Re: Re[4]: Fwd: twe0 interrupt storm

2008-06-18 Thread Stanislav
Dear Daniel,

I have moved two disks to motherboard controller, cvsuped sources to
7.0-STABLE and recompiled kernel and world. gmirrored disks (RAID-1).
Server worked good for about 6 hours, and than again,
interrupt storm detected on irq22:; throttling interrupt source

gans# vmstat -i
interrupt  total   rate
irq1: atkbd0   3  0
irq9: acpi01  0
irq16: ohci0   1  0
irq17: ohci1 ohci3 1  0
irq18: ohci2 ohci4 1  0
irq22: atapci01195017654  55266
cpu0: timer 43244319   1999
cpu1: timer 43244059   1999
Total 1281506039  59265

last pid: 75968;  load averages:  2.54,  2.57,  2.60up 0+06:01:20  20:20:15
292 processes: 1 running, 290 sleeping, 1 zombie
CPU: 21.8% user,  0.0% nice, 12.9% system,  0.6% interrupt, 64.7% idle
Mem: 1033M Active, 435M Inact, 379M Wired, 51M Cache, 210M Buf, 37M Free
Swap: 4128M Total, 296M Used, 3832M Free, 7% Inuse

Maybe there is a way not to move from hetzner? No additional devices
are attached to motherboard.

I remember that server worked on 6.2-STABLE without these storms, but
there were not so much processes.

Best Regards,
Stanislav

2008/6/15 Daniel Gerzo [EMAIL PROTECTED]:
 Hello Stanislav,

 Sunday, June 15, 2008, 12:40:57 PM, you wrote:

 Dear Daniel,

 Please tell me you had this storms only with 3ware 8006-2LP?

 Actually, I do have them on the boxes which do not have the raid
 controller. I have several boxes in hetzner. The ones that does not
 have any additional devices attached to the motherboard seem like not
 having the problem. However most of my machines have at least
 additional NIC device.

 Please note that my machines does not crash even when there is an
 interrupt storm. On the other hand, they suffer from some performance
 problems :/

 The funny thing is that after reboot there's some silent period of
 time, when the interrupts are OK. After the systems becomes to be
 loaded more, the interrupt rate increases unless it reaches around
 350k interrupt rate...

 I have tried to play with the BIOS settings, nothing really helped.

 Because as I remember on MSI motherboard controller, I have no
 problems with FreeBSD 6.2, but when I upgraded to FreeBSD 7.0 that
 storms appeared, but not so often as with 3ware.

 P.S. Maybe I have to ask them to update BIOS? I am so happy with their
 network and service,  don't want to move to another server...

 I have tried to ask them to update the BIOS in the past, AFAIR they
 replied that the BIOS update is up to me...

 --
 Best regards,
  Danielmailto:[EMAIL PROTECTED]


___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: Re[2]: Fwd: twe0 interrupt storm

2008-06-15 Thread Stanislav
Dear Daniel,

Please tell me you had this storms only with 3ware 8006-2LP?
Because as I remember on MSI motherboard controller, I have no
problems with FreeBSD 6.2, but when I upgraded to FreeBSD 7.0 that
storms appeared, but not so often as with 3ware.

P.S. Maybe I have to ask them to update BIOS? I am so happy with their
network and service,  don't want to move to another server...

Kind Regards

2008/6/15 Daniel Gerzo [EMAIL PROTECTED]:
 Hello FreeBSD,

 Saturday, June 14, 2008, 4:38:10 PM, you wrote:

 Be careful with hetzner.de hosting, here is their answer:

 The controller is compatible with the mainboard. With linux there are no
 problems. FreeBSD is also not officially supported by us (only
 Debian/Ubuntu/Suse).

 Yea, I know they do not officialy support FreeBSD, unluckily :-(

 These interrupt storms are the only problem I have with hetzner.de...

 --
 Best regards,
  Danielmailto:[EMAIL PROTECTED]




___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re[4]: Fwd: twe0 interrupt storm

2008-06-15 Thread Daniel Gerzo
Hello Stanislav,

Sunday, June 15, 2008, 12:40:57 PM, you wrote:

 Dear Daniel,

 Please tell me you had this storms only with 3ware 8006-2LP?

Actually, I do have them on the boxes which do not have the raid
controller. I have several boxes in hetzner. The ones that does not
have any additional devices attached to the motherboard seem like not
having the problem. However most of my machines have at least
additional NIC device.

Please note that my machines does not crash even when there is an
interrupt storm. On the other hand, they suffer from some performance
problems :/

The funny thing is that after reboot there's some silent period of
time, when the interrupts are OK. After the systems becomes to be
loaded more, the interrupt rate increases unless it reaches around
350k interrupt rate...

