History log of /3.0.3-GA/ep-engine/ (Results 76 - 100 of 4208)
Revision (<<< Hide revision tags) (Show revision tags >>>)Date Author Comments
(<<< Hide modified files)
(Show modified files >>>)
c5c79c9a15-Aug-2014 Trond Norbye <trond.norbye@gmail.com>

Windows don't support the %zu formatter

Change-Id: Ia36bc151f84af9cfd6c069c3d97c494753d449f3
Reviewed-on: http://review.couchbase.org/40650
Reviewed-by: Chiyoung Seo <chiyoung@couchb

Windows don't support the %zu formatter

Change-Id: Ia36bc151f84af9cfd6c069c3d97c494753d449f3
Reviewed-on: http://review.couchbase.org/40650
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Chiyoung Seo <chiyoung@couchbase.com>

show more ...

0c80349315-Aug-2014 Mike Wiederhold <mike@couchbase.com>

MB-11958: Make sure mutations and snapshot ranges are updated together

There is a race condition where an items seqno is updated and the
flusher runs and persists both the seqno and old

MB-11958: Make sure mutations and snapshot ranges are updated together

There is a race condition where an items seqno is updated and the
flusher runs and persists both the seqno and old snapshot range
before the new snapshot range can be updated.

Change-Id: If124e5ab05c45c68a6956d28cc3898f8c18df6a9
Reviewed-on: http://review.couchbase.org/40642
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

a67794f614-Aug-2014 abhinavdangeti <abhinav@couchbase.com>

MB-11890: Active vb shouldn't initiate add Stream req

Change-Id: Ic902e632d52f426a73d7708f9c69854dc7fab874
Reviewed-on: http://review.couchbase.org/40641
Reviewed-by: Michael Wiederh

MB-11890: Active vb shouldn't initiate add Stream req

Change-Id: Ic902e632d52f426a73d7708f9c69854dc7fab874
Reviewed-on: http://review.couchbase.org/40641
Reviewed-by: Michael Wiederhold <mike@couchbase.com>
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

c21bd1d414-Aug-2014 Chiyoung Seo <chiyoung.seo@gmail.com>

MB-11916 Synchronize the access to thread job logs.

Change-Id: I9551cd70b931b1ed92f806be1df0325d922a239c
Reviewed-on: http://review.couchbase.org/40637
Reviewed-by: abhinav dangeti <

MB-11916 Synchronize the access to thread job logs.

Change-Id: I9551cd70b931b1ed92f806be1df0325d922a239c
Reviewed-on: http://review.couchbase.org/40637
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Tested-by: Chiyoung Seo <chiyoung@couchbase.com>

show more ...

ce5f4f7e14-Aug-2014 Chiyoung Seo <chiyoung.seo@gmail.com>

MB-11945 Remove the assertion check on sequence number in the flusher.

We sporadically observe the sequence number mismatches between the
flusher and Couchstore. This is not a regression

MB-11945 Remove the assertion check on sequence number in the flusher.

We sporadically observe the sequence number mismatches between the
flusher and Couchstore. This is not a regression from any recent
changes, and mostly existed much earlier. We will investigate this
issue separately to see if there are any bugs in ep-engine or
Couchstore.

Change-Id: Ifac2eae0deae09cad09dd7b5664f63bcbf828b24
Reviewed-on: http://review.couchbase.org/40630
Reviewed-by: Michael Wiederhold <mike@couchbase.com>
Tested-by: Chiyoung Seo <chiyoung@couchbase.com>

show more ...

0db302e506-Aug-2014 Mike Wiederhold <mike@couchbase.com>

MB-11299: Don't let replica vbuckets send partial snapshots

Replica vbuckets cannot create their own snapshots and must
rely on the snapshots it received from the active vbucket
when

MB-11299: Don't let replica vbuckets send partial snapshots

Replica vbuckets cannot create their own snapshots and must
rely on the snapshots it received from the active vbucket
when forwarding the snapshots to a downstream consumer. In
order to make sure that this happens properly we will only
stream mutations from the closed checkpoint for non-active
vbuckets. This will guarantee that the above requirement is
met.

