Jump to content

JJack

Hornbill Users
  • Posts

    167
  • Joined

  • Last visited

  • Days Won

    1

Posts posted by JJack

  1. Hi,
    We have an update user import, using a csv file and goDb2HUserImport.
    This includes setting the user's manager to a value based on the csv file.
    The setting of the manager field involves a look up. If the input manager value is blank, the Hornbill account would stay as it is, ie become out of date.
    I'm not aware of any change I can make to get it to set it to a value based on the input file unless that's blank, in which case clear the existing value.
    "ProfileMapping":{
    "Manager":"{{.MGREMPNUM}}",

    "Manager":{
            "Action": "Both"
            , "Value": "{{.MGREMPNUM}}"    
            , "Options": {
                "GetStringFromValue": {
                    "Regex" : ""
                    , "Reverse": false
                }
                , "MatchAgainstDistinguishedName": false
                , "Search": {
                    "Enable": true
                    , "SearchField": "h_user_id"
                }
    Suggestions gratefully received. I'm trying to avoid a special separate extra run prior to our standard run just to clear values.

  2. Sorry for not being clear @Steve Giller

    I want to return, from the linked request to the parent call (the call that raised the linked request), an indicator of what has gone on within the linked request - eg whether a task there has failed or succeeded. I don't care much how I indicate this - it's just I can't see a way to get information out of the linked request back to the call that raised it.

    (Background: we want two things to happen at the same time, as putting them as sequential tasks in a single call builds in a delay. Hence one as a task, one in parallel as a linked call that also raises a task.
    However, we need to know the outcome of each of these tasks to decide what to do next.
    If we didn't need, (because of our ways of working and how our request views are set up), a parent call for each task so that they can be assigned to different teams, we could just have parallel tasks.
    )
     

  3.  Hi,

    Can anyone advise on the following? It's a different aspect of parallel processing.

    On 15/01/2024 at 12:49, Steve Giller said:

    This will wait for all linked Requests to be completed (as defined by the settings

    (from:

    )

    I have a task and a linked request within parallel processing nodes.
    It then has a 'Wait for linked requests completion'.

    Does 'Resolved linked requests are completed [string] : (Yes | No) "status.resolved" request status constitutes completion of linked request (Default: Yes).' MEAN THAT the linked call coming out of the parallel processing with a status of resolved is regarded as 'successful'?
    ie is that what makes 
    Hornbill Automations->Log extra call->outcome == success ? 

    What could an outcome of 'failed' come from?

    I want to know the success of the task and the success of the task completed (or some other indicator) within the linked call. Is this possible?
    I can't see it is possible to update anything in a linked call except the timeline.

    I'd like two different calls so they can be assigned to two separate teams, not just two tasks in one call.
    Hope makes sense.

     


  4. Hi,

    Part of one of our scheduled jobs failed last night, with an 'Unable to connect to Instance':
    2024/01/15 23:45:59 [MESSAGE] Import Defined Page Size 0
    2024/01/15 23:45:59 [MESSAGE] Overridden Page Size 10000
    2024/01/15 23:46:21 [ERROR] Unable to write Import History: Post "https://lon-p01-api.hornbill.com/icthelpme/xmlmc//data/?method=entityAddRecord": dial tcp 212.71.225.68:443: connectex: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.
    2024/01/15 23:46:21 [ERROR] Unable to Connect to Instance

    (That was the final line in the log file).

    The failed part was rerun and ran through okay:
    2024/01/16 03:21:28 [MESSAGE] Import Defined Page Size 0
    2024/01/16 03:21:29 [MESSAGE] Overridden Page Size 10000
    2024/01/16 03:21:29 [DEBUG] Processing Old Log Files Current Retention Set to: 0
    2024/01/16 03:21:29 [DEBUG] Connecting to Database Server: localhost
    2024/01/16 03:21:30 [WARN] [DATABASE] Connection Successful
    2024/01/16 03:21:30 [WARN] [DATABASE] Running database query for Customers. Please wait...

    Does the mention of 'Unable to connect to Instance' indicate a problem at your end or should I check with my colleagues? I can't see what 212.71 / lon-po1 etc is supposed to be.

    Thanks.

