Release Notes - Couchbase Server - Version 2.0 - HTML format

Technical task

  • [MB-7166] - [system test] Resetting rebalance status since it's not really running

Bug

  • [MB-1482] - docs: TAP_CONNECT extras flag size seems to be 4 bytes, not 8
  • [MB-2125] - libconflate/libcurl fix for RHEL 5
  • [MB-2940] - stats program should be able to show the stats without sorting them
  • [MB-3158] - cluster failover/rebalance expected behavior when one or more nodes are out of disk space
  • [MB-3651] - Entire bucket of cluster goes offline (no cache hits) when one node is put offline with ns_memcached002 code
  • [MB-3809] - Stats aggregation consume too much memory and can crash erlang
  • [MB-3848] - ep-engine doesn't eject items from memory when RAM is > 85% but number of item less than 500,000(or <500 per vbucket)
  • [MB-3865] - "/opt/membase/bin/membase server-list" returns incorrect error when no username/pass is supplied
  • [MB-3899] - rest interface allows 4 replicas while the web gui limits it to 3 replicas
  • [MB-3976] - per-server stats not working if servers time is out of sync
  • [MB-3985] - Rpm installations shouldn't install static archives...
  • [MB-3998] - false alert "IP address seems to have changed" in 3-4 nodes cluster under 10k ops/sec set load
  • [MB-4033] - email alerts need to support TLS
  • [MB-4035] - Issue with retaining replicas during aborted rebalance
  • [MB-4047] - autofailover recepients UI section needs tooltip
  • [MB-4059] - Moxi reporting incorrect version
  • [MB-4061] - ep_tap_idle_timeout set to huge value on bucket startup
  • [MB-4065] - Cursor lines up on wrong side of text box when creating dev view
  • [MB-4087] - Erlang breaks when mnesia files are deleted while still running
  • [MB-4111] - rolling upgrade doesn't work if user removes a node , upgrades it and add it back in less than 5 minutes
  • [MB-4138] - Allow TAP to broadcast item persistence
  • [MB-4142] - Editing and saving invalid JSON key in view development screens erases value
  • [MB-4176] - UI seems to truncate (instead of round) memory quota
  • [MB-4210] - Views are no longer automatically compacted
  • [MB-4214] - add and rebalancing a node fails in the first attempt in 2.0 ({gen_fsm,sync_send_event,2}, {menelaus_web,do_handle_rebalance,3}
  • [MB-4222] - Uncaught TypeError: Cannot read property 'swap_total'
  • [MB-4227] - quitting and restarting does not work on osx .app binaries
  • [MB-4229] - "enable auto-failover" section should not be enabled if there are less than 3 nodes in the cluster
  • [MB-4248] - Unsupported CouchAPI calls should fail with indication of being unsupported
  • [MB-4262] - 30% active item resident ratio versus 70% replica item resident ratio in a cluster ( RAM = 260 GB , Data : 150 GB , 180 M items)
  • [MB-4292] - ep-engine does not flush expired replica items if expiry time is 2-3 seconds (memory leak)
  • [MB-4301] - installing couchbase-server 2.0 on an existing membase 1.7.x installation fails
  • [MB-4307] - database compaction does not run at setting 90%
  • [MB-4325] - Unexpected CAS behavior
  • [MB-4340] - Resident ratio shows negative number
  • [MB-4347] - rightscale centos and ubuntu template should have python 2.6
  • [MB-4368] - removed node still see old cluster in rolling upgrade
  • [MB-4378] - ep-engine is printing garbage to the memcached log using -vvv
  • [MB-4385] - stop rebalance button does not work when click during rebalance
  • [MB-4388] - couch.log should go to /var/lib/couch/log instead of /var/log/couch
  • [MB-4391] - servers page doesn't show rebalance progress for the node which is being removed
  • [MB-4394] - memcached crash while rebalancing 15 nodes with 30M items (FATAL: Object returned from mccouch with CAS == 0)
  • [MB-4400] - Unable to create bucket on 2.0.0
  • [MB-4401] - cli commands including cbstats and couchbase-cli don't work in Mac OSX
  • [MB-4407] - failed to restart couchbase server after change bind_address from 127.0.0.1 to 0.0.0.0 in default.ini file
  • [MB-4416] - couchbase 2.0 Web Console does not function properly on Firefox 3.x release
  • [MB-4432] - Replication is failed to send data to remote server
  • [MB-4436] - Unable to listen on 'ns_1@10.169.11.126' on windows
  • [MB-4443] - Remove "Node IP Address" column from XDCR table(s)
  • [MB-4465] - Rest API to set data directory path is broken in 2.0.
  • [MB-4466] - Unable to join 2 nodes on CB-2.0
  • [MB-4467] - Enable core generation by default
  • [MB-4468] - Unstable cluster after rebalance and failover test on windows
  • [MB-4482] - mbcollect info should grab tap and checkpoint stats from each node
  • [MB-4495] - Dashborad Disk Fetches Per Second Graphs does not populate
  • [MB-4505] - deleting and recreating bucket of same name in 2.0 causes views to "reappear" if bucket has millions of items and there are multiple nodes and couchdb indexing process is indexing
  • [MB-4512] - Incorrect total Cluster Storage , summing up index_path and data_path sizes
  • [MB-4530] - failed to add node back after failover in windows
  • [MB-4539] - Unable to drain items in toy-build (399), with couchDB changes.
  • [MB-4544] - Bucket screen reports all buckets being on all nodes incorrectly during rebalance
  • [MB-4565] - Error sending mail through gmail
  • [MB-4567] - No symbols for some threads in memcached core dump
  • [MB-4569] - sqlite doesn't mean python 2.6
  • [MB-4580] - ./cbdumpconfig.escript doesn't run since escript is not in the path
  • [MB-4581] - display a better message for cb* command lines if they need sudo access
  • [MB-4582] - typical server's disk path doesn't fit allocated space and wraps
  • [MB-4584] - Node Ejection did not work due to conflicting vclock config
  • [MB-4591] - delete unused membase icons from program_files\couchbase\server\images
  • [MB-4609] - AdminUI: Incorrect info/text for Active disk queue drain rate stat
  • [MB-4638] - UI error when creaing a view if a node is added but not rebalanced in yet
  • [MB-4652] - the task progress div should hide the edit button
  • [MB-4690] - Couchbase server console has incorrect version info after deploying the server by ./cluster_run
  • [MB-4727] - init script doesn't check for corefile limits when starting
  • [MB-4733] - autofailover 2 minutes delay when failover timeout is set to 30 seconds
  • [MB-4749] - 2.0 source tarball fails when building because libconflate is not bundled in the tar file anymore
  • [MB-4767] - adjusting memory quota on a bucket ejects all data
  • [MB-4783] - overstressed memcached keeps conns in close_wait even if the client has already closed its conns
  • [MB-4797] - querying the view before it is replicated will return "no_db_file" and "not_found,missing" if it is not replicated to the entire cluster
  • [MB-4818] - after a series of TAP dumps, doing a dump of a cache starts to fail
  • [MB-4824] - memcached silently closes connection in auth request [was: ns_server: throws exception in ebucketmigrator_srv during rebalance test]
  • [MB-4828] - rebalancing multiple nodes can hang if a bucket has less than 100k items due to a race condition in tap take-over
  • [MB-4833] - Error in logs from "phone home" service
  • [MB-4835] - UI typo in "delete view" text
  • [MB-4836] - Created a design doc that cannot be deleted
  • [MB-4845] - REST API docs need cleaning up
  • [MB-4862] - upgrading a 1.8.1 couchbase server ubuntu installation might fail to upgrade/migrate the data due to missing openssl dependency
  • [MB-4870] - Spurious failures for memcached t/binary-sasl test
  • [MB-4915] - controller/create_replication rest api call doesn't work properly
  • [MB-4917] - XDCR: Putting a bad document into the _replicator/ causes the server to be unusable
  • [MB-4920] - bucket creation/deletion fails: vbucket map is never ready
  • [MB-4921] - beam.smp taking 100% CPU after rebalance on a large cluster (52 nodes cluster)
  • [MB-4942] - noproc error retrieving bucket topkeys
  • [MB-4947] - ep-engine segfault during heavy write load
  • [MB-4958] - Incorrect stats in UI
  • [MB-4984] - cbstats kvstore is listed twice in usage/help output
  • [MB-4986] - Incorrect backend name from cbstat
  • [MB-4989] - using views after loading data using docloader with sample data ( beer-sample.zip ) on dp4 Mac OSX fails with "from": "local", "reason": "epipe" error
  • [MB-5004] - Automatic Failover page fixes
  • [MB-5047] - invoking flush while indexing and rebalancing causes issues ( MB4594 still a problem in dp4. Undefined set view `default` for `_designXXX` after rebalance )
  • [MB-5071] - moxi make test failures on master branch
  • [MB-5091] - disallow the user to do rpm/dpkg installation if python 2.6 is not installed / default python installer
  • [MB-5095] - Failures on Backing up of Data with Large WAL size on 1.8.1 throws error : Database disk image is malformed
  • [MB-5098] - Failure to restore data from a failed/incomplete backup
  • [MB-5117] - UI and everything is not responsive during failover which may in certain cases take very long time
  • [MB-5119] - Invalid new item id returns when writing a item to Couchstore fails
  • [MB-5129] - After deleting the default bucket and recreating it, it does not show up on the replications drop down menu
  • [MB-5143] - Stability issues - memcached/rest interface not responding
  • [MB-5172] - negative resident ratio after warmup
  • [MB-5174] - ["web request failed", {path,"/pools/default/saslBucketsStreaming"} sometimes at startup
  • [MB-5176] - Rebalancing out a Node from a cluster causes a very high ejection to the disk.