Change-Id: I58151ce96868e34083f87dca0ce1d18b6a12aedf
Reviewed-on: http://review.couchbase.org/40346
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

f97e7c3714-Aug-2014 Sundar Sridharan <sundar.sridharan@gmail.com>

MB-11943: replace map with vector for cachedVBStates

std::map can incur thread-unsafe rb balance operations which
may invalidate concurrent iterators causing asserts or crashes
Pre-a

MB-11943: replace map with vector for cachedVBStates

std::map can incur thread-unsafe rb balance operations which
may invalidate concurrent iterators causing asserts or crashes
Pre-allocated vectors are thread-safe as long as the size is not
altered
Change-Id: I777b0cc7d9c421cc11425db75506fdf90044a93c
Reviewed-on: http://review.couchbase.org/40600
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Chiyoung Seo <chiyoung@couchbase.com>

show more ...

8b75fac614-Aug-2014 Mike Wiederhold <mike@couchbase.com>

MB-11941: Use the latest seqno during tap mutation/deletion cur snapshot

We generate the seqno when inserting the item into the hashtable when
using tap. As a reuslt we need to retrieve

MB-11941: Use the latest seqno during tap mutation/deletion cur snapshot

We generate the seqno when inserting the item into the hashtable when
using tap. As a reuslt we need to retrieve that seqno when updating
the latest snapshot because the byseqno passed into the mutation and
deletion functions for tap is always 0.

Change-Id: I21b7888207d9fea4f46b3d5918d1c8b60b2ea1ff
Reviewed-on: http://review.couchbase.org/40601
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

171ef02214-Aug-2014 Mike Wiederhold <mike@couchbase.com>

MB-11928: Fast path for rollback to 0

Right now we run through all of the headers even if we have to
rollback to 0. This change provides a fast path to rollback to
zero by immediatel

MB-11928: Fast path for rollback to 0

Right now we run through all of the headers even if we have to
rollback to 0. This change provides a fast path to rollback to
zero by immediately resetting the vbucket.

Change-Id: I8e7b62ec711838894c4eda92eaff78704bd94d50
Reviewed-on: http://review.couchbase.org/40595
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

4f74c48113-Aug-2014 abhinavdangeti <abhinav@couchbase.com>

Incorrect parameter for scheduleVBDeletion

This caused completeBeforeShutdown to be set
to false, in resetVBucket/deleteVBucket
scenarios.

Change-Id: I72a1f07f475af49a7f3686

Incorrect parameter for scheduleVBDeletion

This caused completeBeforeShutdown to be set
to false, in resetVBucket/deleteVBucket
scenarios.

Change-Id: I72a1f07f475af49a7f368695a8e83cf2c5df9e28
Reviewed-on: http://review.couchbase.org/40593
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

6ac065a213-Aug-2014 Mike Wiederhold <mike@couchbase.com>

MB-11933: Only set cur snapshot when receving the first snapshot item

We were setting the current snapshot when recieving the snapshot
marker and this is incorrect because if we disconne

MB-11933: Only set cur snapshot when receving the first snapshot item

We were setting the current snapshot when recieving the snapshot
marker and this is incorrect because if we disconnect before recieving
any items for that snapshot then the next time the replica vbucket
tries to connect to the active the high seqno will be less than the
last snapshot start seqno and this will cause the stream request to
fail with invalid arguments.

Change-Id: I75e1f555656c24221b2697011df878c579dce70e
Reviewed-on: http://review.couchbase.org/40583
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

72bcbab613-Aug-2014 Mike Wiederhold <mike@couchbase.com>

MB-11941: Update current vbucket snapshot when receiving tap items

We need to update the vbucket snapshot in order to support upgrades
because we initially have a tap stream and then mov

MB-11941: Update current vbucket snapshot when receiving tap items

We need to update the vbucket snapshot in order to support upgrades
because we initially have a tap stream and then move to a dcp stream.
Since tap doesn't support snapshots we can just set the current snapshot
to be the same as the seqno for that item.

Change-Id: I9da32df9366508276619507cf6cb9a66557e0919
Reviewed-on: http://review.couchbase.org/40582
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

6090dc2613-Aug-2014 Chiyoung Seo <chiyoung.seo@gmail.com>

MB-11952 Lower the default max number of readers.

