remove use of the zpool entirely (e.g. zpool destroy > tank) and do a write test to each disk itself? E.g.: > > dd if=/dev/zero of=/dev/ad8 bs=64k count=100 > dd if=/dev/zero of=/dev/ad10 bs=64k count=100 > dd if=/dev/zero of=/dev/ad12 bs=64k count=100 I don't have free space for move my data from zpool (1,91T) > I don't recommend using large block sizes (e.g. bs=1M, bs=3M). dd if=/dev/zero of=/tank/test.zero bs=64k count=1 1+0 records in 1+0 records out 65536 bytes transferred in 50.294832 secs (13030365 bytes/sec) > If all of the above dds show good/decent throughput, then there's > something strange going on with ZFS. If this is the case, I would > recommend filing a PR and posting to freebsd-fs about the problem, > pointing folks to this thread. > > If all of the dds show bad throughput, then could you please do the > following: > > - Provide vmstat -i output vmstat -i output interrupt total rate irq1: atkbd02368 0 irq6: fdc017 0 irq16: vgapci0 ath+ 1728264100 irq18: uhci2 ehci0* 2183829127 irq19: uhci4+ 427434 24 irq21: uhci1 42295 2 irq23: uhci3 ehci1 18154 1 cpu0: timer 34317326 1997 irq256: hdac01561005 90 irq257: re0 2458465143 cpu1: timer 34316042 1997 cpu3: timer 34316081 1997 cpu2: timer 34316130 1997 Total 145687410 8482 > - Install ports/sysutils/smartmontools and run smartctl -a /dev/ad8, > smartctl -a /dev/ad10, and smartctl -a /dev/ad12 In the first message I wrote that smatmontools installed smartd daily output: Checking health of /dev/ada2: OK Checking health of /dev/ada3: OK Checking health of /dev/ada4: OK In the logs there are no any error messages that the controller and drives smartctl -a /dev/ad8 smartctl 5.39.1 2010-01-28 r3054 [FreeBSD 8.1-STABLE amd64] (local build) Copyright (C) 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net === START OF INFORMATION SECTION === Model Family: Western Digital Caviar Green family Device Model: WDC WD10EADS-00M2B0 Serial Number:WD-WCAV51709425 Firmware Version: 01.00A01 User Capacity:1 000 204 886 016 bytes Device is:In smartctl database [for details use: -P show] ATA Version is: 8 ATA Standard is: Exact ATA specification draft version not indicated Local Time is:Thu Aug 5 18:42:22 2010 KRAST SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x82) Offline data collection activity was completed without error. Auto Offline Data Collection: Enabled. Self-test execution status: ( 25) The self-test routine was aborted by the host. Total time to complete Offline data collection: (20400) seconds. Offline data collection capabilities:(0x7b) SMART execute Offline immediate. Auto Offline data collection on/off support. Suspend Offline collection upon new command. Offline surface scan supported. Self-test supported. Conveyance Self-test supported. Selective Self-test supported. SMART capabilities:(0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability:(0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time:( 2) minutes. Extended self-test routine recommended polling time:( 235) minutes. Conveyance self-test routine recommended polling time:( 5) minutes. SCT capabilities: (0x303f) SCT Status supported. SCT Feature Control supported. SCT Data Table supported. SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 200 200 051Pre-fail Always -
Re: zpool - low speed write
