Return-Path: X-Original-To: apmail-cassandra-user-archive@www.apache.org Delivered-To: apmail-cassandra-user-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id B44B2E77E for ; Mon, 11 Mar 2013 18:21:55 +0000 (UTC) Received: (qmail 70651 invoked by uid 500); 11 Mar 2013 18:21:53 -0000 Delivered-To: apmail-cassandra-user-archive@cassandra.apache.org Received: (qmail 70581 invoked by uid 500); 11 Mar 2013 18:21:52 -0000 Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: user@cassandra.apache.org Delivered-To: mailing list user@cassandra.apache.org Received: (qmail 70571 invoked by uid 99); 11 Mar 2013 18:21:52 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 11 Mar 2013 18:21:52 +0000 X-ASF-Spam-Status: No, hits=-2.3 required=5.0 tests=RCVD_IN_DNSWL_MED,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (nike.apache.org: local policy) Received: from [192.174.58.134] (HELO XEDGEA.nrel.gov) (192.174.58.134) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 11 Mar 2013 18:21:47 +0000 Received: from XHUBB.nrel.gov (10.20.4.59) by XEDGEA.nrel.gov (192.174.58.134) with Microsoft SMTP Server (TLS) id 8.3.245.1; Mon, 11 Mar 2013 12:21:16 -0600 Received: from MAILBOX2.nrel.gov ([fe80::19a0:6c19:6421:12f]) by XHUBB.nrel.gov ([::1]) with mapi; Mon, 11 Mar 2013 12:21:24 -0600 From: "Hiller, Dean" To: "user@cassandra.apache.org" Date: Mon, 11 Mar 2013 12:21:24 -0600 Subject: =?windows-1250?Q?Re:_dynamic_switch_works_with_NetworkSTrategy_NOT_Simple?= =?windows-1250?Q?Strategy..solution_follows=8A.?= Thread-Topic: =?windows-1250?Q?dynamic_switch_works_with_NetworkSTrategy_NOT_SimpleStra?= =?windows-1250?Q?tegy..solution_follows=8A.?= Thread-Index: Ac4ehT0nIEZaruuuRxKGl5QqEnXnPA== Message-ID: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.3.1.130117 acceptlanguage: en-US Content-Type: text/plain; charset="windows-1250" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Virus-Checked: Checked by ClamAV on apache.org Ticket filed. https://issues.apache.org/jira/browse/CASSANDRA-5333 Thanks, Dean From: Edward Capriolo > Reply-To: "user@cassandra.apache.org" > Date: Monday, March 11, 2013 9:16 AM To: "user@cassandra.apache.org" > Subject: Re: dynamic switch works with NetworkSTrategy NOT SimpleStrategy..= solution follows=8A. You should file a JIRA if dsnitch only works with LOCAL_QUORUM something is= very wrong. On Mon, Mar 11, 2013 at 9:58 AM, Hiller, Dean > wrote: Well, we finally have dynamic switch working. It seems to switch nodes to = a Remote node and SimpleSTrategy can not deal with that well. Also, we had= to move to CL=3DQUOROM_LOCAL instead of QUOROM. So for those of you that want cassandra to keep performing well when one no= de starts to get really slow, make sure you have 1. NetworkTopologyStrategy set 2. Have CL=3DQUOROM_LOCAL We had to dive into the code to figure all that out. I hope it helps someo= ne else. Later, Dean From: aaron morton = >> Reply-To: "user@cassandra.apache.org>" >> Date: Monday, March 11, 2013 6:48 AM To: "user@cassandra.apache.org>" >> Subject: Re: has anyone used dynamic snitch at all Check that read_repair_chance on the CF's is 0.1, not the old 1.0 Wait at least 10 minutes for the DynamicSnitch to re-calculate. Use the org.apache.cassandra.db:type=3DDynamicEndpointSnitch MBean to see w= hat scores it has given the nodes. Cheers ----------------- Aaron Morton Freelance Cassandra Consultant New Zealand @aaronmorton http://www.thelastpickle.com On 8/03/2013, at 11:40 AM, Edward Capriolo >> wrote: dynamic_snitch=3Dtrue is the default. So it is usually on "wrapping" other = snitches. I have found several scenarios where it does not work exactly as = your would expect. On Fri, Mar 8, 2013 at 2:26 PM, Hiller, Dean >> wrote: Our test setup 4 nodes, RF=3D3, reads at CL=3DQUOROM and we tried CL=3DTWO Tell the network card to slow down every packet on node 2 After fixing astyanax to not go to node 2 anymore, we are still seeing cass= andra have issues as it seems to be involving node 2 somehow. If we take n= ode 2 down, it all speeds back up. We are trying to get this working such that a slow node in cassandra does n= ot impact our customers. We are in 1.2.2 and added the following properties=85.(our properties show = PropertyFileSnitch though I see the keyspace has org.apache.cassandra.locat= or.SimpleStrategy set probably because it was created through a tool instea= d of CLI=85shucks)=85.anyways, I still expected dynamic snitch to work=85. # controls how often to perform the more expensive part of host score # calculation dynamic_snitch: true dynamic_snitch_update_interval_in_ms: 100 # controls how often to reset all host scores, allowing a bad host to # possibly recover dynamic_snitch_reset_interval_in_ms: 600000 # if set greater than zero and read_repair_chance is < 1.0, this will allow # 'pinning' of replicas to hosts in order to increase cache capacity. # The badness threshold will control how much worse the pinned host has to = be # before the dynamic snitch will prefer other replicas over it. This is # expressed as a double which represents a percentage. Thus, a value of # 0.2 means Cassandra would continue to prefer the static snitch values # until the pinned host was 20% worse than the fastest. dynamic_snitch_badness_threshold: 0.1 Any help appreciated, Thanks, Dean