# Development Tips and Tricks
We've collected a list of tips and tricks to optimize your local quickstart or Getting Started development environment. We invite you to ask questions, collaborate with the community, and share your own favorite practices over on the Entando forum (opens new window).
# Quickstart Management
Here are a few common questions about the quickstart environment. The quickstart environment uses Multipass to launch an Ubuntu VM, where K3s Kubernetes is then installed and from which Entando is deployed.
Q: How can I remove a quickstart environment?
A: If you want to completely remove the VM created by Multipass, you can use
multipass delete YOUR-VM-NAME (where the default YOUR-VM-NAME for a quickstart is
entando) and then
multipass purge to recover the resources. If you just want to shutdown Entando but keep the VM, you can use
multipass shell YOUR-VM-NAME to shell into the VM and then remove the namespace via
sudo kubectl delete namespace entando.
Q: What if the installation fails due to timeout?
A: A Docker Hub policy limiting download bandwidth may cause the quickstart installation to fail with timeout errors. The workaround is a two-step process.
Note: Use the ent CLI to send commands to Kubernetes from the host machine.
- Delete the
ent kubectl delete namespace entando
- Run the following command
ent quickstart "entando" "quickstart" --simple --debug=1 --yes --with-vm --release=v7.2.2
The namespace will be recreated, preserving the images already pulled, so it's unlikely the installation time will exceed the timeout threshold again.
Q: How can I shell into a Multipass VM?
multipass shell YOUR-VM-NAME. If you don't provide YOUR-VM-NAME, Multipass will use the default name
primary, and even launch it for you if it doesn't exist.
Q: How can I adjust the resources dedicated to my Multipass VM?
A: A quickstart VM installed via the Getting Started guide has a minimal configuration with respect to RAM, CPU, and storage. Additional resources may be needed if multiple bundles or microservices are installed. The VM resources can be modified using your local hypervisor or through Multipass itself (opens new window). Typically the VM will need to be shutdown before modifying its configuration.
Q: What do I need to do after restarting my laptop?
A: By default, Multipass is installed as a service and will restart automatically. If Multipass isn't running, you'll need to first initialize this service; then you can start your VM via
multipass start YOUR-VM-NAME. Kubernetes will launch automatically along with any installed pods, including Entando. It can take a few minutes for all of the pods to fully initialize, but you can use
sudo kubectl -n entando get pods --watch to observe the progress.
Q: How can I pause or idle my Entando instance?
A: You can pause with
multipass stop YOUR-VM-NAME, or idle with
multipass suspend YOUR-VM-NAME to preserve the VM state. You can then use
multipass start YOUR-VM-NAME to start the VM.
Q: What else can Multipass do?
A: You can run
multipass help or refer to the Multipass docs (opens new window) for more information on Multipass.
# Entando in Kubernetes
Q: How can I install a new copy of Entando into an existing VM?
A: By default, the quickstart installation deploys Kubernetes resources into a dedicated namespace called
entando. If you want to remove all of the resources in
entando, you can simply delete the namespace with
ent kubectl delete namespace entando. You can then recreate the namespace and reinstall the resources. Alternatively, you can achieve this with
ent quickstart --vm-reuse=true, but you'll need to set other
ent quickstart options, so check the
Q: How can I shell into a running pod or view its logs?
A: You can use the standard Kubernetes commands, e.g.
sudo kubectl exec -it YOUR-POD-NAME -c YOUR-CONTAINER-NAME -- bash or
sudo kubectl logs YOUR-POD-NAME YOUR-CONTAINER-NAME.
Q: What do I do if Entando doesn't fully initialize?
A: The most common cause is a networking problem. See the Network issues section below for details. If all else fails, reach out to the Entando team on Slack.
Q: What do I do if the AppBuilder is lacking its left menu?
A: The most common cause is a failure of the EntandoApp to initialize. This could be caused by networking issues (see the previous point) or configuration issues related to the database, Keycloak, etc. The
appbuilder-menu-bff-deployment will not be created until the EntandoApp installation is successful, and without that service, the left menu will be unable to load.
