Ssh_timeout being ignored? Deployment cancelling after five minutes regardless

Full Disclosure - I’m a newbie with packer!

Hi all,

I’m trying to create a script which will deploy a Debian 10 VM to a vCenter environment with no DHCP.

Fundamentally I have my initial build process working (Debian 10 is installed without any user interaction, public keys added, ssh enabled and open-vm-tools installed) as long as i run packer with either -debug or -on-error=ask.

When I don’t run packer with either of the above the script terminates exactly five minutes after the last key is pressed in the boot_command, at the “Waiting for IP” prompt. When run with -debug or -on-error=ask the script still terminate after 5 minutes; but waits for input from me… but if I let the VM continue what it’s doing, within a couple of minutes the VM is built, rebooted and waiting for a login.

My build process takes around 7 minutes. I have played with ssh_timeout, setting it to 10m and then to 30m… it doesn’t make any difference, the script still terminates after 5 minutes. I wondered (as a newbie) if I was tweaking the wrong variable, but having just set PACKER_LOG=1 and rerunning my packer script it still terminated at 5 minutes, but at least reported it knew it should be terminating at 30m

I’m pretty confident I’m doing something wrong, I just don’t know what! Either that or ssh_timeout is being ignored - but that seems unlikely else more people would be shouting surely?

Any assistance gratefully received!



2020/04/30 08:11:22 [INFO] (telemetry) Starting builder vsphere-iso
==> vsphere-iso: Creating VM…
==> vsphere-iso: Customizing hardware…
==> vsphere-iso: Mounting ISO images…
2020/04/30 08:11:31 packer.exe plugin: Creating CD-ROM on controller ‘&{{{} 200 0xc00063c860 0 } 0 }’ with iso ‘[DS1] ISOs/Linux/debian-10.2.0-amd64-netinst.iso’
2020/04/30 08:11:33 packer.exe plugin: No floppy files specified. Floppy disk will not be made.
==> vsphere-iso: Set boot order temporary…
==> vsphere-iso: Power on VM…
==> vsphere-iso: Waiting 15s for boot…
==> vsphere-iso: Typing boot command…
2020/04/30 08:11:50 packer.exe plugin: Special code ‘’ found, replacing with: CodeEscape
2020/04/30 08:11:51 packer.exe plugin: Waiting 1 second
2020/04/30 08:13:15 packer.exe plugin: Special code ‘’ found, replacing with: CodeReturnEnter
2020/04/30 08:13:15 packer.exe plugin: [INFO] Waiting for IP, up to total timeout: 30m0s, settle timeout: 5s
==> vsphere-iso: Waiting for IP…
Stalls here for 5minutes
==> vsphere-iso: Clear boot order…
==> vsphere-iso: Power off VM…
==> vsphere-iso: Destroying VM…
2020/04/30 08:18:19 [INFO] (telemetry) ending vsphere-iso
Build ‘vsphere-iso’ errored: Post EOF
2020/04/30 08:18:19 machine readable: error-count string{“1”}

==> Some builds didn’t complete successfully and had errors:
==> Some builds didn’t complete successfully and had errors:
2020/04/30 08:18:19 machine readable: vsphere-iso,error string{“Post EOF”}
==> Builds finished but no artifacts were created.
–> vsphere-iso: Post EOF
2020/04/30 08:18:19 [INFO] (telemetry) Finalizing.

==> Builds finished but no artifacts were created.
2020/04/30 08:18:20 waiting for all plugin processes to complete…
2020/04/30 08:18:20 C:\path\to\packer\packer\packer.exe: plugin process exited
2020/04/30 08:18:20 C:\path\to\packer\packer\packer.exe: plugin process exited

Packer version: 1.5.5 [go1.13.7 windows amd64]


Note, credentials for vcenter are being fed in from another config file and passed to packer with -var-file=. I’ve not shared that file for obvious reasons, and do not believe it’ll have any influence on the problem I’m experiencing.

"variables": {
	"version": "1.0",
	"Date": "17/04/2020",

	"vm_name": "test-vm",
	"CPUs": "2",		
	"cpu_cores": "2",
	"RAM": "2048",
	"disk_size": "20480",			

	"sshusername": "redacted",
	"sshpassword": "redacted",
"builders": [
		"type": "vsphere-iso",

		"vcenter_server": "{{user `vcenter_server`}}",
		"username": "{{user `username`}}",
		"password": "{{user `password`}}",
		"cluster": "{{user `cluster`}}",
		"host": "{{user `host`}}",
		"datastore": "{{user `ds`}}",

		"vm_name":"{{user `vm_name`}}",

		"guest_os_type": "debian10_64Guest",

		"shutdown_command": "sudo shutdown -P now",
		"ssh_port": 22,
		"ssh_username": "{{user `sshusername`}}",
		"ssh_password": "{{user `sshpassword`}}",
		"ssh_timeout": "30m",

		"CPUs":  "{{user `CPUs`}}",
		"cpu_cores": "{{user `cpu_cores`}}",
		"RAM": "{{user `RAM`}}",
		"RAM_reserve_all": true,

		"disk_controller_type":  "pvscsi",
		"disk_size": "{{user `disk_size`}}",
		"disk_thin_provisioned": true,

		"network_adapters" : [
				"network" : "Some Network",
				"network_card": "vmxnet3"
				"network" : "Some Network",
				"network_card": "vmxnet3"	

		"iso_paths": [
			"[DS1] ISOs/Linux/debian-10.2.0-amd64-netinst.iso"

		"boot_wait": "15s",

		"boot_command": [
			"install ",
			"debian-installer/locale string en_GB ",
			"debian-installer/language=en ",
			"debian-installer/country=GB ",
			"debian-installer/locale=en_GB.UTF-8 ",
			"clock-setup/utc=true ",
			"time/zone=GB/London ",
			"console-setup/ask_detect=false ",
			"keyboard-configuration/xkb-keymap=gb ",
			"preseed/url= ",
			"netcfg/choose_interface=ens192 ",
			"netcfg/disable_autoconfig=true ",
			"netcfg/get_ipaddress= ",
			"netcfg/get_netmask= ",
			"netcfg/get_gateway= ",
			"netcfg/get_nameservers= ",
			"netcfg/confirm_static=true ",
			"netcfg/get_hostname=test-vm ",
			"netcfg/ ",
"provisioners": [
		"type": "shell",
		"inline": ["ls /"]

In case anyone else runs into similar problems, the issue was I was accessing the vCenter server via a VPN, which also had a deep packet inspection firewall on it.

I didn’t bottom out exactly the underlying issue, but running packer on a bounce box behind the VPN/firewall completed successfully so definitively something networky going on.



1 Like