  • [MB-5184] - Find some way to avoid overwriting old erlang crash dumps when startup fails and dumps erlang crash dump
  • [MB-5185] - incorrect text concatination in delete view confirmation screen
  • [MB-5191] - Streaming bucket details and sasl buckets streaming requests are potentially a CPU & memory hogs
  • [MB-5217] - rebalance hang when remove and add back node
  • [MB-5242] - xdcr unit tests failure
  • [MB-5263] - create replication breaks destination UI
  • [MB-5264] - XDCR: sync speed in xdcr is so slow
  • [MB-5265] - It should not allow to create 2 same replication from bucket A to bucket B
  • [MB-5266] - disallow xdcr replication from/to the same bucket of the same cluster
  • [MB-5268] - XDCR: items don't persist to disk in destination cluster in xdcr
  • [MB-5274] - couchbase server 2.0.0 does not start after install on windows 2008 R2
  • [MB-5276] - [system] On rebooting of a node in a large cluster(14 node) noticed sudden drop in replica_item across all the nodes. With time, the replica item count is gradually improving.
  • [MB-5280] - Failover of a node on large cluster setup takes a long time ( around 8 minutes)
  • [MB-5297] - active item/replica item number mismatch post rebalance on 181 (total replica count is lesser than total active count*replica_factor)
  • [MB-5298] - Rebalance failed with reason {case_clause,{{ok,replica},{ok,replica}}} when rebalancing out a node which was failed over due to netwrok connectivity issues but it re-appears while rebalancing
  • [MB-5303] - couchbase.log inside of collect_info unusable on Windows
  • [MB-5317] - some intermittent test failures on ep-engine unit test with make test TEST_TIMEOUT=120
  • [MB-5320] - IE pops up "Out of memory at line: 1" on UI page
  • [MB-5321] - CouchKVStore memory leaks
  • [MB-5326] - Add more explicit messaging to restoration process
  • [MB-5329] - ep engine build failure on linux 32bit (comparison btw unsigned & signed)
  • [MB-5335] - Memory leak in ep-engine make test
  • [MB-5343] - memcached dropped connections: Rebalance failed due to replicator_died: exited (ns_single_vbucket_mover)
  • [MB-5350] - uninitialized memcached process shows invalid memory stats
  • [MB-5353] - Auto-failover failed over 2 nodes(master node and non-master node) when only non-master node was rebooted in a large cluster setup.
  • [MB-5373] - master.couch file being counted and hashed into the CouchKVStore internal vbucket map
  • [MB-5378] - Rebalance failes because of memcached request timeout
  • [MB-5382] - ep engine unit test has memory leaks
  • [MB-5400] - Rebalance failure due to shutdown_error
  • [MB-5406] - ep-engine drops connections too fast when rebalancing out a node
  • [MB-5422] - Request for documentation clarification: http://www.couchbase.com/docs/couchbase-manual-1.8/couchbase-introduction-architecture-nodes.html
  • [MB-5445] - couchbase server crashing, hanging, whatever it's doing I'm not sure
  • [MB-5449] - lots of compaction crash reports logged during view tests
  • [MB-5460] - dialyzer failures in capi_crud
  • [MB-5470] - incorrect value listener changers being assigned for warmup_min_memory/items_threshold
  • [MB-5511] - rebalance hanging in latest 2.0
  • [MB-5527] - phone home rebalance_failure count does not list the reason why rebalance failed
  • [MB-5530] - cbstat/mcstat cannot access to memcached after rebalance
  • [MB-5534] - keys failing to persist to disk with state "ram_but_not_disk" during rebalance
  • [MB-5548] - ep_warmup_count is always zero when access log is present
  • [MB-5573] - Ep-engine changes aren't in the changes file
  • [MB-5587] - set default values for warmup_min_items_threashold and warmup_min_memory_threashold to max (max=100)
  • [MB-5588] - Include evaluation of data traffic when checking isDegradedMode()
  • [MB-5599] - xdcr performance with small number of items is slow
  • [MB-5607] - docloader fails to run at all in 2.0 builds leading to DP5
  • [MB-5612] - Exceeded retries count trying to get consistent keys/values for config replication.
  • [MB-5615] - NS SERVER must enable data traffic via CMD_ENABLE_TRAFFIC
  • [MB-5616] - Add configuration variable, data_traffic_enabled, and set it to true by default
  • [MB-5623] - ns_server needs to deal efficiently with large number of smart client connections that pull streaming vbucket map updates
  • [MB-5669] - data files are preserved after failover on a node with non-standard db path
  • [MB-5672] - [longevity]Item counts/disk-Write Q dips intermittently on large cluster bucket2[ only activity on the cluster - very small load running on the cluster]
  • [MB-5684] - max number of buckets should be set to a default value but configurable using a RESTful API (max set to 10 by default)
  • [MB-5685] - Memcached core due assert failure in CheckpointManager::queueDirty
  • [MB-5706] - user has reported that memcached process on linux with 10 buckets hangs when one of the buckets has 14M items in disk write queue during bulk loading data
  • [MB-5707] - View index with a big int fails badly at some value threshold
  • [MB-5709] - warmup_key/value_count not increasing during the first and second phases of warmup
  • [MB-5710] - warmup continues to load data even after warm is complete
  • [MB-5726] - cbworkloadgen hangs up and print nothing when the ip is invalid
  • [MB-5763] - refresh page reset Pending Removal nodes(UI)
  • [MB-5778] - Aggregate num-Items per server is less than total num_items displayed on the UI
  • [MB-5787] - Bad link: not pointing to CLI docs
  • [MB-5792] - windows installation stalls on windows server 2008 server core edition because internet explorer is not installed
  • [MB-5794] - View parser returns JSON strings unescaped
  • [MB-5797] - docloader doesn't work with 2.0 build
  • [MB-5807] - windows 2.0 build failure from couchkvstore
  • [MB-5808] - Couchstore couchstore_docinfos_by_sequence() does not callback for deleted doc
  • [MB-5812] - Please ammend "cloud best practices" page to include that specifying an IP address is valid as well as a hostname
  • [MB-5813] - Provide documentation on changing Couchbase quotas
  • [MB-5815] - Ammend cbrestore instructions to provide best practices for efficience
  • [MB-5818] - Empty bucket selection on XDCR
  • [MB-5819] - Deleting a target bucket during replication doesn't throw error or warning
  • [MB-5822] - XDCR: Seeing very slow ( currently 0-4 items) replication to destination cluster on enabling replication from source node.
  • [MB-5824] - Rebalance exited with reason {exited, {'EXIT',<0.7371.51>, {badmatch,[{<19424.31984.25>,killed}]}}}
  • [MB-5828] - rebalance is stuck during 1 hour when memory is almost full and then it's failed
  • [MB-5833] - Rebalance exited with reason replicator_died ( after upgrade?)
  • [MB-5837] - Remove references to "membase" in documentation
  • [MB-5839] - No link to "failover with Couchbase"
  • [MB-5849] - Hard Out Of Memory Error from some of nodes in 1.8.1 longevity large cluster test
  • [MB-5858] - Rebalance hangs with exception "replicator_died"
  • [MB-5863] - cbbackup can not backup all the buckets on 2.0.0 build 1416
  • [MB-5869] - Memcached core due to segfault in CheckpointManager::getOpenCheckpointId_UNLOCKED
  • [MB-5870] - Seeing unusually large number on "curr_temp_items" on xdcr destination node
  • [MB-5871] - Setting access scanner sleeptime caused memcached to crash
  • [MB-5879] - Not seeing "top-keys" on the destination cluster with XDCR replication
  • [MB-5880] - Not seeing "deletes per sec" on destination cluster UI ( items are deleted as expected, but the UI shows no stats change on deletes)
  • [MB-5882] - NRU reference bits were set for warmup loaded items
  • [MB-5884] - Moxi port conflict when starting multiple server on the same node using cluster_run
  • [MB-5886] - During Rebalance-in , the xdcr halts while rebalancing is on and item count fluctuates (> expected count) till xdcr completes the replication of available items.
  • [MB-5894] - UI shows ops/sec , CPU spike-high-low behaviour on the destination node on XDCR
  • [MB-5900] - "docs fragmentation" starts at near 100% with no data
  • [MB-5907] - New 2.0 item pager eviction need engine parameter for external setting
  • [MB-5908] - higher drain rate with larger ep queue size in performance graphs.
  • [MB-5909] - xdcr replication status shows "ongoing" , and replication appears to have stopped after replicating 898k items.