- Examine the status of the EntandoApp:
kubectl get EntandoApp
NAME PHASE OBSERVED GENERATION AGE quickstart error 1 3d
If the phase is not
successful, try restarting the installation process. This can be done by adding an annotation to the EntandoApp resource, e.g.,
- Create a file named
metadata: annotations: entando.org/processing-instruction: force
- Apply the patch to the EntandoApp:
kubectl patch EntandoApp quickstart --type merge --patch-file entandoapp-redeploy.yaml
The EntandoApp phase should change to
started and the Entando Operator will resume the installation process of the EntandoApp.
# Shared Servers
We recommend using Multipass to quickly spin up an Ubuntu VM to host a local Kubernetes cluster for test purposes. A local environment is often useful, but most teams utilize a shared Kubernetes cluster. This shared cluster is managed by an operations team, and installed either on-premise or with a cloud provider for full integration testing, CI/CD, DevOps, etc.
# Network Issues
A local Entando 7 quickstart installation (e.g. what you get if you follow the Getting Started guide) may use a set of local domain names to enable access to Entando services. Your IP address will vary, but may look something like this:
The base domain configured via the ingressHostName (e.g. in your entandoapp.yaml) is based on the IP address that is created during the initial VM installation. This domain is used to generate ingress routes to map incoming URLs to individual services. In production environments, there's generally a dedicated network layer to manage IPs/routing (both on premise and in the cloud), but this is not readily available in most local setups. Below are a couple of common issues that can prevent Entando from initializing in a local environment:
.nip.io isn't allowed
- This could be due to firewall settings or corporate security policies. The simplest workaround is to manually edit your /etc/hosts file and map the domain to the IP of your local virtual machine.
192.168.99.1 quickstart.192.168.99.1.nip.io 192.168.99.1 YOUR-APP.192.168.99.1.nip.io
- If you add microservices to your installation, you may need to add additional mappings for the new ingresses.
- To update files on Windows, see the Windows process below.
The IP address changed after the initial install
- Restarting a Windows computer can cause this (see Windows Hyper-V IP Changes below), and the workaround noted in the
.nip.io isn't allowedsection above also applies (i.e. update your /etc/hosts file). Simply update the IP address in the first column to use the current IP of your virtual machine.
# Windows Development
# Multipass loses control of VMs
Q: What do I do if Multipass cannot access my VMs?
A: The most common symptoms include an
IP=UNKNOWN entry when issuing a
multipass list, and when attempts to stop or shell into the VM consistently fail.
Internet Connection Sharing (ICS) is a Windows service that provides Internet connectivity to virtual machines, and its
hosts.ics file can occasionally get corrupted. Restarting the host laptop or desktop should remedy this, but a quicker and simpler fix is to shutdown any VMs using the hypervisor (Hyper-V or VirtualBox), remove the
hosts.ics file from
Windows/System32/drivers/etc using elevated privileges, and then restart the VM(s). You can examine the
hosts.ics file first to check if it is well-formed, with clean IP to VM-NAME mappings insteaad of spurious numbers or letters.
# Hyper-V IP changes
Q: My Entando installation stops working when I restart Windows. How can I fix this?
A: The basic issue is that Windows Hyper-V makes it difficult to set a static IP for a VM (see this forum post (opens new window) for details). As discussed above, Entando's ingress routes rely on a fixed IP address and will break if the IP address changes after initial installation. Here are a few options to solve this issue, short of modifying your router or network switch settings:
# Option 1: Single host routing
The simplest way to deal with the peculiarities of Hyper-V IP assignment is to avoid it, instead using Windows-specific mshome.net addresses. This allows you to access a VM with an address like
YOUR-VM-NAME.mshome.net. If you set up your enviroment using the Automatic Install instructions, then the ent CLI will select the single host option and the address will be
entando.mshome.net. You can accomplish the same thing yourself using the
ent quickstart script, but see
--help for the current set of options.
# Option 2: Manually update your hosts file
The next simplest option to re-enable external access to your cluster is to update your hosts file after each Windows restart.
You need two pieces of information for this workaround, and you'll also need administrator access.