I have tried to play with the BIOS settings, nothing really helped.

 Because as I remember on MSI motherboard controller, I have no
 problems with FreeBSD 6.2, but when I upgraded to FreeBSD 7.0 that
 storms appeared, but not so often as with 3ware.

 P.S. Maybe I have to ask them to update BIOS? I am so happy with their
 network and service,  don't want to move to another server...

I have tried to ask them to update the BIOS in the past, AFAIR they
replied that the BIOS update is up to me...

-- 
Best regards,
 Danielmailto:[EMAIL PROTECTED]

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: Fwd: twe0 interrupt storm

2008-06-14 Thread Daniel Gerzo
Hello Stanislav,

Thursday, June 12, 2008, 3:45:36 PM, you wrote:

 Hi, I have problem on FreeBSD 6.2, 7.0 with 3ware 8006-2LP RAID-1 controller:

Is it MSI mobo by any cnance?
I am seeing these interrupt storms on a few boxes (irq21 too, twe(4) or
fxp(4) attached to these) with an

Base Board Information
Manufacturer: MICRO-STAR INTERANTIONAL CO.,LTD
Product Name: MS-7368
Version: 1.0

I was unable so far to fix this problem...seems like some hardware
issue

-- 
Best regards,
 Danielmailto:[EMAIL PROTECTED]

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: Fwd: twe0 interrupt storm

2008-06-14 Thread Stanislav
Dear Daniel,

Yes, you're right:
Handle 0x0001, DMI type 1, 27 bytes
System Information
Manufacturer: MICRO-STAR INTERANTIONAL CO.,LTD
Product Name: MS-7368
Version: 1.0
Serial Number: To Be Filled By O.E.M.
UUID: Not Present
Wake-up Type: Power Switch
SKU Number: To Be Filled By O.E.M.
Family: To Be Filled By O.E.M.

Thank you very much!

Kind Regards

2008/6/14 Daniel Gerzo [EMAIL PROTECTED]:
 Hello Stanislav,

 Thursday, June 12, 2008, 3:45:36 PM, you wrote:

 Hi, I have problem on FreeBSD 6.2, 7.0 with 3ware 8006-2LP RAID-1 controller:

 Is it MSI mobo by any cnance?
 I am seeing these interrupt storms on a few boxes (irq21 too, twe(4) or
 fxp(4) attached to these) with an

 Base Board Information
Manufacturer: MICRO-STAR INTERANTIONAL CO.,LTD
Product Name: MS-7368
Version: 1.0

 I was unable so far to fix this problem...seems like some hardware
 issue

 --
 Best regards,
  Danielmailto:[EMAIL PROTECTED]


___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: Fwd: twe0 interrupt storm

2008-06-14 Thread FreeBSD
Be careful with hetzner.de hosting, here is their answer:

The controller is compatible with the mainboard. With linux there are no
problems. FreeBSD is also not officially supported by us (only
Debian/Ubuntu/Suse).

Sorry, but it is not possible to buy a new model only for a few customers.

2008/6/14 Stanislav [EMAIL PROTECTED]:
 Dear Daniel,

 Yes, you're right:
 Handle 0x0001, DMI type 1, 27 bytes
 System Information
Manufacturer: MICRO-STAR INTERANTIONAL CO.,LTD
Product Name: MS-7368
Version: 1.0
Serial Number: To Be Filled By O.E.M.
UUID: Not Present
Wake-up Type: Power Switch
SKU Number: To Be Filled By O.E.M.
Family: To Be Filled By O.E.M.

 Thank you very much!

 Kind Regards

 2008/6/14 Daniel Gerzo [EMAIL PROTECTED]:
 Hello Stanislav,

 Thursday, June 12, 2008, 3:45:36 PM, you wrote:

 Hi, I have problem on FreeBSD 6.2, 7.0 with 3ware 8006-2LP RAID-1 
 controller:

 Is it MSI mobo by any cnance?
 I am seeing these interrupt storms on a few boxes (irq21 too, twe(4) or
 fxp(4) attached to these) with an

 Base Board Information
Manufacturer: MICRO-STAR INTERANTIONAL CO.,LTD
Product Name: MS-7368
Version: 1.0

 I was unable so far to fix this problem...seems like some hardware
 issue

 --
 Best regards,
  Danielmailto:[EMAIL PROTECTED]



___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re[2]: Fwd: twe0 interrupt storm

2008-06-14 Thread Daniel Gerzo
Hello FreeBSD,

Saturday, June 14, 2008, 4:38:10 PM, you wrote:

 Be careful with hetzner.de hosting, here is their answer:

 The controller is compatible with the mainboard. With linux there are no
 problems. FreeBSD is also not officially supported by us (only
 Debian/Ubuntu/Suse).

