History log of /5.5.2/couchdb/etc/couchdb/default_dev.ini.in (Results 1 - 25 of 156)
Revision (<<< Hide revision tags) (Show revision tags >>>) Date Author Comments
# f7298f32 14-Nov-2017 harsha <harsha.hs.81@gmail.com>

MB-26479 Add ipv6 protocol support in views

Change-Id: Ife18b0318d9505772d71dfeb7de71d8f6a1cb043
Reviewed-on: http://review.couchbase.org/86613
Reviewed-by: Aman Kishore Achpal <aman

MB-26479 Add ipv6 protocol support in views

Change-Id: Ife18b0318d9505772d71dfeb7de71d8f6a1cb043
Reviewed-on: http://review.couchbase.org/86613
Reviewed-by: Aman Kishore Achpal <aman.achpal@couchbase.com>
Reviewed-by: Jeelan Basha Poola <jeelan.poola@couchbase.com>
Tested-by: Jeelan Basha Poola <jeelan.poola@couchbase.com>

show more ...


# a69a4ffe 16-Nov-2017 Artem Stemkovski <astemkov@gmail.com>

remove unused authentication handlers and auth cache

Change-Id: I590e9ac9082945215b3613e823029eab0b2e9461
Reviewed-on: http://review.couchbase.org/85620
Tested-by: Artem Stemkovski <

remove unused authentication handlers and auth cache

Change-Id: I590e9ac9082945215b3613e823029eab0b2e9461
Reviewed-on: http://review.couchbase.org/85620
Tested-by: Artem Stemkovski <artem@couchbase.com>
Reviewed-by: Harsha H S
Reviewed-by: Artem Stemkovski <artem@couchbase.com>

show more ...


Revision tags: v4.6.0, v4.5.0
# 49e10b8b 09-Apr-2016 harsha <hhs.couchbase@gmail.com>

MB-8512 Do not load doc when document fields are not indexed

If document is not used by all the view functions defined in design doc, we
can optimize by not loading documents during inde

MB-8512 Do not load doc when document fields are not indexed

If document is not used by all the view functions defined in design doc, we
can optimize by not loading documents during indexing.
esprima and unused java script are used to detect if javascript map
function uses document in its function body. If none of the views
defined on design doc does not index document fields in their map function,
then we prevent loading of documents for that design document during
updater runs. This helps with sync gateway views.

Change-Id: Ic3d50185cb2a5d13b29a2e43290e51b663cebf8c
Reviewed-on: http://review.couchbase.org/54777
Reviewed-by: Abhishek Singh <abhishek@couchbase.com>
Reviewed-by: Volker Mische <volker.mische@gmail.com>
Tested-by: buildbot <build@couchbase.com>

show more ...


# 6460cc0a 01-Aug-2014 harsha <hhs.couchbase@gmail.com>

MB-11831 Maximum size of document that can be indexed is set to 20MB

Couchbase server restricts max size for JSON documents to 20 MB.
The view engine enforced a limit of 1 MB on document

MB-11831 Maximum size of document that can be indexed is set to 20MB

Couchbase server restricts max size for JSON documents to 20 MB.
The view engine enforced a limit of 1 MB on documents that can
be indexed. This limit is now increased to 20 MB to ensure every
document gets indexed and not silently dropped by the view engine
if size of document exceeded previously enforced limit.

Change-Id: I4f357d003cc4de467c9facdf57e9c6f7d8bf55cf
Reviewed-on: http://review.couchbase.org/40175
Reviewed-by: Volker Mische <volker.mische@gmail.com>
Tested-by: buildbot <build@couchbase.com>
Reviewed-by: Harsha H S <hhs.couchbase@gmail.com>

show more ...


# 861be4f8 25-Jul-2014 Nimish Gupta <nimish@couchbase.com>

MB-11838 Replace upr with dcp in logging and configuration

Change-Id: Ia4815f36b6cdb2de26912bfb87931cfabd447c7f
Reviewed-on: http://review.couchbase.org/39865
Reviewed-by: Volker Mis

MB-11838 Replace upr with dcp in logging and configuration

Change-Id: Ia4815f36b6cdb2de26912bfb87931cfabd447c7f
Reviewed-on: http://review.couchbase.org/39865
Reviewed-by: Volker Mische <volker.mische@gmail.com>
Tested-by: buildbot <build@couchbase.com>

show more ...


