Jeff Riechers Posted March 1, 2021 Share Posted March 1, 2021 So I deployed all the latest ESXi 7 updates to my demo lab and for some reason my ADC VPX doesn't seem to work anymore. If I login to the ADC console and attempt to ping out I get ping: sendto: Network is down I deployed a new ADC, set it up, same issue. Anyone else seen this with the new updates? Link to comment Share on other sites More sharing options...
Martin Meier Posted March 1, 2021 Share Posted March 1, 2021 I don´t have a ESXi 7.x in my LAB, but Citrix docs did not look so good: Support matrix and usage guidelines (citrix.com) The last as supported listed version is ESXi 7.0 Update 1 of 2020-10-06. Hopefully this is just not up to date. Link to comment Share on other sites More sharing options...
Jeff Riechers Posted March 1, 2021 Author Share Posted March 1, 2021 Well, I rolled back to 7.0 U1c and it works. Hopefully it is just something with 7.0 U1d with the Intel NUC I utilize. 1 Link to comment Share on other sites More sharing options...
Jeff Riechers Posted March 5, 2021 Author Share Posted March 5, 2021 Well, just found a client with 7.0 U1d and they have the same issue. And they are running on certified hardware, so looks like something with the ADC code and ESXi. Link to comment Share on other sites More sharing options...
Jeff Riechers Posted March 5, 2021 Author Share Posted March 5, 2021 Just received confirmation from support. This is a known issue and engineering is looking into it. So make sure all your vmware hosts hold off on U1d. Link to comment Share on other sites More sharing options...
Christian Pieper Posted March 8, 2021 Share Posted March 8, 2021 I have the same issue. Seems to work with E1000, but not with vmxnet3. Link to comment Share on other sites More sharing options...
Robert McArthur Posted March 9, 2021 Share Posted March 9, 2021 Seeing the same issue on update 1C with a VMXNET3 adapter with a new 13.0 76.29 VPX. Changed it back to an E1000 and it works fine. console shows "vmx0: device enable command failed" Anyone seeing the same thing? Link to comment Share on other sites More sharing options...
FUNDY MUTUAL Posted March 13, 2021 Share Posted March 13, 2021 This may not be the same issue, but my experience is that if the ESXi host is AMD based (i.e. HPE Proliant DL325 or DL385), processes running within NSVPX (i.e. the SSH shell) crash unless you manually roll the VM version back to machine hardware version 7 by editing the .vmx, then using "vim-cmd vmsvc/getallvms" to get the vmid of the NXVPX followed by "vim-cmd vmsvc/reload vmid". dcc Link to comment Share on other sites More sharing options...
Ben Jones1709162621 Posted March 16, 2021 Share Posted March 16, 2021 On 3/9/2021 at 4:32 PM, Robert McArthur said: Seeing the same issue on update 1C with a VMXNET3 adapter with a new 13.0 76.29 VPX. Changed it back to an E1000 and it works fine. console shows "vmx0: device enable command failed" Anyone seeing the same thing? Hi We're seeing exactly the same thing. Running a clean install of vSphere 7 U2 (obviously with vCenter 7 U2) on Dell R7525 (AMD) servers with new install of VPX 13.0.76.29. When changing to VMXNET3, we run into the same error and the IP address although allocated, doesn't respond and we cannot connect to it. When using the default E1000 NIC, we're able to connect to the NSIP and configure the VPX. We also tried playing with the VM versions, from 7.0 to 7.0U1 and 7.0U2 and these didn't change anything for us, so we've left them at default for the time being. Take a snapshot before upgrading the VM version, and you can revert back again if you want to try a different version. One thing to note, is that the only way we can reboot / shutdown / power cycle the current VPX, is to power off or reset it. It doesn't seem to respond to VMTools or within the shell or GUI. It issues the command, (we can see the command through the VMware Console) but the VPX doesn't react to the command. I do appreciate that this isn't a supported VMware platform yet, especially 7.0 U2 as it was only very recently released, and that the E1000 isn't ideal, but hopefully this will help others to get going. We're only using it for load balancing, so can get away with a less performant NIC. Regards Link to comment Share on other sites More sharing options...
FUNDY MUTUAL Posted March 16, 2021 Share Posted March 16, 2021 Humor me. Remove the NSVPX from inventory and roll your HW version back to 7 manually in the .vmx with vi. (virtualHW.version = "7") Then re-add to inventory and see what happens. I've found the AMD machines are fussy with these Citrix VMs.... Link to comment Share on other sites More sharing options...
FUNDY MUTUAL Posted March 16, 2021 Share Posted March 16, 2021 Never mind - I just tested it myself and rolling back to HW v7 doesn't help. :-) Link to comment Share on other sites More sharing options...
Ben Jones1709162621 Posted March 16, 2021 Share Posted March 16, 2021 Thanks for checking. We're still having problems even after the using E1000 NICs. Currently unable to configure HA, which is the simplest task. The VPX's won't sync, so using HA isn't possible, at least not for us. There's some real issues with support for the current versions of VMware. Link to comment Share on other sites More sharing options...