  5. Hi,

    Just wondering if there is an easy way to construct the link (eg to put in a capture) for a portal document in a library, which was created as an external link.
    The actual external target put in the capture directly will work, and I can also make up a link using the 'document id' of the external link, based on what I know works for an uploaded document.

    For uploaded documents, there is an 'Click for bookmarkable link to file' that will provide a usable link but I couldn't see anything similar for an external link document.

    I can see as well as a Document Id there is a 'Filename' but I don't know what that is/does.

    I've made a link up using https://live.hornbill.com/ + instanceid/ + docmanager/ + document/+ view/+ the document id but assume I'm missing the obvious.

    Thanks in advance,
    Jill.

     

  6. Hi @James Ainsworth
    Yes, we could do that, but I thought a bit tedious! I'd have to see if the process could find out what roles they already have, and if it mattered if I tried to remove ones they didn't have. Can't even tell if multiple roles can be specified against the 'Role' option.
    I'll have to experiment, or I might try your suggestion of an import configuration first 🙂

    • Like 1
  7. Hi,

    Thanks @Berto2002, I had missed these two posts in system admin and enhancements.

    For us too we are notified of an individual who needs to become a full user after their account has been created by an import; we cannot easily identify them at the time, as there is no definite area or other value that means they need such a role.
    I'll look into whether we could use a new data import configuration as @James Ainsworth suggests but limiting to an individual. This would at least bring it into service manager rather than handling as a scripted import.
    I'm not sure I see how I could get a process with Update User Account/Remove Role from User/Add Role to User to work, removing whatever current roles they had first, although I do quite like this idea as it potentially allows for more flexibility.

  8. Hi,
    I'm looking for some advice on the best way of changing someone from a basic to a full user, or vice versa, and changing the roles they have.
    I see ways of doing this manually or via an import and account updates but wanted something more automated available directly through service manager. For example, change user type, remove current roles, assign a new set of roles. It seems user templates can only be used when creating new accounts.
    The only way I can think of doing this seems tedious, and I thought it might be a common and simple requirement so hope I'm missing something obvious :- )

    I've seen an old post from @AndyHill but not sure what approach he was using and think I might be wanting to change more things.

     

     

  9. Thanks for the suggestion @Steve Giller. We currently have well over 50 calls specified, so not immediately keen on re-entering all the details.
    We also currently populate extra fields that do not seem obviously catered for in the Scheduled Request functionality, such as customer ids (not necessarily important), various analyst ids for the tickets to be assigned to, and several different due by times to set up within the activities that the process raises.

    We have not had any more errors since the 10th ('11th' above was my mistake), so it just seems to have occurred on three days.

    However, there were failures and 'security error: 1' must be set deliberately somewhere, so I would be interested to know what it means.

  10. Hi,

    1) We have a ticket showing in our 'Completed' processes list, where the call shows the last timeline entry as 'The status has been set to Closed by the Business Process Engine' on 01/09/2023.
    When will it disappear from this list? - I had understood this should be about 30 days.

    2) Where we have a new version of a process, can I assume it is okay to deactivate the previous version, even if it still has calls showing against it, as Running, Suspended, Completed or whatever?
    (I see from other posts probably not okay to delete an old process that still has calls against it, but not sure I fully understand why, as the process associated with a call is its very own version, including progress through it).

    Thanks, Jill.

  11. Thanks @Berto2002, I don't think for us it's the details of the particular item that are the problem. The entry in our csv file that failed yesterday has been set up fine today, with the call associated to its process, but a different entry has failed with the security error.
    I've looked back through the logs and in total we seem to have had three failures, today 11th, yesterday 10th, and an earlier one from the morning of the 3rd (and a different set of call details on that date). 

    The process is mentioned in the log file:
    2023/11/03 07:55:14 [DEBUG] Using Catalog ID 553 [Automatic tickets]
    2023/11/03 07:55:14 [DEBUG] Using BPM is---scheduled---automatic-tickets
    2023/11/03 07:55:14 [DEBUG] New Service Manager Request: SR00017386
    2023/11/03 07:55:14 [DEBUG] Request Social Object Ref Update Successful
    but then fails to get initiated:
    2023/11/03 07:55:30 [DEBUG] Spawning BPM [is---scheduled---automatic-tickets] for [SR00017386]
    2023/11/03 07:55:30 [ERROR] MethodResult not OK: Spawn BPM: Security Error: 1

    As the calls are raised by the RequestImport program, I don't have access to any process logs. 
    Might there be somewhere else I can look for more information?