On Thu, Aug 05, 2010 at 02:09:57PM +0800, Alex V. Petrov wrote: > В сообщении от 5 августа 2010 13:35:04 вы написали: > > Write performance here is abysmal, agreed. This is very odd. > > > > I hate to say this, but can you remove ahci.ko (ahci_load="yes") from > > your loader.conf and reboot? You may need to change filesystem names > > around in /etc/fstab for your OS disk (assuming it's on ada0), but for > > ZFS it should just magically find the disks on adXX. > > > > If you could also provide pciconf -lvc output that would be helpful. > > Thanks. > > dd if=/dev/zero of=/tank/test.zero bs=3M count=1000 > 1000+0 records in > 1000+0 records out > 3145728000 bytes transferred in 485.431690 secs (6480269 bytes/sec) Can you please remove use of the zpool entirely (e.g. zpool destroy tank) and do a write test to each disk itself? E.g.: dd if=/dev/zero of=/dev/ad8 bs=64k count=100 dd if=/dev/zero of=/dev/ad10 bs=64k count=100 dd if=/dev/zero of=/dev/ad12 bs=64k count=100 I don't recommend using large block sizes (e.g. bs=1M, bs=3M). If all of the above dds show good/decent throughput, then there's something strange going on with ZFS. If this is the case, I would recommend filing a PR and posting to freebsd-fs about the problem, pointing folks to this thread. If all of the dds show bad throughput, then could you please do the following: - Provide vmstat -i output - Install ports/sysutils/smartmontools and run smartctl -a /dev/ad8, smartctl -a /dev/ad10, and smartctl -a /dev/ad12 If only one of the dds shows bad throughput, then please: - Install ports/sysutils/smartmontools and run smartctl -a /dev/XXX, where XXX is the disk which has bad throughput - Try making a ZFS pool with all 3 disks, but then do "zpool offline tank XXX" and then re-attempt the following dd: dd if=/dev/zero of=/tank/test.zero bs=64k count=100 And see what throughput looks like. Thanks. -- | Jeremy Chadwick j...@parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB | ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
В сообщении от 5 августа 2010 13:35:04 вы написали: > Write performance here is abysmal, agreed. This is very odd. > > I hate to say this, but can you remove ahci.ko (ahci_load="yes") from > your loader.conf and reboot? You may need to change filesystem names > around in /etc/fstab for your OS disk (assuming it's on ada0), but for > ZFS it should just magically find the disks on adXX. > > If you could also provide pciconf -lvc output that would be helpful. > Thanks. dd if=/dev/zero of=/tank/test.zero bs=3M count=1000 1000+0 records in 1000+0 records out 3145728000 bytes transferred in 485.431690 secs (6480269 bytes/sec) zpool iostat -v 10 10 capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 829G 0 64 49,7K 4,73M ad12 598G 333G 0 22 19,9K 1,71M ad8633G 295G 0 21 13,8K 1,65M ad10 727G 201G 0 19 15,9K 1,36M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 829G 0116 0 5,91M ad12 598G 333G 0 39 0 2,20M ad8633G 295G 0 39 0 2,02M ad10 727G 201G 0 37 0 1,68M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 829G 0140 0 9,09M ad12 598G 333G 0 51 0 3,69M ad8633G 295G 0 45 0 2,94M ad10 727G 201G 0 43 0 2,46M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 829G 0133 0 7,66M ad12 598G 333G 0 46 0 2,84M ad8633G 295G 0 44 0 2,59M ad10 727G 201G 0 43 0 2,23M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 829G 0133 6,39K 5,84M ad12 598G 333G 0 47 6,39K 2,34M ad8633G 295G 0 43 0 1,83M ad10 727G 201G 0 42 0 1,67M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 829G 0113 0 5,54M ad12 598G 333G 0 39 0 1,97M ad8633G 295G 0 37 0 1,98M ad10 727G 201G 0 35 0 1,59M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 829G 0152 0 10,1M ad12 598G 333G 0 52 0 3,41M ad8633G 295G 0 52 0 3,65M ad10 727G 201G 0 47 0 3,06M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 828G 0116 0 5,61M ad12 598G 333G 0 41 0 2,16M ad8633G 295G 0 40 0 1,95M ad10 727G 201G 0 34 0 1,50M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 828G 0176 0 11,1M ad12 598G 333G 0 60 0 3,78M ad8634G 294G 0 60 0 3,95M ad10 727G 201G 0 55 0 3,35M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 828G 0112 0 7,55M ad12 598G 333G 0 39 0 2,73M ad8634G 294G 0 39 0 2,66M ad10 727G 201G 0 33 0 2,15M -- - - - - - - pciconf -lvc hos...@pci0:0:0:0: class=0x06 card=0x50001458 chip=0x2e208086 rev=0x02 hdr=0x00
