Cluster System Info
The System Information pane gives you access to additional settings, as well as information about your cluster, server farm, and client versions.
Cluster Info
In the cluster info pane, you can see the "Product Name" where you can change the branding of your cluster. The "Assigned License Count" area displays licensing details, including the license edition, number of users, cluster ID, and Triofox server version. Here you can upgrade your environment to a higher edition by clicking on the shopping cart next to your edition. Also, you can view and copy the Cluster ID.
Server Farm
In the Server Farm you can set up your External DNS, configure the settings for the Email Service, get the Database Information and view the number of Worker Nodes.
Email Service
In the Triofox solution, there are many places where the user needs to be contacted by email. The email service is used to set up the SMTP email service to send the emails.
By default, it works out of box using the default email service with the Cluster Server's customer support email address as the sender.
It is recommended to set up the SMTP service to use your own SMTP service for sending emails.
If your SMTP service does not require authentication, you can enter a dummy email in the Authenticate User field.
Worker Node Count
Cluster Server Farm has two types of nodes. One is "Worker Node", and the other is "Web Nodes".
Cluster Worker Nodes
This type of node contains services such as the Web Browser Based File Manager, Storage Service Connectors, etc. Again, additional nodes can be added as the load increases. Since cache information resides on each node, users have an affinity for a single node once it is assigned. If the load balancer distributes users evenly across all worker nodes, the cache information can be present on all worker nodes.
Worker Node Settings
There are some settings that apply to all worker nodes. After clicking on the "Settings" icon, the "Advanced Settings" panel is displayed.
Always force SSL on Login
In a production environment, almost 100% of the time you will need to enable the "Always force SSL on Login" option. If this option is checked and Triofox detects that the incoming connection is HTTP, it will do a redirect to HTTPS. If you enable SSL, you must first set up an SSL certificate.
However, if you have SSL-offload, such that SSL is offloaded to a hardware appliance, and after that, the incoming connection is HTTP between the hardware appliance and Triofox. In this SSL-offload case, you will NOT check "Always force SSL on Login" because it will create an infinite redirect loop because the incoming connection is always HTTP as far as the Triofox Server is concerned.
Always force SSL for Native Clients
In a production environment, almost 100% of the time you will need to enable the "Always force SSL for Native Clients" option.
Especially in the case of SSL-Offload, you MUST check "Always force SSL for Native Clients". Otherwise, the Triofox Server may think that the incoming connection is HTTP, so it will continue to encourage the native clients (such as Windows clients) to use HTTP instead of HTTPS.
Note
On iOS devices, Application Transport Security may be enforced by the operating system, and HTTPS must be used for an iOS application to connect to the Cluster Server.
Disable worker-node load balance
If you have your own load balancer, you will disable worker-node load balancing. The Cluster Server has built-in node-affinity load balancing that can be done on a per-user basis. If you have your own load balancer, you can have session affinity or just round-robin, either is fine.
Note
How to add a worker node?Simply install the Cluster Server during the installation and point the Cluster Server to the same database. Once the Cluster Server worker node installation is complete, reboot. The web portal page appears and prompts you to add the worker node to the server farm.
Warning
What happens if you have changed the Cluster Server Host Name?In Windows Server 2012 and newer servers OS, a server that is newly provisioned is usually named similarly in the hostname format (WIN-ABCDEFG). Sometimes it is desirable to change the name in the Control Panel -> Systems. If the Cluster Server is already installed, changing the name will make the Cluster Server add itself again with the new name. So the next time you visit http://localhost on the Cluster Server after the server has been renamed, you will see that the worker node section contains both the node with the old name (which no longer exists) and the node with the new name (which is current and good). In this case, you simply need to remove the worker node with the old name.
Worker Node Properties
You may need to change the worker node properties when you set up SSL and the DNS name for the cluster.
Node Name
The Node Name needs to match the hostname of the worker node. If you rename the Windows hostname (NETBIOS name) of a worker node after installing the Cluster Server, it may happen that the Cluster Server displays a web page after rebooting, asking you to add the new worker node. In this case, you can add the new worker node and then delete the old worker node.
External URL
The External URL needs to match the external URL of the worker node. In a production environment, this is usually in the format https:// and contains the DNS name of the node.
The External URL is an important property for email templates. After the Cluster Server installation is complete, the dashboard displays the warning message "External DNS has not been configured for this worker node. Some functionality may not work properly. Config Now".
The moment that you have finalized on the External DNS name of the Cluster Server, you must come here and configure the External URL property for the Cluster Server.
Internal URL
The Internal URL is the internal URL of the node, usually in the format http://local-ip-address. In later Cluster Server builds, this property is hidden and does not need to be set any more.
