Berto2002

We have an incomplete asset list in Hornbill SM but insist that Assets are added when required for Change; so we need a fast way for Users to add Assets but we need them reviewed so they can be finalised as necessary by someone in that area. A solution to this would be to have a very basic PCF that asks for the Name, Serial and Operational State; and for that to go through a workflow that creates the Asset - to the person can continue to build their RFC but assigns the Request to the Configuration Manager for review and update as necessary. I have checked the Hornbill Automation piece and Assets appears under Requests and allows to Add Customer Assets to the Request, to Get Asset Information or to Update the Properties on existing Assets (which is close because that's useful). It's so very nearly there but nothing for Add New Asset?! At the moment, we think we need to give access to everyone to be able to create raw Assets and that feels a little too open. Thanks for any advice.

Thank you, I will see what I can do with this suggestion. However, do you also see that there is no need for this failure to be the case if only Service Manager would acknowledge the Expired Task and act as if it had expired while created?

Hi both. Thank you, this is indeed what I was missing. I can now differentiate the dates and the Notice works. What I am not sure about yet - but will test - is, if I alter the Scheduled Dates, what I have to do to have the Notice pick-up the new dates in case of Scheduling alterations. However, this issue has thrown-up that I think our timezone was incorrectly implemented on our system. The times shown in the Notice are in GMT which means they are displaying one hour behind the real time in the UK! I have raised a Premier Support ticket for this since messing with timezones might have unintended consequences.

Hello @Steven Boardman, I have seen this topic which helps me understand how I must do a Cloud automation to get a properly formatted date: The principle works but I cannot see how to get two dates through this feature. I tried to use one Cloud Automation for each date: But they both output the SAME OUTCOME and I cannot edit it to be different for each case: So the latter over-writes the former and thus, my Notice looks like this where both of the times are the same: Can you please clarify how I go about getting the TWO dates as you did in your example above on this post ()? Many thanks in advance, Berto2002

Hi Mary. Thanks for pointing that out. Yes, the task does appear in the Expired list, as below screenshot. So, the issue is that the Expiry of that task (presumably immediately after creation) has not triggered the BP to move on as it should (via the Decision with "No Match" route as above) to the "Email End of Change Window" node. However, this BP does move on in that manner when the Expiry occurs when the "Implementer: Start Implementation" task is already created. To cater for this case, I think I need to put-in additional node before that Task that checks the Expiry and if it's already passed, go straight to the Email node. Unless @Mary you know of another way to ensure this does not fail for this reason?

I have a Human Task that is set to expire when the Scheduled End Date of the Change Request is reached. The BP then should assess if no answer has been given and should move-on to the next step for a Change Manager review. This is a screenshot of the executed BP on a test Request which is awaiting the completion of the "Implementer: Start the Implementation": This is a shot of the CR itself which shows the BP did not create the Human Task it states it is waiting for (no Activities). So the CR is stuck and cannot move forward even though my BP provides a case for a Scheduled Date being passed (review and re-schedule). The issue appears to be created when the Scheduled End Date is already in the past. This seems to fail to create the Human Task rather then creating it and immediately expiring it. In short, the BP and the back-end of Service Manager app are not congruent. Although our teams are usually not silly enough to schedule something for the past - and we usually spot it in review - it is possible that it can happen and I think this is a bit of a flaw in the system: to allow it to break CRs right at the end of the process.

Hi @Steven Boardman. A little issuette here. I would have hoped our timezone was congruent but something in that config is posting the Notice time different from the Actual time; like a zone mismatch.

Steve, that is a great option and I've put that into our test flow. Do you know if there is a way to LOCK the fields for Scheduling Change so that beyond a certain point they cannot be edited (such as after authorisation)?

After our Change Requests are authorised, I want everyone who opens the record to see the focus is on the Scheduled Date and Time fields. Else the default is the Update/Comment field. This keeps the Change Window front of mind. How do I do this please? I can see how this is done with the Suspend option but I don't want it to be in suspense during this period.

Hi @Steve G, are you saying that there is no facility at the moment (either through the email aspect or Cloud Automation) to do enable an update to an existing conversation and you will now consider developing that facility in the Cloud Automation piece? Although not promising, do you have a typical time period for this type of thing? I.e. should I forget about it for now or be kinda hopeful for something later this summer?

I can use the email external function in a node to start a Teams conversation. What I REALLY want to be able to do is use Teams as the collaboration workspace and have subsequent updates during the lifecycle of my Change Request that ping updates to the SAME conversation within a Teams Channel. Is there any way to do this? This is what we want: The conversation started with a formatted entry which could be the one initiated by the email receipt from HBSM Our people then REPLY to that thread through the lifecycle to keep the threads for each CR separate. I want HBSM to add to this thread. It should be possible because every Teams conversation has a unique link as per second screenshot.

Victor. My test of this appears to work perfectly but it would be nice to know that this is the officially supported way of releasing to live.

