From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Layton Subject: Re: [ 2375.793397] WARNING: CPU: 0 PID: 1149 at net/netlink/genetlink.c:1037 genl_unbind+0xc0/0xd0() Date: Thu, 15 Jan 2015 07:49:56 -0500 Message-ID: <20150115074956.4e5617cd@tlielax.poochiereds.net> References: <20150114161334.28acf5fc@tlielax.poochiereds.net> <1421275700.1950.34.camel@sipsolutions.net> <1421277946.1950.38.camel@sipsolutions.net> <20150114212039.68c9a5a6@synchrony.poochiereds.net> <1421311071.1962.2.camel@sipsolutions.net> <20150115071426.563573e7@tlielax.poochiereds.net> <1421324985.1962.5.camel@sipsolutions.net> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: Jeff Layton , netdev@vger.kernel.org To: Johannes Berg Return-path: Received: from mail-qc0-f175.google.com ([209.85.216.175]:40067 "EHLO mail-qc0-f175.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751947AbbAOMt7 (ORCPT ); Thu, 15 Jan 2015 07:49:59 -0500 Received: by mail-qc0-f175.google.com with SMTP id p6so11891512qcv.6 for ; Thu, 15 Jan 2015 04:49:58 -0800 (PST) In-Reply-To: <1421324985.1962.5.camel@sipsolutions.net> Sender: netdev-owner@vger.kernel.org List-ID: On Thu, 15 Jan 2015 13:29:45 +0100 Johannes Berg wrote: > On Thu, 2015-01-15 at 07:14 -0500, Jeff Layton wrote: > > > I was able to reproduce it again this morning. This time it generated > > 28 warnings (all with similar stack traces to the one I posted > > yesterday). > > Thanks. I just took a look at this, and it seems that sadly it doesn't > include netlink messages in the log? At least I didn't find them - so I > don't really know what happened. What I was thinking happened would > (mostly) play out over netlink messages. > (re-cc'ing netdev -- I sent Johannes the log in a private email as it was 12M compressed) It may have. I just recently started playing with trinity, so I don't know much about what it does. I do see these sorts of messages in the logs that imply that it's opening netlink sockets: [main] fd[661] = domain:16 (PF_NETLINK) type:0x2 protocol:10 [main] fd[675] = domain:16 (PF_NETLINK) type:0x2 protocol:0 ...and then it does random I/Os on those fds. > > I haven't tested your patchset out yet, but I can try to do that > > later. > > You won't be able to reproduce the warning with it since I removed it > there. I'm reasonably certain though that you triggered one of the two > cases I found, more likely the former that doesn't involve removing > genetlink families (don't see how you could have done that with trinity > unless it randomly loads and unloads modules?) > Ok, sounds good. FWIW, it tries to call delete_module, but I was running this as an unprivileged user so I don't think that can succeed. -- Jeff Layton