tl;dr - this turned out to be me
I don't fully have a repro case for this yet, but wanted to start an issue so we can track it.
Yesterday I was working on JavaPengines. The test server I'm using just starts an http server and loads pengines. I'm not changing the timeout.
I start my Java client back up and sent a /create message and get this back
{"answer":{"data":[{"X":{"args":["taco"],"functor":"a"},"Y":3}],"event":"success","id":"03d18c99-6da2-4434-8eac-4144069f720a","more":true,"projection":["X","Y"],"time":0.000020260000000000243},"event":"create","id":"03d18c99-6da2-4434-8eac-4144069f720a","slave_limit":3}
Strange part is, that query is from yesterday!
I had a bunch of debug turned on, so captured this in the debug monitor. Notably, the pengine in question was destroyed.
So why is the server hanging onto that old query???
),answer(success('03d18c99-6da2-4434-8eac-4144069f720a',[json{'X':a(taco),'Y':3}],2.0260000000000243e-5,true))])
[Thread 13] Sending create('03d18c99-6da2-4434-8eac-4144069f720a',[slave_limit(3),answer(success('03d18c99-6da2-4434-8eac-4144069f720a',[json{'X':a(taco),'Y':3}],2.0260000000000243e-5,true))]), timout: 300
[Thread 9] Got create('03d18c99-6da2-4434-8eac-4144069f720a',[slave_limit(3),answer(success('03d18c99-6da2-4434-8eac-4144069f720a',[json{'X':a(taco),'Y':3}],2.0260000000000243e-5,true))]) from <message_queue>(0x7f634002d920)
[Thread 9] DESTROY? create('03d18c99-6da2-4434-8eac-4144069f720a',[slave_limit(3),answer(success('03d18c99-6da2-4434-8eac-4144069f720a',[json{'X':a(taco),'Y':3}],2.0260000000000243e-5,true))])
[Thread 9] Running hook: header
[Thread 9] Field: [cache_control('no-cache, no-store, must-revalidate'),pragma('no-cache'),expires('0'),content_type('application/json; charset=UTF-8')]
[Thread 9] Transfer-encoding: none
[Thread 9] Running hook: send_header
[Thread 9] Header: [connection('keep-alive'),cache_control('no-cache, no-store, must-revalidate'),pragma('no-cache'),expires('0'),content_type('application/json; charset=UTF-8')]
[Thread 9] Running hook: close
[Thread 9] [35] 200 OK (0.000 seconds; 326 bytes)
[Thread 9] Wiping posted data
[Thread httpd@9900_4] Got job requeue((0x7f6368026e80),(0x7f6368027a70),user:http_dispatch,[protocol(http),peer(ip(127,0,0,1)),pool(client('httpd@9900',user:http_dispatch,(0x7f6368026e80),(0x7f6368027a70)))])
[Thread httpd@9900_4] Waiting for keep-alive ...
[Thread httpd@9900_4] Timeout on keep-alive connection
[Thread httpd@9900_4] Closing connection from ip(127,0,0,1)
[Thread httpd@9900_4] Waiting for a job ...
[Thread 11] '9be79c5f-6050-4081-8477-516588fd186e': 6 = destroy => 1
[Thread 11] Reply to <message_queue>(0x7f634002abd0): destroy('9be79c5f-6050-4081-8477-516588fd186e')
[Thread 11] Sending destroy('9be79c5f-6050-4081-8477-516588fd186e'), timout: 300
[Thread 12] 'c9c19ec8-d5b9-44e2-8bcb-8208b23eada8': 6 = destroy => 1
[Thread 12] Reply to <message_queue>(0x7f634002c490): destroy('c9c19ec8-d5b9-44e2-8bcb-8208b23eada8')
[Thread 12] Sending destroy('c9c19ec8-d5b9-44e2-8bcb-8208b23eada8'), timout: 300