From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga03.intel.com (mga03.intel.com [134.134.136.65]) by dpdk.org (Postfix) with ESMTP id 52713FEB for ; Thu, 1 Mar 2018 09:20:23 +0100 (CET) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga004.fm.intel.com ([10.253.24.48]) by orsmga103.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 01 Mar 2018 00:20:21 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.47,407,1515484800"; d="scan'208";a="33784986" Received: from fmsmsx104.amr.corp.intel.com ([10.18.124.202]) by fmsmga004.fm.intel.com with ESMTP; 01 Mar 2018 00:20:21 -0800 Received: from fmsmsx157.amr.corp.intel.com (10.18.116.73) by fmsmsx104.amr.corp.intel.com (10.18.124.202) with Microsoft SMTP Server (TLS) id 14.3.319.2; Thu, 1 Mar 2018 00:20:19 -0800 Received: from shsmsx102.ccr.corp.intel.com (10.239.4.154) by FMSMSX157.amr.corp.intel.com (10.18.116.73) with Microsoft SMTP Server (TLS) id 14.3.319.2; Thu, 1 Mar 2018 00:20:19 -0800 Received: from shsmsx103.ccr.corp.intel.com ([169.254.4.116]) by shsmsx102.ccr.corp.intel.com ([169.254.2.124]) with mapi id 14.03.0319.002; Thu, 1 Mar 2018 16:19:59 +0800 From: "Tan, Jianfeng" To: "Lazarenko, Vlad (WorldQuant)" , "'users@dpdk.org'" Thread-Topic: Multi-process recovery (is it even possible?) Thread-Index: AdOwxXAvHZ5QGX6aRnmNCp+LjczNoAAaaIMQ Date: Thu, 1 Mar 2018 08:19:59 +0000 Message-ID: References: <790E2AC11206AC46B8F4BB82078E34F8081E0EAB@PWSSMTEXMBX002.AD.MLP.com> In-Reply-To: <790E2AC11206AC46B8F4BB82078E34F8081E0EAB@PWSSMTEXMBX002.AD.MLP.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.239.127.40] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-users] Multi-process recovery (is it even possible?) X-BeenThere: users@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK usage discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 01 Mar 2018 08:20:24 -0000 > -----Original Message----- > From: users [mailto:users-bounces@dpdk.org] On Behalf Of Lazarenko, Vlad > (WorldQuant) > Sent: Thursday, March 1, 2018 2:54 AM > To: 'users@dpdk.org' > Subject: [dpdk-users] Multi-process recovery (is it even possible?) >=20 > Guys, >=20 > I am looking for possible solutions for the following problems that come > along with asymmetric multi-process architecture... >=20 > Given multiple processes share the same RX/TX queue(s) and packet pool(s) > and the possibility of one packet from RX queue being fanned out to multi= ple > slave processes, is there a way to recover from slave crashing (or exits = w/o > cleaning up properly)? In theory it could have incremented mbuf reference > count more than once and unless everything is restarted, I don't see a > reliable way to release those mbufs back to the pool. Recycle an element is too difficult; from what I know, it's next to impossi= ble.=20 To recycle a memzone/mempool is easier. So in your case, you might want to = use different pools for different queues (processes). If you really want to recycle an element, rte_mbuf in your case, it might b= e doable by: 1. set up rx callback for each process, and in the callback, store a specia= l flag at rte_mbuf->udata64. 2. when the primary to detect a secondary is down, we iterate all element w= ith the special flag, and put them back into the ring. There is small chance to fail that , mbuf is allocated by a secondary proce= ss, and before it's flagged, it crashes. Thanks, Jianfeng >=20 > Also, if spinlock is involved and either master or slave crashes, everyth= ing > simply gets stuck. Is there any way to detect this (i.e. outside of data = path)..? >=20 > Thanks, > Vlad >=20 >=20 >=20 > ########################################################## > ######################### >=20 > The information contained in this communication is confidential, may be >=20 > subject to legal privilege, and is intended only for the individual named= . >=20 > If you are not the named addressee, please notify the sender immediately > and >=20 > delete this email from your system. The views expressed in this email ar= e >=20 > the views of the sender only. Outgoing and incoming electronic > communications >=20 > to this address are electronically archived and subject to review and/or > disclosure >=20 > to someone other than the recipient. >=20 > ########################################################## > #########################