Re: zpool - low speed write
On Thu, Aug 05, 2010 at 12:47:42PM +0800, Alex V. Petrov wrote: > > Your ada3 disk is different from the other two. Can you please provide > > the output from the following 3 commands? > > > > camcontrol identify ada2 > > camcontrol identify ada3 > > camcontrol identify ada4 > > > > > vfs.zfs.cache_flush_disable=1 > > > vfs.zfs.zil_disable=1 > > > > I question both of these settings, especially the latter. Please remove > > them both and re-test your write performance. > > I removed all settings of zfs. > Now it default. > > camcontrol identify ada2 > pass2: ATA-8 SATA 2.x device > pass2: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) > > protocol ATA/ATAPI-8 SATA 2.x > device model WDC WD10EADS-00M2B0 > firmware revision 01.00A01 > serial number WD-WCAV51709425 > WWN 50014ee2adf88aae > cylinders 16383 > heads 16 > sectors/track 63 > sector size logical 512, physical 512, offset 0 > LBA supported 268435455 sectors > LBA48 supported 1953525168 sectors > PIO supported PIO4 > DMA supported WDMA2 UDMA6 > > Feature Support EnableValue Vendor > read ahead yes yes > write cacheyes yes > flush cacheyes yes > overlapno > Tagged Command Queuing (TCQ) no no > Native Command Queuing (NCQ) yes 32 tags > SMART yes yes > microcode download yes yes > security yes no > power management yes yes > advanced power management no no 0/0x00 > automatic acoustic management yes no 254/0xFE128/0x80 > media status notification no no > power-up in Standbyyes no > write-read-verify no no 0/0x0 > unload no no > free-fall no no > data set management (TRIM) no > > * > > camcontrol identify ada3 > pass3: ATA-8 SATA 2.x device > pass3: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) > > protocol ATA/ATAPI-8 SATA 2.x > device model WDC WD10EADS-00L5B1 > firmware revision 01.01A01 > serial number WD-WCAU4D726772 > WWN 50014ee238ab988 > cylinders 16383 > heads 16 > sectors/track 63 > sector size logical 512, physical 512, offset 0 > LBA supported 268435455 sectors > LBA48 supported 1953525168 sectors > PIO supported PIO4 > DMA supported WDMA2 UDMA6 > > Feature Support EnableValue Vendor > read ahead yes yes > write cacheyes yes > flush cacheyes yes > overlapno > Tagged Command Queuing (TCQ) no no > Native Command Queuing (NCQ) yes 32 tags > SMART yes yes > microcode download yes yes > security yes no > power management yes yes > advanced power management no no 0/0x00 > automatic acoustic management yes no 254/0xFE128/0x80 > media status notification no no > power-up in Standbyyes no > write-read-verify no no 0/0x0 > unload no no > free-fall no no > data set management (TRIM) no > > * > > camcontrol identify ada4 > pass4: ATA-8 SATA 2.x device > pass4: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) > > protocol ATA/ATAPI-8 SATA 2.x > device model WDC WD10EADS-00M2B0 > firmware revision 01.00A01 > serial number WD-WMAV50095864 > WWN 50014ee014f3265 > cylinders 16383 > heads 16 > sectors/track 63 > sector size logical 512, physical 512, offset 0 > LBA supported 268435455 sectors > LBA48 supported 1953525168 sectors > PIO supported PIO4 > DMA supported WDMA2 UDMA6 > > Feature Support EnableValue Vendor > read ahead yes yes > write cacheyes yes > flush cacheyes yes > overlapno > Tagged Command Queuing (TCQ) no no > Native Command Queuing (NCQ) yes 32 tags > SMART yes yes > microcode download yes yes > security yes no > power management yes yes > advanced power management no no 0/0x00 > automatic acoustic management yes no 254/0xFE128/0x80 >