- Determine the original IP used for your VM. This is included in the
ENTANDO_DEFAULT_ROUTING_SUFFIX, or you can find it included in the ingress names. Run
kubectl -n entando get ingressto see something like this:
NAME CLASS HOSTS default-sso-in-namespace-ingress <none> quickstart.192.168.235.100.nip.io quickstart-ingress <none> quickstart.192.168.235.100.nip.io
- Determine the current IP using
hostname -Iin the VM, or by running
multipass listfrom Windows:
$ multipass list Name State IPv4 Image primary Running 172.31.118.12 Ubuntu 18.04 LTS
- As a Windows administrator, edit your hosts file
(C:\Windows\System32\drivers\etc\hosts)to map any required URLs from the old IP to the new IP. This will bypass .nip.io lookups.
172.31.118.12 quickstart.192.168.235.100.nip.io 172.31.118.12 YOUR-APP.192.168.235.100.nip.io
- You should now be able to access your Entando URLs via the new IP. If your Entando installation stalled during startup, it should continue initializing as soon as the external address is functional again.
# Option 3: Add a Windows route
This option is initially a little more involved, but future repairs to your network settings can then be made very easily. You'll need to choose a static IP, configure a Windows route to map it to the Hyper-V interface, and claim the IP in the Ubuntu VM via a netplan entry.
When implementing this option for the first time, all steps must be executed before installing Entando. Subsequent Windows restarts require steps #1 and #2, only.
Determine an IP that is unused on your local network (e.g. via ping). The following steps assume that IP 192.168.99.1 is selected.
Determine the interface address to Hyper-V (e.g. 32 below). Use cmd
route printand find the Interface entry for Hyper-V:
Interface List 32...00 15 5d 86 45 20 ......Hyper-V Virtual Ethernet Adapter
- Using elevated privileges, add a persistent route to map your IP to the Hyper-V interface:
route -p add [YOUR-IP] mask 255.255.255.255 0.0.0.0 IF [HYPER-V-INTERFACE] route -p add 192.168.99.1 mask 255.255.255.255 0.0.0.0 IF 32
Verify the route was added via
route print 192.168.99.1. This command is useful after restart to check if the route needs to be created again.
Configure your VM to claim the same address. Shell into the VM using
winpty multipass shell YOUR-VM-NAME.
Change to the root user to make the following steps simpler:
Determine your network adapter name via
ip link, e.g. eth0. It's often second in the list after the loopback adapter.
ubuntu@primary:~$ ip link 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000 link/ether 00:15:5d:00:1a:0c brd ff:ff:ff:ff:ff:ff
Navigate to your netplan directory:
Create a netplan entry starting with 0 (so it's indexed and loaded first):
network: version: 2 renderer: networkd ethernets: YOUR-NETWORK-ADAPTER: dhcp4: no addresses: - YOUR-IP/24
network: version: 2 renderer: networkd ethernets: eth0: dhcp4: no addresses: - 192.168.99.1/24
Apply the changes with
From the VM, verify connectivity via
ping 192.168.99.1. You should receive a response and not a timeout.
(Optional) Run a Python server to verify you can access the VM from the host at
http://192.168.99.1:8000.It may take a minute or so before the server is ready.
python3 -m http.server 8000
- You should now be able to install Entando using the static IP. If your Entando installation stalled during startup, and was previously configured with a static IP, it should continue initializing as soon as the external address is functional again.
# Option 4: Reinstall Entando
We're including this option because it works and requires no additional configuration. If you work with Entando regularly, we recommend developing in a centralized and shared Kubernetes instance rather than running a full stack locally. If you require a local cluster, we recommend using option 1 or 2.
# Multipass with VirtualBox
Q: How do I run Multipass with VirtualBox?
A: Multipass supports the use of VirtualBox on Windows as an alternative to Hyper-V. Refer to the Multipass documentation for VirtualBox configuration instructions.
For Entando to work correctly with VirtualBox, you will need to add a port forwarding rule to access Entando from your host system.
- Create your VM within Multipass
- Go to the Oracle VM VirtualBox Manager to edit the
Networksettings for the VM
- Go to the
Advancedoptions and click
Port Forwarding Rules
- Add a new rule
Name: your choice
Host IP: leave this blank
Host Port: 80
Guest IP: leave this blank
Guest Port: 80
- Click OK
- Any requests to port 80 on your localhost should be forwarded to the VM.
- Use the IP of your host to configure the
ENTANDO_DEFAULT_ROUTING_SUFFIXin your YAML file, e.g.
192.168.64.25.nip.io. You must use the host ID and not the non-routable address identified from within the guest VM, e.g. 10.0.2.15.