Change-Id: I43dbcc4d606eea07276327ec7b9ed4f9bf3fc0e4
Reviewed-on: http://review.couchbase.org/40579
Reviewed-by: Sundararaman Sridha

MB-11952 Lower the default max number of readers.

Change-Id: I43dbcc4d606eea07276327ec7b9ed4f9bf3fc0e4
Reviewed-on: http://review.couchbase.org/40579
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Tested-by: Chiyoung Seo <chiyoung@couchbase.com>

show more ...

4f8f59f113-Aug-2014 Sundar Sridharan <sundar.sridharan@gmail.com>

MB-11642 Batch load dcp mutations if replication

Instead of loading 1 mutation per vbucket send out 10
dcp mutations per dcp step() if it is a intra-cluster
replication stream.
T

MB-11642 Batch load dcp mutations if replication

Instead of loading 1 mutation per vbucket send out 10
dcp mutations per dcp step() if it is a intra-cluster
replication stream.
This is done so that intra-cluster replication is always
faster than XDCR and views.
Noticeable improvement seen in intra-cluster replication
Change-Id: I15fc0081852e8e11027a20febe3224acdf21b661
Reviewed-on: http://review.couchbase.org/40392
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Sundararaman Sridharan <sundar@couchbase.com>

show more ...

60cb687f06-Aug-2014 Mike Wiederhold <mike@couchbase.com>

MB-11299: Allow rollbacks to a partial snapshot

If we roll back to a partial snapshot we need to make sure to get
the snapshot start and end seqnos for the partial snapshot in order

MB-11299: Allow rollbacks to a partial snapshot

If we roll back to a partial snapshot we need to make sure to get
the snapshot start and end seqnos for the partial snapshot in order
to make sure that we don't accidentally miss any items when
reconnecting.

Change-Id: If95d0476d8d582d6b57786387212f03b2aafc7ba
Reviewed-on: http://review.couchbase.org/40340
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

6dbd231f12-Aug-2014 Chiyoung Seo <chiyoung.seo@gmail.com>

MB-11912 Monitor workload pattern for dynamic compaction scheduling.

This change allows us to monitor workload pattern (i.e.,
read_heavy, write_heavy, mixed) at runtime and use the monit

MB-11912 Monitor workload pattern for dynamic compaction scheduling.

This change allows us to monitor workload pattern (i.e.,
read_heavy, write_heavy, mixed) at runtime and use the monitored
workload pattern to adjust the number of concurrent compaction
tasks dynamically at runtime. If the workload pattern is read-heavy,
then only one compaction task is executed at a time.

Note that we plan to keep improving this area for better I/O
scheduling.

Change-Id: Ibbb27da0a7e9f3951571ad8068570ebbf9abb318
Reviewed-on: http://review.couchbase.org/40520
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Tested-by: Chiyoung Seo <chiyoung@couchbase.com>

show more ...

91d3524d12-Aug-2014 Sriram Ganesan <sriram@couchbase.com>

Add id to the logs of checkpoint persistence

Add the id to the logs of checkpoint persistence so that we will have the
sequence number/checkpoint id in case a timeout occurs on checkpoin

Add id to the logs of checkpoint persistence

Add the id to the logs of checkpoint persistence so that we will have the
sequence number/checkpoint id in case a timeout occurs on checkpoint
persistence

Change-Id: Ifd3c70c9d40de43f0b3f1fdc1cf495ee0dc95d60
Reviewed-on: http://review.couchbase.org/40513
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Sriram Ganesan <sriram@couchbase.com>

show more ...

a4fa7dd206-Aug-2014 Mike Wiederhold <mike@couchbase.com>

MB-11299: Use the snapshot start seqno when creating a failover entry

Change-Id: I7a2c0adf51a761a491d0f12720f2abef8127e5c3
Reviewed-on: http://review.couchbase.org/40341
Reviewed-by:

MB-11299: Use the snapshot start seqno when creating a failover entry

Change-Id: I7a2c0adf51a761a491d0f12720f2abef8127e5c3
Reviewed-on: http://review.couchbase.org/40341
Reviewed-by: Michael Wiederhold <mike@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

007528bb07-Aug-2014 Mike Wiederhold <mike@couchbase.com>

MB-11887: Properly send the snapshot start and end seqnos

Currently we always set the snapshot start and end seqno to be the
same as the start seqno. This is incorrect if we recently wer

MB-11887: Properly send the snapshot start and end seqnos

Currently we always set the snapshot start and end seqno to be the
same as the start seqno. This is incorrect if we recently were
disconnected and need to resume from a partial snapshot.