  • [MB-5910] - xdcr - "Replication failed " while replicating between 2 clusters - unidirectional replcation. Replicated 1/3rd items and replication failed/stopped
  • [MB-5911] - xdcr-UI should not allow duplicate replication between same set of clusters
  • [MB-5913] - cbrestore can not execute on 2.0.0 build 1447-rel
  • [MB-5914] - cbbackup has a running error on build 2.0.0-1451-rel
  • [MB-5928] - user should not be able to create a xdcr replication on a mixed cluster ( 1.8/2.0) or on a cluster where nodes are down
  • [MB-5929] - active resident ratio in one node is zero after warmup in 1.8.0 cluster
  • [MB-5933] - Edit Document doesn't work if design doc has a % character
  • [MB-5934] - Rebalance exited with reason {noproc, {gen_server,call, [{'capi_set_view_manager-bucket4'ns_1@10.3.2.22'}, {wait_until_added,0}, infinity]}}"
  • [MB-5937] - 2.0 performance under heavy DGM scenario (access phase)
  • [MB-5938] - UI needs to deal with very slow docloader (alternatively, docloader is too slow)
  • [MB-5939] - Memcached losing connection intermittently fails with "bad match" error [Changed bucket password, and started load on the cluster]
  • [MB-5941] - Seeing no data replicated with XDCR replication from cluster2 to cluster1 [cluster2 is loading data]. Node on cluster1 is down after 2-3 minutes.
  • [MB-5942] - Access.log replaces Access.log.old because of which ep_warmup_estimated_warmup_count messed up
  • [MB-5944] - ep_warmup_key_count becomes zero in case of corrupted access.log.old file and no access.log file
  • [MB-5950] - XDCR: need to check stats before verification of all buckets
  • [MB-5954] - Re-arrange suggestion for UI stats
  • [MB-5956] - host:11210 port get down in case of only access.log.old (corrupted one)
  • [MB-5957] - remove some unnecessary data in core dump logs
  • [MB-5958] - beam.smp using 400 MB on a node with couple of connections open to do queries
  • [MB-5962] - Source node unable to restart(on a reboot) when unidirectional replication is on-going and the destination cluster is rebalancing-in 1 new node.
  • [MB-5963] - Seeing few items not drained to disk on destination cluster ( Seeing 6 items in disk-write queue for over an hour).
  • [MB-5966] - Item pager should be able to evict active items even if checkpoint is not done yet
  • [MB-5969] - healthcheck tool need to check whether all active or replica vbucket are on the same node and warn the user
  • [MB-5970] - couchbase 1.8 mac release does not use tcmalloc for memory accounting
  • [MB-5972] - Cannot call method 'concat' of undefined At: http://orange:8091/js/analytics.js:229
  • [MB-5973] - 2.0 cbbackup can't handle large CAS numbers
  • [MB-5979] - [longevity] noticed that on a cluster running for few days that disk_commit time 90th percentile takes 10m when index is building
  • [MB-5989] - backup/restore - better error msg when user/pswd was needed
  • [MB-5998] - cbbackup/cbrestore can not import module pump_csv on build 2.0.0-1482-rel
  • [MB-6000] - UI crashed during rebalance in few nodes
  • [MB-6001] - couchbase server should validate that all vbucket maps supplied are 1024 when config is requested
  • [MB-6008] - kvtiming stats are not available for couchstore backend
  • [MB-6010] - Memcached constantly crashes (exit with139) on destination cluster, during xdcr replication from source cluster and one node on source is rebooted.
  • [MB-6011] - couchbase-cli node-init delete default bucket when set data path
  • [MB-6012] - couchbase server 2.0 set up wizard should show memcached bucket type as the second option in the list or maybe even hide it and have a link that shows more
  • [MB-6014] - Replication status shows "Failed" but data is replicated to destination cluster, on rebooting a node on source cluster
  • [MB-6018] - cbupgrade emits dots instead of detailed progress status
  • [MB-6019] - replication backfill remaining overflow
  • [MB-6022] - Replication fails on unidirectional replication from cluster1 to cluster2, [loading 22M items at source, seeing only 15M items on destination.]
  • [MB-6024] - Port server memcached on node 'ns_1@10.3.121.31' exited with status 134." during xdcr
  • [MB-6025] - rebalance failed due to control connection to memcached disconnected
  • [MB-6026] - mystery of why bg_fetches rises when compaction kicks in
  • [MB-6027] - When backup detects rebalance and halts, backup exits misleadingly with a "done" message
  • [MB-6037] - couchbase-cli docs need more info and updating
  • [MB-6041] - XDC replication keeps on replicating even after replication document is removed
  • [MB-6046] - cbbackup continues when node failover happens in the cluster
  • [MB-6052] - Control connection to memcached on server disconnected: function_clause, during unidirectional replication load from cluster1 to cluster2.
  • [MB-6060] - Adding a node to cluster fails with error incompatible_cluster_version
  • [MB-6064] - Moxi takes 100% CPU with a multi-get of 1M keys
  • [MB-6070] - xdcr- Source node crashes while replicating 12M items on unidirectional replication.
  • [MB-6073] - XDCR - UI fails to show replication progress or past replications. But replicating is going on as expected.
  • [MB-6082] - On windows, couchstore.py should be installed under bin directory
  • [MB-6085] - Build fails on windows for project bucket_engine
  • [MB-6088] - Move cbtransfer from bin/tools back to bin directory
  • [MB-6092] - Json documents with an attribute starting with _ gets stored as type base64
  • [MB-6095] - as a user I would like to have a link to open UI consoles of remote clusters
  • [MB-6107] - query performance degrades over time on spinning HDD
  • [MB-6110] - ops drop during rebalance
  • [MB-6115] - Mismatch on Expiry flags on expired-update items during unidirectional replication from source to destination cluster.
  • [MB-6133] - Mcsoda doesn't load the exact number of items when item count is a small number
  • [MB-6136] - For build 1542 cbtransfer still need the -x try_xwm=0 flag
  • [MB-6149] - It's impossible to set false boolean parameters for view queries from ui
  • [MB-6150] - cbtransfer does not recognize a temporary folder storing 1.8.1 sqlite files as a valid source
  • [MB-6151] - cbbackup doesn't terminiate and throw a Broken pipe error when a node is already down before backup
  • [MB-6166] - View query with a filter on key doesn't work if key has a "+" sign in it
  • [MB-6175] - master node is still orchestrating the rebalance operation even after its failed over
  • [MB-6176] - cbbackup/cbrestore doesn't backup and restore item flag correctly
  • [MB-6179] - Preview a Random Document inconsistent
  • [MB-6185] - Initiator node stops responding while deleting/recreating bucket
  • [MB-6190] - xdcr - MAX_CONCURRENT_REPS_PER_DOC has little effect
  • [MB-6192] - User build on mac os x fails to create new bucket
  • [MB-6195] - backup/restore - need to handle exception during SASL auth
  • [MB-6196] - cbtransfer will throw ValueError for deleted items
  • [MB-6197] - cbtransfer throw errors "AttributeError: 'unicode' object has no attribute 'get'" for build 1566
  • [MB-6211] - There should be only one access scanner task running at any time.
  • [MB-6212] - cbdocloader continues to work and give an unhelpful error message when -s RAM quota > actual RAM quota for the bucket
  • [MB-6217] - badarith in ns_doctor's task aggregation code
  • [MB-6219] - items are not marked as deleted/expired in couchstore after they expire (View query results with stale=false include expired items)
  • [MB-6221] - Rebalance failed with reason replicator_died
  • [MB-6238] - disk commit failed with error : cannot save CouchDB docs for vbucket = 77 when loading 100M items blobs
  • [MB-6239] - membase-cli breaks due to missing timeout module
  • [MB-6240] - XDCR replication stops replication ( only 2.3M items of 3.5M are replicated) while/after rebalancing-In one node at Destination cluster.
  • [MB-6244] - No title for one page in the REST API docs
  • [MB-6245] - cbworkloadgen can't load items to a bucket with a dedicated port
  • [MB-6247] - cbworkloadgen gives no message about Thread when set -r as 0
  • [MB-6254] - Add doxygen style comments to CouchKVStore class
  • [MB-6255] - cbepctl --help do not print the whole command usage
  • [MB-6257] - cbrestore in latest membase-cli pakage can not restore design doc from build 2.0.0-1592
  • [MB-6259] - Refactor CouchKVStore module
  • [MB-6264] - Rebalance fails due to timeout error. Rebalancing in 3 nodes on a 1 node cluster ( with 6 buckets).
  • [MB-6271] - Index compaction termination was seen in the logs after rebalacing in nodes
  • [MB-6272] - Slow first rebalance with multiple empty buckets
  • [MB-6274] - Docloader failures prints rest password in cleartext and failures are not marked in UI
  • [MB-6277] - User-triggered compaction of view `beer-sample/_design/beer` failed: {badmatch, {error, initial_build}}
  • [MB-6279] - Incomplete/missing information in ns_server logs
  • [MB-6313] - erlang prints out "{badrpc,nodedown}" when user attempts to shut down or stop couchbase-server if its already down
  • [MB-6315] - service fails to start sometimes [was: cluster is broken when reboot all nodes at the same time]
  • [MB-6322] - memcached events are being logged to ns_server logs, memcached logs are empty
  • [MB-6326] - Xdcr - Replication has halted at destination, but seeing 14-15k ops/sec on destination nodes.
