This is continuation of story, which started last year in March. For following months we were fighting, fixing and adjusting this interface.
Here are some more issues which should have been considered at the beginning of this project, but everybody knows that the wisest you are after the facts;):
- there should be a mapping between workflows of Tickets in both of the systems (finally we have adjusted it after 4th trial). Some of the statuses were clustered together, some were mapped and simplified
- there should be exchange of attachments of Tickets between the systems (in Omnitracker finally proper script must have been created)
- To avoid duplicate comments – special set of regular expressions was used in Omnitracker to parse unparsable Worklog combination with Comments from ServiceNow
- and the best was left for the end:
- synchronization between ResponsibleGroups, CMDBs between both of the systems:
- of course it could have been done by simple import/export of data – but if the both systems are in totally different subnets and you have no possibility to send data via REST API, queues or even ftp then I must admit, it is very challenging. The solution in our case was: sending email from Omnitracker to ServiceNow support team every time unknown data was detected. Then this data was manually added to the system.
I still don’t know whether it is the end of this project – next part will be taken over by my team and I will take care of some more forward-looking technology.
Eventually it will be replaced by ServiceNow…and all of this interface will be deactivated.