Re: [FUG-BR] FreeBSD+Raid+Hp ML350

2009-02-06 Por tôpico William David FUG-BR
Só uma dica  use o iozone para você verificar  a velocidade do I/O
com diversos tamanhos de arquivos  tem o ports ele.





-- 
- = - = - = - = - = - = - = - = - = -
.  Of course it runsWilliam David Armstrong
|==   Bio Systems Security Networking
'  FreeBSD   MSN / GT  biosystems  gmail . com
 http://biosystems.ath.cx:8080/  http://biosystems.broker.freenet6.net/
--
-
Histórico: http://www.fug.com.br/historico/html/freebsd/
Sair da lista: https://www.fug.com.br/mailman/listinfo/freebsd


Re: [FUG-BR] FreeBSD+Raid+Hp ML350

2009-01-25 Por tôpico Matheus Cucoloto
Pessoal... Sério mesmo que ninguem ai usa FreeBSD em um Hp ML350?



 On Thu, Jan 22, 2009 at 9:48 AM, Matheus Cucoloto
 matheuscucol...@gmail.com wrote:
  Pessoal
 
  To com o FreeBSD 7.1-stable em um HP ML 350 g5 e estou enfrentando
  lentidão extrema no IO da controladora.
 
 
  Com Raid 0 consigo o seguinte IO em um dump.
 
  marte# iostat -w 1 -c 1000 da0
   tty da0 cpu
   tin tout  KB/t tps  MB/s  us ni sy in id
09 30.41  10  0.29   1  0  0  4 95
0  129 64.00 153  9.56   0  0  0  3 97
0   43 64.00 153  9.55   0  0  1  5 94
0   43 64.00 153  9.55   0  0  1  4 95
0   43 63.37 152  9.40   0  0  1  6 93
0   43 63.37 152  9.40   0  0  0  5 95
0   43 64.00 153  9.55   0  0  2  6 92
0   43 64.00 153  9.55   0  0  0  4 96
0   43 63.69 154  9.57   0  0  1  4 95
0   43 63.38 154  9.52   0  0  0  4 95
0   43 62.80 155  9.50   0  0  1  3 96
0   43 64.00 153  9.55   0  0  0  2 98
0   43 63.69 154  9.57   0  0  1  6 94
0   43 61.85 156  9.41   0  0  1  2 98
 
  Em Raid1 consigo o seguinte () em um dump.
 
  marte# iostat -c 1000 -w 1 da0
   tty da0 cpu
   tin tout  KB/t tps  MB/s  us ni sy in id
1  688  6.48  71  0.45   0  0  1  0 99
0  561 55.12 114  6.13   0  0  1  0 98
0  459 37.02 114  4.12   0  0  1  0 99
0  425 34.52 115  3.87   0  0  1  0 99
0  396 40.32 114  4.49   0  0  1  0 99
0  353 25.21 119  2.93   0  0  1  0 99
0  420 39.19 114  4.36   0  0  0  0 100
0  435 36.80 113  4.05   0  0  1  0 99
0  454 40.60 117  4.64   0  0  1  0 99
0  360 34.14 116  3.86   0  0  0  0 99
0  442 39.00 116  4.42   0  0  0  0 99
0  353 34.69 118  3.99   0  0  0  0 99
0  541 42.55 112  4.65   0  0  1  0 99
0  408 32.27 117  3.68   0  0  0  0 99
0  418 35.75 113  3.94   0  0  0  0 100
0  285 44.03 116  4.98   0  0  0  0 100
0   96 28.09 112  3.07   0  0  2  0 98
0   43  7.52 126  0.92   0  0  0  0 100
 
  Em uma maquina normal isso é superado umas 30x mais
 
 
  Ja dei uma pesquisada mas nada...
 
  Pergunto...
 
  Alguem ai ja teve problemas com este hardware?
 
  Como eu posso reportar o time de desenvolvimento do driver sobre este 
  problema.
 
