{"thread":{"id":"50439","subject":"[Hang] t5562 subtest 8 on NonStop","startedAt":"2019-02-08T12:23:27Z","lastAt":"2019-02-09T20:22:47Z","messageCount":2,"participants":["Randall S. Becker"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"368833","messageId":"002301d4bfa9$158716e0$409544a0$@nexbridge.com","threadId":"50439","inReplyTo":null,"subject":"[Hang] t5562 subtest 8 on NonStop","fromName":"Randall S. Becker","fromEmail":"rsbecker@nexbridge.com","sentAt":"2019-02-08T12:23:17Z","receivedAt":"2019-02-08T12:23:27Z","isPatch":false,"sender":{"key":"randall.becker@nexbridge.ca","avatar":"https://avatars.githubusercontent.com/u/28956764?v=4"},"body":"Hi All,\n\nWe have suddenly encountered a hung git-http-backend in t5562 in the NonStop\nport. This is a new problem not seen before on the platform, surprisingly. I\nam wondering whether this is a result of not actually having an apache2\nserver on-board. Is that a possibility and can that sub-test be bypassed if\nno apache2 is detected?\n\nWe also had subtest 15 and I am investigating, but it may depend on 8's data\n(I had to kill the git-http-backend process, so maybe that's why).\n\nThanks,\nRandall\n\n"},{"id":"368943","messageId":"000b01d4c0b5$358b6b50$a0a241f0$@nexbridge.com","threadId":"50439","inReplyTo":"002301d4bfa9$158716e0$409544a0$@nexbridge.com","subject":"RE: [Hang] t5562 subtest 8 on NonStop","fromName":"Randall S. Becker","fromEmail":"rsbecker@nexbridge.com","sentAt":"2019-02-09T20:22:36Z","receivedAt":"2019-02-09T20:22:47Z","isPatch":false,"sender":{"key":"randall.becker@nexbridge.ca","avatar":"https://avatars.githubusercontent.com/u/28956764?v=4"},"body":"On February 8, 2019 7:23, I wrote:\n> We have suddenly encountered a hung git-http-backend in t5562 in the\n> NonStop port. This is a new problem not seen before on the platform,\n> surprisingly. I am wondering whether this is a result of not actually\nhaving an\n> apache2 server on-board. Is that a possibility and can that sub-test be\n> bypassed if no apache2 is detected?\n> \n> We also had subtest 15 and I am investigating, but it may depend on 8's\ndata\n> (I had to kill the git-http-backend process, so maybe that's why).\n\nI'm back to investigating subtest 8 hanging after dealing with the /dev/zero\nthing. What we end up with two git processes, a git-http-backend, a perl,\nand a few shell processes. The act.err file had:\n\nfatal: request ended in the middle of the gzip stream\n\nWhich went through die() and should have unwound the whole thing, but\ndidn't. The git-http-backend is waiting to clean up children and each git is\npolling for input. It seems like SIGCHLD did not get back through the\nprocess that started it, or its parent - but I don't quite follow the flow\nin this test. I have seen the perl port not like SIGCHLD sometimes, so that\nmight be what is going on. Is it possible to skip the hanging subtest during\na whole test run? We don't really have anyone using the http backend on\nplatform (SSH is the generally preferred method to get to git on box), so I\nam tempted to ignore the ones that hang as conditional known breakages, if\nthat is possible, so that we can get our CI builds back in operation.\nThoughts? Advice?\n\nThanks,\nRandall\n\n-- Brief whoami:\n NonStop developer since approximately 211288444200000000\n UNIX developer since approximately 421664400\n-- In my real life, I talk too much.\n\n\n\n"}]}