Gerry

@Ieuan Payne IPO This should be a simple case of using the iBridge, assuming you want to trigger the CI/CD pipeline from a BPM at a specific point in that process. You also have the option of triggering this sort of thing from an AutoTask/Custom Button, and if you are using ITOM, also from Runbooks... so yes there are lots of options. There is ready made integrations in iBridge so all you really need to do is point and click a few things and you will be good to go https://wiki.hornbill.com/index.php?title=Azure_DevOps Gerry

@Adrian Simpkins @Michael Sharp Thanks for your feedback also, and the suggestion. I think the problem is, what you are asking for leads to this... And I am sure you will see that if we made such a list, weather intended or not, people would then use it as a work list, that would shortly be followed by "oh can can we just have this one extra field"... and before you know it, we would be re-creating a request list of sorts. This is what I am nervous about, it would dilute the concentrate of functionality and create yet more stuff that users would need to understand. I think it would be far better to solve this issue in another way, ideally using something that is more geared to that need. If its a case that you need your techs to prioritise based on ageing, then it would seem sensible to set up the request list in that way. Another option might be to limit the history count in the notifications list precisely to stop them from using the list as a work list. I can't help thinking this question has come up because maybe there is a shortcoming in the way the request list works, or the way in which the system facilities bringing to the fore the important aging stuff, where age is playing a part. I would encourage you to think about that and see what might already be possible - and/or drive changes to Service Manager request list to get a better-aligned working practices. I would say though, please do keep the questions/suggestions coming, but I will do my best to try and expand on our thinking so it doe snot just come over as us saying No.. Thanks Gerry

@Michael Sharp @Adrian Simpkins There has been some internal discussions around this request and for now at least, this is not looking like something we would change. The Notification List is designed under the hood for its functionality a reliance on the fact that data is sorted in chronologically descending order. To change this would be to change the nature of its behaviour which could have quite aa knock-on effect. For example, notifications on timelines are detected on chronology, and not an "unread-flag" per post, this is because unlike email there is no such thing as opening a post to read it/mark it a read. Philosophically, what you are describing is using notifications list as a "work list" and really thats not what its for. Surely this is what the request list should be used for? Gerry

@Michael Sharp Its always worth mentioning that we have some customers that actively use two different login schemes, where they have some users logging in using SSO and others that log into Hornbill directly as a Hornbill user, so to them, the second option is not just a maintenance-only button. So while in your case it makes sense, for others it would not. Glad you got it sorted for your users Gerry

@Aaron Carter Not currently, I can tell you that it is substantially implemented already but still needs work around the parallel processing. You should not have to do anything, its been designed to work with existing BPM's without you having to make any changes. What it brings to the table is mostly a big maturity step-change. * A clean design based on the last 5 years of evolution. * Significant performance improvements * Better enablement for our application developers * Improvements in the way the BPM engine handles parallel processes and asynchronous behaviour. * Vastly improved per-process error logging and reporting. * A vastly improved design-rule-checking and error reporting. * Extended design to support some enterprise features around object lifecycle management. As for timing, it really should have been last year, but other priorites etc. Its already well baked so subject to the outstanding changes and some fairly rigorous testing it most definitely should see the light of day in the next 6-8 weeks I would have thought. Gerry

@Aaron Carter Thanks for your post. There is a stream of work where we are looking at a re-work of the current BPM engine, and some work around parallel processing is happening. I am not sure about this particular use case, although from memory I think there is something in the design of the current implementation that makes this a limitation. I expect BPM 2.0 (which is the very imaginative internal name for this work) will address this. For now though, its a bit of a limitiation Gerry

@Michael Sharp I could be wrong here, but I am sure the login pages are fully translatable so really, you can customise the buttons to anything you like. Gerry

@Chris Winship Hi Chris, As I read your requirement, from what I can read, I think you are wanting to use Tasks as small tickets. In other words, you want the ability to open a task, update the task (notes) by which I think you mean like the timeline on the request but just for that individual task and ultimately complete that task? We do get asked to add "ticket-like" functionality to tasks but we see tasks as "a human task" and a task as an instruction to complete a specific task. If more interaction that that is required, then we would see tickets (like incidents, problems etc..) would be the right alternative to create rather than tasks. I guess what I am saying is we don't want tasks to become tickets in their own right. How you are using service requests is exactly right, like a "container" for the running fulfilment processes and a top-level view, that makes sense. So I wanted to ask, why not use Incidents (or other ticket types) rather than tasks? Gerry

@Kelvin Thanks, have PM'd you Gerry

@Paul Alexander Thanks for the note, and Happy New Year to you and yours also. Hope you and the team have had a great Christmas. Yes ITOM is out of Technical Preview and is in Early Adopter stag. I will PM you an EA pricing matrix and we can perhaps have a chat. Gerry

