Simple SQS Examples in Python

Helper scripts

• purge_queues.py - purge all messages from test standard queue and test.fifo queue

Standard Queues

• send_standard.py - send a single message to a standard queue
• receive_standard.py - receive a single message from the queue queue and delete it
• send_multiple_standard.py - send 1000 messages to a standard queue

FIFO Queues

• send_fifo.py - send a single message to a fifo queue
• send_multiple_fifo.py
• receive_fifo.py - receive a single message from a fifo queue
• receive_multiple_fifo.py - receive a batch of 10 messages from a fifo queue and delete them
• send_multiple_standard.py - send 1000 messages all with the same group id
• send_fifo_dup.py - send a single message with a static duplication id, send this as many times as you like then run receive_fifo.py to see there is only one due to deduplication
• send_multiple_fifo_groupid.py - send 100 messages with the same group id then 20 more per group id, up to group id of 5. This loads one groupid more than others to experiement with.
• receive_multiple_fifo_delay.py - receive a batch of 10 messages from a fifo queue then wait 3 seconds before deleting them. Run this with another recieve_multiple_fifo.py in paralell to test behaviour of FIFO SQS.
• receive_multiple_fifo_threads.py - using 5 threads, receive a batch of 10 messages from a fifo queue waiting 2 seconds between batch before deleting them, demonstrating behaviour between threads/workers picking up messages.
• receive_multiple_fifo_threads_but_fail.py - as above, but don't delete the messages after to simulate a fail in processing. Messages are not able to be processed by other workers within those group ids until the default visability timeout (30 seconds) has expired.
• send_multiple_standard_throttled.py - use the bucket throttling algorithm to throttle requests to avoid overwhelming a queue if downstream system fails
• send_multiple_standard_limiter_module.py - same as above but using an external module called limiter