Re: zpool - low speed write
> Your ada3 disk is different from the other two. Can you please provide > the output from the following 3 commands? > > camcontrol identify ada2 > camcontrol identify ada3 > camcontrol identify ada4 > > > vfs.zfs.cache_flush_disable=1 > > vfs.zfs.zil_disable=1 > > I question both of these settings, especially the latter. Please remove > them both and re-test your write performance. I removed all settings of zfs. Now it default. camcontrol identify ada2 pass2: ATA-8 SATA 2.x device pass2: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) protocol ATA/ATAPI-8 SATA 2.x device model WDC WD10EADS-00M2B0 firmware revision 01.00A01 serial number WD-WCAV51709425 WWN 50014ee2adf88aae cylinders 16383 heads 16 sectors/track 63 sector size logical 512, physical 512, offset 0 LBA supported 268435455 sectors LBA48 supported 1953525168 sectors PIO supported PIO4 DMA supported WDMA2 UDMA6 Feature Support EnableValue Vendor read ahead yes yes write cacheyes yes flush cacheyes yes overlapno Tagged Command Queuing (TCQ) no no Native Command Queuing (NCQ) yes 32 tags SMART yes yes microcode download yes yes security yes no power management yes yes advanced power management no no 0/0x00 automatic acoustic management yes no 254/0xFE128/0x80 media status notification no no power-up in Standbyyes no write-read-verify no no 0/0x0 unload no no free-fall no no data set management (TRIM) no * camcontrol identify ada3 pass3: ATA-8 SATA 2.x device pass3: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) protocol ATA/ATAPI-8 SATA 2.x device model WDC WD10EADS-00L5B1 firmware revision 01.01A01 serial number WD-WCAU4D726772 WWN 50014ee238ab988 cylinders 16383 heads 16 sectors/track 63 sector size logical 512, physical 512, offset 0 LBA supported 268435455 sectors LBA48 supported 1953525168 sectors PIO supported PIO4 DMA supported WDMA2 UDMA6 Feature Support EnableValue Vendor read ahead yes yes write cacheyes yes flush cacheyes yes overlapno Tagged Command Queuing (TCQ) no no Native Command Queuing (NCQ) yes 32 tags SMART yes yes microcode download yes yes security yes no power management yes yes advanced power management no no 0/0x00 automatic acoustic management yes no 254/0xFE128/0x80 media status notification no no power-up in Standbyyes no write-read-verify no no 0/0x0 unload no no free-fall no no data set management (TRIM) no * camcontrol identify ada4 pass4: ATA-8 SATA 2.x device pass4: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) protocol ATA/ATAPI-8 SATA 2.x device model WDC WD10EADS-00M2B0 firmware revision 01.00A01 serial number WD-WMAV50095864 WWN 50014ee014f3265 cylinders 16383 heads 16 sectors/track 63 sector size logical 512, physical 512, offset 0 LBA supported 268435455 sectors LBA48 supported 1953525168 sectors PIO supported PIO4 DMA supported WDMA2 UDMA6 Feature Support EnableValue Vendor read ahead yes yes write cacheyes yes flush cacheyes yes overlapno Tagged Command Queuing (TCQ) no no Native Command Queuing (NCQ) yes 32 tags SMART yes yes microcode download yes yes security yes no power management yes yes advanced power management no no 0/0x00 automatic acoustic management yes no 254/0xFE128/0x80 media status notification no no power-up in Standbyyes no write-read-verify no no 0/0x0 unload no no free-fall no no data set management (TRIM) no * dd if=/dev/zero of=/tank/test.zero bs=3M co