@samwoo Ok I see what you are asking for I think. Let me first clarify, all content in Hornbill is considered "secure content" that is, there are access mechanisms to control access to documents, images and other content, and access is generally bound to a user session. In other words, there is no default or public access to almost any content, at the point you access a document, for example when you are viewing a list of documents being presented to you in Document Manager, at the point of you clicking the link to view/download the system will verify that you have access through whichever security mechanisms are protecting that content, and then issue an access token, which our serves will recognise and serve the content of the access token authorises them to do so. That is the essence of how the security model works. Now what I think you are asking for is a way of managing content such as images., documents, files and other stuff, and then create share/publically accessible links which you can subsequently use in other content such as web pages, wiki markup etc... Is that what you are looking for? I think what you are looking for is a Digital Asset Management (DAM) solution - or at least that class of functionality, tools/platforms like these: https://www.capterra.co.uk/sem/compare/directory/30203/digital-asset-management/software Is that the sort of thing you are looking for? That does fall into the category of a different application, its certainly something that the Hornbill platform could support, its a form of reimagine of Document Manager is how I would see it, but I guess at this point there is very limited demand for such a capability, it would be quite a lot of engineering effort to build such an application and given the plethora of cloud-based file-sharing solutions that exist out there already, many of which are free to use, it does not appear commercially viable at this time for Hornbill to build such an application. Given there is such a diverse assortment of file sharing applications already out there, including things like o365, Google Docs, Sharepoint and the list goes on - what do you see Hornbill brining to the table? I guess I am asking what are your motivations when you look to Hornbill for a solution to the file sharing need? (I am just curious about why Hornbill should be considering building such a solution... which is why I am asking that particular question) Thanks, Gerry

@samwoo Thanks for the clarifications. For static content like documents, XLS and general text documents, Document Manager does this reasonably well already, it allows you to maintain versions, organise with tags and provide descriptions etc... FAQ's also let you include static images, and a link to media such as Youtube or Vimeo video content. With regards to *storing video* the storage is really only a small part of the equation, you also need to deal with streaming, and that means you have to deal with encoding/transcoding when you upload video to put it into a streamable format, and then you need the web server infrastructure that can understand and manage the video streaming requests. This collectively is known as Video on Demand VoD and is essentially what YouTube and Vimeo do. Having VoD capability on our platform is something I have looked at previously, there is clearly a need for secure video hosting/streaming which is what a platform like Hornbill could offer over and above what something like YouTube offers. The problem with this is, its a very expensive service to provide, storage an bandwidth are huge, any CDN provider will be charging for both storage and streaming bandwidth and overall I very much doubt there is enough demand in our customer base, who would be also willing to pay for such a service, and so its not made any priority at this time, I think Video-On-Demand is probably out of scope for any type of Servicedesk or CRM solution. In terms of the static content though, I think you can already achieve what you want? Gerry

Hi @samwoo As its now possible to share documents from Document Manager, what are the nature of the images? Are these images you want to include in FAQ's? I ask because its also possible to add images to FAQ's so with these two things, can this meet your need? Gerry

@AndyGilly I am late to the table here with a response, but just to round this out, you probably are aware that this has now been implemented and is Live as of a couple of weeks ago. Please post back if you have seen it/tried it/like it (or maybe not) Gerry

I *think* (but I am not 100 sure) there is a way in a custom expression to reference a specific task outcome using the Node ID. I would need to get someone who knows how to do this for sure to confirm, but I am sure that in the execution context we keep task outcomes against their specific Node ID. Its just knowing how to reference the outcomes (or even knowing of that task has been invoked and completed yet, for example the flow might not have got there in the process yet, then you inadvertently try to reference it by its ID, and it will error, most likely in a way that we could not handle very well. This is why the node ID's are generally hidden. So probably not officially supported, but I expect its possible. Maybe we need to consider a way where we can do this in a more supportable manor. The background to why this has not already been done is, outcomes were originally intended to be flow navigation pointers, but it seems some people do use them as a fast way to collect some basic information. Gerry

Hello Adnan, I expect these particular users are Hornbill users (i.e. not basic users), so for them they probably have their default view set to the News Feed, which you can do from the user profile dropdown menu at the top right of the screen Gerry

Thanks everyone, it seems the fix is confirmed. Once again, my sincerest apologies for the inconvenience caused, I am glad we finally got to the bottom of the issue. Thank you all for your patience and support. Gerry

