Re: ESXi 5 Doesn't Like My LSI MegaRAID

EddieA · ‎08-29-2011

Hi,

I have an LSI MegaRAID SATA 300-8XLP PCI-E controller. This worked perfectly in ESXI 4 and 4.1.

I just tried upgrading to 5.0, and the device is shown in the Storage Adapters list, and correctly recognized. But the bottom box, has no devices or paths listed. This means I can't get to to my main VMFS store.

The device is still listed on the HCL as supported, so what can I do to get my main VMFS store back.

Cheers.

Dave_Mishchenko · ‎08-30-2011

A couple of things

1) if you run fdisk -l do you see any datastores (vmfs partitions)?

2) if you click Rescan All and then check the vmkernel log file, are any errors recorded regarding storage?

admin · ‎08-31-2011

can you try this.

list the devices in /vmfs/devices/disks if it list's the disk then can you partedUtil get <disk> and partedUtil getptbl <disk>.

resteves · ‎08-31-2011

Were you able to fix the problem? I'm going to buy this card, so i realy need to know if it's still compatible with ESXi 5.0.

EddieA · ‎08-31-2011

OK, I reinstalled 5 to a USB stick, so as to preserve my (recovered ) 4.1 system, which is on an IDE drive.

Here's the results from the questions above:

~ # fdisk -l

Disk /dev/disks/t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________: 81.9 GB, 81964302336 bytes
64 heads, 32 sectors/track, 78167 cylinders
Units = cylinders of 2048 * 512 = 1048576 bytes

                                                                               Device Boot      Start         End      Blocks Id System
/dev/disks/t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________p1             5       900    917504    5 Extended
/dev/disks/t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________p2           901      4995   4193280    6 FAT16
/dev/disks/t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________p3          4996     78168 74928384   fb VMFS
/dev/disks/t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________p4   *         1         4      4080    4 FAT16 <32M
/dev/disks/t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________p5             5       254    255984    6 FAT16
/dev/disks/t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________p6           255       504    255984    6 FAT16
/dev/disks/t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________p7           505       614    112624   fc VMKcore
/dev/disks/t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________p8           615       900    292848    6 FAT16

Partition table entries are not in disk order
Found valid GPT with protective MBR; using GPT

fdisk: Sorry, can't handle GPT partitions, use partedUtil
~ #
~ #
~ # ls /vmfs/devices/disks/
mpx.vmhba32:C0:T0:L0
mpx.vmhba32:C0:T0:L0:1
mpx.vmhba32:C0:T0:L0:5
mpx.vmhba32:C0:T0:L0:6
mpx.vmhba32:C0:T0:L0:7
mpx.vmhba32:C0:T0:L0:8
t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________
t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________:1
t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________:2
t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________:3
t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________:4
t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________:5
t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________:6
t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________:7
t10.ATA_____Maxtor_6Y080P0__________________________Y2473MLE____________:8
vml.0000000000766d68626133323a303a30
vml.0000000000766d68626133323a303a30:1
vml.0000000000766d68626133323a303a30:5
vml.0000000000766d68626133323a303a30:6
vml.0000000000766d68626133323a303a30:7
vml.0000000000766d68626133323a303a30:8
vml.010000000059323437334d4c452020202020202020202020204d6178746f72
vml.010000000059323437334d4c452020202020202020202020204d6178746f72:1
vml.010000000059323437334d4c452020202020202020202020204d6178746f72:2
vml.010000000059323437334d4c452020202020202020202020204d6178746f72:3
vml.010000000059323437334d4c452020202020202020202020204d6178746f72:4
vml.010000000059323437334d4c452020202020202020202020204d6178746f72:5
vml.010000000059323437334d4c452020202020202020202020204d6178746f72:6
vml.010000000059323437334d4c452020202020202020202020204d6178746f72:7
vml.010000000059323437334d4c452020202020202020202020204d6178746f72:8

~ #
~ #

~ # partedUtil getptbl "/vmfs/devices/disks/mpx.vmhba32:C0:T0:L0"
gpt
243 255 63 3913664
1 64 8191 C12A7328F81F11D2BA4B00A0C93EC93B systemPartition 128
5 8224 520191 EBD0A0A2B9E5443387C068B6B72699C7 linuxNative 0
6 520224 1032191 EBD0A0A2B9E5443387C068B6B72699C7 linuxNative 0
7 1032224 1257471 9D27538040AD11DBBF97000C2911D1B8 vmkDiagnostic 0
8 1257504 1843199 EBD0A0A2B9E5443387C068B6B72699C7 linuxNative 0

I can only see entries for the USB stick and the IDE drive. Nothing for the MegaRAID.