# 1e205d5a 22-Jun-2014 Sarath Lakshman <sarathlakshman@slynux.com>

Disable updater writer throttling by default

We have identified few bugs that blocked compaction performance.
The compactor can now run as fast as updater and hence writer throttling

Disable updater writer throttling by default

We have identified few bugs that blocked compaction performance.
The compactor can now run as fast as updater and hence writer throttling
is useless for now.

Change-Id: I925e2d17ec6363596dd344f81933e723e08564b4
Reviewed-on: http://review.couchbase.org/38649
Reviewed-by: Volker Mische <volker.mische@gmail.com>
Tested-by: buildbot <build@couchbase.com>
Reviewed-by: Sarath Lakshman <sarath@couchbase.com>
Tested-by: Sarath Lakshman <sarath@couchbase.com>

show more ...


# dddb3490 13-Jun-2014 Nimish Gupta <nimish@couchbase.com>

MB-11416 Make control buffer size configurable for QE tests

Make buffer size configurable, so that QE can configure it to lower
value and verify that unacked_bytes are less then 20 % of

MB-11416 Make control buffer size configurable for QE tests

Make buffer size configurable, so that QE can configure it to lower
value and verify that unacked_bytes are less then 20 % of buffer size.

Change-Id: I75fd8bbcc4333916bd6cfb09c76b3432553fddb8
Reviewed-on: http://review.couchbase.org/38227
Tested-by: buildbot <build@couchbase.com>
Reviewed-by: Volker Mische <volker.mische@gmail.com>

show more ...


# c6a29006 17-Mar-2014 Trond Norbye <trond.norbye@gmail.com>

Build using CMake

Change-Id: If0f29acdd08ff8b148a5362f3a6fb4d6a5be901c
Reviewed-on: http://review.couchbase.org/32799
Reviewed-by: Chris Hillery <ceej@couchbase.com>
Tested-by: T

Build using CMake

Change-Id: If0f29acdd08ff8b148a5362f3a6fb4d6a5be901c
Reviewed-on: http://review.couchbase.org/32799
Reviewed-by: Chris Hillery <ceej@couchbase.com>
Tested-by: Trond Norbye <trond.norbye@gmail.com>

show more ...


# 3615009f 26-Feb-2014 Filipe David Borba Manana <fdmanana@gmail.com>

Add throttle option for incremental index updates

Currently the updater seems to proceed significantly faster than the
compactor, which has a retry phase not very efficient when there's

Add throttle option for incremental index updates

Currently the updater seems to proceed significantly faster than the
compactor, which has a retry phase not very efficient when there's many
small updates during the initial compaction phase, which makes the
compactor retry phase apply too many small log files to the new index,
causing fragmentation on the new index to be high by the time compaction
finishes. This in turn causes the compaction scheduler to schedule yet
another compaction shortly after a compaction finishes.

So for now make the updater sleep for a short period before applying each
batch if compaction is currently running and in the retry phase. Not ideal
and this is a temporary workaround which can ne disabled or adjusted at
any time by setting the sleep interval (couch_config setting throttle_period).

A better solution is to make the retry phase more efficient, faster and
causing much less fragmentation. Amongst other possibilities, merging log
files into a single batch and applying it all at once makes a big difference
in regards to efficiency and speed - basically restoring what we had until
revision 1969a700dd9dfe072608c7b5fcf324eac8ab008d but in a fully correct
way.

Change-Id: I374cbbc0b7fb01257457c4b3f81b72a332afb6ff
Reviewed-on: http://review.couchbase.org/33991
Tested-by: buildbot <build@couchbase.com>
Reviewed-by: Volker Mische <volker.mische@gmail.com>
Reviewed-by: Sarath Lakshman <sarath@couchbase.com>
Tested-by: Filipe David Borba Manana <fdmanana@gmail.com>

show more ...


