http://www.couchbase.com/issues This file is an XML representation of an issue en-us 5.2.4 845 26-12-2012 http://www.couchbase.com/issues/browse/MB-4366 Couchbase Server screenshot attached <br/> <br/> NOTE: we&#39;re converting this to main &#39;named tap issues&#39; ticket. <br/> <br/> So what&#39;s not safe about reusing named taps as of 1.8.0? <br/> <br/> If something happened to destination node after tap was disconnected. And if that something affected data for vbuckets replicated as part of named tap, then subsequent reuse of named tap will incorrectly assume that we can continue sending stuff instead of re-negotiating which data needs to be resent. <br/> MB-4366

ns_server is reusing tap names unsafely which causes data loss or inconsistency in replication when a node is removed and added back

Bug Blocker Closed Fixed Public Aleksey Kondratenko Farshid Ghods 1.8.1-release-notes Wed, 19 Oct 2011 19:20:35 -0500 Wed, 9 Jan 2013 22:59:21 -0600 Thu, 12 Apr 2012 22:44:34 -0500 1.7.2 1.8.0 1.8.1 ns_server 0 3 another screenshot : 5 minutes after stopping the rebalance tap stream only stops if there is no item added to the backlog <br/> if the user keeps the load running this tap stream remains alive forever Farshid, cannot make sense of this screenshots. Can you elaborate? basically that means there is still one tap_rebalance stream open and running even after rebalance was stopped. <br/> <br/> we seem to be stopping most of the streams except one waiting 5 minutes will not work if there are ongoing mutuations in the cluster because this tap stream only times out after 5 minutes of inactivity so it&#39;s ep-engine issue then ? I mean we close tap streams as much as possible in ns_server. Named tap streams are kept alive by ep-engine. If there&#39;s anything ns_server can do to really stop those tap producers, I&#39;ll be happy to do that. this is the main ticket for the named tap approach/fix is this a blocker for 1.8.1? Yes, because this may be causing data loss in some conditions. <br/> <br/> Farshid, I believe there are a few other tickets where this is the underlying problem. Can you reference them here for completeness? Thanks <a href="http://review.couchbase.org/14555">http://review.couchbase.org/14555</a> fixes it on 1.8.1.<br/> <br/> 1.8 and master have a bit different code in this area so this work still needs some forward-porting. fix is in gerrit (but more work still needed to enable 1.8.2) let&#39;s keep this open for now. While I&#39;ll adapt it for 1.8.2 I may have to change 1.8.1 code to enable forward-compatibility with 1.8.2 and master Aliaksey, code complete is friday and we need to merge everything in by then. <br/> What changes need to be made to ensure forward-compatibility? Minor. I&#39;ll be doing that tomorrow first-priority. I&#39;ve found no further changes to 1.8.1 are needed. 1.8.2 implementation is here <a href="http://review.couchbase.org/14827">http://review.couchbase.org/14827</a> Integrated in github-ns-server-2-0 #333 (See [<a href="http://qa.hq.northscale.net/job/github-ns-server-2-0/333/">http://qa.hq.northscale.net/job/github-ns-server-2-0/333/</a>])<br/> &nbsp;&nbsp;&nbsp;&nbsp;only reuse tap name when changing vbucket filter.<a href="http://www.couchbase.com/issues/browse/MB-4366" title="ns_server is reusing tap names unsafely which causes data loss or inconsistency in replication when a node is removed and added back"><strike>MB-4366</strike></a> (Revision 61bf78355e64fff2e807939fea385862ca6919d5)<br/> reimplemented named tap fix for branch-18. <a href="http://www.couchbase.com/issues/browse/MB-4366" title="ns_server is reusing tap names unsafely which causes data loss or inconsistency in replication when a node is removed and added back"><strike>MB-4366</strike></a> (Revision e3b833480ceb5b7832e22131ed5d3fb532e6ea83)<br/> <br/> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Result = SUCCESS<br/> Aliaksey Artamonau : <br/> Files : <br/> * src/ns_server_cluster_sup.erl<br/> * src/ebucketmigrator_srv.erl<br/> * src/ns_vbm_sup.erl<br/> <br/> Aliaksey Artamonau : <br/> Files : <br/> * src/ns_vbm_new_sup.erl<br/> * src/ns_vbm_sup.erl<br/> * src/ebucketmigrator_srv.erl<br/> * src/ns_server_cluster_sup.erl<br/> * src/cb_gen_vbm_sup.erl<br/> Integrated in github-ns-server-2-0 #337 (See [<a href="http://qa.hq.northscale.net/job/github-ns-server-2-0/337/">http://qa.hq.northscale.net/job/github-ns-server-2-0/337/</a>])<br/> &nbsp;&nbsp;&nbsp;&nbsp;fixed typo in start_vbucket_filter_change. <a href="http://www.couchbase.com/issues/browse/MB-4366" title="ns_server is reusing tap names unsafely which causes data loss or inconsistency in replication when a node is removed and added back"><strike>MB-4366</strike></a> (Revision 5db3c35e8a5ff6a5885271df4466b30c5369fa38)<br/> <br/> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Result = SUCCESS<br/> Steve Yen : <br/> Files : <br/> * src/ebucketmigrator_srv.erl<br/> Dependency Date of First Response Thu, 22 Dec 2011 07:01:55 -0600 Rank 6073 Sprint Priority 7.0 Time In Status