Re: zpool - low speed write
On Wed, Aug 04, 2010 at 08:08:24PM +0800, Alex V. Petrov wrote: > ada2 at ahcich2 bus 0 scbus2 target 0 lun 0 > ada2: ATA-8 SATA 2.x device > ada2: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) > ada2: Command Queueing enabled > ada2: 953869MB (1953525168 512 byte sectors: 16H 63S/T 16383C) > ada3 at ahcich3 bus 0 scbus3 target 0 lun 0 > ada3: ATA-8 SATA 2.x device > ada3: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) > ada3: Command Queueing enabled > ada3: 953869MB (1953525168 512 byte sectors: 16H 63S/T 16383C) > ada4 at ahcich4 bus 0 scbus4 target 0 lun 0 > ada4: ATA-8 SATA 2.x device > ada4: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) > ada4: Command Queueing enabled Your ada3 disk is different from the other two. Can you please provide the output from the following 3 commands? camcontrol identify ada2 camcontrol identify ada3 camcontrol identify ada4 > vfs.zfs.cache_flush_disable=1 > vfs.zfs.zil_disable=1 I question both of these settings, especially the latter. Please remove them both and re-test your write performance. -- | Jeremy Chadwick j...@parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB | ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
On Wed, Aug 04, 2010 at 10:13:36PM -0400, Joshua Boyd wrote: > On Wed, Aug 4, 2010 at 10:13 AM, Alex V. Petrov wrote: > > > interesting results: > > > > From single-UDF-disk to pool: > > $ dd if=petrovs-disk1.iso of=/tank/petrovs-disk1.iso bs=1M > > 3545+1 records in > > 3545+1 records out > > 3718002688 bytes transferred in 438.770195 secs (8473690 bytes/sec) > > > > From single-UDF-disk to null: > > $ dd if=petrovs-disk1.iso of=/dev/null bs=1M > > 3545+1 records in > > 3545+1 records out > > 3718002688 bytes transferred in 83.304575 secs (44631435 bytes/sec) > > -- > > Alex V. Petrov > > ___ > > freebsd-stable@freebsd.org mailing list > > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > > To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org" > > > > What controllers are you using? > > What's the results of dd if=/dev/ada4 of=/dev/null bs=1M count=100 ? His problem is with writes, not reads. I strongly doubt his problem is with the controller (Intel ICHxx and ESBxx controllers are heavily tested on FreeBSD, both with and without AHCI, including ahci.ko). -- | Jeremy Chadwick j...@parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB | ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
> > What controllers are you using? > > What's the results of dd if=/dev/ada4 of=/dev/null bs=1M count=100 ? > > Have you tried switching to the ad driver? Maybe ada is buggy on your > hardware. $ dd if=/dev/ada4 of=/dev/null bs=1M count=100 100+0 records in 100+0 records out 104857600 bytes transferred in 1.399283 secs (74936655 bytes/sec) Intel® ICH10 motherboard Gigabyte GA-EP43-DS3 (rev. 1.0) P43 / Socket 775 CPU: Intel(R) Core(TM)2 Quad CPU Q8200 @ 2.33GHz (2335.41-MHz K8-class CPU) switch to ad I'll try later - Alex V. Petrov ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