# 33844fff 31-Jan-2014 Volker Mische <vmx@couchbase.com>

Set UPR port to 11209

As Alk suggested on the last port change [1], the indexer
should use the internal UPR/TAP port and not the one that
is used by the normal clients.

Here

Set UPR port to 11209

As Alk suggested on the last port change [1], the indexer
should use the internal UPR/TAP port and not the one that
is used by the normal clients.

Here's what Alk proposed:

"btw you should use 11209.
It's exact same thing as 11210. But with "connections quota"
specifically reserved for internal uses.
I.e. so that clients' connections exhausting "general" connections
limit of 10k cannot DoS our own internal interaction with memcached.
It works because there's dedicated 1k connections limit for
port 11209 which normal 11210 connections cannot exhaust."

[1] http://review.couchbase.org/32605

Change-Id: I747dae1a1fe269dee05b1741200aa40a621733ef
Reviewed-on: http://review.couchbase.org/33067
Reviewed-by: Pratap Chakravarthy <pratapsrc@gmail.com>
Tested-by: Volker Mische <volker.mische@gmail.com>
Reviewed-by: Aliaksey Kandratsenka <alkondratenko@gmail.com>
Reviewed-by: Filipe David Borba Manana <fdmanana@gmail.com>

show more ...


# bd4315ae 17-Jan-2014 Volker Mische <vmx@couchbase.com>

Set UPR port to the default 11210

./cluster_run was updated to set the UPR port accordingly, hence
the port can be changed to the default one for Couchbase
installations.

Ch

Set UPR port to the default 11210

./cluster_run was updated to set the UPR port accordingly, hence
the port can be changed to the default one for Couchbase
installations.

Change-Id: I4539922a99ac054166a9d08de51a5c17507ef83b
Reviewed-on: http://review.couchbase.org/32605
Tested-by: Volker Mische <volker.mische@gmail.com>
Reviewed-by: Filipe David Borba Manana <fdmanana@gmail.com>

show more ...


# eddce01d 07-Jan-2014 Volker Mische <vmx@couchbase.com>

Use 12000 as UPR port

For development Couchbase is normally started via ./cluster_run.
There the UPR port 12000. Hence change to this port for now.

Change-Id: If6abe5678cb03fead

Use 12000 as UPR port

For development Couchbase is normally started via ./cluster_run.
There the UPR port 12000. Hence change to this port for now.

Change-Id: If6abe5678cb03fead2d922846bebd4b0bce50335
Reviewed-on: http://review.couchbase.org/32323
Tested-by: buildbot <build@couchbase.com>
Reviewed-by: Filipe David Borba Manana <fdmanana@gmail.com>

show more ...


# 9a8a6c07 12-Nov-2013 Volker Mische <vmx@couchbase.com>

MB-9085: Don't hard-code UPR port

Instead of hard-coding the UPR port, use a configuration
setting for it. This makes it also possible to start the
fake UPR server on a random port (

MB-9085: Don't hard-code UPR port

Instead of hard-coding the UPR port, use a configuration
setting for it. This makes it also possible to start the
fake UPR server on a random port (assigned by the
operating system), when the value is set to "0".

Hence it's now possible to run the set_view_tests in
parallel again.

Change-Id: I6c319dbb6fcac8b4f20f38ce8a44c6abed24658d
Reviewed-on: http://review.couchbase.org/29404
Reviewed-by: Volker Mische <volker.mische@gmail.com>
Tested-by: Volker Mische <volker.mische@gmail.com>

show more ...


# 239b2861 22-Aug-2013 Volker Mische <vmx@couchbase.com>

MB-8903: First version of couch_upr

See test/02-couch-upr.t for the API usage.

This commit also adds the fake UPR server.

Change-Id: I670f8e36196057399b4f942c7ffebfbdb92950

MB-8903: First version of couch_upr

See test/02-couch-upr.t for the API usage.

This commit also adds the fake UPR server.