I've attached the tail of the /var/log/vmkernel.log, which shows a bunch of errors, but all for other devices, as the MegaRAID adapter is: vmhba3.

**UPDATE**

Further scouting in the log found this:

2011-08-31T18:37:08.268Z cpu0:2615)Loading module megaraid_mbox ...
2011-08-31T18:37:08.268Z cpu0:2615)Elf: 1862: module megaraid_mbox has license G
PL
2011-08-31T18:37:08.269Z cpu0:2615)module heap: Initial heap size: 8388608, max
heap size: 22339584
2011-08-31T18:37:08.269Z cpu0:2615)vmklnx_module_mempool_init: Mempool max 22339
584 being used for module: 33

2011-08-31T18:37:08.269Z cpu0:2615)vmk_MemPoolCreate passed for 2048 pages

2011-08-31T18:37:08.269Z cpu0:2615)module heap: using memType 2
2011-08-31T18:37:08.269Z cpu0:2615)module heap vmklnx_megaraid_mbox: creation succeeded. id = 0x410003000000
2011-08-31T18:37:08.269Z cpu0:2615)<6>megaraid cmm: 2.20.2.7 (Release Date: Sun Jul 16 00:01:03 EST 2006)
2011-08-31T18:37:08.269Z cpu0:2615)<6>megaraid: 2.20.5.1-6vmw (Release Date: Thu Nov 16 15:32:35 EST 2006)
2011-08-31T18:37:08.269Z cpu0:2615)PCI: driver megaraid is looking for devices
<6>megaraid: probe new device 0x1000:0x409:0x1000:0x3008: 2011-08-31T18:37:08.269Z cpu0:2615)bus 10:slot 14:func 0
2011-08-31T18:37:08.269Z cpu0:2615)DMA: 524: DMA Engine 'vmklnxpci-0:10:14.0' created.
2011-08-31T18:37:08.269Z cpu0:2615)DMA: 524: DMA Engine 'vmklnxpci-0:10:14.0' created.
2011-08-31T18:37:08.291Z cpu0:2615)IDT: 991: 0x98 <megaraid> sharable (entropy source), flags 0x10
2011-08-31T18:37:08.291Z cpu0:2615)VMK_VECTOR: 137: Added handler for shared vector 152, flags 0x10
2011-08-31T18:37:08.334Z cpu0:2615)<5>megaraid: fw version:[815C] bios version:[H432]
2011-08-31T18:37:08.334Z cpu0:2615)DMA: 524: DMA Engine 'vmklnxpci-0:10:14.0' created.
2011-08-31T18:37:08.334Z cpu0:2615)DMA: 569: DMA Engine 'vmklnxpci-0:10:14.0' destroyed.
2011-08-31T18:37:08.334Z cpu0:2615)LinPCI: LinuxPCI_DeviceIsPAECapable:532: PAE capable device at 0000:0a:0e.0
2011-08-31T18:37:08.334Z cpu0:2615)VMK_PCI: 684: Device 000:010:14.0 name: vmhba3
2011-08-31T18:37:08.334Z cpu0:2615)DMA: 524: DMA Engine 'vmhba3' created.
<6>scsi[7]: scanning scsi channel 0 [Phy 0]2011-08-31T18:37:08.334Z cpu0:2615) for non-raid devices
2011-08-31T18:37:08.624Z cpu3:2615)PCI: driver megaraid claimed device 0000:0a:0e.0
2011-08-31T18:37:08.624Z cpu3:2615)PCI: driver megaraid claimed 1 device
2011-08-31T18:37:08.624Z cpu3:2615)ScsiNpiv: 1525: GetInfo for adapter vmhba3, [0x4100090d2180], max_vports=0, vports_inuse=0, linktype=0, state=0, failreason=0, rv=-1, sts=bad0020
2011-08-31T18:37:08.624Z cpu3:2615)Mod: 4015: Initialization of megaraid_mbox succeeded with module ID 33.
2011-08-31T18:37:08.624Z cpu3:2615)megaraid_mbox loaded successfully.

Cheers.

EddieA · ‎09-07-2011

OK, so I guess no-one has any ideas why a fully supported device doesn't work: http://www.vmware.com/resources/compatibility/detail.php?deviceCategory=io&productid=12417&deviceCat...

I've ensured I'm on the latest BIOS for the card, and even rebuilt the logical drive. ESXi still sees the adapter, but no attached devices.

Is there any way to bring this to VMWare's notice, without forking out $250 bucks for a support ticket. Remember, this *is* a supported device.