On Wed, Aug 4, 2010 at 10:13 AM, Alex V. Petrov wrote: > interesting results: > > From single-UDF-disk to pool: > $ dd if=petrovs-disk1.iso of=/tank/petrovs-disk1.iso bs=1M > 3545+1 records in > 3545+1 records out > 3718002688 bytes transferred in 438.770195 secs (8473690 bytes/sec) > > From single-UDF-disk to null: > $ dd if=petrovs-disk1.iso of=/dev/null bs=1M > 3545+1 records in > 3545+1 records out > 3718002688 bytes transferred in 83.304575 secs (44631435 bytes/sec) > -- > Alex V. Petrov > ___ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org" > What controllers are you using? What's the results of dd if=/dev/ada4 of=/dev/null bs=1M count=100 ? Have you tried switching to the ad driver? Maybe ada is buggy on your hardware. -- Joshua Boyd JBipNet E-mail: boy...@jbip.net http://www.jbip.net ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
В сообщении от 4 августа 2010 22:40:22 вы написали: > Try booting with the following on /boot/loader.conf: > vfs.zfs.vdev.max_pending="10" > vfs.zfs.txg.write_limit_override=268435456 > > And remove setting: > vfs.zfs.cache_flush_disable=1 > > Then try a dd from /dev/zero. OK. dd if=/dev/zero of=/tank/test.zero bs=3M count=1000 1000+0 records in 1000+0 records out 3145728000 bytes transferred in 394.974934 secs (7964374 bytes/sec) During execution dd: zpool iostat -v 10 10 capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 819G 0 52 53,5K 4,09M ada4 601G 330G 0 18 20,2K 1,47M ada2 637G 291G 0 18 17,3K 1,45M ada3 730G 198G 0 16 15,9K 1,18M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 821G 13125 1,08M 9,22M ada4 600G 331G 4 43 390K 3,25M ada2 636G 292G 5 43 371K 3,32M ada3 729G 199G 3 38 345K 2,64M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 821G 0183 0 15,2M ada4 600G 331G 0 63 0 5,28M ada2 636G 292G 0 64 0 5,58M ada3 729G 199G 0 55 0 4,37M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 821G 0177 0 14,6M ada4 600G 331G 0 62 0 5,29M ada2 636G 292G 0 60 0 5,14M ada3 729G 199G 0 53 0 4,17M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 821G 0193 0 15,6M ada4 601G 330G 0 65 0 5,41M ada2 636G 292G 0 68 0 5,60M ada3 729G 199G 0 59 0 4,60M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 821G 0178 0 12,8M ada4 601G 330G 0 61 0 4,45M ada2 636G 292G 0 63 0 4,73M ada3 729G 199G 0 53 0 3,65M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 821G 0190 0 14,1M ada4 601G 330G 0 68 0 5,24M ada2 636G 292G 0 64 0 4,88M ada3 729G 199G 0 57 0 3,95M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 821G 4178 269K 14,4M ada4 601G 330G 1 64 122K 5,47M ada2 636G 292G 1 60 77,2K 4,86M ada3 729G 199G 0 53 70,3K 4,03M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 820G 5128 327K 7,26M ada4 601G 330G 1 39 128K 2,02M ada2 636G 292G 1 46 109K 2,95M ada3 730G 198G 1 41 90,5K 2,29M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 820G 9181 541K 10,3M ada4 601G 330G 3 65 181K 4,07M ada2 636G 292G 2 59 192K 3,44M ada3 730G 198G 2 56 168K 2,78M -- - - - - - - - Alex V. Petrov ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
Try booting with the following on /boot/loader.conf: vfs.zfs.vdev.max_pending="10" vfs.zfs.txg.write_limit_override=268435456 And remove setting: vfs.zfs.cache_flush_disable=1 Then try a dd from /dev/zero. Cheers, mm Dňa 4. 8. 2010 16:13, Alex V. Petrov wrote / napísal(a): > interesting results: > > From single-UDF-disk to pool: > $ dd if=petrovs-disk1.iso of=/tank/petrovs-disk1.iso bs=1M > 3545+1 records in > 3545+1 records out > 3718002688 bytes transferred in 438.770195 secs (8473690 bytes/sec) > > From single-UDF-disk to null: > $ dd if=petrovs-disk1.iso of=/dev/null bs=1M > 3545+1 records in > 3545+1 records out > 3718002688 bytes transferred in 83.304575 secs (44631435 bytes/sec) > -- > Alex V. Petrov > ___ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org" ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