Change-Id: I670f8e36196057399b4f942c7ffebfbdb9295066
Reviewed-on: http://review.couchbase.org/29402
Reviewed-by: Filipe David Borba Manana <fdmanana@gmail.com>
Tested-by: Volker Mische <volker.mische@gmail.com>

show more ...


# 5297f10d 04-Nov-2013 Filipe David Borba Manana <fdmanana@apache.org>

MB-9467 Enforce max total KV size that can be emitted per doc/view

To avoid consuming too much memory while running a single map function,
enforce a maximum limit for the total amount of

MB-9467 Enforce max total KV size that can be emitted per doc/view

To avoid consuming too much memory while running a single map function,
enforce a maximum limit for the total amount of data that can be emitted
per document and per view. This limit is configurable via the couch_config
option "max_kv_size_per_doc", section "mapreduce", and corresponds to the
sum of the byte size of all emitted keys and values.

A value of 0 for this new setting, disables the limit (meaning unlimited,
as before this change).

So far, before any memory blowup could happen, the function execution was
aborted due to the trigger of the timeout, such as in the following example
case:

function(doc, meta) {
for (var i = 0; i < 1000000; i++) {
for (var j = 0; j < 1000000; j++) {
emit(doc.value, null);
}
}
}

However, one user was able to get V8 heap exhaustion while running a map
function. Despite unsuccessful attempts so far to reproduce the heap
exhaustion issue, this change allows to enforce sane limits.

Change-Id: I4af03208027868a461c568dcd20ff49bec27ed76
Reviewed-on: http://review.couchbase.org/30055
Tested-by: buildbot <build@couchbase.com>
Reviewed-by: Volker Mische <volker.mische@gmail.com>
Tested-by: Filipe David Borba Manana <fdmanana@gmail.com>

show more ...


# 84711665 04-Nov-2013 Filipe David Borba Manana <fdmanana@apache.org>

MB-9467 Don't process documents with too large bodies

Too large documents slow down indexing and make the JavaScript engine
use too much cpu time (JSON parsing/validation), causing the w

MB-9467 Don't process documents with too large bodies

Too large documents slow down indexing and make the JavaScript engine
use too much cpu time (JSON parsing/validation), causing the whole
system to be inneficient during rebalance and failover operations or
not to be able to accomplish such operations at all.

Therefore limit the maximum size of documents passed to the JavaScript
engine, configurable via the internal parameter "indexer_max_doc_size".

Change-Id: I8e130d4c419c63f495452dab10f3e5ea96dd0db1
Reviewed-on: http://review.couchbase.org/30053
Tested-by: buildbot <build@couchbase.com>
Reviewed-by: Volker Mische <volker.mische@gmail.com>
Tested-by: Filipe David Borba Manana <fdmanana@gmail.com>

show more ...


# 9e9f1dc2 22-Jul-2013 Volker Mische <vmx@couchbase.com>

MB-8582: Specify indexer in couch_set_view:start_link/2

Change the signature of couch_set_view:start_link/1 to /2 in
order to include the indexer. This way the start_link function
ca

MB-8582: Specify indexer in couch_set_view:start_link/2

Change the signature of couch_set_view:start_link/1 to /2 in
order to include the indexer. This way the start_link function
can also be used for the spatial indexer.

Change-Id: I5e71134b1f55660560a9a7b21ae31443898dad83
Reviewed-on: http://review.couchbase.org/27612
Reviewed-by: Filipe David Borba Manana <fdmanana@gmail.com>
Tested-by: Filipe David Borba Manana <fdmanana@gmail.com>

show more ...


# a84ccbb7 01-Jul-2013 Volker Mische <vmx@couchbase.com>

MB-8531: Implement initial spatial index build

The initial index build for spatial indexes is now using the
new C code.

Change-Id: I8edd00cfb5341907369b468da6a9aca4ed13c21c

MB-8531: Implement initial spatial index build

The initial index build for spatial indexes is now using the
new C code.

Change-Id: I8edd00cfb5341907369b468da6a9aca4ed13c21c
Reviewed-on: http://review.couchbase.org/27220
Tested-by: buildbot <build@couchbase.com>
Reviewed-by: Filipe David Borba Manana <fdmanana@gmail.com>

show more ...


