
- Subscribe to RSS Feed
- Mark as New
- Mark as Read
- Bookmark
- Subscribe
- Printer Friendly Page
- Report Inappropriate Content
Farewell...
Our company said a farewell to Helsinki on Saturday... and received with open arms Istanbul.
A troubled relationship we had with Helsinki, a 6 month period since October full of major incidents, pain and rejection.
From day 1 we had a major incident with the cache and many performance related issues the following week.
Our application nodes were doubled...
Investigations led to the route cause being Helsinki's new AMB (Asynchronous Message Bus)
Our instance being a large footprint and heavily used pushed Helsinki to it's limit and it just fell over.
All the new functionality we had to turn off..
- User Presence
- Live forms
- Connect
- Service Portal
- History
Hello Istanbul...
6 months later the work that ServiceNow has done in making the product more operationally stable is clear to see.
Many of our reported issues that were tuned into problems have been fixed and included in Istanbul Patch 3.
We have worked with ServiceNow to include our instance when testing new versions prior to early access.
Sharing Our Upgrade Experience...
We have 24x7 Global Service Management Processes and taking down the platform causes a large disruption to the business.
For the first time in our ServiceNow history (Since 2012) we planned to upgrade the platform without taking it out of service.
During testing we found that mid server related orchestrations would fail due to conflicting versions so we removed our store and identity features from service only.
We also found we had to apply workarounds for the SLA engine and mid server polling time as there were some performance issues with Powershell Orchestration.
During the upgrade we observed;
- There were 200+ users on each node as the upgrade started
- Users were NOT kicked out of sessions
- Users were NOT required to log back in
- Browser display remained open until you clicked another button on the UI.
- Once clicked there would be a short pause as a new session was established on another node and we could carry on working
- The count of sessions moving to the upgraded nodes as users were doing the above
- There were no visible performance issues, forms were loading in acceptable times and all functionality up
A huge step forward in upgrade experience this time round and great improvement on stability and upgrade progress.
Looking forward to enabling all the functionality we had to turn off in October very soon!
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.