  Com outro SO a taxa de transferencia é extraordinariamente maior.
 
  marte# pciconf -vl
  hos...@pci0:0:0:0:  class=0x06 card=0x31fd103c chip=0x25d08086
  rev=0xb1 hdr=0x00
 vendor = 'Intel Corporation'
 device = '5000Z Chipset Memory Controller Hub'
 class  = bridge
 subclass   = HOST-PCI
  pc...@pci0:0:2:0:   class=0x060400 card=0x chip=0x25f78086
  rev=0xb1 hdr=0x01
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset PCIe x8 Port 2-3'
 class  = bridge
 subclass   = PCI-PCI
  pc...@pci0:0:3:0:   class=0x060400 card=0x chip=0x25e38086
  rev=0xb1 hdr=0x01
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset PCIe x4 Port 3'
 class  = bridge
 subclass   = PCI-PCI
  pc...@pci0:0:4:0:   class=0x060400 card=0x chip=0x25e48086
  rev=0xb1 hdr=0x01
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset PCIe x4 Port 4'
 class  = bridge
 subclass   = PCI-PCI
  pc...@pci0:0:5:0:   class=0x060400 card=0x chip=0x25e58086
  rev=0xb1 hdr=0x01
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset PCIe x4 Port 5'
 class  = bridge
 subclass   = PCI-PCI
  hos...@pci0:0:16:0: class=0x06 card=0x31fd103c chip=0x25f08086
  rev=0xb1 hdr=0x00
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset Error Reporting Registers'
 class  = bridge
 subclass   = HOST-PCI
  hos...@pci0:0:16:1: class=0x06 card=0x31fd103c chip=0x25f08086
  rev=0xb1 hdr=0x00
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset Error Reporting Registers'
 class  = bridge
 subclass   = HOST-PCI
  hos...@pci0:0:16:2: class=0x06 card=0x31fd103c chip=0x25f08086
  rev=0xb1 hdr=0x00
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset Error Reporting Registers'
 class  = bridge
 subclass   = HOST-PCI
  hos...@pci0:0:17:0: class=0x06 card=0x31fd103c chip=0x25f18086
  rev=0xb1 hdr=0x00
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset Reserved Registers'
 class  = bridge
 subclass   = HOST-PCI
  hos...@pci0:0:19:0: class=0x06 card=0x31fd103c chip=0x25f38086
  rev=0xb1 hdr=0x00
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset Reserved Registers'
 class  = bridge
 subclass   = HOST-PCI
  hos...@pci0:0:21:0: class=0x06 card=0x31fd103c chip=0x25f58086
  rev=0xb1 hdr=0x00
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset FBD Registers'
 class  = bridge
 subclass   = 

Re: [FUG-BR] FreeBSD+Raid+Hp ML350

2009-01-25 Por tôpico Jean Carlos Zanuzo


On Sun, 25 Jan 2009 18:57:51 -0200, Matheus Cucoloto
matheuscucol...@gmail.com wrote:
 Pessoal... Sério mesmo que ninguem ai usa FreeBSD em um Hp ML350?
 
 
 
 On Thu, Jan 22, 2009 at 9:48 AM, Matheus Cucoloto
 matheuscucol...@gmail.com wrote:
  Pessoal
 
  To com o FreeBSD 7.1-stable em um HP ML 350 g5 e estou enfrentando
  lentidão extrema no IO da controladora.
 
 
  Com Raid 0 consigo o seguinte IO em um dump.
 
  marte# iostat -w 1 -c 1000 da0
   tty da0 cpu
   tin tout  KB/t tps  MB/s  us ni sy in id
09 30.41  10  0.29   1  0  0  4 95
0  129 64.00 153  9.56   0  0  0  3 97
0   43 64.00 153  9.55   0  0  1  5 94
0   43 64.00 153  9.55   0  0  1  4 95
0   43 63.37 152  9.40   0  0  1  6 93
0   43 63.37 152  9.40   0  0  0  5 95
0   43 64.00 153  9.55   0  0  2  6 92
0   43 64.00 153  9.55   0  0  0  4 96
0   43 63.69 154  9.57   0  0  1  4 95
0   43 63.38 154  9.52   0  0  0  4 95
0   43 62.80 155  9.50   0  0  1  3 96
0   43 64.00 153  9.55   0  0  0  2 98
0   43 63.69 154  9.57   0  0  1  6 94
0   43 61.85 156  9.41   0  0  1  2 98
 