Disable management functionality
You can create an internal facing worker node (that doesn't have an external URL) and allow management functionality only on this worker node. This is a security feature.
Edit Cloud Monitor Setting
Enable Storage Scan
Enables or disables storage scan on the worker node. There is a Cloud Monitor service on the Worker Node. The service performs background monitoring and scans the storage from time to time to correct quota calculation and perform other maintenance tasks.
Scan Starts Hour
Normally, you set the start time for the scan sometime in the early morning, around 1 AM.
Scan End Hour
Typically, you set the end time for the scan to be sometime in the morning, such as 8 AM, before everyone gets to work. The idea is to use the idle time (when people are not at work) for scanning.
Scan User Storage Every (n) Days
Typically, you can set this to every week or every other week. So a number between 7 and 15 is appropriate.
Enable Change Monitor
If you enable Change monitor, the attached local storage, e.g., storage from file server network share, will be monitored and notification of file changes will be reported to remotely connected clients. This is typically required if your users modify documents both directly from the backend attached network share and from the front-end cluster access clients.
Index External Storage
This setting indexes storage services added through the "Storage Manager". The index is written to the files table in the database.
Enable Storage Purge of Deleted User
When a user is deleted from the system, the user's home directory is not immediately removed. In many cases, you may not want to delete it at all. For example, a user is deleted from the Cluster Server, but the user can still use the files and folders directly from the network.
Process Background Task
Specifies whether this particular node should process background tasks.
Enable Change Monitor for Home Drive
When Active Directory Home Drive integration is enabled, this allows Cluster Server to monitor changes on the home drive and notify remote client agents that the files/folders have been changed.
Send daily scan email
When storage scan is enabled, a daily scan email is sent to the cluster administrator about the scan result.
Cluster Web Nodes
Note
For a small deployment, it is not necessary to have web nodes. You can go directly to worker nodes, since worker nodes are also web nodes by default.
The Account Management, Sign-in and Load Balancing services are installed on this physical (or virtual) machine. Depending on the load, you may need 1 to N such nodes. Typically, we recommend using 10+ worker nodes for each web front node. For small deployments, you can omit web front nodes and combine them into worker nodes. All the installation work is the same. If you do not need web front node, you do not need to assign them in Cluster Manager.
Example:
ACME Corporation deploys two web front nodes node1.acme.com and node2.acme.com. Each node runs a copy of the Cluster Server connected to the same SQL database.
ACME Corporation acquires a domain name (DNS) of cloud.acme.com, which is load balanced to node1.acme.com and node2.acme.com.
When users point their browsers to https://cloud.acme.com, they are directed to the login page of one of the nodes.
Note
NOTE 1: If you have hardware load balancing available, you do not need to use web nodes at all.NOTE 2: Windows 2012/R2 comes with Network Load Balancing (NLB). If you use NLB, you do not need web nodes at all.
If you have an existing load balancer, you can basically omit the web nodes.
Zones
The concept of a zone is to connect your worker nodes to the location of storage. When you think about zones, you will first think about the storage location.
For example, I have storage in LA, so I have a LA zone. I also have storage in NY, so I have a NY zone.
You can also have worker nodes from different zones and assign users to a specific zone. If the user's home directory is from the LA zone, the user must be assigned to the LA zone.
Client Versions
Windows Client
For Windows Client, Mac Client and Windows Server Agent, there is an automatic client update feature. Each upgrade package contains the updated clients. By clicking on the Publish button, the newer package can be published to the clients.
Each new Cluster Server upgrade includes the newer Windows client, Windows Server Agent, and Mac client. Cluster users can get the clients that are included in Cluster Server via the manual download. However, for existing users with clients already installed, these older clients are not automatically upgraded until the newer client packages are published.
Daily Upgrade Limit
This is a per worker node setting. For example, if you have 2 worker nodes and set the daily upgrade limit to 100, a maximum of 200 clients will be upgraded per day.
Apply to Users
This option is usually used for testing purposes before the client is published.
Do Not Apply to Users
This option is usually used for testing before the client is published and to exclude certain users.
Note
The Windows client out there has a process that runs as a Windows service in the background. The service checks in regular intervals of about 1-2 hours if a newer upgrade is available. Once a newer client package is published and discovered, the newer package is downloaded. However, if the client is still actively running, the replacement and upgrade will not occur until the client application is stopped and restarted. This usually happens when the user logs out of Windows or restarts their desktop.If the Windows client software is actively running, a message may appear in the system tray asking the user if they want to restart the client software and get the newer version.
Once a client is published for automatic client upgrade, you can use Unpublish to stop the automatic client upgrade.
