From owner-freebsd-questions@freebsd.org Thu Mar 10 12:32:16 2016 Return-Path: Delivered-To: freebsd-questions@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id A117DACA823 for ; Thu, 10 Mar 2016 12:32:16 +0000 (UTC) (envelope-from konstantin@schukraft.org) Received: from server949-han.de-nserver.de (server949-han.de-nserver.de [77.75.250.185]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 404191A60 for ; Thu, 10 Mar 2016 12:32:15 +0000 (UTC) (envelope-from konstantin@schukraft.org) Received: (qmail 19840 invoked from network); 10 Mar 2016 13:32:02 +0100 X-Fcrdns: Yes Received: from schukraft.it (HELO [10.0.1.109]) (80.147.5.44) (smtp-auth username konstantin@schukraft.org, mechanism plain) by server949-han.de-nserver.de (qpsmtpd/0.92) with (ECDHE-RSA-AES256-SHA encrypted) ESMTPSA; Thu, 10 Mar 2016 13:32:02 +0100 Subject: Re: unresponsive process issue To: Travis Parker , freebsd-questions@freebsd.org References: From: Konstantin Schukraft X-Enigmail-Draft-Status: N1110 Message-ID: <56E1693C.4070603@schukraft.org> Date: Thu, 10 Mar 2016 12:31:56 +0000 User-Agent: "" MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit X-User-Auth: Auth by konstantin@schukraft.org through 80.147.5.44 X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 10 Mar 2016 12:32:16 -0000 -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Hi, On 03/10/2016 01:32, Travis Parker wrote: > I've twice now had a process get into a stuck state that I don't > believe should be possible: stopped (ps reports 'T', top reports > 'STOP'), but unresponsive to any signal, including even CONT (KILL > followed by CONT isn't clearing it). It's a redis process, in a > jail, that is listening on a streaming unix domain socket. I see the same thing happening here with long running processes in jails. In my case it's firefox and thunderbird. At first I thought it's yet another firefox bug that somehow creeped into the thunderbird codebase (I don't know how disjunct those are). When it happens, they completely stop responding to signals (including a kill -9) and their graphical UI stops responding at least partly. E.g. I sometimes can still select other tabs in firefox, but it only updates the window title, not the actual content. > For the moment I have switched it over to TCP on localhost and I'll > have to wait and see if that works around whatever got it into this > state (it takes a few days to occur). I can confirm this timescale of a few days. > I wanted to reach out to this list in case there's something > obvious I'm missing before mailing freebsd-bugs. I've found a few > descriptions of a similar issue (STOP state unresponsive to CONT) > from googling, but only back around 2004 and always resolved. Since we seem to have the same core problem, please send that bug report. I can at least corroborate your findings, if nothing else. All the best, Konstantin -----BEGIN PGP SIGNATURE----- iQIcBAEBCAAGBQJW4Wk2AAoJEH3raMNeVmMFKcIP/3xnE5OcUF8SBfKcSP4d2Vye Ybb5KQWjRsAaj/jmyalXHEDExPJa1PlFdFyhy0WR3LL31txdabPEYu4Wv/PeK1dZ rLAO9SZdI2cuLqJAseHn58MTI8EtYFgZ39LNsvgBh5abXO0A0cCJqZO8p6FjrFZf p37rrBS4TVM6a23+OXpwT4RD6hKKZnEtUqT4O02PqTc9mevEC+VrqZ8seExt4jrI no67/KETC4lnB+YPZ6KZGwPohn+Mh0HhbKotjm122V+AbnNv0/vyqgLqWfh25zQI EX7GyZGqpeT3DMKVrFEZ5WNzOxHdzFFByZmUYH910RjbEqzOpEi/9zp02NccOuVU /7tU1A1AzmpB13qa/c5HyDSFdte+dKH2nphzf2hA8LR5t5DnQefEW/wmbkdtzB4w WBAxzBs6s7p75yu1nD2ydrFxEQcm6bfvrDPatAmC/K+GqpyqyWXR0F1LwAM77xG/ Fwi3uZLIJaS/UWAJo/abNxHUQOhb5JFSHNvexWTBQ5dKgZ3FCzNaW0lDlPM77hsJ jaubV3QHv7SZ6RAuQqy4tBuO7QLSJVM/hKzbYRAVadx7emlIqemdpH9AXLMexJwJ QJuvsvjBDLisgisVicQKHXFxhjbgi9alOkBWKdEFSZNAMWrNQ9eLfV5UmnaOc1Zq pIlppf5UfreyjTUfu5N8 =WNJK -----END PGP SIGNATURE-----