All, As a wrap up to this, here is the best I can do to explain the problem and give you some form of RCA. So part of the solution to the improvements we have made to SSO was to maintain backwards compatibility with customers existing SSO profiles. In order to achieve this what we were doing was setting a cookie client-side and then using the PHP endpoint to essentially do a re-direct. For reasons we still do not understand, for some users, and for no obvious reason we can find, the browser would sometimes *not send* the cookie to the PHP endpoint even though it had all the right security settings and options. It seems to be specific to Chrome more than other browsers... and we are even more stumped by the fact that clearing the application cookies actually resolved the issue, albeit temporarily in most cases.... its all very odd. To resolve this issue we have changed the PHP redirect code so it no longer needs that cookie, but instead does a zoneinfo lookup to get the information required. I apologise for the inconvenience this has caused, for all who were affected. We were honestly stumped and could not re-create the issue ourselves, all of our testing did not show the issue either. We will of course continue to monitor this just to make sure its doing exqctly what we expect but it looks like we have finally found a fix. Thanks again to @matthew.payne for being kind enough to let us observe the problem first hand, from that we were able to confirm what we had theorised about, which really shortcut the list of things we *could have tried* so ultimately led to the speedy fix this afternoon. If anyone else who was having this problem can confirm they are now ok I would appreciate that, thank you to everyone else for reporting the issue and answering the questions above, all of which helped us better understand what we were dealing with. Gerry

@matthew.payne Thanks for the confirmation, thats good to know. Hopefully this is now also working for everyone else too Gerry

@matthew.payne et'al We have now applied that hot-fix, we believe this solves the problem. Can you please check that you are no longer seeing the issue please? Thanks, Gerry

@Dave Longley Its SaaS so everyone gets the benefit of the same HotFix. Gerry

@matthew.payne Just a quick word of thanks for joining us on a call and allowing us to interactively diagnose the problem. We confirmed that one of Trev's theories seems to be the issue, it relates somehow to how the browser is dealing with the redirect cookie, its odd behaviour and strange how it only impacts some users. In any case, seeing it first hand was tremendously helpful and we have a way forward now. We are looking at a workable solution now and will hot-fix before the end of today, I will post back here when that has been done. Thank you (and 'all) for your patience, this is not a problem we were anticipating, or up until just a view minutes ago did not understand. Special thanks to Matthew for making the time for us. Gerry

@matthew.payne Hi Matt, Would you mind if one of our devs sets up a call with you? Can you do today? Any time you would like to suggest, ideally before 5PM if you can do it? Thanks, Gerry

@HHH @matthew.payne @AlexOnTheHill @LifeOfJonny @RobW Hi All, we have been trying to track this problem down for a few days now. It seems that most customers are not seeing this issue, but a small number of you are. What we are unclear os, is what the extent of this issue is, and what patterns that might emerge from this line of questioning. So I have a bit of an explanation for you and then some questions, if you can all read and answer the questions it would really help. So in the more recent releases we have changed the way SAML requests are handled. Previously, a SAML response (Assertion) from your iDP (Active Directory for most) would redirect to code in PHP on the front-end servers, which would make an API call to your instance via an API. This was inefficient and we have been systematically removing the use of PHP from the front end servers. To that end, we have implemented a new API endpoint which directly handles the SAML Assertion response. However, to maintain backwards compatibility with your existing SSO profiles (so you do not have to re-create them as we roll out these changes) we have left the original PHP endpoint in place, but instead of processing the SAML assertion and calling the API, it simply tells the browser it should re-direct to the new endpoint with a 307-moved permanently redirect response. As I say, this is working for the vast majorly of customers and has been for the last couple of weeks. We cannot see any technical reason for this, and have added more logging to try and understand what is happening, and its also very odd that once you clear your cookies this starts to work again. What we need is some clarity around the circumstances when this issue occurs. If you could each answer the following this would be really helpful for us. * When the issue occurs, is the only way to make it work for you is by you clearing the cookies? * When the problem occurs, you press the login button, and it redirects you back to the login page. If you press the login button again, does it just keep on re-directing back to the login page? * How many of your users are affected by this? is it all of your users? * Are you using the Service Portal? And are you seeing any issues there? * When you clear the cookies, and login successfully, do you see the problem again the next day? * If the answer to the above is yes, when you have logged in successfully, if you logout do you see the problem again? or does it work? * Same as above, but when you log in successfully after clearing the cookies, if you close the browser and then open it again, do you see the same issue? or are you able to log in? * How re-creatable is this issue, and if you can create this at will, are you willing to jump on a Zoom call and share you screen with one of our devs so we can investigate further with you? We are sorry this remains a problem, we are very keen to fix this, but with the information we have so far we are pretty stumped at this point, any held would be much appreciated. Gerry

@Jeremy Quick question here, why are you asking for someone to identify their line manager, you should be able to look that up from their profile? Gerry