Jeff Riechers Posted March 17, 2021 Author Share Posted March 17, 2021 Not sure what happend. I just rebooted my ADC on my 1c lab and now it is not working again on this build. What's going on? Link to comment Share on other sites More sharing options...
Ben Jones1709162621 Posted March 18, 2021 Share Posted March 18, 2021 For those interested, a new update of 13.0 Build 76.31 was released a day or so ago ... I'll save you the trouble of upgrading however ... We deployed a new install rather than upgrade. Although there's a good amount of updates, the VMXNET3 adapter still doesn't work at all, and neither does VMTools, meaning we're still stuck with the E1000 and the only way to gracefully reboot or shutdown the VPX is to simply "Reset it" to reboot it, or "Power Off" through vCenter. Link to comment Share on other sites More sharing options...
Robert McArthur Posted March 19, 2021 Share Posted March 19, 2021 17 hours ago, Ben Jones1709162621 said: For those interested, a new update of 13.0 Build 76.31 was released a day or so ago ... I'll save you the trouble of upgrading however ... We deployed a new install rather than upgrade. Although there's a good amount of updates, the VMXNET3 adapter still doesn't work at all, and neither does VMTools, meaning we're still stuck with the E1000 and the only way to gracefully reboot or shutdown the VPX is to simply "Reset it" to reboot it, or "Power Off" through vCenter. Thanks for letting us know. We actually don't have the vmware tools power management issue you're experiencing. I'm able to issue a shutdown or reboot through the vsphere client which then occur normally. I'm also able to initiate reboots from the CLI of the netscaler. Link to comment Share on other sites More sharing options...
Robert McArthur Posted March 23, 2021 Share Posted March 23, 2021 Our case has been updated by Citrix who have confirmed this is a known issue with ADC and ESXi 7.0 update 1c (and presumably newer). The issue is expected to be fixed in version 13.0 79+ which should be released around the 30th March. Link to comment Share on other sites More sharing options...
Owe Imerslund-Kvisler1709162065 Posted April 6, 2021 Share Posted April 6, 2021 On 3/23/2021 at 10:31 AM, Robert McArthur said: Our case has been updated by Citrix who have confirmed this is a known issue with ADC and ESXi 7.0 update 1c (and presumably newer). The issue is expected to be fixed in version 13.0 79+ which should be released around the 30th March. Can confirm that 13.0 76.31 is not working with ESXi 7.0 update 1d (7.0.1, 17551050) either. Has anyone seen crashes or problems running E1000? Is it supported? Waiting for new firmware... Link to comment Share on other sites More sharing options...
Robert McArthur Posted April 6, 2021 Share Posted April 6, 2021 1 hour ago, Owe Imerslund-Kvisler1709162065 said: Can confirm that 13.0 76.31 is not working with ESXi 7.0 update 1d (7.0.1, 17551050) either. Has anyone seen crashes or problems running E1000? Is it supported? Waiting for new firmware... We're in the middle of a new deployment of 8 x VPX. We're using E1000 to enable us to carry on testing and have had no issues so far. Obviously, we'd like to have VMXNET adapters before going in to production though. Link to comment Share on other sites More sharing options...
Owe Imerslund-Kvisler1709162065 Posted April 8, 2021 Share Posted April 8, 2021 New firmware released, 13.0 79.64. Can't faind anything in the release notes about vmware compatibility though. Has anyone tried it? Link to comment Share on other sites More sharing options...
Michael Weishaar Posted April 8, 2021 Share Posted April 8, 2021 1 hour ago, Owe Imerslund-Kvisler1709162065 said: New firmware released, 13.0 79.64. Can't faind anything in the release notes about vmware compatibility though. Has anyone tried it? I upgraded a minute ago and it is working on VMware ESXi 7.0.2 Build 17630552 and vmxnet3. Best Regards, Michael Link to comment Share on other sites More sharing options...
Owe Imerslund-Kvisler1709162065 Posted April 8, 2021 Share Posted April 8, 2021 1 minute ago, Michael Weishaar said: I upgraded a minute ago and it is working on VMware ESXi 7.0.2 Build 17630552 and vmxnet3. Best Regards, Michael Awesome. Thank you for sharing Link to comment Share on other sites More sharing options...
Jeff Riechers Posted April 13, 2021 Author Share Posted April 13, 2021 Can confirm, this fixed it. Is 12.1 going to get fixed as well though? Link to comment Share on other sites More sharing options...
Morten Kallesoslashe Posted April 21, 2021 Share Posted April 21, 2021 On 4/13/2021 at 5:58 PM, Jeff Riechers1709152667 said: Can confirm, this fixed it. Is 12.1 going to get fixed as well though? great question :) lets cross fingers. Link to comment Share on other sites More sharing options...
Alex Henstra Posted April 28, 2021 Share Posted April 28, 2021 Disabling the DirectPath I/O in the Vmware VM setting for the VMXnet3 NIC, seems a workaround for 12.1 Link to comment Share on other sites More sharing options...
Patrick Adamczak Posted May 11, 2021 Share Posted May 11, 2021 On 4/28/2021 at 10:09 AM, Alex Henstra said: Disabling the DirectPath I/O in the Vmware VM setting for the VMXnet3 NIC, seems a workaround for 12.1 Unfortunately not 1 Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now