If I change a health check profile, which Backend Services are affected?

Short answer

All the Backend Services and Backend Groups to which the profile is assigned, and therefore all the vServices that use them, are affected; the change is applied the moment it is saved. The affected ones are seen in the Used in tab in the profile preview or with the Health filter in the Backend Services list. After the change, the status of the Backend Services that use the profile returns to Preparing, and they do not receive traffic until Required Successes consecutive successful checks; for this reason, the change is made at a low-traffic time or with a gradual transition to a new profile.

How Does It Work?

1. The profile runs as a separate check for each Backend Service to which it is assigned (in a Backend Group, for each server in the group).
2. When the profile is saved, the settings of these checks are renewed and the previous result count is reset; the status becomes Preparing.
3. After Required successes consecutive successful checks, the status becomes OK and the Backend Service returns to traffic. With the defaults (Period 1 s, Required successes 3), this takes a few seconds.
4. Since all the Backend Services that use the profile return to Preparing at the same time, all the servers of a vService can be out of traffic during this time.

Seeing the Affected Ones

Method	Where
Used in tab	Load Balancer → Profiles → Health Checks → profile row → preview → Used in; the tab is displayed only if the profile is used, and the badge shows the usage count
Filter / Usage Filter	Filters the profiles in the Health Checks list by where they are used; Unused entries for the unused ones
Health filter	Load Balancer → Backend Services → Filter → the profile is selected in the Health card; the Backend Services that use that profile are listed
Backend Groups	Load Balancer → Backend Groups → Filter

The profile may also have been selected to monitor the default gateway of a route table (Gateway Monitoring); this usage does not appear in the Used in tab and is checked separately from the Network → Interfaces screens.

In TR7

• Used in is checked first: The profile may be used on more Backend Services than expected.
• Gradual transition: A separate profile is created with the new settings and the Backend Services are moved to the new profile one by one; each time, only one Backend Service returns to Preparing (How should I set the Period, Timeout, Required Successes and Required Failures values?).
• The type does not change: Check Type cannot be changed after saving; a new profile is required for a different type.
• Deletion: An assigned profile can be deleted without a warning; the Backend Services that use the deleted profile remain without a check (Delete).
• Cluster: By default, the change is transferred to the peer device automatically and is applied there in the same way; if Manual Sync Mode is on, it is valid only on the device where the change was made until it is sent with Synchronize.

Common Misconceptions

• "The change affects only the selected Backend Service." The profile is shared; all the Backend Services that use the profile are affected.
• "The change does not affect traffic at all." The Backend Services that use the profile do not receive traffic until Required successes successful checks.
• "Used in shows all usages." The usage in the gateway monitoring of a route table does not appear in Used in.

How it is done in the documentation

• Health Checks · The "Used in" tab must be checked
• Health Checks · Edit
• Health Checks · Delete
• Backend Services · Filter

Related questions

• How should I set the Period, Timeout, Required Successes and Required Failures values?
• How many seconds does it take for a health check result to affect traffic?
• How do I see which vServices use a Backend Service ("Used in")?