В сообщении от 4 августа 2010 20:45:26 вы написали: > on 04/08/2010 15:08 Alex V. Petrov said the following: > > Hi All! > > > > $ dd if=/dev/random of=/tank/test bs=3M count=1000 > > /dev/random is slow. On single disk (UFS) speed of write faster: $ dd if=/dev/random of=/home/alex/temp/test bs=3M count=1000 1000+0 records in 1000+0 records out 3145728000 bytes transferred in 75.778427 secs (41512184 bytes/sec) Result for /dev/zero: $ dd if=/dev/zero of=/tank/test bs=3M count=1000 1000+0 records in 1000+0 records out 3145728000 bytes transferred in 113.110421 secs (27811124 bytes/sec) In Krusader speed copying files from a single disk to an pool of about 8 MB/s Fragment of `systat -v 1` in proces copying: Disks ada0 ada1 ada2 ada3 ada4 KB/t 0,00 122 44,35 32,89 48,50 tps 054565354 MB/s 0,00 6,42 2,42 1,70 2,55 %busy 011 7 482 - Alex V. Petrov ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
В сообщении от 4 августа 2010 21:09:23 вы написали: > Guckux Alex > > The first output of zpool iostat shows you only a statistical output for a > specific time (unknown to me now). > Have you tried s.th. like: > zpool iostat -v 10 10 > shows you the first statistic summary, followed 9 output with 10sec delay > and a statistic average to the last output. > Result for 1 proces copying (Krusader) single HDD to pool zpool iostat -v 10 10 capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 826G 98 5 12,0M 206K ada4 599G 332G 30 1 3,69M 74,4K ada2 634G 294G 31 1 3,90M 72,5K ada3 728G 200G 36 1 4,44M 59,6K -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 826G 0172 0 7,96M ada4 599G 332G 0 58 0 2,88M ada2 634G 294G 0 58 0 2,70M ada3 728G 200G 0 55 0 2,38M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 826G 0166 0 8,06M ada4 599G 332G 0 56 0 2,99M ada2 634G 294G 0 56 0 2,81M ada3 728G 200G 0 53 0 2,26M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 826G 0162 0 7,61M ada4 599G 332G 0 54 0 2,64M ada2 634G 294G 0 56 0 2,80M ada3 728G 200G 0 52 0 2,18M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 826G 0145 0 6,27M ada4 599G 332G 0 47 0 2,14M ada2 634G 294G 0 51 0 2,29M ada3 728G 200G 0 46 0 1,84M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 826G 0130 0 6,37M ada4 599G 332G 0 45 0 2,52M ada2 634G 294G 0 44 0 2,09M ada3 728G 200G 0 40 0 1,77M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 826G 0143 0 6,68M ada4 599G 332G 0 48 0 2,43M ada2 634G 294G 0 49 0 2,39M ada3 728G 200G 0 45 0 1,86M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 826G 0147 0 7,03M ada4 599G 332G 0 49 0 2,63M ada2 634G 294G 0 49 0 2,39M ada3 728G 200G 0 48 0 2,02M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 826G 0176 0 7,61M ada4 599G 332G 0 58 0 2,89M ada2 634G 294G 0 60 0 2,61M ada3 728G 200G 0 56 0 2,10M -- - - - - - - capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,92T 826G 0142 0 7,27M ada4 599G 332G 0 47 0 2,54M ada2 634G 294G 0 50 0 2,67M ada3 728G 200G 0 45 0 2,06M -- - - - - - - In time dd if=/dev/zero of=/tank/test.zero bs=3M count=1000 1000+0 records in 1000+0 records out 3145728000 bytes transferred in 91.863862 secs (34243368 bytes/sec) zpool iostat -v 10 10 capacity operationsbandwidth pool used avail read write read write -- - ---
Re: zpool - low speed write
interesting results: From single-UDF-disk to pool: $ dd if=petrovs-disk1.iso of=/tank/petrovs-disk1.iso bs=1M 3545+1 records in 3545+1 records out 3718002688 bytes transferred in 438.770195 secs (8473690 bytes/sec) From single-UDF-disk to null: $ dd if=petrovs-disk1.iso of=/dev/null bs=1M 3545+1 records in 3545+1 records out 3718002688 bytes transferred in 83.304575 secs (44631435 bytes/sec) -- Alex V. Petrov ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
