Have you ever wondered what actually happens during your instance when it is being upgraded? If so, please read on to learn about some of the high level activities and how they may impact you.

THE INSTANCE UPGRADE PROCESS

Records and record creation will not be impact during the upgrade process. Only Scripts and the Schema will be modified during the upgrade.

The upgrade progress is managed by one node. The nodes are first upgraded and, once upgraded, trigger the upgrade for the Schema and Database. In addition, there is parallel upgrade activity within the instance on the other nodes. Once the nodes have finished upgrading, they are restarted thereby resetting connections. During this resetting period, there may be minimal connection impact to active users (see below).

After the nodes have been upgraded, there is a delta between the upgraded nodes and the database version which takes slightly longer than upgrading the nodes and is a subsequent activity after the nodes have been upgraded. The database upgrade will then be completed whereby the nodes and database will be the same version.

The MID Server will continue to run during the database upgrade but discovery schedules will not be running. However, although no additional work will be directed to the MID, it will complete jobs already within the queue. Once the instance has been upgraded, the MID Servers will start to upgrade to match the current version running on the Instance. The MID server will power-cycle and discovery schedules will not be ran during this period. This typically takes a few minutes to complete.

In addition, only “Upgrade Safe” scheduled jobs will run during the database upgrade (identified by the “upgrade_safe” flag). Any custom sys_triggers will be marked as not being upgrade safe and will be ran following the completion of the upgrade.

Although users will be minimally impacted during the upgrade, it is ServiceNow’s recommendation to limit access as this allows full resources to be allocated to the upgrade process.

HOW ARE MY USERS IMPACTED?

During an upgrade, users can still log into the environment and work as usual. There will be minimal impact and user connection will be reset as the node the user is currently assigned is finished upgrading and reset.

There is no impact to users creating records or to existing records created.

WHAT ABOUT MY DATA?

Currently, there is not an option for an ad-hoc backup. However, ServiceNow does take backups for Primary and Secondary databases (where available). This includes 4 weekly full backups with delta backup occurring for the past 6 days (e.g. 28 days of backups).

If in the rare circumstance production data is corrupted, we can opt for a “Point in Time” restoration of the production data into a temporary instance. The corrupted data then can be migrated over to back to the production instance.