  • [MB-6327] - XDCR - UI fails to show replication status "Replicating/Failed/Cancelled" during any replication actions. Seeing no past replications on the UI as well.
  • [MB-6329] - XDCR - Error on delete replication on source . Source node goes into pending state.
  • [MB-6336] - Missing replica items (replica items < active items) on idle cluster
  • [MB-6337] - cbcollect_info does not collect memcached logs
  • [MB-6338] - cbepctl usage on buckets with dedicated port
  • [MB-6342] - User-triggered compaction of view `default/_design/dev_view4` failed: shutdown, when rebalancing out a server node
  • [MB-6361] - Active tasks has stale information about view compaction
  • [MB-6362] - Rename "TAP" dispatcher to the more generic name
  • [MB-6364] - Aggregate TAP queue related stats through a single method
  • [MB-6365] - cbbackup/cbrestore doesn't handle external ip address
  • [MB-6366] - Development indexes (stock couch indexes) remain on disk after bucket deletion
  • [MB-6367] - XDCR : Replicating 100k items from source to destination, replication halts with errors -error "db_not_found"
  • [MB-6371] - Failed to start user triggered index compaction
  • [MB-6372] - Unable to get active_tasks listing due to json_encode error
  • [MB-6376] - UUID for key and value are treated as JSON--at least in _all_docs
  • [MB-6381] - We need clearly defined, clean and forward-compatible replications API
  • [MB-6382] - Bucket deletion fails to delete on-disk data and indexes
  • [MB-6391] - max 1.5K ops even set parameter ops_per_sec to 10000
  • [MB-6392] - Indexing 11 ddocs might push system into swap pretty quickly
  • [MB-6407] - memcached does not wait until all items are drained when asked to shut down the engine (Data loss on shutdown)
  • [MB-6410] - Memcached is not releasing memory post bucket deletion
  • [MB-6411] - Failed to load beer-sample bucket on single node
  • [MB-6412] - XDCR stats get populated without xdcr being setup
  • [MB-6415] - Failure to open view group when rebalance is ongoing
  • [MB-6418] - Mac build broke due to collectd tool being gone
  • [MB-6419] - Rebalance exited with reason failed no_vbucket_states_set
  • [MB-6423] - Allow users to set index_path post upgrade to 2.0
  • [MB-6428] - Rebalance failed due to bulk_set_vbucket_state_failed with exit noproc
  • [MB-6429] - Unable to issue rebalance due to bad_return_value,no_active_nodes_left
  • [MB-6435] - Don't use recursive make
  • [MB-6436] - Incorrect license and support address
  • [MB-6437] - libconflate should ignore config.cache
  • [MB-6438] - Use multiple sub directories for libconflate
  • [MB-6440] - queries return "error":"no_active_vbuckets","reason":"Cannot execute view query since the node has no active vbuckets" during rebalance in 1->2
  • [MB-6444] - strange symbol appears in the endkey field for Documents Filter
  • [MB-6446] - couchbase server does not start on mac osx if there is an stale epmd process waiting on different port
  • [MB-6447] - About Couchbase Server dialog on mac
  • [MB-6448] - Unexpected error on couchbase when trying to read documents from admin console
  • [MB-6451] - Remove the invalid vbucket table remover
  • [MB-6453] - [longevity] UI does not show compaction process
  • [MB-6455] - Incorrectly updating the new item counter within PersistenceCallback
  • [MB-6457] - Failed to execute localRandomKey with exit badarg
  • [MB-6458] - impossible to stop bucket compaction, UI errors - Invalid post received: / errMsg: Object Not Found, statusCode: 404,
  • [MB-6461] - cbtransfer sometimes could not read couchstore file
  • [MB-6462] - Race between set_vbucket_state commands and the TAP replication establishment
  • [MB-6469] - get several ns_doctor call error in log page
  • [MB-6470] - cbrestore error: design sink nodes missing
  • [MB-6471] - 2.0-1649 rebalancing is two times slower than on 1.8.1-942( default bucket, 1M items, rebalance in 3 nodes)
  • [MB-6473] - stale=ok\update_after queries are failed with timeout error rebalance out 10->9 nodes
  • [MB-6479] - Couchbase Server versions from "About Couchbase Server" should match the About from admin console
  • [MB-6480] - Slow initial indexing performance on 2.0
  • [MB-6481] - Rebalance failed due to the timeout in bulk_set_vbucket_state operation with continuous view queries
  • [MB-6485] - restore of design docs that use built-in reduce functions with extra spaces (from 2.0DP4) does not work
  • [MB-6487] - cbtransfer is not compatible without Python 2.4
  • [MB-6490] - Rebalance failed with reason "Partition 687 not in active nor passive set" in add in node rebalance
  • [MB-6495] - Queries don't show any results on Windows when cluster has more than 1 node, collation error
  • [MB-6499] - remove "edit" button from production views since user is not allowed ( or should be allowed to edit them )
  • [MB-6500] - UI - Clicking on the document key hyperlink shown in view results, doesn't load it in sample document pane
  • [MB-6503] - query for just created view returns {"error":"error","reason":"inconsistent_state"} after warming up a node
  • [MB-6517] - deleting development ddocs can delete production index files
  • [MB-6519] - cbtransfer can not recover data from couchstore files in build 2.0.0-1677-rel
  • [MB-6520] - [consistent views enabled] Rebalance hungs after failover 2 nodes
  • [MB-6522] - Warmup key count does not match with Hot keys
  • [MB-6523] - rebalance failed after crasher: initial call: ebucketmigrator_srv and then crashes one by one : erlang:apply, ns_vbucket_mover, ns_memcached, ns_port_server...
  • [MB-6526] - 1.8.1 and 2.0 mixed cluster rebalance is broken
  • [MB-6538] - In rare cases CRC codes dont match when reading data from couch file
  • [MB-6539] - memcached does not build with clang/clang++
  • [MB-6544] - Couchstore should not segfault when couchstore_close_db is passed NULL
  • [MB-6547] - Big performance drop when doing small proportion of bgfetches
  • [MB-6549] - Need to address license issue with 3rd party source code supervisor2.erl added to source tree
  • [MB-6552] - couchbase-server unable to restart due to an erlang crash
  • [MB-6560] - we cannot serve ddocs list on node that doesn't host bucket yet [was: 500 Internal Server Error in request of pools/default/buckets/bucket-0/ddocs?_=1347042398100 during rebalance]
  • [MB-6564] - observing 1000 items on a node takes 1 second on average
  • [MB-6565] - Rebalance exited with reason "Couldn't open database `default/0`, reason: {not_found,no_db_file}"
  • [MB-6568] - Error occured reading set_view _info of ddoc ddoc1: {"error":"error","reason":"{db_open,<<\"default/master\">>,{not_found,no_db_file}}"} -----crasher: initial call: couch_set_view_group
  • [MB-6569] - Unidirectional Replication - Mismatches seen in CAS and sequence numbers in automated tests with just create and delete tasks
  • [MB-6570] - Web UI alert if compaction has crashed and not compacting any more
  • [MB-6573] - [longevity] rebalance failed due to error "Resetting rebalance status since it's not really running" when there are major page faults on some of the nodes in the cluster
  • [MB-6574] - Bidirectional Replication: Automated test failing during data validation (Replication with ddoc compaction)
  • [MB-6576] - Seeing very frequent compaction on clusters, during xdcr
  • [MB-6579] - swap Rebalance exited with reason {{bulk_set_vbucket_state_failed, couldnt_connect_to_memcached, with a list of crashes before: ebucketmigrator_srv,erlang:apply/2,ns_single_vbucket_mover
  • [MB-6582] - erl_crash occurred on node after removing max number of buckets(31) in the cluster and rebalance out the cluster/ eheap_alloc: Cannot allocate 5568010120 bytes of memory (of type "heap").- incremental_rebalance_in_out_with_max_buckets_number
  • [MB-6584] - Remove update_seq filter from "Filter Results" drop down in UI on Views Tab
  • [MB-6585] - Rebalance fails while doing swap rebalance with bidirectional replication
  • [MB-6590] - Memcached logs are not gzipped anymore?
  • [MB-6591] - some issue processing \uFFFF in query string in our binary packages [was: garbage chars in query string when Looking up doc via UI ]
  • [MB-6593] - [windows] node becomes unreachable, port 11210 is in CLOSE_WAIT state already 15 min
  • [MB-6594] - Server error during processing: ["web request failed", {path,"/pools/default/tasks"}, {type,error}, {what, {badmatch, {false, [{cancelURI, when add ONGOING REPLICATIONS with non default bucket
  • [MB-6597] - couchbase server is terminated on one of the existing nodes after rebalancing 2 node to 3 node cluster with 1 bucket and 5 ddocs/1 view per ddoc
  • [MB-6598] - Very choppy stats with Bi directional xdcr with moderate/low load.