Change-Id: I7c9430e0b24ef01cc427cb03b518f4c41a55d875
Reviewed-on: http://review.couchbase.org/40276
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

c8f54d3e08-Aug-2014 abhinavdangeti <abhinav@couchbase.com>

After APPEND/PREPEND to json, set datatype to binary

This is because appending or prepending anything
to a JSON document breaks the JSON structure.

Change-Id: I0acd99b0c43a54e0b

After APPEND/PREPEND to json, set datatype to binary

This is because appending or prepending anything
to a JSON document breaks the JSON structure.

Change-Id: I0acd99b0c43a54e0b34ce21f3a69c41709e3817e
Reviewed-on: http://review.couchbase.org/40390
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

7fb8133707-Aug-2014 abhinavdangeti <abhinav@couchbase.com>

MB-11741: Better scheduling of checkpointRemoverTask

Waking up the checkpoint Remover Task once the cursors
complete parsing a checkpoint till the end, helps
prevent retaining blob i

MB-11741: Better scheduling of checkpointRemoverTask

Waking up the checkpoint Remover Task once the cursors
complete parsing a checkpoint till the end, helps
prevent retaining blob instances for a longer time
than necessary.

Change-Id: Ieb1435dafac2e9b9984664218d6761c6c807333a
Reviewed-on: http://review.couchbase.org/40355
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

5dcdd64206-Aug-2014 Sundar Sridharan <sundar.sridharan@gmail.com>

fix copy paste bug to set max non-io at runtime

Change-Id: I097e5fc5d926387dff8874427d373e41e383f64e
Reviewed-on: http://review.couchbase.org/40317
Tested-by: Sundararaman Sridharan

fix copy paste bug to set max non-io at runtime

Change-Id: I097e5fc5d926387dff8874427d373e41e383f64e
Reviewed-on: http://review.couchbase.org/40317
Tested-by: Sundararaman Sridharan <sundar@couchbase.com>
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>

show more ...

6e3f9a3305-Aug-2014 Trond Norbye <trond.norbye@gmail.com>

MB-11872: Compile failure on (some?) Debian 7 systems

Building from a Debian 7 setup under SmartOS works out of the box,
but one of our builders does not work for some reason. From looki

MB-11872: Compile failure on (some?) Debian 7 systems

Building from a Debian 7 setup under SmartOS works out of the box,
but one of our builders does not work for some reason. From looking
at the preprocessed file it doesn't define linux like our other
systems do. Extend the test to work for __linux__ and __linux as
well.

Change-Id: I4a7cf304ab8bfddb6cc117e9064165e1eb16b66e
Reviewed-on: http://review.couchbase.org/40297
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Chris Hillery <ceej@couchbase.com>

show more ...

4960b2c505-Aug-2014 abhinavdangeti <abhinav@couchbase.com>

MB-11884: Consider min_item_threshold in full eviction

In this case, as warmed up keys always matches warmed
up values, we can mark the min item threshold using
warmup's estimated it

MB-11884: Consider min_item_threshold in full eviction

In this case, as warmed up keys always matches warmed
up values, we can mark the min item threshold using
warmup's estimated item count.

Change-Id: Ibd1d58edd15be877c24ef794615fe0d56f9fc675
Reviewed-on: http://review.couchbase.org/40308
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

34d2bfbe05-Aug-2014 abhinavdangeti <abhinav@couchbase.com>

MB-11847: [Warmup] Estimated value count with full eviction

In loadingKVPairs phase, set estimated value count equal
to the estimated item count, as no keys have been warmed
up yet,

MB-11847: [Warmup] Estimated value count with full eviction

In loadingKVPairs phase, set estimated value count equal
to the estimated item count, as no keys have been warmed
up yet, when in full eviction mode.

Change-Id: Id74237c96e0c144d9a9b94e0670f4975739e7914
Reviewed-on: http://review.couchbase.org/40304
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

12345678910>>...169