Revision tags: 2.1.1r, 2.1.0r, 2.0.2r
# 789b84c1 22-May-2013 Volker Mische <vmx@couchbase.com>

MB-8295: Start a couch_set_view instance for dev views

Development views should use the partition aware set views,
but with a single partition. This commit starts the manager
for tho

MB-8295: Start a couch_set_view instance for dev views

Development views should use the partition aware set views,
but with a single partition. This commit starts the manager
for those development views. It shouldn't intefere with
the existing ones.

More code will be moved to the new dev views in subsequent
commits.

Change-Id: I0f1f99552a5dc9dfe323aaf1961f5160c2e5cd7d
Reviewed-on: http://review.couchbase.org/26476
Reviewed-by: Filipe David Borba Manana <fdmanana@gmail.com>
Tested-by: Filipe David Borba Manana <fdmanana@gmail.com>

show more ...


# c71fd2fb 16-May-2013 Volker Mische <vmx@couchbase.com>

MB-8295: Parametrise Set View ETS tables

ETS tables a global. With this commit it's possible
to start couch_set_view with a paramter ('prod' or
'dev') and the ETS tables will be set

MB-8295: Parametrise Set View ETS tables

ETS tables a global. With this commit it's possible
to start couch_set_view with a paramter ('prod' or
'dev') and the ETS tables will be set accordingly.
They won't share the tables, but have their own ones.

This commit is just a refactoring, so it doesn't
really change the functionality or the external APIs.

This commit needs to be merged together with:
http://review.couchbase.org/26383

Change-Id: I4ed8093b402c041ba3f0e9842887b2547fd4cdc7
Reviewed-on: http://review.couchbase.org/26344
Reviewed-by: Filipe David Borba Manana <fdmanana@gmail.com>
Tested-by: Filipe David Borba Manana <fdmanana@gmail.com>

show more ...


# 725362b8 08-May-2013 Filipe David Borba Manana <fdmanana@apache.org>

Remove not used anymore parameters in ini config

These are not used anymore for a long time. They're just
cluttering the ini file and wasting memory in couch_config.

Change-Id:

Remove not used anymore parameters in ini config

These are not used anymore for a long time. They're just
cluttering the ini file and wasting memory in couch_config.

Change-Id: I33abfc9686e48d6d723a66ceddb2520b0e5d94ba
Reviewed-on: http://review.couchbase.org/26198
Tested-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Volker Mische <volker.mische@gmail.com>

show more ...


# 9d3190a8 07-May-2013 Filipe David Borba Manana <fdmanana@apache.org>

MB-8029 More efficient and simpler index build

Instead of accumulating small buffers in memory, sort them
and write them to disk while in parallel other KV batches are
being merged (

MB-8029 More efficient and simpler index build

Instead of accumulating small buffers in memory, sort them
and write them to disk while in parallel other KV batches are
being merged (on disk), keep the process simpler by just writing
all KV pairs into an unsorted file, and at the end sort those
files with the C file sorter in couchstore (couch_view_file_sorter
program). This approach is faster in C, but was attempted in the
past in Erlang and verified to be slower than current approach.

The goal of this change is mostly to move some work out from Erlang
to C, and is ground work for more incoming changes that will allow
moving all of the index build/update to C.
Another positive side effect of this change is that overall CPU and
memory consumption was smaller than before (Erlang!).

One of the tests that was done to measure performance follows.

Building an index for a design document with 4 views, from
one of the evperf tests, for a dataset of 20M documents:

Before this change:

Indexing time: 44m55.362s
Final file size: 2.3Gb

After this change:

Indexing time: 33m13.003s
Final file size: 2.3Gb

Change-Id: Ia7b3a0114e03d0110af2c749787fec7187d46ca0
Reviewed-on: http://review.couchbase.org/26194
Tested-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Volker Mische <volker.mische@gmail.com>

show more ...