  • [MB-6600] - Node went down during initial indexing of 100M x 2K items
  • [MB-6601] - cbenable_core_dumps does not enable core dumps for all users in centos 5.x
  • [MB-6602] - rename replication tab in web console to XDCR
  • [MB-6603] - [longevity] erlang crash during swap rebalance test
  • [MB-6605] - unexecpted swap usage on node while loading data
  • [MB-6608] - cbrestore doesn't support --data or -d but it appears on Fun Spec on wiki
  • [MB-6609] - got error when running healthchecker report script
  • [MB-6614] - open_doc_with_docinfo failure due to an invalid docinfo being passed
  • [MB-6616] - Missing keys while loading data from the access log during warmup
  • [MB-6618] - beam.smp eats up all memory on node beyond the membase quota and causes oom killer
  • [MB-6620] - Preview Random Document in the View tab doesn't show the actual values for rev, expiration, flags
  • [MB-6623] - Example in docs is out of date (setting -name option in couchbase-server)
  • [MB-6624] - [32-bit] Disk write failed for item when loading on 32-bit machines with McSoda
  • [MB-6629] - Slow incremental indexing and view compaction on idle cluster
  • [MB-6636] - possible to leave view for new when it was recreated
  • [MB-6655] - memcached exited with status 134 in during bucket deletion
  • [MB-6656] - [windows] write commit message during rebalance 1->2
  • [MB-6659] - rebalance failed with reason {badmatch,wrong_rebalancer_pid}
  • [MB-6664] - [windows 7] bucket creation on windows 7 sometimes timeout if bucket is created immediately after another bucket was deleted
  • [MB-6670] - unexepcted {badrpc,nodedown} error on uninstall
  • [MB-6678] - [consistent views enabled] Rebalance 2->3 failed {db_open_error,<<"default/688">>, {not_found,no_db_file},
  • [MB-6682] - [consistent views enabled]queries during rebalance out returns no_db_file error
  • [MB-6684] - remove memcachetest tool from 2.0
  • [MB-6685] - [consistent view enabled] view compaction failed with no_db_file error
  • [MB-6687] - on an idle cluster there were too many I/O requests by beam.smp ( compaction possibly ? )
  • [MB-6688] - test observe performance duirng rebalance
  • [MB-6690] - XDCR replication page should show ip/bucket for replication created.
  • [MB-6694] - pump_gen.py should return total ops as total msgs for pump engine
  • [MB-6697] - 2.0 Beta Readme.txt file contains reference to "developer preview", should be "Beta"
  • [MB-6698] - regular crashing on 2.0 beta on debian/ubuntu 32-bit
  • [MB-6699] - cannot cancel compaction from web console, possible browser request problem
  • [MB-6701] - base64-encoded documents for non-json content is displayed in encoded form on document and view GUI pages
  • [MB-6703] - Incorrect warning logged by fatbuf_get
  • [MB-6705] - Slight change to the Documentation page, possibly?
  • [MB-6706] - [system test] rebalance hang when add nodes to cluster
  • [MB-6707] - [system test] rebalance does not stop when click on "Stop Rebalance" button
  • [MB-6716] - Extend look back of logs
  • [MB-6733] - [system test] query view returns error {"error":"unknown_error","reason":"badarg"}
  • [MB-6745] - [system test] rebalance failed with error "Partition 67 not in active nor passive set" with consistent view enable
  • [MB-6750] - [system test] web request failed "/pools/default/buckets/default/localRandomKey"}, {type,error}, {what,badarg},
  • [MB-6752] - [system test] compaction_daemon crashed due to {not_found, no_db_file}
  • [MB-6754] - [system test] database compaction did not run in some nodes
  • [MB-6758] - ubuntu-32-2.0-single-node CreateMembaseBucketsTests.test_default_on_non_default_port failed on build 1776
  • [MB-6760] - swaprebalance.SwapRebalanceFailedTests.test_failed_swap_rebalance get stuck for build 1776 while rebalancing
  • [MB-6761] - XDCR Error Logging for Flush bucket and Delete bucket on source/destination
  • [MB-6763] - XDCR Error Logging
  • [MB-6766] - XDC Num Documents replicated consistently drops during rebalance at destination cluster with ongoing load/mutations.
  • [MB-6770] - [system test] disk write commit failed during warmup
  • [MB-6771] - Swap Rebalance failed on build 1779 for Ubuntu 32 bit
  • [MB-6773] - JSON docs with formerly reserved keys display incorrectly in the UI
  • [MB-6775] - moxi does not handle errors during append correctly
  • [MB-6782] - [windows] rebalance 1->2 failed reason {badmatch,{error,closed}
  • [MB-6785] - spatial views needs to be documented as experimental feature in 2.0 documentation
  • [MB-6791] - [system test] view query hang during rebalance
  • [MB-6794] - cpu_util reported in the UI might not include utilization from all cores
  • [MB-6795] - [windows] no_db_file error appears when trying to get ddoc info in adding node during rebalance in
  • [MB-6797] - When storing a document with a property starting with an underscore, the document is stored as an attachment.
  • [MB-6800] - [system test] updater stopped with reason: {updater_error, shutdown}
  • [MB-6802] - CbDocloader regression
  • [MB-6803] - [windows] A mistery node appeared during rebalance
  • [MB-6807] - couchbase server crashes in mapDoc at mapreduce.cc:184 (kernel: beam.smp[4552]: segfault ) if map view function produces run time errors
  • [MB-6808] - beta tester reported seeing UI errors : "web request failed", {path,"/pools/default"} when there are multiple buckets
  • [MB-6809] - Disable Flush with XDCR on the source cluster
  • [MB-6812] - CLONE - multi-key get sometimes, transiently returns duplicate values during rebalance
  • [MB-6818] - [windows] After upgrade from 1.8.0 to 2.0(1791) all data is lost
  • [MB-6819] - Rebalance exited with reason {not_all_nodes_are_ready_..}, after upgrade 1.8.0->2.0 when add 2.0 node to mix cluster
  • [MB-6820] - master node is in pending state after upgrade 1.8.0r -> 2.0
  • [MB-6826] - "maybe" dynamically configure index_aware after bucket creation - internal setting
  • [MB-6830] - [system test] disk write commit failed in idle cluster
  • [MB-6831] - Display queue stats in-memory for xdcr queue on the source cluster
  • [MB-6832] - There is no way today for end-user to know from the Web-Console , whether XDCR is up/running/failed/terminated.
  • [MB-6834] - Rebalance fails with reason {noproc, {gen_server,call, [{'janitor_agent-default','ns_1@10.2.2.60'} when rebalance is stopped/resumed
  • [MB-6835] - upgrade 1.8.0r-55 -> 2.0.0-1810-rel all data is lost
  • [MB-6837] - Add missing details to documentation regarding automated index updates
  • [MB-6841] - XDCR fails when admin password contains special characters
  • [MB-6842] - long values greater than 2^53 need to be stored as string
  • [MB-6843] - Correctly handle deleting cluster reference that's used by currently defined replications
  • [MB-6848] - Incorrect stats curr_items/total_items after flushing a memcached bucket
  • [MB-6852] - debian package allows downgrade but fails whereas rpm package detect the version difference
  • [MB-6853] - In Mixed cluster of 1.8.1 and 2.0.0 it allows to create views on 2.0.0 node
  • [MB-6854] - Broken look of memcached bucket on list of buckets
  • [MB-6855] - Disk space alerts don't check index_path
  • [MB-6856] - Fatal error in persisting SET
  • [MB-6857] - Reduce the logging on 2.0 (or increase the default logging params i.e num_files, size of each log)
  • [MB-6858] - Provide simple walkthrough-style documentation to replicate operations exposed in the web UI
  • [MB-6859] - The collect_info logs is missing the mapreduce_errors log file.
  • [MB-6860] - [system test] Index file descriptor leaks
  • [MB-6862] - Stats vkey returns item_deleted for evicted keys
  • [MB-6863] - Default data path restored after upgradation from 1.8.1 to 2.0.0
  • [MB-6864] - On windows Mixed cluster of 1.8.1 and 2.0.0 it allows to create views on 2.0.0 node
  • [MB-6865] - Memcached client doesn't receive TMPFAIL error when deleting data while bucket flush is in progress
  • [MB-6869] - Sometimes node being ejected fails to reset its config
  • [MB-6871] - stale=false including recently deleted items
  • [MB-6873] - exhausted limit of open fds during system tests
  • [MB-6874] - turn off debug level logs for 2.0 product
  • [MB-6877] - Exit on non-updater process: config_change
  • [MB-6878] - Server error during viewing stats by the hour/day/week as "{what,badarith},"
  • [MB-6879] - After Upgrade from 1.8.1.x to 2.0.0.x Keys do not get expired after expired time elapsed
  • [MB-6880] - view queries fails with error 'missing' during rebalance in
  • [MB-6884] - Sharp drop in replica resident ratios on Destination clusters only, XDCR.
