History log of /3.0.3-GA/ep-engine/ (Results 26 - 50 of 4208)
Revision (<<< Hide revision tags) (Show revision tags >>>)Date Author Comments
(<<< Hide modified files)
(Show modified files >>>)
c4be44ac23-Oct-2014 Mike Wiederhold <mike@couchbase.com>

MB-12305: Don't update the snap start/end seqno in tap mutation

We do this in the queueDirty function so this code is not needed. It
also appeared to cause a race in updating the snapsho

MB-12305: Don't update the snap start/end seqno in tap mutation

We do this in the queueDirty function so this code is not needed. It
also appeared to cause a race in updating the snapshot start and end
sequence numbers so removing it should solve the bug linked above.

Change-Id: Ia8fa36df958be9147ea208ba9ebd78496048ebb4
Reviewed-on: http://review.couchbase.org/42503
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

fde80eed23-Oct-2014 Trond Norbye <trond.norbye@gmail.com>

MB-12398: log incorrect timeout values

The loglevel DEBUG won't appear in any logs, so customers using
this in production won't get the information anywhere (the server
just change t

MB-12398: log incorrect timeout values

The loglevel DEBUG won't appear in any logs, so customers using
this in production won't get the information anywhere (the server
just change their requested lock time without telling the user).

I guess a better behavior here would be to return ERANGE in
this case, but that would potentially break user applications
and isn't something we should do in a patch revision.

Change-Id: Ib6a37deceb2755b9fea53b0542a13fb2e16a3261
Reviewed-on: http://review.couchbase.org/42387
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

a8b244a220-Oct-2014 Mike Wiederhold <mike@couchbase.com>

Hide the -a argument in cbstats

The -a argument is used by cbcollectinfo so we can't just
get rid of it.

Change-Id: I0f04c2d454c117271b44176df904e90a938965e9
Reviewed-on: ht

Hide the -a argument in cbstats

The -a argument is used by cbcollectinfo so we can't just
get rid of it.

Change-Id: I0f04c2d454c117271b44176df904e90a938965e9
Reviewed-on: http://review.couchbase.org/42297
Reviewed-by: Michael Wiederhold <mike@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

a5bb2d5815-Oct-2014 Sriram Ganesan <sriram@couchbase.com>

MB-12226: Handle ENOENT error during unlink

If the file is not found during an unlink, it shouldn't be added
to the pending file deletions queue.

Change-Id: Ief306277bbbc946ae18

MB-12226: Handle ENOENT error during unlink

If the file is not found during an unlink, it shouldn't be added
to the pending file deletions queue.

Change-Id: Ief306277bbbc946ae18e39dd4819f811f12ea76c
Reviewed-on: http://review.couchbase.org/42173
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Sriram Ganesan <sriram@couchbase.com>

show more ...

138a67c714-Oct-2014 Mike Wiederhold <mike@couchbase.com>

MB-12271: Set the default dcp producer noop interval to 20 seconds

This is a fix to support backwards compatibility between 3.0 and
3.0.1+ versions of Couchbase. The problem is that 3.0

MB-12271: Set the default dcp producer noop interval to 20 seconds

This is a fix to support backwards compatibility between 3.0 and
3.0.1+ versions of Couchbase. The problem is that 3.0 has a noop
interval of 20 seconds by default and 3.0.1 has a default noop
interval of 200 seconds. In 3.0.1 the consumer explicitly sets
the noop interval of the producer so in 3.0.1+ clusters the 20
second default will be overriden by the consumer, but in the
scenario when a 3.0.1+ node connects to a 3.0 node the 20 second
default will remain.

Change-Id: I2e18e9ad68037f3a82abe5167f2bca89f381f318
Reviewed-on: http://review.couchbase.org/42136
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

0b7c515c13-Oct-2014 Mike Wiederhold <mike@couchbase.com>

MB-11454: Add an option to output json in cbstats

Change-Id: I0adcb16adde82ddf3671e4edc0118061d684f2d4
Reviewed-on: http://review.couchbase.org/42084
Reviewed-by: abhinav dangeti <ab

MB-11454: Add an option to output json in cbstats

Change-Id: I0adcb16adde82ddf3671e4edc0118061d684f2d4
Reviewed-on: http://review.couchbase.org/42084
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

6aa436f111-Oct-2014 Mike Wiederhold <mike@couchbase.com>

MB-12271: Set noop interval individually for dcp connections

We need to do this because it might be the case that two servers
have their respective noop intervals set to different values

MB-12271: Set noop interval individually for dcp connections

We need to do this because it might be the case that two servers
have their respective noop intervals set to different values. If
they are this can cause the connections to be disconnected because
each side is expecting to see a noop at different times.