Server Agent
The Windows Server Agent can be published separately for automatic upgrade.
Mac Client
The Mac client can be published separately for automatic upgrade.
Administrators
The "Administrators" option allows a Triofox administrator to change the cluster administrator's default email address, reset the password, and add additional administrators.
Cluster Branding
You can access the cluster branding under System Info.
In Cluster Branding, you can change the logo, bitmaps, and other branding-related information. There are two branding supports. One is the built-in self-service branding, which is fully controlled by the "Cluster Branding" settings in the "System Info". The other is the full branding service. Both rely on "Cluster Branding" to change the appearance of the web portal.
Built-in branding works with white-label clients, which upon the first connection to the cluster, will download the branding-related information and use the branding-related information. Compared to the full-branding service, full-branding clients burn artwork, logo bitmaps, and related information into the client binaries.
General
The General tab lets you specify the name and other settings as described below.
Product Name
Here you specify what you want to call the product. This is the name that users will see when they log in to either the web portal or client applications. To access the branding settings, click the branding icon (1), then click "EDIT" (2), and then change the setting you want (3). Do not forget to save your settings. You can also choose a color theme that you want your users to see when they log in to the portal. You can choose a color theme that matches your company's colors.
Home Page URL
This is the URL of your "Home Page" page.
Copyright Statement
This is the content of your "Copyright Statement".
Web Portal
You can find Web Portal section under Cluster Branding.
Note
In previous builds, the icons worked best when the icon files were on the same server and the icons were referenced by a relative link.For example, you can create a subfolder under the Cluster Server installation folder, such as the root/imagetest folder. The dimensions of all the icons for each setting in the web portal should match what is displayed for each setting. The branding of the icons and images require the icons and images have the same width/height as specified or the same aspect ratio if the resolution is higher.
In later builds, the icons used are what-you-see-is-what-you-get and you can upload those icon sets.
Application Icon
In the Web Portal section of the cluster branding, you can change the application icon. This is the image that appears next to the product name in the web portal.
Drive Icon
This is the icon used for the Triofox drive. For example, in the tree view of the web portal.
Logo URL & Login Page Left Image
Please follow the same steps for branding settings for "Login Background Image", "File Share Stamp Icon", "IOS Client App ID", "Login Page Note", "Change Password URL", "Tutorial Page URL".
Client Download
You can find Client Download section under Cluster Branding. You can also choose not to show the download link for some clients.
Mobile Clients Download Links
Once you have branded your own iOS client and/or Android client, you can point the download link to your own AppStore and Google Play.
Windows Client
You can find Windows Client section under Cluster Branding.
Here you can specify the URLs for the application icon and the drive icon. You can also enter your company name under "Manufacturer Name" along with the "Contact Info" email. You also have the option to create your own branded MSI Windows client here. You can also use your own code signing certificate to digitally sign the MSI package. The advantage of creating your own MSI client package is that when users download and install the Windows client you provide, they will see your company name along with your branding during the client installation.
The Windows client supports multiple languages. Some language packs are included and shipped with Triofox. If you want to run the Windows client under a different language, you can set the UI language there.
After clicking the "Edit" button to edit the branding information of the Windows client, you can specify the EULA (End User License Agreement) and Code Signing Certificate.
EULA
The input is in an RTF file format.
Code Signing Certificate
You can acquire a code signing certificate from your code signing certificate vendor. Most SSL providers also offer code signing certificates. Make sure that you use SHA 256 (SHA2) as the hash algorithm for your digital signing certificate.
If your Code Signing certificate is already installed, you can also use the option - Sign using cert in certificate store.
MAC Client
You can configure the MAC client and MAC client installation package branding under here.
Client Branding
Emails
There are many places in Cluster Manager where users need to be contacted via email. So, the Emails tab is used to set up the email templates used for contacting users via email.
Welcome Email for New Team User
The team user is a regular user in a cluster. This is the email template that is sent to the user when the user account is created.
Welcome Email for New Guest User
A guest user is a regular user in a tenant that does not have a home directory associated with it. Therefore, the guest user can only work in the files and folders shared by other regular users. This is the email template that is sent to the guest user when the guest user's account is provisioned.
Email for File/Folder Share
This is the email that is sent to a user when the user is about to receive file/folder shares.
Request a File
This is the email sent to a user when the user is about to receive an invitation to upload a file.
Notify external user that shared file changed
When a shared file/folder has been modified, this email is sent to the user who receives the shared files/folders.
Admin Reset User Password Email
This is the email that is sent to a user when their password is reset.
User Reset Password Email
This is the email that sent to a user when the user resets the password for himself/herself.
