CraigP

@samwoo Sorry if this is obvious, but in the function you posted you have the parameter called $customfields but then feed in the variable called $jsonObject instead. Is $jsonObject being set somewhere else? It looks like the whole function to me.

I was quite excited to finally see a Get Asset Information node for the BPM, so thank you for adding it. However I see it only searches by Asset ID so I'm curious what the intent is for how to best use it? Seems unlikely that one would know the asset ID without searching for the asset by it's name/tag in the first place. Using it to retrieve linked asset information seems difficult unless you can guarantee the Get Linked Assets is only returning one asset ID. My main use case for this was to check that the Asset Name/Tag provided by our analysts when issuing a device was correct before automating assignment etc, but they will only know the asset name/tag, they wouldn't know what ID the asset was assigned in Hornbill. The above would however be more elegantly solved by having a "Hornbill Asset Picker" option in task Capture Fields (like the existing group and user pickers), is this something we can expect down the road please? One assumes the reason for not being able to search on Name/Tag for Get Asset Information is because the node won't know what to do if multiple objects are returned, and thus it is not likely this will be an option?

Hi @Ann-MarieHolloway This has been a frustration for our organisation too. However this thread brought it to the forefront of my mind again and I had an idea that I'm now currently trialling. I'm not sure if you are using these for something else, but there are custom attribute fields on the user accounts. I've used one of these fields to store the group ID of the "default team" for users who have multiple teams. Now in the BPM where the owner is assigned, I've first included a "Get User Custom Attributes" node (SM > Application > Users) and then a decision node with a branch checking if the "default team" Attribute field is set. If it is set then it assigns to the user and the team using the group ID in the attribute field, otherwise it just assigns to the user as normal (which will default to their first team alphabetically). Obviously you need to make sure you don't put anything other than a group ID in that attribute field on any user accounts, and I suppose it's possible (as Steve pointed out above) that the team you have in the attribute field doesn't support the service the request is in, but for us that is not currently a problem because we don't have any users in multiple teams which support different services.

Just to confirm, the Advanced Request Task Completer role doesn't let you complete tasks assigned to other Teams, only users. It would be very helpful if it it did allow you to complete tasks assigned to Teams!

Funnily enough I've just had to do this for someone who has left our organisation. It seems you re-assign the tasks by ticking them in the list then using the box in the top right to find and select who you want to re-assign to. However I agree in thinking the UI is confusing, my first instinct was to do the same thing and click on the task title which navigated to my personal task view and displayed a blank task.

Great, thank you for the swift response @Steve G!

I'm investigating using iBridge to automate Azure processes. To start I'd simply like to be able to add/remove users from Azure AD groups, but I don't see this in the list, am I missing something? This thread suggests there used to at least be a "Users" subheading.

Hi @Estie, not sure if this helps, but I've used this statement in the Database Direct section to retrieve a user's active tasks in the past: SELECT h_obj_ref_urn,h_title,h_created_on,h_assigned_to,h_assigned_to_group FROM h_sys_tasks WHERE h_State = "1" and h_assigned_to = "urn:sys:user:<username>" The h_obj_ref_urn contains the associated request ID.

I'm having troubling finding the Hornbill Automation version of this too. I'm curious to know why the Hornbill Automation version is considered "much better in the long run"? Are iBridge automations not reliable?

I've just tried to create a "Function" under a "Department" and it seems this is also considered a "Functional group", despite the documentation suggesting otherwise. I can't add basic users, and the "Allow Task View" and "Allow Task Action" columns are visible on the "Associate Users" screen. Am I misunderstanding the documentation?

I seem to recall the variable option here defaults to looking for users if you don't provide the full urn, but I have used the variable option in the BPM to assign a human task to a group. The format is "urn:sys:group:<groupID>". I haven't tried with roles, but looking at the "Assigned to" field in the h_sys_tasks table suggests to me that it follows the same format (urn:sys:role:<role id>).

I noticed this on our instance too. Unfortunately one of the few "Raise on Behalf" requests we have is Long Term Absence, and it did result in some accounts being disabled that shouldn't have been! It was especially frustrating because as far as I can tell, the Update Customer node still returns "Success" when this happens. So this was happening even though I already had a failsafe decision node that checked the Update Customer outcome before it moved on in the process. It was only brought to my attention because our Service Desk had multiple reports of accounts being disabled this way. I included a task to manually update the customer for a little bit to keep an eye on them and I noticed it wasn't even happening every time, but couldn't work out a reason for this. I ended up just removing the customer swapping aspect of the request and having to update the rest of the process to accommodate. Apologies if I missed it somewhere, but shouldn't Hornbill notify their customers directly when a defect like this is discovered? Something like this could potentially cause major issues like our disabled accounts example above, or the wrong accounts getting software or access rights they shouldn't etc.

Yes, that has done the trick! Thank you Armando!

The description for this API call says "This operation is used to link one or more Assets to a Request". How do I go about feeding it multiple assets? I've tried separating them with commas, semi colons and spaces, but none of these were accepted.

From what I understand the bug was that if you had certain access rights you could re-assign anything you wanted, regardless of your team, and this was not intended. I can understand wanting to lock that up if not intentional (but obviously there should be a role/system setting that allows this because there will be situations where tasks need to be re-assigned where the assignees aren't available to do so - I still haven't had a response to my question about how Hornbill intend for their customers to handle this situation). But now it sounds like you're implying that people should not even be able to re-assign their own team's tasks to another team at all? What would the purpose of that even be, considering you can re-assign tasks assigned to you individually, and tasks assigned to a role you're in.