Hi Victor, that sounds interesting. Perhaps we've missed this piece of education. I understand the concept here. A couple of Q's please: 1) Are there any articles referring to this in case there's more to know about it? I could not find reference in Wiki 2) It feels to me that I would be going live on an untested BP but are you saying this is the 'official' way to do this and that the download/upload process carries-over 100% of the the config and can be relied upon for this as the release mechanism? Thanks, Berto

I am afraid the way this works means the Rename option is almost completely pointless since there is almost never a time, after a process has been set into motion, that it will ever not have at least one ticket using the lifecycle; and therefore can never be renamed. Even in test we have loads of tickets so our test processes cannot be renamed either. It's bad. It should work that a new Request takes a snapshot of the BP at that time and works on that copy for it's lifecycle. If the main BP then alters (such as a name) that's fine, the next Request that comes along will grab the latest version/name. This facility is already in place with the versioning of the BPs in the editing window. As it stands, if I want to make a major change to a BP, I have to copy the Live BP and give it new (permanent) name. Then, when I have finished with the testing and want to go live, i am now FORCED to locate and edit all the Catalogue items that use the current live BP and edit them and re-publish them to use the new BP. Effect: I can never have a release system that means the live BP's are all called "[Business process Name] - Live" and all the test BPs are called "Test". I have to name my test BPs as if they are Live and then somehow tell everyone they are not! Really suggest this needs some re-thinking at HB. At least provide a one-click facility to find and address all the existing Requests preventing the renaming so we can clear the way. I can search for them but the links are not active so i have to manually open them one by one...

When our Hornbill Service Manager was put in place, our consultant set-up the PCF Questions to capture all the information like Implementation Plan, Rollback Plan, etc. But then we found out such information could not be edited later in the lifecycle. Our last internal Administrator then moved to capturing the information using Human Tasks for each 'Plan' and posting them to the timeline. But now we find that not only does this mean the parts of the plan are scattered down the Timeline but editing them is doing just that and not actually altering some underlying content that can then be used later. All we seek is a way for all the core fields relating to building the Change Request to be stored and editable through the lifecycle (i.e. if the Peer Reviewer asks for more info, or the Change Manager does or CAB does. They would then be fixed at the point of submitting for approval/authorisation. I cannot help but think that neither of the ways we've seen are very good. The requirement here is very simple and should be easy. How can the Requester be given a set of fields to use to build-out the details of a Change, editing when required? What facility are we missing in Service Manager for this?!

I couldn't resolve a ticket. It looked like there was an Assignee and I could not click "Assign" but Service Manager claims No Owner. I had to refresh the ticket before the Assignee disappeared and I could then assign again to resolve.

#Facepalm Steve, I've been shown as a rookie again and did not see that Variable. Thank you. I will try this. Any way to do the other option I suggested, for future reference? I.e. to set-up parallel processing with two streams - one with a Human Task and the other an Await Expiry node - wherein the completion of either node automatically completes the other node (i.e. over-rides the wait or completes the HT); and thus satisfies the end of parallel processing so I can trigger the second, escalation HT Or to allow parallel processing to operate a first past the post system and be satisfied to continue even with fewer than all parallel streams being satisfied?

I have a logic problem I cannot solve. Can you help, please? I want a Human Task that Service Desk can act upon if they receive Leavers' hardware returned early but to have that task 'disappear' and be replaced by a different Human Task (escalation) if the Line Manager does not return hardware by an expiry date derived from the PCF completed when the Request was logged. The logical ways I can conceive of this working but cannot work out how to configure are: to expire a Human Task based on a variable. This would then directly replace one HT with another after a period of time. There is an option to expire on a set period since creation of the Task but not refer to an external value as there is in Hornbill Automations or Expressions to set-up parallel processing with two streams - one with a Human Task and the other an Await Expiry node - wherein the completion of either node automatically completes the other node (i.e. over-rides the wait or completes the HT); and thus satisfies the end of parallel processing so I can trigger the second, escalation HT At present, I have had to be satisfied with having a wait step (with a Notice to Service Desk to wait, and suspend of all actions) and the Human Task for the action only appearing after expiry. Any help appreciated. Thank you, Rob

I skimmed them and then Harry disappeared. But I am being asked questions on the content so where do I find those updates again to read and pass-on?

Reference: IN00167847

OK thank you. We will raise a Premier Support request

Hi Victor. I have this basic flow: In the Human Task I have this Field Capture Then I map the data to customer_g: Then I Get Request Information and trigger an email. I am testing with only a few chars so should be within the tolerance for custom_g: "VARCHAR custom fields a - o each suitable for holding up to 255 characters of any type." The email then has that custom syntax in it and it doesn't ever give data; in either the summary or the body. Am I doing anything incorrectly? Cheers, Rob

Hi, I am having a similar issue. before I ask a stupid question, can you please point me to where on the Wiki I can learn about inserting what I call active/dynamic fields into the text? I am having trouble with the insert in yellow not working. The green one does work. If I can see a Wiki page which has all the references to this I may be able to work it out.

@Mohamed I need to tidy-up now so I'll go ahead and make the rest of the multi-ticket updates. We have fixed our workflow now so we won't get this recurring on Incident but I will be back to do the Service Requests in due course so if you need anything more from me, let me know. At present, I think there is something in the software that's stopping it...

I believe there is a Hornbill design flaw here. This is my thinking: When I put a request on hold, I get a box that asks me to enter a Reason and that has a visibility setting where I can select Customer. However, this DOES NOT send an email to the customer. However, when I place an normal update on the request using the Comments feature, and I select Customer, Hornbill DOES send an email: These two GUI features should invoke the same options for sending. My Business Case is that my Users want to receive emails for manifest updates (like going on hold) and my Service Desk do not want to have to enter an update in TWO places. My workaround is that I am going to disable the Reason field on the On Hold sub-status which will then ensure the Service Desk use the Comments to update the customer. Unfortunately this is a little bit of a productivity hit since they have to update the request in TWO places when putting it on hold. My request is that there be an option introduced to enable the email customer feature from the sub-status box (or it invokes the same send options as for Comments). Rob