  • [MB-6885] - [system test] ebucketmigrator_srv:confirm_sent_messages:679]Got error while trying to send close confirmation: {error,enotconn}
  • [MB-6886] - Complete data loss on node restard
  • [MB-6887] - Fix documentation on Couchbase Network Ports page
  • [MB-6889] - Data loss after restart when cluster started by cluster_run
  • [MB-6890] - couchdb make dist fails because 031-doc-from-binary.t is not removed from etap tests
  • [MB-6891] - Fix sporadic failure in warmup test (bug3522 test)
  • [MB-6892] - Change in Data path default location causing data lost during upgrade on ubuntu cluster as well
  • [MB-6893] - Change in Data path default location causing data lost during upgrade on Windows cluster as well
  • [MB-6894] - Developer Guide : Invalid link to 1.8 documentation instead of 2.0 (chapter 1.5)
  • [MB-6896] - prevent UI from trying to load samples during rebalance (was: Loading sample db beer-sample during rebalance gives "Unexpected server error, request logged")
  • [MB-6897] - UI - Views Page is not automatically refreshed, can show incorrect information
  • [MB-6899] - nodes/self doesn't return couchApiBase
  • [MB-6903] - cbbrowse_logs gives an error when /bin/sh is symlink to dash
  • [MB-6905] - typo in healthchecker collector.py
  • [MB-6907] - Bucket deletion was stuck by repeative reconnection to mccouch
  • [MB-6908] - Links from 2.0 release to 1.8 or older SDK
  • [MB-6909] - Small doc inconsistency between Developer Guide and Server Manual in the introduction
  • [MB-6910] - Typos in Developer Guide 2.0
  • [MB-6912] - cbflushctl tool is not renamed to cbepctl
  • [MB-6913] - Incomplete documentation about cbepctl tool
  • [MB-6914] - erlang core dump in test rebalarebalance_in_with_ddoc_compaction with rebalance in 1 node and compaction ( ubuntu 10_64)
  • [MB-6915] - occasional evperf test query thread crash on startup
  • [MB-6916] - flush is not completely synchronous
  • [MB-6917] - Upgrade fails on centos5.4 from 1.8.1 to 2.0 with error : "var/lib/couchbase/config/config.dat already exists while copying"
  • [MB-6918] - [system test] unable to save custom data path in UI
  • [MB-6919] - Possibly incorrect OUTBOUND XDCR OPERATIONS seen on source cluster, very slow replication (very less number of sets) observed with number of items > 200M
  • [MB-6920] - System testing for XDCR: XDC Replication queue is continuosly growing under moderate load.
  • [MB-6921] - Sporadic failures in set_with_meta exp persisted
  • [MB-6922] - ns_server should send 302 reply when view query hits node without any active vbuckets
  • [MB-6924] - XDCR: stat "XDCR replication queue" in summary section should be changed to "XDCR docs to replicate"
  • [MB-6925] - Avoid unnecessary IO in compaction daemon's view cleanup phase
  • [MB-6926] - Remove unused per-node uuid
  • [MB-6927] - Fix sporadic failure in duplicate items on disk test
  • [MB-6928] - cbworkloadgen doesn't work with both python24 and python26 installed on one machine.
  • [MB-6929] - Ep-engine error retrieving doc infos (was: Queries having parameters startkey\endkey, debug=true and stale=false have lost keys
  • [MB-6930] - On the Documentation page (www.couchbase.com/docs), make emailIDs of editors not scannable
  • [MB-6933] - [system test] failed add node back in rebalance out
  • [MB-6934] - Displaying XDCR Replication error messages/warnings.
  • [MB-6935] - "inconsistent_state" in query result after rebalance
  • [MB-6936] - We don't log database and index dir changes at all
  • [MB-6937] - Smart clients connection/disconnection messages are spamming our debug logs
  • [MB-6938] - After we upgrade a node from 181 to 200, the value of a key might change.
  • [MB-6939] - XDC queue grows and checkpoint commit failures in bi-directional XDCR with front-end workload
  • [MB-6940] - When no_active_vbucket error is received a non user-friendly error appears on UI
  • [MB-6941] - UI - Active/replica item counts could overlap rebalance progress bars
  • [MB-6944] - Very small pending XDC queue on source cluster( less than 1 M) is not replicated.
  • [MB-6945] - DocInfo.rev_seq is declared as 64-bit, but only 32 bits are persisted
  • [MB-6947] - set_view_outdated errors from couchdb should not be logged with log level error
  • [MB-6949] - ebucketmigrator mis-handles situation when all vbuckets to replicate are not ready
  • [MB-6950] - Node unable to restart after reboot on cluster
  • [MB-6951] - ns_pubsub debug messages about subscribing/unsubscribing are not too useful and can spam logs
  • [MB-6952] - Reference bit (NRU) of newly replicated items on the dest. node should not be set
  • [MB-6953] - [system test] rebalance hang when doing swap rebalance
  • [MB-6955] - [system test] Rebalance failed with reason "Partition x not in active nor passive set" in swap rebalance
  • [MB-6956] - [windows]Memcached exited and rebalance failed with reason {unexpected_exit, wait_checkpoint_persisted_failed,bucket-0
  • [MB-6957] - [windows] Rebalance exited with reason {{badmatch, {failed, [{'ns_1@10.3.2.137',no_vbucket_states_set}]}},
  • [MB-6958] - collect_info doesn't work on Mac
  • [MB-6960] - "manually" running innocent-looking mapreduce code causes erlang crash
  • [MB-6961] - result of INCR operations should show up in views as JSON number, not base64 string
  • [MB-6962] - Memcached killed by O/S (oomkiller) even without any load or replication going on (erlang too much memory)
  • [MB-6963] - Make sure all stats are listed in stats.org file
  • [MB-6964] - Erlang using up a lot of memory on idle node (a node where memcached was killed (oomkiller) but the node recovered)
  • [MB-6965] - [system test] compaction progress does not log in couchdb log
  • [MB-6967] - dev-views index seems not to be built (at least for _count reduce) during rebalance with consistent view (1DD, 4 views) - build 1868
  • [MB-6969] - various errors when building couchbase-server from source tarball distribution
  • [MB-6970] - users which install couchbase server 2.0 on centos 6 and ubuntu 12.04 they need to install missing dependencies
  • [MB-6971] - users which install couchbase server 2.0 on centos 6 and ubuntu 12.04 they need to install missing dependencies
  • [MB-6973] - update setup wizard to picks 60% as the bucket size by default (to leave room for indexing pages in the cache)
  • [MB-6975] - ops drop for mixed workload
  • [MB-6977] - sample datasets are not loaded when user loads the bucket data through the setup wizard
  • [MB-6978] - inconsistency on the UI on the view screen between beer-sample and default bucket views
  • [MB-6980] - erlang core dump in incremental rebalancein with queries / test signal 11, Segmentation fault( Centos 5.7)
  • [MB-6986] - UI issue in displaying documents returned by _all_docs
  • [MB-6990] - User is able to save uncompilable map function for spatial view
  • [MB-6991] - Segmentation fault in ep-engine shutdown
  • [MB-6992] - rebalance hangs after failing over disconnected node
  • [MB-6994] - xdcr tasks status aggregation code depends in a hard way on too many currently unused xdcr stats
  • [MB-6995] - set_db_and_ix_paths and ns_server:restart are raceful which may cause couch_db to fail to start
  • [MB-6996] - offline upgrade from 2.0 beta to 2.0 is not supported [was: Missing data/error message after upgrading 2.0beta to latest build]
  • [MB-6998] - Server error while viewing the bucket stats w/ XDCR - "{what, {timeout, {gen_server,call,[<0.30480.0>,target]}}},"http://ec2-184-169-227-56.us-west-1.compute.amazonaws.com:8091/
  • [MB-7000] - [system test] online upgrade failed due to "Conflicting configuration changes to field {node,'ns_1@10.6.2.39', config_version}"
  • [MB-7002] - use sync/scheduled threads and 512 GC for windows
  • [MB-7003] - [system test] couchbase server of 2.0 node does not start after added to 1.8.1 cluster
  • [MB-7004] - XDCR - Document Queue, Queue size UI stats broken, does not display correct Doc in Queue/ Queue Size
  • [MB-7005] - In outgoing XDCR stats, number of successful checkpoints issued is displayed, would be useful if number of failed checkpoints is displayed as well.
  • [MB-7008] - Reduce the vbucket reference overhead in the flusher loop
  • [MB-7009] - Inserting keys with \0, \n with MemcachedAsciiClient using moxi port returns error -1
  • [MB-7010] - [windows] Rebalance exited with reason "Partition 236 not in active nor passive set"
  • [MB-7012] - Server failed to start after upgrade from 1.8.1 to 2.0.0 need to do it manually
  • [MB-7014] - nodes are not cleaned within a minute after removing from cluster
  • [MB-7017] - XDCR: extremely high number of connections in bidirectional 4->4 nodes performance tests
  • [MB-7019] - cbcollect_info clutters diag output with curl progress bar
  • [MB-7021] - We should undo now unused broadcasting of persisted checkpoints into mccouch_events
  • [MB-7022] - Clean up the empty flush queues for vbuckets that don't exist anymore
  • [MB-7023] - There is a race condition between vbucket deletion and recreation.
