we recently buy a Dell R640 server with ESXi
and one Broadcom NetXtreme BCM5720 dual port Gigabit Ethernet
and one Broadcom BCM57416 NetXtreme-E dual port 10GBASE-T
i create a Vswitch0 with the two gigabit ethernet card for management
i create a Vswitch1 with the two 10GBASE-T card for all virtuals machines (actually just one vm)
firmware ok, driver ok, ESXi version ok, linkspeed auto-negotiation enable
everything seems to be ok but i can't ping my only virtual machine
thx for your ideas
For sure FAULTY FW for the Broadcom NICs. Dont use the Dell FW Complete Bundle.... pick the latest Broadcom NetExtreme package from dell.com/support and upgrade NIC FW trough iDRAC. Downgrade will also solve the problem. If you call Dell support they will tell you.
Same happend on HPE Proliant Gen.10 but there we got incomming Traffic but trafficlos and interrupts on outgouing.
If you are using the 214.x.xxx.x firmware version, there are major bugs with it affecting ESXi.
It appears Dell has released updated firmware:
Broadcom NetXtreme-E Network Device Firmware 21.4 | Driver Details | Dell US
I have HPE servers using the same chip and their 214.x.xxx.x firmware, and after a couple of weeks HPE has acknowledged there are issues, but they have not released a fix and have advised I revert to 212.x.xxx.x firmware, which I already tried succcesfully on my own. In my case, the hosts appear to be fine for a short time, but soon drop off the network randomly, especially when trying to send data.
I can confirm the BUGs in HPE Nic FW as well because one of my customer tried to upgrade FW by it self and got all problems you have described. We downgrade the Nic FW to the old version and all was fine again. It was some hassle because vMotion does not work anymore and we need to cable the unused 1GbE nics on the hosts.
Were you able to find that older 212 firmware without HPE support? I can't find the older 212 versions... Just the new.
Glad I was not alone with this shitty issue. I found the firmware here: https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX-7a372c2f702d42298dda63d1a3#tab3
I searched on the specific firmware listed on VMWare's compat list, crazy that's the only way to find it at HPE...
New HP firmware posted a few days ago.
Has anyone tried it yet?
NVM Image version 214.0.224000
Notes don't mention the issues in this thread about disconnections.
It's also optional.
I have applied both new firmware and driver to two of our effected hosts. Will be testing for several days before putting into production. I have had an open ticket with HPE for over a month regarding this issue. They stated this release will address it.
I'm running into the same issue. After aplying the SPP 2019.09 to one of my HPE DL380 Gen10 the host lost connectivity to storage on the Broadcom 10GbE NIC.
SPP 2019.09 install Firmware 214.0.203000.
Can you confirm that 214.0.224000 firmware is working?
Firmware 214.0.224000 seems to be working fine. I didn't see any issues after installing this firmware by now.
Hi - I have a M620 running ESXi6.5 and latest Dell BIOS 2.8.0
I downloaded the latest firmware for M620 Esxi6.5 from
Support for PowerEdge M620 | Drivers & Downloads | Dell US
The file is a .BIN format
When I tried to upload VIA DRAC I get an error:
RAC0613: The uploaded file is invalid. Make sure to upload the required file in the correct format and retry the operation.
How do I apply this update ??
Sounds weird but you have to download the FW Update in the Windows Format and not Linux or BIN or $whatever. You can upgrade any FW trough iDRAC by uploading the Windows package.
Thank You for the info, the firmware update solved the my problem on my two DL380 G10.
There is also a new SPP from HPE released on 06.01.2020.
I was doubtful if this was true.
Here is link, Solved: Poweredge BIOS update for ESX 6.7 - Dell Community