Troubleshooting & Problem Resolution Flashcards
7 cards from real COA practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 7 Troubleshooting & Problem Resolution flashcards as text
After a failed 'openstack overcloud deploy', the Heat stack shows 'CREATE_FAILED'. Which command retrieves the specific resource that caused the failure?
Answer: openstack stack failures list overcloud
'openstack stack failures list' provides a concise summary of which resources failed and why, designed specifically for post-deploy troubleshooting.
An instance's console log (openstack console log show) displays kernel panic messages. What does this indicate about the troubleshooting path?
Answer: The guest OS has a problem unrelated to OpenStack infrastructure
A kernel panic in the console log indicates an issue within the guest OS itself — hardware emulation, disk, or OS configuration — not an OpenStack infrastructure problem.
A Swift object store is returning 503 errors sporadically. What is the most probable cause indicated by the proxy-server log showing 'insufficient storage'?
Answer: Not enough object storage nodes have the required replica count available
Swift returns 503 when it cannot satisfy the minimum write quorum (replica count) because too few storage nodes are healthy or reachable.
Which command would you use to determine why the Nova scheduler is rejecting all compute nodes for a new instance request?
Answer: nova-scheduler --debug (check scheduler.log for NoValidHost)
Enabling debug logging on the Nova scheduler and examining scheduler.log reveals which filters rejected each host and the reason for NoValidHost.
A Cinder volume shows 'in-use' in the database but the Nova instance it was attached to has been deleted. How do you safely reset the volume state?
Answer: openstack volume set --state available
'openstack volume set --state available' resets an orphaned 'in-use' volume to available without affecting any data, resolving the stale attachment record.
An operator notices that the RabbitMQ message queue is growing without bound and Nova services are lagging. Which configuration parameter limits the number of concurrent RPC workers processing messages?
Answer: executor_thread_pool_size in oslo.messaging
The 'executor_thread_pool_size' in oslo.messaging controls how many threads a Nova service uses to process incoming RPC messages concurrently.
After a network partition healed, some Neutron DHCP namespaces on a network node have stale port bindings. Which command cleans up and resynchronizes the DHCP agent state?
Answer: neutron-netns-cleanup --force
'neutron-netns-cleanup --force' removes stale network namespaces and allows the DHCP agent to rebuild clean state on restart.