# 3cefe098 10-Apr-2013 Filipe David Borba Manana <fdmanana@apache.org>

MB-8029 Use native file sorter for incremental index updates

The native file sorter, part of couchstore, is a C program to
sort files containing view KV records. Instead of doing this in

MB-8029 Use native file sorter for incremental index updates

The native file sorter, part of couchstore, is a C program to
sort files containing view KV records. Instead of doing this in
Erlang, we now do it with the external file sorter, which is more
efficient and faster.

Sample test:

For an index with 4 views (from evperf lucky8 test) and with
3.2M items per view, adding more 3.2M documents and triggering
an index update, the time it took to update the index and the
final index file size were measured.

Before this change:

Index update time: 10m21.888s
Final file size: 4.0Gb

After this change:

Index update time: 7m15.144s
Final file size: 3.2Gb

Change-Id: I91248d25f5200dac8ffe9a1ce28f6a73a85c0eb3
Reviewed-on: http://review.couchbase.org/25592
Reviewed-by: Volker Mische <volker.mische@gmail.com>
Tested-by: Filipe David Borba Manana <fdmanana@gmail.com>

show more ...


Revision tags: 2.0.1-linux
# fb6316e6 17-Dec-2012 Volker Mische <vmx@couchbase.com>

MB-7406: Add updater barrier for spatial indexer

Currently there's one spatial indexer for every vBucket,
hence a lot of updaters are running at the same time.
This commit adds a bar

MB-7406: Add updater barrier for spatial indexer

Currently there's one spatial indexer for every vBucket,
hence a lot of updaters are running at the same time.
This commit adds a barrier, so that you can set the
maximum number of updaters that should run at the same
time. The default value is 4. The MapReduce indexer
already uses such a barrier.

Those two commits should me merged at the same time:
http://review.couchbase.org/24830
http://review.couchbase.org/24831

Change-Id: Ie1e2a5b2f50cfa1137ec24ed414e9d8a4adac4d2
Reviewed-on: http://review.couchbase.org/24830
Tested-by: Volker Mische <volker.mische@gmail.com>
Tested-by: Fulu Li <fulu@couchbase.com>
Reviewed-by: Fulu Li <fulu@couchbase.com>

show more ...


# e64bba5b 28-Dec-2012 Filipe David Borba Manana <fdmanana@apache.org>

MB-7413 More efficient and faster incremental updates

At any time, insert larger batches of kv pairs into the
btrees. These batches are on disk sorted files, and while
they're applie

MB-7413 More efficient and faster incremental updates

At any time, insert larger batches of kv pairs into the
btrees. These batches are on disk sorted files, and while
they're applied to the btrees, the next batch is prepared
in parallel (maps, conversion to binaries, sorting, etc).
This reduces index update time, slows down fragmentation
and requires less disk space.

This has a very significant positive impact if the number
of changes to apply to an index are large, such as the case
when adding a group of new partitions (active, passive of
replica) to an index.

Small/medium scale performance test descriptions follows.

Test 1:

Initially 3.2M documents, design document with 4 views (both
from evperf lucky8 tests), index was built and later more
3.2M documents were added (new 512 partitions added to the
index as active).

Initial build took: 11m17.911s
Index file size after initial build: 380Mb

** Before this change

Incremental update took: 75m43.575s
Index file size after incremental update: 30.0Gb

** After this change

Incremental update took: 18m36.762s
Index file size after incremental update: 4.0Gb

Test 2:

Initially 3.2M documents (from lucky8 evperf test), design
document with a single view (emits meta.id key and null value
for each document), index was built and later more 3.2M documents
were added (new 512 partitions added to the index as active).

Initial build took: 6m23.199s
Index file size after initial build: 134Mb

** Before this change

Incremental update took: 21m31.615s
Index file size after incremental update: 8.6Gb

** After this change

Incremental update took: 7m13.394s
Index file size after incremental update: 1.8Gb

Change-Id: I66fde0adeb0f7a810c125bf7cb42cdd17dd03d9d
Reviewed-on: http://review.couchbase.org/23674
Tested-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Volker Mische <volker.mische@gmail.com>

show more ...


1234567