jaytee75
Enthusiast
Enthusiast

Crucial MX300 perfomance issue / WARNING: NMP: nmp_DeviceRequestFastDeviceProbe

Jump to solution

Hi,

did anyone has an idea for this erro of dmesg (host):

2016-12-30T16:15:32.222Z cpu0:65555)WARNING: NMP: nmp_DeviceRequestFastDeviceProbe:237: NMP device "t10.ATA_____Crucial_CT525MX300SSD1__________________________163413BAA738" state in doubt; requested fast path state update...

I have poor perfromance on vsphere 6.5

and some more errors:

2016-12-30T17:49:51.037Z cpu1:65556)ScsiDeviceIO: 2927: Cmd(0x4395009d1880) 0x2a, CmdSN 0x9172b from world 182413 to dev "t10.ATA_____Crucial_CT525MX300SSD1__________________________163413BAA738" failed H:0x2 D:0x0 P:0x0 Invalid sense data: 0x75 0x70 0x$

2016-12-30T17:49:51.054Z cpu1:65670)vmw_ahci[0000001f]: scsiUnmapCommand:Unmap transfer 0x18 byte

2016-12-30T17:49:51.331Z cpu0:65555)ScsiDeviceIO: 2927: Cmd(0x4395009fbc00) 0x2a, CmdSN 0x91735 from world 65545 to dev "t10.ATA_____Crucial_CT525MX300SSD1__________________________163413BAA738" failed H:0x2 D:0x0 P:0x0 Invalid sense data: 0x0 0x0 0x0.

2016-12-30T17:49:51.573Z cpu1:65670)vmw_ahci[0000001f]: scsiUnmapCommand:Unmap transfer 0x18 byte

2016-12-30T17:49:51.902Z cpu1:65556)ScsiDeviceIO: 2927: Cmd(0x4395009bb000) 0x2a, CmdSN 0x91748 from world 65545 to dev "t10.ATA_____Crucial_CT525MX300SSD1__________________________163413BAA738" failed H:0x2 D:0x0 P:0x0 Invalid sense data: 0x0 0x0 0x0.

2016-12-30T17:49:52.075Z cpu0:65555)ScsiDeviceIO: 2927: Cmd(0x4395009da480) 0x2a, CmdSN 0x91749 from world 182413 to dev "t10.ATA_____Crucial_CT525MX300SSD1__________________________163413BAA738" failed H:0x2 D:0x0 P:0x0 Invalid sense data: 0x0 0x0 0x0.

SSD broken?

I tried this: LINK

but not better

2016-12-30T19:58:21.330Z cpu0:65754)<6>ahci_scsi_abort: cmd 0x85 (0x4395055c1fc0), entering...

2016-12-30T19:58:21.330Z cpu0:65754)<7>ata2: ahci_port_reset, entering...

2016-12-30T19:58:21.330Z cpu0:65754)<6>ata2: reset in progress...

Tags (3)
0 Kudos
1 Solution

Accepted Solutions
jaytee75
Enthusiast
Enthusiast

I think, I got it.

Don't use AHCI in HP gen8 MicroServer! I switch to Raid Controller, and no more errors in dmesg.

Thanks for assist.

Smiley Happy

View solution in original post

0 Kudos
6 Replies
wila
Immortal
Immortal

Hi,

As you mention poor performance on ESXi6.5, have a look at this blog post:

http://www.nxhut.com/2016/11/fix-slow-disk-performance-vmwahci.html

--

Wil

| Author of Vimalin. The virtual machine Backup app for VMware Fusion, VMware Workstation and Player |
| More info at vimalin.com | Twitter @wilva
0 Kudos
jaytee75
Enthusiast
Enthusiast

Thanks for your link!

Are the Errors normal?

0 Kudos
jaytee75
Enthusiast
Enthusiast

I got sill troubles:

esxcli storage core adapter stats get

vmhba0

   Successful Commands: 1174605

   Blocks Read: 16466433

   Blocks Written: 229934646

   Read Operations: 477974

   Write Operations: 679705

   Reserve Operations: 3033

   Reservation Conflicts: 0

   Failed Commands: 46031

   Failed Blocks Read: 0

   Failed Blocks Written: 0

   Failed Read Operations: 3250

   Failed Write Operations: 37953

   Failed Reserve Operations: 0

   Total Splits: 0

   PAE Commands: 0

also on normal hdd. I checked the disk outside the server, and seems okay.

Also I have freeze of the client system. Most time if I am on ssh and change e.g. smb.conf.

Any hints for me?

I use a HP gen8 Microserver 8 GB RAM / AHCI mode.

0 Kudos
wila
Immortal
Immortal

Hi,

Sorry for the delay, got a bit busy.

No the errors are not normal.

For comparison on a system that I have which also has a Crucial SSD I get the following stats:

esxcli storage core adapter stats get

vmhba0

   Successful Commands: 506241886

   Blocks Read: 8145501146

   Blocks Written: 5260614764

   Read Operations: 330918587

   Write Operations: 172433241

   Reserve Operations: 1381695

   Reservation Conflicts: 0

   Failed Commands: 10552112

   Failed Blocks Read: 0

   Failed Blocks Written: 0

   Failed Read Operations: 0

   Failed Write Operations: 0

   Failed Reserve Operations: 0

   Total Splits: 0

   PAE Commands: 0

No Failed read/write operations at all.

Freezing can have other reasons, I have seen this for example when there was some DNS issues.

--

Wil

| Author of Vimalin. The virtual machine Backup app for VMware Fusion, VMware Workstation and Player |
| More info at vimalin.com | Twitter @wilva
0 Kudos
jaytee75
Enthusiast
Enthusiast

I think, I got it.

Don't use AHCI in HP gen8 MicroServer! I switch to Raid Controller, and no more errors in dmesg.

Thanks for assist.

Smiley Happy

0 Kudos
mphodge
Enthusiast
Enthusiast

I have the exact same problem with Crucial CT525MX300 drives in my HP Proliant Microserver Gen7, getting lots of these messages:

2017-01-11T08:54:01.662Z cpu0:65555)NMP: nmp_ThrottleLogForDevice:3617: Cmd 0x2a (0x4395009e6c00, 65545) to dev t10.ATA_____Crucial_CT525MX300SSD1__________________________164814DCA70F" on path "vmhba0:C0:T0:L0" Failed: H:0x2 D:0x0 P:0x0 Invalid sense $

Rather than switch from AHCI to RAID in the BIOS, I switched drivers from vmw_ahci to ahci and now I can see all 5 SATA controllers rather than just 1, and so far no more errors!

Not sure if there's any performance difference between the two BIOS settings though.

0 Kudos