Hello!
I am using ESXi 6.7.0. Is it possible to increase file descriptor limit on the dedicated server?
ulimit -Hn1024
ulimit -Sn256
I am afraid I have exceeded it as our dedicated server produces quite weird network network problems that made us think it is caused by reaching the limit of file descriptors.
I have tested this with my CIM client software (Adaptec Storage Manager) and with curl and I can only authenticate to CIM using the root account with the administrator role assigned. I would like to use a service account with limited access. I have tried using a different role with all permissions assigned just to test but it stills fails 401 unauthorized. I have seen some posts about this being a bug from years ago but i am on ESXI 6.7u2 so I am confused if this is still a bug or not. Any help would be appreciated.
I ran into something today that I think could be a bug.
Trying to mount an NFS 4.1 share with :
esxcli storage nfs41 add -H 10.10.10.1 -s /data/nfstest -v nfstest
results in:
Unable to complete Sysinfo operation. Please see the VMkernel log file for more details.: Timeout
Running tcpdump from the NFS server shows that ESXi issues an arp request, but never follows up with communication.
I can ping the NFS server from the host and furthermore:
[root@esx:~] nc -z 10.10.10.1 2049
Connection to 10.10.10.1 2049 port [tcp/nfs] succeeded!
vmkernel.log shows the following relevant lines:
2020-05-12T19:06:44.598Z cpu3:2099718 opID=997a8ef8)World: 12458: VC opID esxcli-0f-b668 maps to vmkernel opID 997a8ef8
2020-05-12T19:06:44.598Z cpu3:2099718 opID=997a8ef8)NFS41: NFS41_VSIMountSet:403: Mount server: 10.10.10.1, port: 2049, path: /data/nfstest, label: nfstest, security: 1 user: , options: <none>
2020-05-12T19:06:44.598Z cpu3:2099718 opID=997a8ef8)StorageApdHandler: 976: APD Handle Created with lock[StorageApd-0x431d75a03540]
2020-05-12T19:06:44.598Z cpu3:2099718 opID=997a8ef8)NFS41: NFS41_ConnectionLookup:785: Created new connection for address tcp 10.10.10.1.8.1
2020-05-12T19:06:59.601Z cpu12:2107491)WARNING: SunRPC: 3936: fail all pending calls for client 0x430944c01370 IP 10.10.10.1.8.1 (socket disconnected)
2020-05-12T19:07:14.600Z cpu3:2099718 opID=997a8ef8)WARNING: NFS41: NFS41FSWaitForCluster:3702: Failed to wait for the cluster to be located: Timeout
2020-05-12T19:07:14.600Z cpu3:2099718 opID=997a8ef8)WARNING: NFS41: NFS41_FSMount:4814: NFS41FSDoMount failed: Timeout
2020-05-12T19:07:14.600Z cpu6:2098147)NFS41: NFS41_ConnectionRemove:1071: Connection: 0x431d75a042f0 [55] addr: tcp 10.10.10.1.8.1
2020-05-12T19:07:14.600Z cpu3:2099718 opID=997a8ef8)StorageApdHandler: 1062: Freeing APD handle 0x431d75a03540 []
2020-05-12T19:07:14.600Z cpu3:2099718 opID=997a8ef8)StorageApdHandler: 1146: APD Handle freed!
2020-05-12T19:07:14.600Z cpu3:2099718 opID=997a8ef8)WARNING: NFS41: NFS41_VSIMountSet:411: NFS41_FSMount failed: Timeout
2020-05-12T19:07:14.601Z cpu6:2098147)SunRPC: 1104: Destroying world 0x202863
Note the IP 10.10.10.1.8.1. Where did it get .8.1 from?? Between seeing the additional numbers on the address and not seeing communication (besides a single arp request) via tcpdump on the nfs server, it looks to me like NFS might be buggy on ESXi 7. Unless I'm missing something, but I have had good luck with NFS on previous versions of ESXi.
Updated before posting
I attempted using esxcli storage nfs add -H 10.10.10.1 -s /data/nfstest -v nfstest with a result of:
2020-05-12T19:14:50.383Z cpu12:2099721 opID=58765395)World: 12458: VC opID esxcli-4a-b6a0 maps to vmkernel opID 58765395
2020-05-12T19:14:50.383Z cpu12:2099721 opID=58765395)NFS: 162: Command: (mount) Server: (10.10.10.1) IP: (10.10.10.1) Path: (/data/nfstest) Label: (nfstest) Options: (None)
2020-05-12T19:14:50.383Z cpu12:2099721 opID=58765395)StorageApdHandler: 976: APD Handle 05893e1b-b8b24cee Created with lock[StorageApd-0x431d23c028b0]
2020-05-12T19:15:00.551Z cpu15:2099721 opID=58765395)SunRPC: 3306: Synchronous RPC abort for client 0x430944c01370 IP 10.10.10.1.0.111 proc 3 xid 0x5066cf32 attempt 1 of 3
2020-05-12T19:15:11.551Z cpu15:2099721 opID=58765395)SunRPC: 3306: Synchronous RPC abort for client 0x430944c01370 IP 10.10.10.1.0.111 proc 3 xid 0x5066cf32 attempt 2 of 3
2020-05-12T19:15:22.551Z cpu15:2099721 opID=58765395)SunRPC: 3306: Synchronous RPC abort for client 0x430944c01370 IP 10.10.10.1.0.111 proc 3 xid 0x5066cf32 attempt 3 of 3
2020-05-12T19:15:22.551Z cpu15:2099721 opID=58765395)SunRPC: 1104: Destroying world 0x202877
2020-05-12T19:15:22.551Z cpu15:2099721 opID=58765395)StorageApdHandler: 1062: Freeing APD handle 0x431d23c028b0 [05893e1b-b8b24cee]
2020-05-12T19:15:22.551Z cpu15:2099721 opID=58765395)StorageApdHandler: 1146: APD Handle freed!
