Environment

• YugabyteDB Anywhere - 2.20.x and later

Issue

YugabyteDB universe upgrade fails with the following error:

CheckFollowerLag, timing out after retrying 11 times for a duration of 649613ms, greater than max time out of 600000ms for node: <node-name>. Failing....

Cause

The error occurs because there is a follower lag in the universe. The upgrade process requires the universe to be in a healthy state.

Resolution

Follow these steps to bypass the follower lag check temporarily:

1. Go to YBA UI > Admin > Advanced.

2. Under Global Configuration, search for yb.checks.follower_lag.enabled.

3. Disable this setting by editing the configuration.

4. Save the changes.

5. Retry the upgrade from the YBA UI.

6. Once the upgrade is complete, revert the flag to its original state.

Additional Information

Before initiating an upgrade, ensure the universe is healthy by following these steps:

• Confirm no active alerts or health errors are present.

• Run pre-checks using the YBA UI to detect any issues.

YBA applies the following pre-checks before performing edit and upgrade operations:

• Under-replicated Tablets Check (for rolling operations)

• Node Safety Check (ensures nodes are safe to take down)

• Leaderless Tablets Check

• Universe Consistency Check

• Cluster Consistency Check

• Follower Lag Check

For more details, refer to the documentation on running pre-checks before edit and upgrade operations.