First Attempt at Image Build Times Out

[jjaswanson4]

Image Builder builds a local cache of packages the first time a compose is submitted, which takes a few minutes.

Currently, the 'Start compose' task will fail on the first try due to not waiting long enough for this timeout:

An exception occurred during task execution. To see the full traceback, use -vvv. The error was: socket.timeout: timed out
fatal: [edge-manager-aws]: FAILED! => {"changed": false, "module_stderr": "Shared connection to 18.191.233.197 closed.\r\n", "module_stdout": "Traceback (most recent call last):\r\n  File \"/home/ec2-user/.ansible/tmp/ansible-tmp-1673036945.4251916-430574-120449954195479/AnsiballZ_start_compose.py\", line 107, in <module>\r\n    _ansiballz_main()\r\n  File \"/home/ec2-user/.ansible/tmp/ansible-tmp-1673036945.4251916-430574-120449954195479/AnsiballZ_start_compose.py\", line 99, in _ansiballz_main\r\n    invoke_module(zipped_mod, temp_path, ANSIBALLZ_PARAMS)\r\n  File \"/home/ec2-user/.ansible/tmp/ansible-tmp-1673036945.4251916-430574-120449954195479/AnsiballZ_start_compose.py\", line 48, in invoke_module\r\n    run_name='__main__', alter_sys=True)\r\n  File \"/usr/lib64/python3.6/runpy.py\", line 205, in run_module\r\n    return _run_module_code(code, init_globals, run_name, mod_spec)\r\n  File \"/usr/lib64/python3.6/runpy.py\", line 96, in _run_module_code\r\n    mod_name, mod_spec, pkg_name, script_name)\r\n  File \"/usr/lib64/python3.6/runpy.py\", line 85, in _run_code\r\n    exec(code, run_globals)\r\n  File \"/tmp/ansible_infra.osbuild.start_compose_payload_pje5d7f1/ansible_infra.osbuild.start_compose_payload.zip/ansible_collections/infra/osbuild/plugins/modules/start_compose.py\", line 288, in <module>\r\n  File \"/tmp/ansible_infra.osbuild.start_compose_payload_pje5d7f1/ansible_infra.osbuild.start_compose_payload.zip/ansible_collections/infra/osbuild/plugins/modules/start_compose.py\", line 259, in main\r\n  File \"/tmp/ansible_infra.osbuild.start_compose_payload_pje5d7f1/ansible_infra.osbuild.start_compose_payload.zip/ansible_collections/infra/osbuild/plugins/module_utils/weldrapiv1.py\", line 176, in post_compose\r\n  File \"/tmp/ansible_infra.osbuild.start_compose_payload_pje5d7f1/ansible_infra.osbuild.start_compose_payload.zip/ansible/module_utils/urls.py\", line 1557, in open\r\n  File \"/usr/lib64/python3.6/urllib/request.py\", line 223, in urlopen\r\n    return opener.open(url, data, timeout)\r\n  File \"/usr/lib64/python3.6/urllib/request.py\", line 526, in open\r\n    response = self._open(req, data)\r\n  File \"/usr/lib64/python3.6/urllib/request.py\", line 544, in _open\r\n    '_open', req)\r\n  File \"/usr/lib64/python3.6/urllib/request.py\", line 504, in _call_chain\r\n    result = func(*args)\r\n  File \"/tmp/ansible_infra.osbuild.start_compose_payload_pje5d7f1/ansible_infra.osbuild.start_compose_payload.zip/ansible/module_utils/urls.py\", line 677, in http_open\r\n  File \"/usr/lib64/python3.6/urllib/request.py\", line 1352, in do_open\r\n    r = h.getresponse()\r\n  File \"/usr/lib64/python3.6/http/client.py\", line 1365, in getresponse\r\n    response.begin()\r\n  File \"/usr/lib64/python3.6/http/client.py\", line 320, in begin\r\n    version, status, reason = self._read_status()\r\n  File \"/usr/lib64/python3.6/http/client.py\", line 281, in _read_status\r\n    line = str(self.fp.readline(_MAXLINE + 1), \"iso-8859-1\")\r\n  File \"/usr/lib64/python3.6/socket.py\", line 586, in readinto\r\n    return self._sock.recv_into(b)\r\nsocket.timeout: timed out\r\n", "msg": "MODULE FAILURE\nSee stdout/stderr for the exact error", "rc": 1}

Increasing the timeout should resolve this issue. It may need to be as high as 10-15 minutes - building the cache can take a hot minute depending on connection speed, system utilization, how the CDN is feeling that day, etc.

[jjaswanson4]

Upon further testing: it seems like the socket is getting closed, yet the build does actually start happening:

[root@edge-manager-aws ~]# composer-cli compose list
424b1575-096c-4613-b357-b6a17372941a RUNNING rhde-image 0.0.1 edge-commit

This can kinda be worked around by just re-running the task (or role) right now, but it queues up another build. Not the worst thing in the world, but probably should be fixed.