From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jason Gunthorpe Subject: Re: Hang in ipoib_mcast_stop_thread Date: Mon, 6 Jun 2016 10:57:42 -0600 Message-ID: <20160606165741.GA6413@obsidianresearch.com> References: <57553DE7.2060009@kyup.com> <57556866.8040703@kyup.com> <575575B5.9010504@kyup.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <575575B5.9010504-6AxghH7DbtA@public.gmane.org> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Nikolay Borisov Cc: Erez Shitrit , Doug Ledford , "linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , SiteGround Operations List-Id: linux-rdma@vger.kernel.org On Mon, Jun 06, 2016 at 04:08:05PM +0300, Nikolay Borisov wrote: > Be that as it may, but the queue seems to never be unlocked by the > driver. Even after restart I still continue to get such messages. One of > the network admins told me this could happen since the infiniband switch > port has to be reset after an outage. I find it hard to believe that > even after physical restart of the server the ib interface cannot connect. I've also randomly seen apparently hung VL0 credit that could only be solved by a switch reset. I also was never able to pin down a root cause. VL15 would work fine.. Jason -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html