Infrastructure Scenario Tests

Azure AD Connect enters a synchronization loop where the delta sync cycle never completes, continuously restarting. Password hash synchronization stops working, and on-premises changes are not reflecting in Azure AD.

The ADFS token signing certificate expires, breaking all federated SSO authentication. Users cannot sign into Office 365, SaaS applications, or any relying party trusts configured to use ADFS for authentication.

DFS Replication backlog between two domain controllers reaches critical levels, with SYSVOL replication lagging by thousands of files. Group Policy is inconsistent across the domain, and some users receive stale GPOs depending on which DC they authenticate against.

Active Directory inter-site replication breaks after a network change removes the IP link between two AD sites. The Knowledge Consistency Checker (KCC) cannot generate a replication topology, and AD objects created in one site are not replicating to the other.

The robotic tape library experiences a mechanical jam in the tape picker mechanism, preventing all tape load/unload operations. Backup jobs queue indefinitely as no tapes can be mounted, and offsite tape rotation is halted.

The S3-compatible object storage (MinIO) reaches its configured quota limit. Application uploads fail, log shipping stops, and backup jobs writing to object storage all fail simultaneously.

Multiple Ceph OSDs fail simultaneously on a storage node after a power supply unit failure, triggering a massive data rebalancing operation. The cluster enters HEALTH_WARN state and client I/O is severely impacted during recovery.

An iSCSI storage target becomes unreachable on the primary path after a switch failure. Multipath failover engages but the secondary path is congested, causing severe performance degradation for all connected initiators.

A ZFS storage pool enters degraded state after a drive failure in a RAIDZ2 vdev. The pool remains operational but with reduced redundancy. A second drive in the same vdev is showing SMART warnings, indicating imminent failure.

The active controller on a Pure Storage FlashArray fails, triggering an automatic failover to the standby controller. During the failover, I/O is briefly paused and disk queue depth spikes, causing latency-sensitive applications to experience errors.

Thinly provisioned LUNs on an EMC Unity storage array exceed physical capacity. The storage pool runs out of space, causing write failures on all LUNs in the pool. VMware datastores become read-only and VMs begin crashing.

A NetApp ONTAP volume goes offline due to an aggregate running out of space, causing all LUNs and NFS exports on that volume to become unavailable. Multiple application servers lose access to their primary storage.

The internal package repository mirror becomes corrupted after a failed rsync, causing all package installation and update operations across the Linux fleet to fail. Servers cannot install security patches or new application dependencies.

An automated SSH key rotation process replaces host keys on 20 servers but fails to update the known_hosts files on the Ansible control node and monitoring servers. All SSH-based automation, configuration management, and monitoring breaks simultaneously.

The cron daemon on a critical infrastructure server crashes and enters a crash loop due to a corrupted crontab file. All scheduled jobs including log rotation, backup scripts, and health checks stop executing. The issue goes undetected for 48 hours.

SELinux enforcing mode blocks a newly deployed application from binding to its configured port and accessing its data directory. The application fails to start with permission denied errors, and the audit log fills with AVC denial messages.

An NFS mount on a production application server becomes stale after the NFS server becomes unreachable due to a network switch failure. All processes attempting to access the NFS mount hang in uninterruptible sleep, and the application server becomes partially unresponsive.

A Linux server runs out of inodes on the root filesystem despite having 40% disk space free. Millions of small session files created by a PHP application consumed all available inodes. No new files can be created, causing application failures and log rotation to break.

A misconfigured systemd unit file creates a circular dependency between three services, causing them to enter a restart loop. Each service depends on another in the cycle, and systemd cannot satisfy all dependencies simultaneously.

A Linux production server experiences kernel panic after the OOM killer is invoked repeatedly, killing critical processes including the database and web server. The system becomes unresponsive after the OOM killer targets the init process.