Change-Id: I6ff475ccba407547e7285fa431b86ad9bf9cdc24
Reviewed-on: http://review.couchbase.org/42049
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

09f697e613-Oct-2014 abhinavdangeti <abhinav@couchbase.com>

Account for incorrect entries through cbepctl

Context: access_scanner_enabled
Relates to: http://review.couchbase.org/#/c/40884/

Change-Id: I35134c91ed03f6ba6093cfd71270484beca1

Account for incorrect entries through cbepctl

Context: access_scanner_enabled
Relates to: http://review.couchbase.org/#/c/40884/

Change-Id: I35134c91ed03f6ba6093cfd71270484beca1a4cd
Reviewed-on: http://review.couchbase.org/42083
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

16cc72d423-Sep-2014 Mike Wiederhold <mike@couchbase.com>

MB-11642: Change the priority based on the type of dcp connection

Change-Id: I1b9a6846879385308bee3920bfa182fc41e39b4f
Reviewed-on: http://review.couchbase.org/41758
Reviewed-by: Sun

MB-11642: Change the priority based on the type of dcp connection

Change-Id: I1b9a6846879385308bee3920bfa182fc41e39b4f
Reviewed-on: http://review.couchbase.org/41758
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

1de9ee7129-Sep-2014 Sundar Sridharan <sundar.sridharan@gmail.com>

use optparse instead of arg_parse cbvdiff for compatibility

Change-Id: I604354823c71b91a167646c01e2bb6ea9d8c8822
Reviewed-on: http://review.couchbase.org/41752
Tested-by: Sundararama

use optparse instead of arg_parse cbvdiff for compatibility

Change-Id: I604354823c71b91a167646c01e2bb6ea9d8c8822
Reviewed-on: http://review.couchbase.org/41752
Tested-by: Sundararaman Sridharan <sundar@couchbase.com>
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

690c6cae29-Sep-2014 abhinavdangeti <abhinav@couchbase.com>

MB-12268: Adding a debug log, when clusterConfig is updated

Change-Id: I0cbb70789f84bddc644bdc4ee59c1e86e1821220
Reviewed-on: http://review.couchbase.org/41746
Reviewed-by: Sundarara

MB-12268: Adding a debug log, when clusterConfig is updated

Change-Id: I0cbb70789f84bddc644bdc4ee59c1e86e1821220
Reviewed-on: http://review.couchbase.org/41746
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

3c05d1fa23-Sep-2014 Sriram Ganesan <sriram@couchbase.com>

MB-12160: Use appropriate CAS for comparison with locked items

In the case of setWithMeta/deleteWithMeta commands, the locked item's
CAS value needs to be compared with the incoming muta

MB-12160: Use appropriate CAS for comparison with locked items

In the case of setWithMeta/deleteWithMeta commands, the locked item's
CAS value needs to be compared with the incoming mutation's CAS.

Change-Id: Id12a3c4717b18bc41c3d4b7dded99ea215179e9d
Reviewed-on: http://review.couchbase.org/41594
Tested-by: Sriram Ganesan <sriram@couchbase.com>
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>

show more ...

11bd93b424-Sep-2014 abhinavdangeti <abhinav@couchbase.com>

Account for keylength while allocating buffer for allKeys

While increasing the buffersize (when needed) when the
AllKeysAPI is invoked, we will need to account for two
additional byt

Account for keylength while allocating buffer for allKeys

While increasing the buffersize (when needed) when the
AllKeysAPI is invoked, we will need to account for two
additional bytes while allocating the key, for its length.

Change-Id: Iba68c4ae7bccae20f97d4e98350d5105093c3487
Reviewed-on: http://review.couchbase.org/41595
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

30a1fcf424-Sep-2014 abhinavdangeti <abhinav@couchbase.com>

MB-12117: Release hashtable locks before disk IO

While in access log generation, we'll need to release
all hashtable partition locks before we create new
entries in the mutation log.

MB-12117: Release hashtable locks before disk IO

While in access log generation, we'll need to release
all hashtable partition locks before we create new
entries in the mutation log.

Change-Id: Ic3dd0a02452b51ee742e30a0f268b86f9ab6205b
Reviewed-on: http://review.couchbase.org/41572
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

4101daac18-Sep-2014 abhinavdangeti <abhinav@couchbase.com>

MB-11999: Preferential loading of active/replica vbuckets

During warmup, order the vbucket list such that
active vbuckets get 60% preference while replica
vbuckets get 40% preference

MB-11999: Preferential loading of active/replica vbuckets

During warmup, order the vbucket list such that
active vbuckets get 60% preference while replica
vbuckets get 40% preference.

Example:
In a 4 node cluster (DGM), approx. resident ratios:
1. Before warmup
active: 36% 34% 34% 41%
replica: 33% 35% 35% 27%
2. After warmup
active: 40% 42% 41% 46%
replica: 31% 29% 29% 24%

Change-Id: I60e0427bca58530247086d730135ebb4be70bb84
Reviewed-on: http://review.couchbase.org/41468
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

128a21d517-Sep-2014 Mike Wiederhold <mike@couchbase.com>

Merge remote-tracking branch 'gerrit/2.5.0' into 3.0.1

Change-Id: I59c45877fc783e30c86dcfd7d303c731cde433fc


1e3ce0f611-Sep-2014 Mike Wiederhold <mike@couchbase.com>

Fix deadlock in checkpoint persistence command

We need to release the hpChkMutex before notifying memcached in
order to prevent a deadlock. The deadlock occurs when the flusher
trys

Fix deadlock in checkpoint persistence command

We need to release the hpChkMutex before notifying memcached in
order to prevent a deadlock. The deadlock occurs when the flusher
trys to notify memcached of checkpoint persistence at the same
time a memcached worker thread is trying to add a new checkpoint
persistence request.