    I haven't found anything on what 'security error: 1' might mean.

  12. Hi,
    Any suggestions on why a 'Spawn BPM: Security Error: 1' might have suddenly appeared?
    We have a csv file of 'ticket details' which we process via a scheduled run of goHornbillRequestImport.exe, controlled by a json file (Hornbill Service Manager Request Import Utility V1.12.0); neither of these files have changed since October but last night's run failed for one out of 32 tickets, where there were no failures on the previous night's run. A ticket was raised, but not associated with the process.
    The csv file includes service id and catalog item id, and is the same for all the records. The catalog item specifies a process.

    2023/11/09 07:55:43 Records Found: 32
    2023/11/09 07:55:43 Requests Logged: 32
    2023/11/09 07:55:43 Requests Skipped: 0
    2023/11/09 07:55:43 Requests with available BPM Workflows: 32
    2023/11/09 07:55:43 BPM Workflows Spawned: 31
    2023/11/09 07:55:43 BPM Workflows Associated to Requests: 31
    previous night:
    2023/11/08 07:55:44 Records Found: 32
    2023/11/08 07:55:44 Requests Logged: 32
    2023/11/08 07:55:44 Requests Skipped: 0
    2023/11/08 07:55:44 Requests with available BPM Workflows: 32
    2023/11/08 07:55:44 BPM Workflows Spawned: 32
    2023/11/08 07:55:44 BPM Workflows Associated to Requests: 32
    The error was:
    2023/11/09 07:55:31 [DEBUG] Spawning BPM [is---scheduled---automatic-tickets] for [SRnnnnn390]
    2023/11/09 07:55:31 [ERROR] MethodResult not OK: Spawn BPM: Security Error: 1

    A successful association shows as:
    2023/11/09 07:55:32 [DEBUG] Spawning BPM [is---scheduled---automatic-tickets] for [SRnnnnn391]
    2023/11/09 07:55:32 [DEBUG] BPM Spawned: BPM20231109000024
    2023/11/09 07:55:32 [DEBUG] BPM [BPM20231109000024] Associated to Request [SRnnnnn391]
    Thanks,
    Jill.

     

  13. We have a call, raised via self-service, where the owner has changed without this being generally visible in the timeline. I'm assuming this is because the visibility was limited. However I'm struggling to find what set-up allows the visibility to be limited to, say, Owner. (This would have been a manual change, not via a process).
    I can see for one call, that I raised not via self service, that Owner is an option if I go to reassign it but this is not the case for my other calls and  I don't understand the principle that's at play here. Can anyone help?

  14. I'd like to extract and act on information from one or more linked requests.
    I can see a business process can create, wait for, and also, as a group, update and resolve linked requests.
    Is this all or is there a way to get a linked request reference number that could be used as a manual request id to a Get Request Information node?

    I can't see a way to read timeline information either.

  15. Thanks for this thread.
    I had to read very carefully to realise that an error saying 'Stage CheckLIST Validation Failed' was referring to a stage checkPOINT.
    I do see that a checkpoint node says 'Set Check List Item Node' but it still seems to me that a checklist is one of those things you have to fill out before you can complete an activity - from the Manage Checklist within Task Options 😞 

  16. Thanks Steve. Sorry if I wasn't clear, it's two records in the incoming file with the same id, but differences in other fields, eg position status or manager name; there's only one person and we only want one Hornbill account.
    I'm not saying it's a good idea to have multiple records but wanted to know what Hornbill does when it comes across them.

  17. We have a user import (XMLMC SQL Import Utility V3.1.0) processing a csv file.
    Sometimes there is more than one record for the same person, identified by their user id.
    I thought it might process each in turn but from the log file I can only see references to it processing the second of the records.

    Does this make sense? and could this behaviour be controlled.

    Thanks, Jill.

×
×
  • Create New...