2020-05-12T19:15:22.551Z cpu15:2099721 opID=58765395)NFS: 173: NFS mount 10.10.10.1:/data/nfstest failed: Unable to connect to NFS server.
In this one it shows IP 10.10.10.1.0.111. NFSv3 is enabled on the server side. Running tcpdump on the NFS server shows a similar lack of layer 3 communication after the arp requests.
NFS seems to load successfully on boot and my network interfaces are on the HCL.
2020-05-12T04:44:12.366Z cpu11:2097688)Activating Jumpstart plugin nfs.
2020-05-12T04:44:12.366Z cpu12:2097716)Activating Jumpstart plugin nfs41.
2020-05-12T04:44:12.448Z cpu23:2097649)Loading module nfsclient ...
2020-05-12T04:44:12.471Z cpu23:2097649)Elf: 2048: module nfsclient has license VMware
2020-05-12T04:44:12.474Z cpu23:2097649)FSS: 1153: Registered fs nfs, module 4e, fsTypeNum 0xb00f
2020-05-12T04:44:12.474Z cpu23:2097649)VProbe: 802: Loaded 3 static probes from: nfsclient
2020-05-12T04:44:12.474Z cpu23:2097649)Mod: 4997: Initialization of nfsclient succeeded with module ID 78.
2020-05-12T04:44:12.474Z cpu23:2097649)nfsclient loaded successfully.
2020-05-12T04:44:12.497Z cpu11:2097688)Jumpstart plugin nfs activated.
2020-05-12T04:44:12.516Z cpu23:2097649)Loading module nfs41client ...
2020-05-12T04:44:12.541Z cpu23:2097649)Elf: 2048: module nfs41client has license VMware
2020-05-12T04:44:12.549Z cpu23:2097649)VProbe: 802: Loaded 5 static probes from: nfs41client
2020-05-12T04:44:12.549Z cpu23:2097649)Mod: 4997: Initialization of nfs41client succeeded with module ID 79.
2020-05-12T04:44:12.549Z cpu23:2097649)nfs41client loaded successfully.
2020-05-12T04:44:12.573Z cpu12:2097716)Jumpstart plugin nfs41 activated.
2020-05-12T04:44:25.332Z cpu3:2097682)Activating Jumpstart plugin restore-nfs-volumes.
2020-05-12T04:44:25.351Z cpu3:2097682)Jumpstart plugin restore-nfs-volumes activated.
I believe you are onto something.
Cannot find a way to add the boot options...
Enter SHIFT + O when the ESXi boots up
I think this is what a.p. is suggesting (Shift + O):
Then appending ignoreHeadless=TRUE before pressing Enter to boot:
That is never presented. Once the post completes, it is either a blank screen or a single solid cursor.
I am able to export the DCUI and all of the Web UI works fine.
It is detected within ESXi and vCenter.
You beat me to it while I was collecting screenshots for my reply
Gotcha! I'm not versed well in the ways of the vCenter, but if you SSH to the host you may be able to use this to take care of it:
esxcli system settings kernel set -s ignoreHeadless -v TRUE
Hi opened a ticket with Broadcom, and this is the result:
"The host id is controlled by VMware. The controller and drive status pre 6.7 U3 was unknown
but VMware fixed it. That is about as good as it will get for now. VMware made some changes
for 6.7 which unfortunately affect the ways the controller and drives are detected.
You should use storcli and LSA as monitoring tools."
I have a NUC8I5BEH with StandAlone esxi 7(free version) installed on a USB-stick.
Every host-config is lost during and the host is returned to state of installation.
I have also tried to add a USB-Network-interface but it seems like connected USB-devices is not shown in vSphere.
CLI "lsusb" show all connected devices.
I guess it may is some variant of this case => https://serverfault.com/a/557950
How can I solve this?
That allowed me to apply the configuration change. Now I have a blank screen with a solid (non-flashing) white cursor.
I have tried both the HP and GA installations.
This is weird...
Is there a basic VGA mode?
try the autobackup.sh script and then check if state.tgz gets an updated timestamp - monitor the commandline forerror messages
Finally found it.
Had to set the BIOS to Legacy for video (not EFI). Now DCUI works perfectly. Weird.
Hurrah! That is indeed an odd one. Hopefully if anyone else comes across it they'll find your solution here.