  Em Raid1 consigo o seguinte () em um dump.
 
  marte# iostat -c 1000 -w 1 da0
   tty da0 cpu
   tin tout  KB/t tps  MB/s  us ni sy in id
1  688  6.48  71  0.45   0  0  1  0 99
0  561 55.12 114  6.13   0  0  1  0 98
0  459 37.02 114  4.12   0  0  1  0 99
0  425 34.52 115  3.87   0  0  1  0 99
0  396 40.32 114  4.49   0  0  1  0 99
0  353 25.21 119  2.93   0  0  1  0 99
0  420 39.19 114  4.36   0  0  0  0 100
0  435 36.80 113  4.05   0  0  1  0 99
0  454 40.60 117  4.64   0  0  1  0 99
0  360 34.14 116  3.86   0  0  0  0 99
0  442 39.00 116  4.42   0  0  0  0 99
0  353 34.69 118  3.99   0  0  0  0 99
0  541 42.55 112  4.65   0  0  1  0 99
0  408 32.27 117  3.68   0  0  0  0 99
0  418 35.75 113  3.94   0  0  0  0 100
0  285 44.03 116  4.98   0  0  0  0 100
0   96 28.09 112  3.07   0  0  2  0 98
0   43  7.52 126  0.92   0  0  0  0 100
 
  Em uma maquina normal isso é superado umas 30x mais
 
 
  Ja dei uma pesquisada mas nada...
 
  Pergunto...
 
  Alguem ai ja teve problemas com este hardware?
Amigo Matheus, tenho uma controladora igual a sua num HP DL 180 G5.

# pciconf -l -v
ci...@pci0:7:8:0:   class=0x010400 card=0x3212103c chip=0x3238103c
rev=0x00 hdr=0x00
vendor = 'Hewlett-Packard Company'
device = 'Smart Array E200/E200i Controller'
class  = mass storage
subclass   = RAID


Em um teste parecido com o seu, num servidor em produção obetive com um
dd if=/dev/random of=/montagem/xxx.teste o seguinte resultado:

# iostat -c 1000 -w 1 da1
  tty da1 cpu
 tin tout  KB/t tps  MB/s  us ni sy in id
   04 16.70  21  0.35   2  0  1  0 97
   0  128 52.78 677 34.91   3  0 27  1 69
   0   43 62.82 478 29.33   1  0 26  0 73
   0   43 62.57 506 30.92   1  0 26  1 72
   0   43 62.91 503 30.90   2  0 26  1 72
   0   43 61.33 574 34.40   3  0 26  0 71
   0   43 60.32 585 34.47   4  0 25  1 70
   0   43 61.83 549 33.16   3  0 27  1 70
   0   43 57.02 591 32.94   2  0 24  1 73
   0   43 63.29 508 31.41   4  0 25  0 71
   0   43 58.98 537 30.90   2  0 25  1 72
   0   43 60.71 573 33.96   3  0 27  2 68
   0   43 54.15 723 38.21   3  0 24  0 72
   1  165 62.91 403 24.76   2  0 21  1 77
   0   43  8.86  14  0.12   2  0  1  0 97
^C


 
  Como eu posso reportar o time de desenvolvimento do driver sobre este
 problema.
 