New Sing-in Action Email
This is the email notification sent to the user when the user logs in from a specific machine.
Settings
This is where the reply email address is set. Normally, the email will be sent using the SMTP service set. However, if the reply address is different, you can set it here.
Cluster Settings
Cluster Settings
Hide Login Failure Message
If this option is enabled, the "Login failed" message will be replaced by a very generic "Login failed" message. If the option is not enabled, a more meaningful login error can be issued, such as user-not-found, authentication-error and so on. This is a security feature if you do not want to reveal too much information so that hackers can guess the reason for the failed authentication.
Hide build number from login page
This controls the build number on the login page of the web portal.
Hide support button
This hides the floating support icon.
Hide "Forgot your password" link on login
This option is most often used when Active Directory integration is enabled. The user then needs to forget and change his password in the normal Active Directory way and not in the way Triofox provides. In this case, it is recommended to hide the "Forgot your password" link.
Don't retry when login failed
This option is often used when the Active Directory user has a low failed-count on lock-out policy. If the user's password is incorrect, a few retries may lock the user's Active Directory account. The retry feature can be used when there is no Active Directory lockout or when the number of lockouts is high.
Show "purge storage option" when delete user
By default, when a user is deleted, the user's home directory storage contents are not touched for later use or review. If it is desired to delete the user's contents when the user is deleted, this can show the purge option.
Don't send email notification to user when purge deleted content
When the user deletes files, they are not actually deleted immediately. The purge is asynchronous and scheduled for a later time. This setting controls the notification.
Don't send email notification to admin when purge deleted content
When the user deletes files, they are not actually deleted immediately. The purge is asynchronous and scheduled for a later time. This setting controls the notification to the administrator.
Retrieve avatar from third party service (i.e. Google)
This is a usability feature that allows the user's image to be retrieved from Google.
Hide file extension in web file browser
This setting hides the file extension.
Disable Windows Client Auto-Logon
This is a security feature. The result is that every time the Windows client is closed, and the user tries to log in the next time, it does not remember the login token and the user has to re-enter the credentials to log in.
Allow personal data tagging
Attach local folder using in place versioned folder
When synchronizing folders from remote PC/Mac to Triofox, using in place versioned folder will make the folder keep the same folder structure as the folder that is being uploaded. Otherwise, Triofox manages the folder contents on the server side in its own way.
Only allow access performance information from local host
Performance data can only be accessed from http://localhost and not from an external URL.
Performace and Throttling
Don't show file icon preview if file size is larger than (KB, 0-No icon preview)
This is used to control the generation of iconview thumbnail in the files and folders view of the web browser. Thumbnail generation takes CPU power from the Cluster Server. For large files, thumbnail generation can affect system performance. Therefore, it is recommended to limit the feature to a certain image size.
Cluster Wide Upload Bandwidth Limit (Per Worker Node, KB/Sec, 0-No Limit)
This limits upload bandwidth.
Cluster Wide Download Bandwidth Limit (Per Worker Node, KB/Sec)
This limits download bandwidth.
Size limit for folder download (MB, 0-no limit)
This is to prevent a user from downloading a very large folder and consuming all the resources of the Cluster Server.
Timeouts and Limits
Languages
This section sets up the web portal languages and the client application languages for Windows client. We have automated the translation and provide the resource files that you can use to localize the web portal and clients in the language of your choice.
Change Log
Keep file change log for n days
This is a cluster-wide retention policy for the file change log.
The file change log resides in the SQL database. For deployments that use SQL Express, there is a size limit on the database. In the deployment guide, there is an option to split the file change log into a MySQL database or into another SQL database. This option is usually used to keep the SQL database size small.
Note
After the Cluster Server has been running in production mode for a while, we recommend that you check the file change log database table and the file index table to see how large these tables are.
Email Address to Receive Cloud Monitor Messages
From time to time, the Cluster Monitor service may send an email about status and alerts.
Logging DB Connection String
This is to split the file change log, device table, file index table, and audit trace table out of the main database and into a secondary database. The secondary database can be a Microsoft SQL Server or a MySQL Community Server.
The Cluster Server database is divided into a core part and a logging part. The core part can store the DB connection string that connects to the secondary database. This setting used to be in the web.config file.
Application Manager
You can also configure web apps under the "Application Manager" tab in Cluster Settings. This allows users to edit documents using the web apps. The apps here apply only to web portal-based editing.
Once an application is activated, you can see the context menu entry in the web-based file and folder manager view.
Languages
Under "Languages" you can set the language of your choice.
Anti-Virus
Under "Anti-Virus" you can activate your own Anti-Virus.