Yea, I know they do not officialy support FreeBSD, unluckily :-(

These interrupt storms are the only problem I have with hetzner.de...

-- 
Best regards,
 Danielmailto:[EMAIL PROTECTED]

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Fwd: twe0 interrupt storm

2008-06-12 Thread Stanislav
Hi, I have problem on FreeBSD 6.2, 7.0 with 3ware 8006-2LP RAID-1 controller:

 Model 8006-2LP
 Serial #  L018501C7271467
 Firmware  FE8S 1.05.00.068
 Driver1.50.01.002
 BIOS  BE7X 1.08.00.048
 Monitor   ME7X 1.01.00.040
 Memory Installed  512 kB
 # of Ports2
 # of Drives   2
 # of Units1

Jun 11 21:38:00 gans kernel: twe0: 3ware Storage Controller. Driver
version 1.50.01.002 port 0xe800-0xe80f mem
0xfebffc00-0xfebffc0f,0xfe00-0xfe7f irq 20 at device 2.0
on pci3
Jun 11 21:38:00 gans kernel: twe0: [GIANT-LOCKED]
Jun 11 21:38:00 gans kernel: twe0: [ITHREAD]
Jun 11 21:38:00 gans kernel: twe0: 2 ports, Firmware FE8S 1.05.00.068,
BIOS BE7X 1.08.00.048

/var/log/messages
Jun 10 14:51:36 gans kernel: interrupt storm detected on irq21:;
throttling interrupt source
Jun 10 14:52:07 gans last message repeated 31 times
Jun 10 14:54:08 gans last message repeated 121 times
Jun 10 15:04:09 gans last message repeated 598 times
Jun 10 15:14:10 gans last message repeated 599 times
Jun 10 15:24:11 gans last message repeated 600 times
Jun 10 15:34:12 gans last message repeated 598 times
Jun 10 15:44:13 gans last message repeated 598 times
Jun 10 15:54:14 gans last message repeated 600 times
Jun 10 16:04:15 gans last message repeated 600 times
Jun 10 16:14:16 gans last message repeated 599 times
Jun 10 16:24:17 gans last message repeated 598 times

[EMAIL PROTECTED] /home/sg]# vmstat -w 5
 procs  memory  page   disk   faults cpu
 r b w avmfre   flt  re  pi  pofr  sr tw0   in   sy   cs us sy id
 8 3 0   5389M   220M  1410  23   5   3   809 1762   0 11200 1377 3451 26 15 59
 7 5 0   5505M   139M 16909 188   1   0 11025   0  81 565387 182791
30440 68 18 14
 1 6 0   5468M   158M 19244  35   1   0 19469   0  42 563246 217685
7941 75 22  3
 6 3 0   5276M   235M 23078   7   1   0 27426   0 135 555198 118844
73807 33 20 47
 0 0 0   5177M   283M  9636   3   1   0 20972   0 108 564052 144113
6643 29 14 57
 5 0 0   5176M   284M  8968   7   0   0  8607   0  22 567849 143837
6767 27 12 60
 9 1 0   5199M   268M 14417   0   0   0 25905   0  31 567551 249480
6951 48 22 29
 1 3 0   5341M   201M  8095  16  25   0  5087   0 112 565233 63826 6777 28  8 63
 1 4 0   5356M   184M 30195  11   2   0 38720   0  95 560968 137901
52156 71 22  7
 1 3 0   5266M   225M 13461   2   1   0 15518   0  46 564340 101664
9105 33 13 54
[EMAIL PROTECTED] /home/sg]# vmstat -i
interrupt  total   rate
irq1: atkbd0   4  0
irq16: ohci0   1  0
irq17: ohci1 ohci3 1  0
irq18: ohci2 ohci4 1  0
irq21: twe0  38629583391 101980
cpu0: timer757587412   1999
cpu1: timer757587403   1999
Total40144758213 105980

I have recompiled kernel to 6.2-STABLE, 7.0-RELEASE, 7.0-RELEASE-p1,
7.0-STABLE but error persists.

System work good for some time (few days) and than vmstat -i irq20;
rate increases from 60-90 to 10+ and than system hangs.
After reboot system work good for some time.

I have replaced sever, moved disks to another server with same
hardware (and controller), and error persists.

Kind Regards
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]