Hi folks,
I am reasonably new to PowerShell - PowerCLI and attepting to export a VM from ESXi to OVA format. All of the required CLI syntax works with exception of "Export-VApp"; I get the following error below and looking for a resolution please.
Export-vApp PowerShell Error
Export-vApp : 13/05/2020 9:24:38 AM Export-VApp Current license or ESXi version prohibits execution of the requested operation.
At line:1 char:51
+ ... iance_v1200_dhcp_1" | Export-vApp -Destination "C:\Utils" -Format OVA
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ CategoryInfo : NotSpecified: (:) [Export-VApp], RestrictedVersion
+ FullyQualifiedErrorId : Client20_VappServiceImpl_ExportVApp_Error,VMware.VimAutomation.ViCore.Cmdlets.Commands.ExportVApp
My ESXi Environment:
Client version: 1.33.4
Client build number: 14093553
ESXi version:6.7.0
ESXi build number: 14320388
Thank you!
Thanks;
state.tgz is updated with new timestamp when I do a auto-backup.sh.
Some cli-output
[root@localhost:~] df -h
Filesystem Size Used Available Use% Mounted on
VMFS-6 931.2G 273.5G 657.7G 29% /vmfs/volumes/DataStore11 => Local storage
VMFS-6 1.8T 1.4G 1.8T 0% /vmfs/volumes/DataStore12 => Local storage
VMFS-5 1.2T 1.2T 29.0M 100% /vmfs/volumes/VM31 => iSCSI storage
VMFS-6 349.8G 27.7G 322.0G 8% /vmfs/volumes/VMSTORE32 => iSCSI storage
VMFS-6 3.6T 1.5G 3.6T 0% /vmfs/volumes/DS02VMStore30 => iSCSI storage
[root@localhost:~] lsusb
Bus 001 Device 001: ID 0e0f:8003 VMware, Inc. Root Hub
Bus 001 Device 002: ID 0951:1665 Kingston Technology Digital DataTraveler SE9 64GB => Bootdevice for eSXI7.=======> =======> Wrong size detected, is actually a DTSE9H/16GB-stick.
Bus 001 Device 003: ID 8087:0aaa Intel Corp.
Bus 001 Device 004: ID 13b1:0041 Linksys Gigabit Ethernet Adapter => Not working USB-NIC
[root@localhost:~] auto-backup.sh
Files /etc/vmware/dvsdata.db and /tmp/auto-backup.530430//etc/vmware/dvsdata.db differ
Saving current state in /bootbank
Clock updated.
Time: 04:32:36 Date: 05/13/2020 UTC
[root@localhost:~] ls -l /bootbank/
total 64
-rw-r--r-- 1 root root 96 May 12 18:45 jumpstrt.gz
-rw-r--r-- 1 root root 57876 May 13 04:32 state.tgz
[root@localhost:~]
I buy a dedicated server on Hetzner and setup VMware 7.x
Then I order one additional IP and /29 subnet.
I created a support ticket and say I want redirect my subnet to my additional IP to they.
Created a a vSwitch and Port on VMware.
Created a new Router Ubuntu 20.04 machine and edit /etc/networks/interfaces and /etc/sysctl.conf files.
Created a new Windows Server 2019 machine and used my new subnet but can't reach internet.
What is the problem?
Thanks again.
I follow these tutorials:
https://community.hetzner.com/tutorials/install-and-configure-routervm-for-vmware-vsphere
https://nickcharlton.net/posts/configuring-esxi-6-on-hetzner.html
Ubuntu config:
auto eth0
iface eth0 inet static
address 95.217.138.109
gateway 95.217.138.111
netmask 255.255.255.248
dns-nameservers 213.133.98.98 213.133.99.99
auto eth1
iface eth1 inet static
address 95.217.54.105
netmask 255.255.255.240
| 95.217.54.104 / 29 | 100 | 500 | 2 | ||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||
Update, and Solution
Since other clients have used NFS successfully on the network I didn't think twice about the switch being an issue. For my lab I am using an Aruba/HP 2530-24G Switch (J9776A). While looking at the switch configuration I came across an interesting line towards the top:
(Switch)# show config
...
encrypt-credentials
dos-filter
no cdp run
aruba-central disable
...
According to HP:
Denial-of-service (DoS) attack filtering: automatically filters and drops common DoS attack traffic types
Well this seemed interesting so I ran :
(Switch)# show dos-filter
Denial of Service packet filtering: Enabled
Maybe it's seeing NFS between the NFS server and esxi host as DoS traffic? I'll disable it and try to mount the NFS datastore again
(Switch)# no dos-filter
It worked!!! The esxi host was succesfully able to mount the NFS share.
I have no idea why the switch seemed to allow other NFS connections, and it doesn't appear to be configurable (it just 'automatically' filters things it doesn't like). I am also still curious about the IP 10.10.10.1.8.1 and IP 10.10.10.1.0.111 in vmkernel logs on the host, but for now I can sleep happily knowing that the two are talking.
Maybe the problem is VPN? Which one do you use? Not a pro, though I can recommend an article on cooltechzone, there's a list of different apps, maybe it might help. Good luck anyway.