There are several limitations in GLPI 10 regarding cable & socket management:

1. You cannot define sockets with both front and rear contacts (this is required by patch panels)
2. Impact analysis does not consider cables (it considers connections only)
3. PDU plugs cannot be connected to assets (computers, network devices, etc)
4. Lines cannot be connect

So here is my proposal:

1. Sockets always have front & rear ports
2. Currently connections can only be established between
   • Network ports
   • USB devices and computers
     This needs to be expanded to allow
   • Network ports to socket (specifying if rear or front port is used)
   • Lines to network ports & sockets
   • Socket to socket (this is the case of connecting patch panels with rooms)
   • PDU plug to asset
   • PDU plug to PDU plug (to connect a power rail to a UPS plug)
3. Cables should be optional attributes of a connection (for those having numbered cables)
4. Impact analysis should show socket names, port numbers, and cable names (probably when clicking the connection line)

This can be done by other tools (such as netbox), but having it integrated in GLPI would be superb.

Also reported in forum https://forum.glpi-project.org/viewtopic.php?id=292630

Hello dear friends.
Many thanks to you for the development and improvement of the GLPI.

GLPI provides very convenient templates for solution and comments in tickets.

It is very convenient and simplifies and speeds up the work with tickets.
But it is very inconvenient that the template functions are not provided for sending a validation request.

It would be nice if you could add the ability to use templates for validation requests similar to follow-up or ticket solutions.
It would also be convenient if you could specify a specific approver for each validation request template.
We, as well as all GLPI users, would be very grateful if you could implement this template function in a validation request.

I created a profile so that technicians can only see tickets assigned to their respective group.
I have a rule that assigns the ticket to the correct group based on the category of the ticket and sets the status to new.
Tickets assigned to a group with the "new" status are visible to everyone because they have ability to assign, is that possible to hide the new tickets for this users ?

The thing is, when you share a link copying URL, you expect to share link with title:

I think can be created some interfaces to give data on each class, it has to look like this:

<meta property="og:title" content="Example Page">
<meta property="og:image" content="https://example.com/images/image.jpg">
<meta property="og:description" content="This is just an example page.">
<meta property="og:url" content="https://example.com/page.html">

And then link sharing will be looks like:

Add a different smtp server per entity in order to send email notification on différents domains.

For example: A common GLPI server for two companies with differents domains

Best regards

When opening a ticket and selecting the Requester, the "plus" button refers to "Import A User" when it would be more practical to be "Add User", so you can add a new user without leaving the ticket. 

Have the ability to manage in profile configuration so that standard users (who have the simplified profile interface) can see changes, problems, and projects. (ITIL object) Without being able to modify.

This idea suppose this another one Improve Search Options of IT Assets - Improvement #1 - Add denial of first conditional have been taken into account.

We all know that it is impossible to make certain queries with GLPI today. One of these queries could be the following

P11: Computers with processor Intel(R) Core(TM) i5
P12: Computers with operating system Microsoft Windows XP
P13: Computers with processor Intel(R) Core(TM) i3
P14: Computers with processor system Microsoft Windows 7
               Q11: ( P11 * P12) + ( P13 * P14)

and this question is impossible to make because Search Options doesn't allow to associate logic operator explicitly. That happens when we try to do something like that.

Example #1

We could think that A + B give us a set of 1584 computers but this is that happens

There are more examples of impossible queries but only one of them is enough to show them exists.

There are other situations where the queries are possible but they are tedious and difficult of build because you must to know Boole algebra. The following is an example of that.

P21: Computers with processor Intel(R) Core(TM) i5
P22: Computers with operating system Microsoft Windows XP
P23: Computers with processor system Microsoft Windows 7
                      Q21: P21 * !(P22 + P23)

This question is impossible to make for the same reason as before; Search Options doesn't allow to associate logic operator explicitly. That happens when we try to do something like that.

Example #2

We know that Q21 is a set of 123 computers but how can we do the query. unlike the previous situation we can transform Q21 into this equivalent query: P21 * !P22 * !P23.

And this is the current state of Search Options, although there are also situations where logical operators are associated in a capricious way giving curious results. :/

To resolve these problems I suggest the following:

This new operator '>', and its opposing party '<', will manage the associative search criterion in a explicit way. This is the idea:

This approach would allow two new features:

• Associate logical operators
• Create nested conditionals up to a fixed level

These would be the solutions to Example #1 and Example #2 with this new approach.

Example #1

Example #2

Simple, isn't it? ;)

Regards,
 

When a power supply is assigned to an phone (only possible from components, not from phone) it will not shown components)