  • [MB-7024] - Deleting a bucket via the REST API fails with (500) Internal Server Error on Windows
  • [MB-7025] - Not able to setup replication , getting a Server error during create replication {what,{badmatch,{not_found,no_db_file}}},
  • [MB-7026] - Drop disk backfilled items that were already transmitted by the memory backfill
  • [MB-7028] - Unit test for "Drop disk backfilled items filled by memory backfill"
  • [MB-7029] - spatial views: error message when try to save incompilable view is not user friendly
  • [MB-7032] - Doc content is not displayed if content is non-json string
  • [MB-7034] - beam.smp segmentation fault when running access phase of view performance tests
  • [MB-7036] - Rebalance gets stuck while waiting for checkpoint being persisted
  • [MB-7037] - Stopping rebalance and then doing failover loses some data (i.e. reliable replica building is supposed to prevent that)
  • [MB-7039] - When opening index, re-create and configure replica index file if it is missing
  • [MB-7040] - Couchbase 2.0 manual does not include complete details about where log files are located and their contents
  • [MB-7041] - sample buckets are loaded but views are not created via settings or set up wizrd
  • [MB-7042] - unable to fully delete bucket
  • [MB-7043] - Rebalance exited with reason not_all_nodes_are_ready_yet for stop_start rebalance test
  • [MB-7044] - unable to add node back to cluster:Failed to add node 10.3.3.121:8091 to cluster. This node cannot add another node ('ns_1@10.3.3.121') because of cluster version compatibility mismatch
  • [MB-7045] - duplicate results when retrieving all_docs due to duplicate in sequence tree on the database file
  • [MB-7046] - Connection between ep-engine and erlang mccouch is constantly being timeout and reset on windows build 1908
  • [MB-7047] - Document the guideline for reduce function
  • [MB-7049] - cbcollect_info orders memcached.log.* files incorrectly
  • [MB-7050] - EP Engine background fetch task unnecessarily goes to snooze
  • [MB-7051] - [system test] compaction stopped right after it started in idle cluster
  • [MB-7052] - memcached.log.txt under C:\Program Files\Couchbase\Server\var\lib\couchbase\logs is empty on Windows installation
  • [MB-7053] - Expired items are not excluded from production/dev views until expiry pager runs ( which deletes the items from database permanently )
  • [MB-7054] - cbcollect_info doesn't gather stats info
  • [MB-7055] - unknown_error, badarg when map function doesn't emit anything with a reduce function present but view called with reduce=false
  • [MB-7056] - Memcached core caused rebalance in (crash on node that was being rebalanced in) to fail on cluster with bidirectional and unidirectional replication set up
  • [MB-7057] - [system test] Rebalance failed with reason "Partition x not in active nor passive set"
  • [MB-7059] - [system test] beam.smp is running at node 43 but all other nodes saw this node down
  • [MB-7063] - "OUTGOING XDCR OPERATIONS" section missing with XDCR still going on (although noticed in nodes that are in "Pend" state)
  • [MB-7064] - duplicate rows in view results when mapping doc id to null
  • [MB-7066] - Add collect_info tools to collect logs to the logging docs
  • [MB-7067] - UI should limit documents list skip to 1000 to avoid overloading view merger
  • [MB-7070] - Compaction stops on the node with zero draining rate (DWQ > 1M) after data loading
  • [MB-7073] - Cannot install 2.0 beta on Mac OSX (build 1723 and 1870)
  • [MB-7078] - warmup can fail stopping data loading even when it reaches the low water mark
  • [MB-7082] - extra items appears in query results during rebalance
  • [MB-7083] - Rebalance exited with reason bulk_set_vbucket_state_failed EXIT' {{{{unexpected_reason,{timeout,{gen_server,call, m silence_upstream]
  • [MB-7085] - freshly created bucket doesn't update vbucket map history causing swap rebalance not to work (was: [system test] swap rebalance only works if added node is the old node of cluster)
  • [MB-7086] - windows - Error encountered when trying to run cbworkloadgen
  • [MB-7087] - Corrupted database files (was: initial indexing constantly fails because of "doc loader error error: {file_corruption,<<"file corruption">>}")
  • [MB-7088] - Memcached cpu util goes up in recent build
  • [MB-7092] - Json encoder/decoder throws Server error during processing:{what,{ucs,{bad_utf8_character_code}}},mochijson2,json_encode_array,2},
  • [MB-7096] - Beam.smp generates a core - node down after 3 days runtime, with XDCR (both bidirectional and unidirectional)
  • [MB-7099] - cbtransfer gen: generates JSON with body field called "min-value-size"
  • [MB-7108] - Rebalance exited with reason {{bulk_set_vbucket_state_failed after crash in tap_replication_manager:init/1 registered_name: 'tap_replication_manager-bucket-1' exception exit: {timeout, {gen_server ,{start_vbucket_filter_change
  • [MB-7109] - unable to get spatial index results on a windows and ubuntu cluster
  • [MB-7113] - windows - constant restarts of mb_master during small scale performance tests
  • [MB-7115] - Rebalance operation failed repetitively while trying to rebalance in 5 nodes and rebalance out 3 nodes on a 5 node cluster, reason possibly because: "Unable to listen" to one of the nodes that was being rebalanced out.
  • [MB-7117] - Query parameter validation should return an explicit error when skip value is a negative integer
  • [MB-7119] - UI unable to handle error generated when view query has negative value for skip
  • [MB-7121] - crash of Mac OS X launcher during attempted shutdown
  • [MB-7123] - clicking on show results for the first time on a small dataset returns empty results on development views
  • [MB-7124] - view editing U/I incorrectly capitalizes document ID's
  • [MB-7126] - add if guards to the views for sample databases
  • [MB-7128] - increase ns_server memcached outer_heavy and inner timeout
  • [MB-7129] - Multiple nodes go down with erlang crash (crash.dump available) in a 10:10 XDCR setup and erlang possibly hung in a couple of nodes as well, all in the same cluster
  • [MB-7132] - Change "docs to replicate" tab in outgoing XDCR operations, to "mutations to replicate" to avoid an ambiguous interpretation
  • [MB-7133] - Low XDCR replication rate, seen with high number of gets but very low number of sets
  • [MB-7135] - Include kvtimings in cbcollect_info
  • [MB-7137] - [windows] cbbrowse_logs incorrectly doesn't collect all logs
  • [MB-7139] - enable replica index in the UI setup wizrd
  • [MB-7141] - /diag doesn't send stats log
  • [MB-7142] - Big docs caused the browser to crash when tried loading json files of size averaging between 2 - 4MB using cbdocloader
  • [MB-7146] - Update supported platforms on Couchbase 2.0 Documentation
  • [MB-7147] - swap rebalance slowness
  • [MB-7159] - views aren't "live" during app development for dev views when full_set is not true
  • [MB-7161] - under higher sets per second clients dont get all view results after observing for all mutations
  • [MB-7162] - Can not restore through moxi to 181 cluster
  • [MB-7169] - missed items after rebalance out 2 failover nodes with replica=2 in 5 nodes cluster
  • [MB-7175] - able to delete a bucket via couchbase-cli without specifying the user and password
  • [MB-7180] - [windows] Initial index never finishes if keep loading data
  • [MB-7186] - couchbase-cli server-add option fails now when adding a node even though user is passing the right credentials
  • [MB-7187] - UI saves extra junk to design docs when publishing them
  • [MB-7190] - 'timings' statistics not consistent between 1.8.1 and 2.0
  • [MB-7194] - cbbackup fails with no design source node error
  • [MB-7197] - reducing number of schedulers from 128 to 16
  • [MB-7198] - Wrong number of checkpoints seen as issued
  • [MB-7200] - cbcollect_info should capture ddocs/view map\reduce separately (not in the usual log files)
  • [MB-7206] - Slow XDCR performance
  • [MB-7208] - Unable to view stats on failed-over-node during xdcr+rebalance, server_error on call for capi_set_view_manager-biXDCR_bucket',
  • [MB-7210] - ops drop during thruput test when compaction kicks in
  • [MB-7213] - add documentation for moving from one recent build to a more recent or an earlier version
  • [MB-7223] - Small UI regression in look of stats
  • [MB-7227] - Unable to run cbstats on mac
  • [MB-7229] - Item flag changed after backup and restore from 1954 to 1954
  • [MB-7237] - couchbase-cli: cluster was not rebalanced after launching rebalance
  • [MB-7238] - ns_server is still validating ip address in ip file even if erlang already has node name defined (was: 2.0 Build 1941: Couchbase Server does not start after a change in IP, server is looking for the old IP even after the hostname resolves to the new one.)
  • [MB-7240] - tools - recovery of 2.0 couchstore files corrupts item flags
  • [MB-7242] - cbcollect_info stores output file in wrong folder
  • [MB-7243] - Need to remove /opt/couchbase if upgrading from 2.0beta/2.0c to 2.0
  • [MB-7244] - For online/offline upgrade the documentation suggests to shutdown couchbase before running cbbackup
  • [MB-7247] - rebalance-in failed with beam.smp resident set size ~20GB
  • [MB-7252] - remove cbclusterstats from 2.0 (cbclusterstats doesn't work with python version less then 2.6)
  • [MB-7254] - error when launch cbdump-config or ebucketmigrator : /usr/bin/env: escript: No such file or directory
  • [MB-7255] - [windows] cbcollect_info doesn't work: tool is trying to access linux path
  • [MB-7256] - cbrestore, cbtransfer exit with ImportError: No module named zlib if python installation on the box does not have zlib library ( when python is built from source )
  • [MB-7257] - All data will be deleted when change Data Files path on online server
  • [MB-7267] - tools - cbtransfer fails to upgrade 1.8.1 sqlite db files into 2.0 couchstore files when data flag is non zero
  • [MB-7270] - Rebalance constantly exited with reason bad_replicas after rebalance with wamup node( Bad replicators after rebalance: Missing = [{'ns_1@10.3.121.113','ns_1@10.3.121.112',205},,, )
  • [MB-7271] - provide documentation for users on how to backup/restore from mac_osx->centos or vice versa ( 64->1024 or 1024->64 vbuckets)
  • [MB-7272] - memcached/ep-engine crashes in flusher or other paths when it receives a shutdown message from ns-server
  • [MB-7273] - Update "Handling Ip Address Change" documentation to include changing the /opt/couchbase/var/lib/couchbase/ip* file.