Cheers.

Riplakish · ‎09-12-2011

"Me too" - I'm having problems with a number of 300-8XLPs.

I'm trying to dig up the process for invoking our service agreement. I'll let you know what I find, if anything.

chrisrode · ‎09-30-2011

Another "me too". I've got the Intel-branded version of this card, and have gotten the installer to recognize it and load the megaraid_mbox driver for it, but no joy seeing the storage attached to it.

Has anybody made any progress with this?

EddieA · ‎09-30-2011

From PMs I've been trading with Riplakish, it appears he has opened a support ticket with VMWare, and is working with them on this issue.

Cheers.

dik2048 · ‎10-16-2011

And me too. I have the same problrm with intel srcs28x controller.

Does anybody solve this prolem?

Olli11 · ‎11-09-2011

I've the same problem. The Controller is shown but there is no attached storage.

Why is the type of the controller SCSI? The LSI MegaRAID SATA 300-8XLP PCI-E controller is connected via SATA?

dik2048 · ‎11-09-2011

I have server motherboard intel s3000AH, controller intel srcs28x (analog LSI MegaRAID SATA 300-8XLP) is connected via PCI-E slot. It's SATA 8 ports controller. It works fine with ESX-i 4.1 but not recognise LUN under ESX-1 5.0. I've customized my installation (corrected OEM.tgz) and nothing :-(((

2011/11/9 Olli11 <communities-emailer@vmware.com<mailto:communities-emailer@vmware.com>>

VMware Communities<http://communities.vmware.com/index.jspa>

ESXi 5 Doesn't Like My LSI MegaRAID

reply from Olli11<http://communities.vmware.com/people/Olli11> in VMware ESXi 5 - View the full discussion<http://communities.vmware.com/message/1860259#1860259

trelum · ‎11-21-2011

I have this same issue. Controller shows up fine but does not list devices or paths. Was anyone ever able to solve this? I looked at the ESXi 5 patches that have been released up to this point but it doesn't appear that this issue has been solved that way. Thanks.

Riplakish · ‎11-21-2011

I had opened a support case via my work contract, but it got tossed because I'm not using 'supported hardware' throughout the entire system.

The need to tear apart a second working system to prove the controller driver isn't doing the right thing, coupled with having VMs consistently hang up for no apparent reason, requiring routine restarts and lost data on machines where I could get it to install, caused me to just stick ESXi 5.0 aside. The only real advantage I was going to gain was > 2TB data stores, and that simply isn't enough to jump through hoops to get it working when 4.1 is doing a good enough job.

dik2048 · ‎11-21-2011

Hello.

I did not resolve this problem. I asked in VMUG community Russian forum, and other sources and nothing ☹. I’ve been applied all esxi patches – problem is not resolved.

dik2048 · ‎12-27-2011

if set vShere clientt -> advanced settings -> disk ->Disk.UseReport:UN=0 - controller begins see LUN-s and assign device and path. but!!!!!

Host boots very very slow. (40 min). It hangs after load multiextent module, after vmfs3 module and during starting usb arbitrator service.

esxcfg-rescan vmhba2 (itd my controller) runs 20 min. Does anybody has any ideas? I'm emty. I'm yet ready to bye new controller or recompile driver megaraud_mbox. 😞

webtronyx · ‎01-05-2012

Did you ever get this resolved? I have been beating my head against the wall trying to get this fixed on a megaraid 84016e controller on esxi 5.

dik2048 · ‎01-05-2012

No. And I have beet my head. I will compile driver. I downloaded OSS from vmware for ESXi5 and Device driver development.kit. Im installing CentOS now. I spend a lot of time. Its sad. Its like wall responses from guru “use hardware from HCL”. My HW is from HCL. But its old. ☺

EddieA · ‎01-05-2012

I can confirm that dik2048's earlier suggestion does actually work. But, as pointed out, the boot process now takes much longer than before, which, in itself, would not be such an issue.

However, it makes *ALL* disk access, to both the LUN on the RAID, and another IDE drive inordinately long. Even for something simple like browsing a DataStore, it took minutes. Because of this I didn't even attempt to start any VMs.

And dik2048's last post also echoes what I pointed out earlier in this thread, the device in question *IS* in the HCL.

Cheers.

webtronyx · ‎01-21-2012

I actually contacted VMware support on the same issue and their end answer was the array was too large for esxi 5 to handle and it needs to be split up. I did split it up on the local mahcine to 4tb arrays and that still did not work.

I did get it to see 1 RAID stripe by making a single 4tb array but that was all that I could do with it.