Guckux Alex > zpool iostat -v >capacity operationsbandwidth > pool used avail read write read write > -- - - - - - - > tank1,91T 829G101 4 12,4M 148K > ada4 597G 334G 31 1 3,81M 53,4K > ada2 633G 295G 32 1 4,03M 51,8K > ada3 727G 201G 37 1 4,59M 42,9K > -- - - - - - - only this command? The first output of zpool iostat shows you only a statistical output for a specific time (unknown to me now). Have you tried s.th. like: zpool iostat -v 10 10 shows you the first statistic summary, followed 9 output with 10sec delay and a statistic average to the last output. Bye Stefan ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
On Wed, Aug 04, 2010 at 03:45:26PM +0300, Andriy Gapon wrote: > on 04/08/2010 15:08 Alex V. Petrov said the following: > > Hi All! > > > > $ dd if=/dev/random of=/tank/test bs=3M count=1000 > > /dev/random is slow. For comparing, try to see what happens with /dev/zero. :) -- Als het niet gaat zoals het moet, dan moet het zoals het gaat. ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: zpool - low speed write
on 04/08/2010 15:08 Alex V. Petrov said the following: > Hi All! > > $ dd if=/dev/random of=/tank/test bs=3M count=1000 /dev/random is slow. -- Andriy Gapon ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
zpool - low speed write
Hi All! $ dd if=/dev/random of=/tank/test bs=3M count=1000 1000+0 records in 1000+0 records out 3145728000 bytes transferred in 298.153293 secs (10550707 bytes/sec) What, i think, very-very low :-( FreeBSD alex.super 8.1-STABLE FreeBSD 8.1-STABLE #76: Mon Aug 2 20:19:09 KRAST 2010 a...@alex.super:/usr/obj/usr/src/sys/ALEX amd64 real memory = 4294967296 (4096 MB) avail memory = 4098732032 (3908 MB) zpool iostat -v capacity operationsbandwidth pool used avail read write read write -- - - - - - - tank1,91T 829G101 4 12,4M 148K ada4 597G 334G 31 1 3,81M 53,4K ada2 633G 295G 32 1 4,03M 51,8K ada3 727G 201G 37 1 4,59M 42,9K -- - - - - - - zpool status -v pool: tank state: ONLINE scrub: scrub completed after 7h12m with 0 errors on Tue Aug 3 04:54:14 2010 config: NAMESTATE READ WRITE CKSUM tankONLINE 0 0 0 ada4 ONLINE 0 0 0 ada2 ONLINE 0 0 0 ada3 ONLINE 0 0 0 errors: No known data errors zpool history History for 'tank': 2009-07-16.19:46:24 zpool create tank ad12 2009-12-13.14:58:46 zpool add tank ad8 2010-04-24.01:59:41 zpool upgrade tank 2010-05-09.02:16:34 zpool add tank ada3 2010-05-25.17:57:12 zpool scrub tank 2010-06-27.16:02:45 zpool scrub tank 2010-08-02.21:41:53 zpool scrub tank ada2 at ahcich2 bus 0 scbus2 target 0 lun 0 ada2: ATA-8 SATA 2.x device ada2: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) ada2: Command Queueing enabled ada2: 953869MB (1953525168 512 byte sectors: 16H 63S/T 16383C) ada3 at ahcich3 bus 0 scbus3 target 0 lun 0 ada3: ATA-8 SATA 2.x device ada3: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) ada3: Command Queueing enabled ada3: 953869MB (1953525168 512 byte sectors: 16H 63S/T 16383C) ada4 at ahcich4 bus 0 scbus4 target 0 lun 0 ada4: ATA-8 SATA 2.x device ada4: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) ada4: Command Queueing enabled smartd daily output: Checking health of /dev/ada2: OK Checking health of /dev/ada3: OK Checking health of /dev/ada4: OK /boot/loader.conf: ahci_load="YES" sem_load="YES" snd_hda_load="YES" nvidia_load="YES" linux_load="YES" wlan_xauth_load="YES" vboxdrv_load="YES" atapicam_load="YES" coretemp_load="YES" aio_load="YES" vfs.zfs.prefetch_disable=0 vfs.zfs.cache_flush_disable=1 vfs.zfs.zil_disable=1 Any ideas? - Alex V. Petrov ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"