  • [MB-7279] - XDCR error: "database out of sync" seen on source after there's been a cluster change (failover/rebalance) on the destination
  • [MB-7283] - update documentation for upgrading from 1.8.x node to 2.0 when init script is modified for host names
  • [MB-7285] - Occasional badmatch in index updater if it performs cleanup (was [system test] view manual compaction failed with error "updater_died, {updater_error, {badmatch, {ok, <<0,0,58,)
  • [MB-7289] - [windows] Upgrade fails on cluster from 181 to 2.0 ( node referenced using hostnames) even after changing "ip" file
  • [MB-7297] - ubuntu 10 & 12.04 32-bit unable to upgrade 1.8.1 to 2.0 latest build ( 1969 ) because cbtransfer is unable to open database file
  • [MB-7299] - [system test] database compaction crashed during and after rebalance
  • [MB-7303] - unable to update centos 32-bit 1.8.1 to 2.0 latest build due to issues when executing cbupgrade
  • [MB-7304] - Need explanation of Beer DB sample views
  • [MB-7305] - Screenshots renders very large
  • [MB-7306] - Memory leak in couchstore's lookup_callback()
  • [MB-7307] - Couchbase Console disk overview doesn't deref data dir symlink
  • [MB-7312] - memcached does not start after upgrading 1.8.1 to 2.0 build because ns-server is unable to start mcccouch ( this happened after upgrading a 2 node cluster where one node was upgrade with rebalancing )
  • [MB-7313] - memcached.log file does not exist on centos 32-bit and 64-bit 2.0 when upgraded from 1.8.1 ( or maybe even for new installations )
  • [MB-7330] - 2.0 node not being treated as a 2.0 node in a 2.0 cluster, XDCR and Views tabs have vanished on that particular node as well.
  • [MB-7342] - Touch with expiry time 0 does not work
  • [MB-7343] - user should be warned that he needs to restart the server when installer detects that libssl0.9.8 doesn't exist during upgrade on ubuntu 12( installation will be stop and give instructions to install the missing libssl0.9.8)
  • [MB-7344] - Changing data path deletes all buckets
  • [MB-7346] - Documentation outdated in page http://www.couchbase.com/docs/couchbase-devguide-2.0/filter-extract-data.html
  • [MB-7357] - TAP deletion events should be throttled in a destination node if it is highly overloaded
  • [MB-7358] - Queue TAP deletion events received during TAP backfill phase into the backfill queue in a destination node
  • [MB-7362] - Rebalance exited with reason {{bulk_set_vbucket_state_failed, after Port server memcached on node 'ns_1@10.3.2.157' exited with status 139 (test_add_back_failed_node)
  • [MB-7364] - Rebalance exited with reason unexpected_exit ( rebalancing cluster with xdcr and views )
  • [MB-7367] - ns_server doesn't recover from osmon crash
  • [MB-7368] - Bucket data lost on changing index_path on 2.0 cluster.
  • [MB-7542] - seq_id set to 0 under some scenarios
  • [MB-7634] - items lost after doing an offline upgrade of a cluster which has xdcr replication stream to another 2.0 cluster
  • [MB-7675] - Auto failover took 5 mins complete from the time of failure of a node
  • [MB-7908] - "Viewing Internal XDCR Settings" documentation specifies redundant -X flag twice
  • [MB-7992] - Fix race condition in stats code path
  • [MB-8784] - Command to change the per-node RAM quota is incorrect
  • [MB-9546] - Duplicate line in manual failover example
  • [MB-9562] - Docs: high_water_mark needs updating
  • [MB-9611] - Doc: For 2.2, the Rest API information for "create/edit buckets" redirects to "Web console instructions" instead,
  • [MB-9641] - /pools/default/logs REST endpoint does not exist
  • [MB-9733] - Broken link in Data Buckets section of Developer guide
  • [MB-10098] - Syntax for URI of sending a test email returns "not found"

Epic

Improvement

  • [MB-3043] - max_size should default to a sane value
  • [MB-3254] - Add rebalance "time remaining" estimate
  • [MB-4155] - Force init script to wait for "proper" memcached shutdown before returning success
  • [MB-4161] - Extend mbrestore to allow for wildcard matching of items
  • [MB-4267] - Change "membase" bucket type to "couchbase"
  • [MB-4270] - Make Development Views the default
  • [MB-4277] - The "keys" field in the Filter Results UI needs to be more user friendly or state that it's contents must be an array
  • [MB-4574] - cbbackup-incremental reimplements tap handling
  • [MB-4776] - Explicitly state that setting a TTL will invalidate the data from the entire system...not just from the cache
  • [MB-4848] - possible active/replica index optimizations
  • [MB-4919] - In the info document for XDCR continuous replication we should have an "up to date status"
  • [MB-5167] - Documentation confusion
  • [MB-5170] - new bucket flush feature
  • [MB-5284] - moxi docs unclear on how to specify multiple Couchbase hosts
  • [MB-5469] - Remove defunct vbucket_batch_count value change listener and priority
  • [MB-5473] - Add missing C++ STL namespace
  • [MB-5477] - start EP Engine read-only dispatcher at the end of initialization
  • [MB-5538] - Add a way to disable Werror in ep-engine build
  • [MB-5553] - Create a make coverage target that runs gcov and reports with lcov
  • [MB-5554] - Changes checked into gerrit are automatically unit tested on linux and ubuntu
  • [MB-5585] - 2.0 Warmup has warmup_value_count & warmup_key_count
  • [MB-5589] - Add new binary cmd CMD_ENABLE_TRAFFIC for warmup
  • [MB-5611] - XDCR: UI should show the reason why replication failed
  • [MB-5756] - XDCR: tests need function to make sure replication is done before verification.
  • [MB-5912] - Reduce the reference counter tracking overhead in the flusher loop
  • [MB-6031] - collect and sample disk drain rate metrics directly from memcached
  • [MB-6035] - log system errno after encounter open/write/read failures on couchstore db files
  • [MB-6098] - couchbase quota vs. cluster quota is confusing
  • [MB-6445] - Remove sqlite from ep-engine
  • [MB-6482] - Pass the compress option for a doc body to couchstore
  • [MB-6548] - Memory management of DocInfo structs is annoying to clients
  • [MB-6554] - gmake clean-xfd should use quiet option...
  • [MB-6632] - XDCR: ep_engine should be able to process a new_checkpoint command and notify this event to the external components.
  • [MB-6651] - Set a Windows 64bit toy builder
  • [MB-6976] - Make checkpoint_flush_timeout adjustable at runtime
  • [MB-6993] - ns_server should log process' IDs even for processes having registered name
  • [MB-7020] - Include node name into the directory name generated by cbcollect_info
  • [MB-7030] - Improve performance of view engine for rebalance with consistent views enabled, and make it use less resources (CPU, IO)
  • [MB-7048] - ns_server diag should include memcached logs
  • [MB-7061] - Document using the REST API to retrieve stats for a bucket on an individual node
  • [MB-7065] - Docs for "remove server by cli" use failover
  • [MB-7106] - Log index monitoring requests and replies (debug level)
  • [MB-7130] - Refactor btree statistics code and make it more complete (percentiles, key/value sizes, etc)
  • [MB-7131] - Tweak btree node splitting method
  • [MB-7188] - Improve incremental indexing and cleanup performance, make fragmentation grow significantly more slowly, and overall better performance
  • [MB-7219] - Enable rest api for XDCR parameters
  • [MB-7228] - Add to release notes - Frequent "IP address seems to have changed. Unable to listen to node"
  • [MB-7291] - Create 2.0 legacy rightscale templates for Enterprise and Community Edition (non-chef)
  • [MB-7356] - Update the network ports page in the docs to add more details about ports that need to be opened for XDCR
  • [MB-7531] - Add .Net Bulk Load and Exponential Backoff
  • [MB-7715] - Failed to Flush Bucket: Save request failed because of timeout
  • [MB-7765] - Need "engineered" way of determining whether certain hotfixes have been applied
  • [MB-7783] - Documentation needed for work-around on ElasticSearch Plugin
  • [MB-8113] - [DOC] Add a REST interface to list the design document of a bucket

Story

  • [MB-4231] - Bucket list should use a blue triangle (similar to what's in Analytics) for displaying Bucket info
  • [MB-4232] - Put per-bucket node status pie charts back and remove the large one in the header
  • [MB-4338] - As a Couchbase Admin, I want to manually trigger compaction when I feel it's needed.
  • [MB-6798] - XDCR: add test of failover or rebalance at destination into simple-test
  • [MB-7145] - XDCR: need better logging for xdcr

Task

  • [MB-10211] - Fix typos in "Observing Documents" section

Edit/Copy Release Notes

The text area below allows the project release notes to be edited and copied to another document.