  Com outro SO a taxa de transferencia é extraordinariamente maior.
 
  marte# pciconf -vl
  hos...@pci0:0:0:0:  class=0x06 card=0x31fd103c chip=0x25d08086
  rev=0xb1 hdr=0x00
 vendor = 'Intel Corporation'
 device = '5000Z Chipset Memory Controller Hub'
 class  = bridge
 subclass   = HOST-PCI
  pc...@pci0:0:2:0:   class=0x060400 card=0x chip=0x25f78086
  rev=0xb1 hdr=0x01
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset PCIe x8 Port 2-3'
 class  = bridge
 subclass   = PCI-PCI
  pc...@pci0:0:3:0:   class=0x060400 card=0x chip=0x25e38086
  rev=0xb1 hdr=0x01
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset PCIe x4 Port 3'
 class  = bridge
 subclass   = PCI-PCI
  pc...@pci0:0:4:0:   class=0x060400 card=0x chip=0x25e48086
  rev=0xb1 hdr=0x01
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset PCIe x4 Port 4'
 class  = bridge
 subclass   = PCI-PCI
  pc...@pci0:0:5:0:   class=0x060400 card=0x chip=0x25e58086
  rev=0xb1 hdr=0x01
 vendor = 'Intel Corporation'
 device = '5000 Series Chipset PCIe x4 Port 5'
 class  = bridge
 subclass   = PCI-PCI
  hos...@pci0:0:16:0: class=0x06 card=0x31fd103c chip=0x25f08086
  rev=0xb1 hdr=0x00
 vendor = 'Intel Corporation'
 device = '5000 Series 

Re: [FUG-BR] FreeBSD+Raid+Hp ML350 (corrigido)

2009-01-22 Por tôpico Matheus Cucoloto
Na verdade é HP ML350 e nao ML360.

On Thu, Jan 22, 2009 at 9:48 AM, Matheus Cucoloto
matheuscucol...@gmail.com wrote:
 Pessoal

 To com o FreeBSD 7.1-stable em um HP ML 360 e estou enfrentando
 lentidão extrema no IO da controladora.


 Com Raid 0 consigo o seguinte IO em um dump.

 marte# iostat -w 1 -c 1000 da0
  tty da0 cpu
  tin tout  KB/t tps  MB/s  us ni sy in id
   09 30.41  10  0.29   1  0  0  4 95
   0  129 64.00 153  9.56   0  0  0  3 97
   0   43 64.00 153  9.55   0  0  1  5 94
   0   43 64.00 153  9.55   0  0  1  4 95
   0   43 63.37 152  9.40   0  0  1  6 93
   0   43 63.37 152  9.40   0  0  0  5 95
   0   43 64.00 153  9.55   0  0  2  6 92
   0   43 64.00 153  9.55   0  0  0  4 96
   0   43 63.69 154  9.57   0  0  1  4 95
   0   43 63.38 154  9.52   0  0  0  4 95
   0   43 62.80 155  9.50   0  0  1  3 96
   0   43 64.00 153  9.55   0  0  0  2 98
   0   43 63.69 154  9.57   0  0  1  6 94
   0   43 61.85 156  9.41   0  0  1  2 98

 Em Raid1 consigo o seguinte () em um dump.

 marte# iostat -c 1000 -w 1 da0
  tty da0 cpu
  tin tout  KB/t tps  MB/s  us ni sy in id
   1  688  6.48  71  0.45   0  0  1  0 99
   0  561 55.12 114  6.13   0  0  1  0 98
   0  459 37.02 114  4.12   0  0  1  0 99
   0  425 34.52 115  3.87   0  0  1  0 99
   0  396 40.32 114  4.49   0  0  1  0 99
   0  353 25.21 119  2.93   0  0  1  0 99
   0  420 39.19 114  4.36   0  0  0  0 100
   0  435 36.80 113  4.05   0  0  1  0 99
   0  454 40.60 117  4.64   0  0  1  0 99
   0  360 34.14 116  3.86   0  0  0  0 99
   0  442 39.00 116  4.42   0  0  0  0 99
   0  353 34.69 118  3.99   0  0  0  0 99
   0  541 42.55 112  4.65   0  0  1  0 99
   0  408 32.27 117  3.68   0  0  0  0 99
   0  418 35.75 113  3.94   0  0  0  0 100
   0  285 44.03 116  4.98   0  0  0  0 100
   0   96 28.09 112  3.07   0  0  2  0 98
   0   43  7.52 126  0.92   0  0  0  0 100

 Em uma maquina normal isso é superado umas 30x mais


 Ja dei uma pesquisada mas nada...

