Hello,
On our VSAN 6.2 environment, with SAS SSD (Toshiba) and everything VSAN HCL Compliant, the vROPS Storage pack (v6.0.5 with VSAN Support) shows all the stats for the HDD's, and things like cache-hit-ratios for the SSD's but no Media wearout indicators for the SSD's.
The node's HBA's are LSI SAS9207-8i controllers. Again, everything is on the HCL. Any idea why the Wear-indicators don't work? It's standard S.M.A.R.T information.
Kind regards,
Steven Rodenburg
I looked up the controller and determined it was supported for pass-through. What type of SSD drives are you using? Is this a hybrid deployment or all-flash?
Doug Arcidino
Hello,
6 node setup, VSAN 6.2, Hybrid, 1 SSD per host of this type:
SanDisk SDLKOEDM-200G-5CA1
Device Type:SAS
Capacity: 200 GB
Interface Speed: 6 Gbps
Form Factor: 2.5"
Endurance: 3650 TBW
Endurance Class: Endurance Class C >=3650 TBW
Performance Class: Class E: 30,000-100,000 writes per second
Series: Optimus Ascend™
Flash Technology: MLC SSDVirtual SAN All Flash Caching Tier
Virtual SAN Hybrid Caching Tier
Virtual SAN All Flash Capacity Tier
ESXi 6.0 U2
ESXi 6.0 U1
ESXi 6.0
ESXi 5.5 U3
ESXi 5.5 U2
ESXi 5.5 U1
I'm using the same exact SSD for my hybrid deployment in my home lab. I also get no indicators but I am doing RAID0 and not pass through. This could be part of the issue that the RAID0 configuration doesn't allow this information to passed forward.
Bump, any update on this?
As I stated in my previous reply, the only thing I can say for certain is that I see this issue in environments where RAID0 is in use for passing the disks through an array controller. Do you have an environment using pass-through for the disks and you are still not showing the media wear out indicators? Since devices are passed through in ESXi as a RAID0 logical volume, ESXi doesn't even know the disk is local or an SSD and you have to mark it as such. That would signal to me that a lot of the disk information is being lost. This is something that should be brought up to your host or array controller manufacturer. Open a feature request(With Dell, HPE, Supermicro, etc) and ask them to write a firmware and driver package with pass-through capability and submit it and the appropriate hardware to VMware for testing and re-certification.
My disks are in pass-through modes.
I was checking via CLI and found that even though my drives are on the HCL they show VAAI is not supported on them. Could that be part of the problem? it's not gathering the information because the drive isn't enhanced with VAAI capabilities?
SMART also shows many NA
esxcli storage core device vaai status get -d naa.51402ec00
005cb46
naa.51402ec00005cb46
VAAI Plugin Name:
ATS Status: unsupported
Clone Status: unsupported
Zero Status: unsupported
Delete Status: unsupported
esxcli storage core device smart get -d naa.51402ec00005cb46
Parameter Value Threshold Worst
---------------------------- ----- --------- -----
Health Status OK N/A N/A
Media Wearout Indicator N/A N/A N/A
Write Error Count N/A N/A N/A
Read Error Count 200 2 200
Power-on Hours 99 0 99
Power Cycle Count N/A N/A N/A
Reallocated Sector Count 100 5 100
Raw Read Error Rate 200 2 200
Drive Temperature 70 0 50
Driver Rated Max Temperature N/A N/A N/A
Write Sectors TOT Count N/A N/A N/A
Read Sectors TOT Count N/A N/A N/A
Initial Bad Block Count N/A N/A N/A
What on earth does VAAI have to do with any of this...
<Constructive commentary mode>
VAAI has nothing to do with harddrives / flashdrives as such
Please read this: VMware vSphere Storage APIs Array Integration (VAAI)
</Constructive commentary mode>
That it does not work in RAID-0 mode is completely obvious.
But that is not what my environment uses. It's pass-through, as you identified and stated yourself already, so ESXi sees the SSD's as local and native.
In other words, it should work...