It wasn't just a bug fix though was it? The assignment UI was changed, and as such it is now even less intuitive than it was before. It is not obvious at all what someone needs to do to re-assign a call to a different team (it literally just looks like it is locked out because the drop down is greyed out), and even if someone manages to assign the task to themselves, then they're expected to know that they need to re-open the task and then assign it back to their team, otherwise the rest of the team loses sight of it in their task view.

@Berto2002 - just to confirm, in your screenshot, I'm assuming "Service Request Assignee" is a role, not a team? As far as I can tell if the task is currently assigned to a team then the team drop down (the one on the right) is always greyed out now (even if you are in said team). It is strange that tasks assigned to roles seem to behave differently from tasks assigned to teams when it comes to re-assigning. I tried adding myself to "Advanced Task Completer" SM role and a custom role I created with Update Tasks like you have there. However this still doesn't let me re-assign tasks assigned to other teams. However, I seem to be able to re-assign tasks assigned to roles I'm not in. I'm glad you've found something that works for you but one hopes Hornbill aren't expecting all their customers to have to update all their BPMs to use roles rather than teams. This is going on almost 3 months, why has there been no update?

Thanks for clarifying that it won't send a notification if you put it on hold yourself, that makes sense. To what extent does having the Request View open stop the notification from sending? Does the request itself need to be on the list you can see on screen, or if you just have a view that would include that request open on a tab somewhere in your browser would this stop the notification? However, I've tried putting it on hold for a few minutes and closing all my Service Manager tabs until it is due to come off hold, but I still haven't received the off-hold notification. Are there any system settings that would stop this type of notification?

I'm not sure who else uses this notification option, but I have ticked it on to test it myself. I set an Incident where I am owner to "On Hold" for 5 minutes but didn't receive any notifications when I put it on hold or when it went back in progress (no email nor Hornbill notification despite both being ticked). Just to confirm, if an incident comes off hold automatically once it has reached its "pause until" time, the owner should receive a notification if they have these notifications ticked? Is there anything else that needs to be configured in a certain way for this to work?

A colleague mentioned last week that they are no longer receiving email notifications for calls assigned to them coming off hold. They have the "Sub-status Change (Owner Notification)" ticked for both Hornbill and Email in their notification settings. I've checked the sent items in the mailbox and it appears to be sending other notifications (e.g. the most recent email in sent items at time of writing is an "Assignment (Team Notification)" sent two minutes ago). Has anyone else noticed this not working?

Thanks. As I mentioned in the original post in this thread, I did manage to get it to work after that post last year. The "relatively simple way to build the required JSON object" I found at the time was to run Questions:queryExec on an existing request of the type I was trying to raise and change the answers as required. It did work at the time, but now upon revisiting it I'm getting this error. Just to confirm, I've tried removing the questions parameter and it runs successfully (without populating the question form obviously), so it has to be the information/format of the information I'm providing there. I've also tried running DEV tools while raising a request in app as per that old thread, but found it raised the request using pc2LogProcapture instead, which seemingly requires a significantly more complex JSON object for its "payload" parameter. If there is a way to get the app to trigger ServiceRequests:logServiceRequest (or Incidents:logIncident) instead, please let me know.

Thanks Gareth. I don't believe the Service has a default BPM, but I have already set the bpmName parameter. For reference I have the following parameters set: requestType serviceId questions catalogId catalogName bpmName I've copied these details as they appear for the catalog item in the h_itsm_service_catalog table

On a side note, I've now noticed that every time I've run my script to test, despite returning an exception called "failedToLog" it has in fact still been logging blank requests on the system. This makes troubleshooting this quite frustrating.

Hi @Steve Giller Sorry but I don't think that has anything to do with my query. That parameter is optional (one assumes it is redundant to tell the "logServiceRequest" API that you want to log a service request), and the error I'm getting is referring to the entity type, not request type. I have now added the requestType parameter anyway, but as expected it has not changed the error. It seems that I've either provided the wrong h_entity_type (but again, I copied the question information returned by the questions queryexec API on an existing request, which returns "request" as the h_entity_type for each question), or the format in which I've provided the questions is wrong. I appreciate if this is not your area of expertise, but then please have you passed this on to a colleague who can answer my questions?

I'm attempting to raise a request using the logServiceRequest API call in PowerShell. I had managed to do this successfully when I was investigating this last year, but I am now getting the failedToLog error - "XmlWriter::add: Unsupported value for \"h_entity_type\"". This implies to me that either "request" is not the correct entity type (even though this is what is returned when I run Questions:queryExec on an existing request of the type I'm trying to raise), or I may need to provide the "questions" parameter with something other than JSON (perhaps XML?). Is this correct? Or if JSON is suitable for this, can someone please have a look at the below example JSON and tell me where I may be going wrong? [{ "h_entity_type": "request", "h_question_id": "h_custom_b", "h_question": "Origin of Request", "h_field_type": "radioset", "h_form_id": "requestOrigin", "h_language": "en-GB", "answer": "Automated Alert" }, { "h_entity_type": "request", "h_question_id": "field_Name", "h_question": "Name", "h_field_type": "text", "h_form_id": "CustomForm", "h_language": "en-GB", "answer": "TestName" }, { "h_entity_type": "request", "h_question_id": "field_Description", "h_question": "Description", "h_field_type": "textarea", "h_form_id": "CustomForm", "h_language": "en-GB", "answer": "TestDesc" }] Secondly, can someone please provide example JSON for the questionFieldMap parameter? The API documentation does not give an example of what this needs to contain.