 Pergunto...

 Alguem ai ja teve problemas com este hardware?

 Como eu posso reportar o time de desenvolvimento do driver sobre este 
 problema.

 Com outro SO a taxa de transferencia é extraordinariamente maior.

 marte# pciconf -vl
 hos...@pci0:0:0:0:  class=0x06 card=0x31fd103c chip=0x25d08086
 rev=0xb1 hdr=0x00
vendor = 'Intel Corporation'
device = '5000Z Chipset Memory Controller Hub'
class  = bridge
subclass   = HOST-PCI
 pc...@pci0:0:2:0:   class=0x060400 card=0x chip=0x25f78086
 rev=0xb1 hdr=0x01
vendor = 'Intel Corporation'
device = '5000 Series Chipset PCIe x8 Port 2-3'
class  = bridge
subclass   = PCI-PCI
 pc...@pci0:0:3:0:   class=0x060400 card=0x chip=0x25e38086
 rev=0xb1 hdr=0x01
vendor = 'Intel Corporation'
device = '5000 Series Chipset PCIe x4 Port 3'
class  = bridge
subclass   = PCI-PCI
 pc...@pci0:0:4:0:   class=0x060400 card=0x chip=0x25e48086
 rev=0xb1 hdr=0x01
vendor = 'Intel Corporation'
device = '5000 Series Chipset PCIe x4 Port 4'
class  = bridge
subclass   = PCI-PCI
 pc...@pci0:0:5:0:   class=0x060400 card=0x chip=0x25e58086
 rev=0xb1 hdr=0x01
vendor = 'Intel Corporation'
device = '5000 Series Chipset PCIe x4 Port 5'
class  = bridge
subclass   = PCI-PCI
 hos...@pci0:0:16:0: class=0x06 card=0x31fd103c chip=0x25f08086
 rev=0xb1 hdr=0x00
vendor = 'Intel Corporation'
device = '5000 Series Chipset Error Reporting Registers'
class  = bridge
subclass   = HOST-PCI
 hos...@pci0:0:16:1: class=0x06 card=0x31fd103c chip=0x25f08086
 rev=0xb1 hdr=0x00
vendor = 'Intel Corporation'
device = '5000 Series Chipset Error Reporting Registers'
class  = bridge
subclass   = HOST-PCI
 hos...@pci0:0:16:2: class=0x06 card=0x31fd103c chip=0x25f08086
 rev=0xb1 hdr=0x00
vendor = 'Intel Corporation'
device = '5000 Series Chipset Error Reporting Registers'
class  = bridge
subclass   = HOST-PCI
 hos...@pci0:0:17:0: class=0x06 card=0x31fd103c chip=0x25f18086
 rev=0xb1 hdr=0x00
vendor = 'Intel Corporation'
device = '5000 Series Chipset Reserved Registers'
class  = bridge
subclass   = HOST-PCI
 hos...@pci0:0:19:0: class=0x06 card=0x31fd103c chip=0x25f38086
 rev=0xb1 hdr=0x00
vendor = 'Intel Corporation'
device = '5000 Series Chipset Reserved Registers'
class  = bridge
subclass   = HOST-PCI
 hos...@pci0:0:21:0: class=0x06 card=0x31fd103c chip=0x25f58086
 rev=0xb1 hdr=0x00
vendor = 'Intel Corporation'
device = '5000 Series Chipset FBD Registers'
class  = bridge
subclass   = HOST-PCI
 hos...@pci0:0:22:0: class=0x06 card=0x31fd103c chip=0x25f68086
 rev=0xb1 hdr=0x00
vendor = 'Intel Corporation'
device = '5000 Series