Change-Id: Ida313f5b39ef0e063dee9882410cd0a19ce55292
Reviewed-on: http://review.couchbase.org/41363
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

256732cc10-Sep-2014 Mike Wiederhold <mike@couchbase.com>

Merge remote-tracking branch 'gerrit/3.0' into 3.0.1

Change-Id: I195452c3d0684dff5ff673efd00a18b9d70da3cf


Revision tags: v3.0.0
7026621209-Sep-2014 Mike Wiederhold <mike@couchbase.com>

MB-12137: Don't update the current snapshot during persistence

If we update the current snapshot only when we persist items to
disk then we may miss out on updates to the current snapsho

MB-12137: Don't update the current snapshot during persistence

If we update the current snapshot only when we persist items to
disk then we may miss out on updates to the current snapshot
when there are vbucket state changes.

Change-Id: Iea3139afb669bdf32e6b4d98e8b3515cafe578dc
Reviewed-on: http://review.couchbase.org/41289
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

1915b03f09-Sep-2014 Sriram Ganesan <sriram@couchbase.com>

Fix build breakage on Linux

Remove usage of constant strings for maintaining file deletion queue

Change-Id: I1992343ff65f923b6d3ffbf7a25932b936e3ff7c
Reviewed-on: http://review.

Fix build breakage on Linux

Remove usage of constant strings for maintaining file deletion queue

Change-Id: I1992343ff65f923b6d3ffbf7a25932b936e3ff7c
Reviewed-on: http://review.couchbase.org/41291
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Tested-by: Sriram Ganesan <sriram@couchbase.com>

show more ...

9665c15005-Sep-2014 Sriram Ganesan <sriram@couchbase.com>

Handle unlink/remove failures

The unlink/remove function can fail if there is another process that
has an open file handle on that file. In such cases, we need to retry
the unlink pe

Handle unlink/remove failures

The unlink/remove function can fail if there is another process that
has an open file handle on that file. In such cases, we need to retry
the unlink periodically in the flusher task until we get rid of those
files.

Change-Id: I4bfcf29b3fa866ec4946db658a245c722f3725ce
Reviewed-on: http://review.couchbase.org/41268
Tested-by: Sriram Ganesan <sriram@couchbase.com>
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>

show more ...

6c265bbe04-Sep-2014 Sundar Sridharan <sundar.sridharan@gmail.com>

remove warmup wait dead code to prevent accidental deadlock

Since we should never have a front-end thread wait for background
thread, having a warmup wait logic will ensure that we do no

remove warmup wait dead code to prevent accidental deadlock

Since we should never have a front-end thread wait for background
thread, having a warmup wait logic will ensure that we do not use
it accidentally in the future and cause a deadlock

Change-Id: I28ec96dceaffa02810415402226df20f8eb44cc7
Reviewed-on: http://review.couchbase.org/41206
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Sundararaman Sridharan <sundar@couchbase.com>

show more ...

5be13bc205-Sep-2014 Chiyoung Seo <chiyoung.seo@gmail.com>

MB-12063 Don't delete expired items during warmup.

When a node is added back to the cluster, its vbucket states
can be different from its previous states (e.g., active -> replica).
T

MB-12063 Don't delete expired items during warmup.

When a node is added back to the cluster, its vbucket states
can be different from its previous states (e.g., active -> replica).
Therefore, we should not remove expired items from active vbuckets
during warmup because the states of those vbuckets can be changed
after the warmup.

Change-Id: Ifec4fda80e424a8afa74bae66e17de77b2e0c36b
Reviewed-on: http://review.couchbase.org/41219
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Chiyoung Seo <chiyoung@couchbase.com>

show more ...

31cef4cd04-Sep-2014 Mike Wiederhold <mike@couchbase.com>

Merge remote-tracking branch 'gerrit/3.0' into 3.0.1

Change-Id: I678740458ccc13cbf79be847369cf6561b5b678b


cbda1c9f03-Sep-2014 Mike Wiederhold <mike@couchbase.com>

MB-12100: Don't expire items for replica vbuckets during warmup

Change-Id: I81095beb69c0996d819158e3cadc451a83653d2f
Reviewed-on: http://review.couchbase.org/41177
Reviewed-by: Sunda

MB-12100: Don't expire items for replica vbuckets during warmup

Change-Id: I81095beb69c0996d819158e3cadc451a83653d2f
Reviewed-on: http://review.couchbase.org/41177
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Reviewed-by: Chiyoung Seo <chiyoung@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

12345678910>>...169