Non-technical Description

SMS messages are sent for two actions

• Verifying a phone number during SMS MFA setup

• Signing in with SMS MFA

A monthly spend cap must be set on our account, this was previously raised to $200 per month but on this occasion we exceeded that limit and SMS messages could no longer be sent.

As a result the two actions noted above were not possible until the limit was raised, approximate impact:

• 10 hours of SMS downtime

• 96 users (based on phone number)

• 307 total failed SMS messages

Trigger

• SMS limit exceeded

Detection

• Identified when TIS team member was unable to sign in to TSS

Resolution

• Increase monthly SMS limit from $200 to $300

• Modify the monitoring to alert at $270

Timeline

BST unless otherwise stated

• 25 Jun 2022 01:10 - Alert sent to #monitoring channel on Slack that we had reached 90% of our SMS limit

• 28 Jun 2022 12:59 - SMS limit exceeded, messages no longer being sent

• 28 Jun 2022 20:17 - TIS dev team member unable to sign in using SMS MFA - no message received, thought to be a phone verification issue

• 28 Jun 2022 22:22 - TIS dev team member unable to verify phone number to set up SMS MFA - problem identified as SMS limit

• 28 Jun 2022 22:37 - SMS limit increased to $300 - verified able to receive messages again

• 28 Jun 2022 22:41 - All clear notification on #monitoring Slack channel

Root Cause(s)

• The SMS costs exceeded the configured limits

  • The limit was not set appropriately

    • Data based limits not yet set, still using guesswork

  • The alert for reaching 90% of the limit was not seen

    • Noisy #monitoring channel due to tis-log-size alerts?

      • Placement sync log spam

      • Prod → Stage DB sync log spam

      • Alarm not configured appropriately - e.g. using more datapoints would help avoid the yoyoing of the alert status which causes slack spam

  • The daily cloudwatch alarm reminders were not checked/actioned

    • ???

Action Items

Lessons Learned

• Monitoring alerts are useless unless they are noticed/actioned