History log of /3.0.3-GA/ep-engine/ (Results 251 - 275 of 4208)
Revision (<<< Hide revision tags) (Show revision tags >>>)Date Author Comments
(<<< Hide modified files)
(Show modified files >>>)
60f8f80223-Jun-2014 Mike Wiederhold <mike@couchbase.com>

MB-11502: Fix incorrect snap_start/end seqno's in stream retry

If we set the snap_send_seqno to always be equal to the
snap_start_seqno then it is possible that the snap_end_seqno will

MB-11502: Fix incorrect snap_start/end seqno's in stream retry

If we set the snap_send_seqno to always be equal to the
snap_start_seqno then it is possible that the snap_end_seqno will
be less than then start seqno and this will cause the stream
creation to fail on the producer side.

Change-Id: I73f1fdda6cf0f979c8b128f01a9366ecb6321b20
Reviewed-on: http://review.couchbase.org/38703
Reviewed-by: Sriram Ganesan <sriram@couchbase.com>
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

ecf226e923-Jun-2014 Mike Wiederhold <mike@couchbase.com>

MB-11085: Always generate a failover entry after an unclean shutdown

Previously we would not generate a new failover entry after an
unclean shutdown if the persisted seqno did not change

MB-11085: Always generate a failover entry after an unclean shutdown

Previously we would not generate a new failover entry after an
unclean shutdown if the persisted seqno did not change between
when the server was originally started and when is crashed. This can
cause issues because it is possible that mutations were received in
memory and streamed to another node and then the server crashes. If
we do not generate a new failover entry in this case then we will
not properly tell the node that received the mutations from memory
to rollback.

Change-Id: I6a97c12f2eb300984aaa81d7f9fdd0eb3272233d
Reviewed-on: http://review.couchbase.org/38698
Reviewed-by: David Liao <david.liao@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

4cf01bc023-Jun-2014 Trond Norbye <trond.norbye@gmail.com>

Free message from StreamEndResponse

Change-Id: I60c407870fef619d6fd56abc06f8e7064bf3b247
Reviewed-on: http://review.couchbase.org/38665
Reviewed-by: Dave Rigby <daver@couchbase.com>

Free message from StreamEndResponse

Change-Id: I60c407870fef619d6fd56abc06f8e7064bf3b247
Reviewed-on: http://review.couchbase.org/38665
Reviewed-by: Dave Rigby <daver@couchbase.com>
Tested-by: Trond Norbye <trond.norbye@gmail.com>

show more ...

5af91ec923-Jun-2014 Trond Norbye <trond.norbye@gmail.com>

Free SnapshotMarkers

> ::findleaks
BYTES LEAKED VMEM_SEG CALLER
12288 12 fffffd7ffc35f000 MMAP
12288 1 fffffd7ffcf7c000 MMAP

Free SnapshotMarkers

> ::findleaks
BYTES LEAKED VMEM_SEG CALLER
12288 12 fffffd7ffc35f000 MMAP
12288 1 fffffd7ffcf7c000 MMAP
8192 1 fffffd7ffcf6d000 MMAP
12288 1 fffffd7ffcd7d000 MMAP
12288 1 fffffd7ffcd6d000 MMAP
8192 1 fffffd7ffc96d000 MMAP
12288 1 fffffd7ffc95c000 MMAP
8192 1 fffffd7ffc76e000 MMAP
12288 1 fffffd7ffc75d000 MMAP
8192 1 fffffd7ffc56f000 MMAP
12288 1 fffffd7ffc55e000 MMAP
12288 1 fffffd7ffc36f000 MMAP
------------------------------------------------------------------------
Total 12 oversized leaks, 131072 bytes

CACHE LEAKED BUFCTL CALLER
0000000000485028 1 00000000089eb2a0 ep.so`MemoryTracker::MemoryTracker+0x6b
000000000048a028 512 0000000008a6cb60 libstdc++.so.6.0.17`operator new+0x1e
------------------------------------------------------------------------
Total 513 buffers, 32776 bytes
> 0000000008a6cb60::bu
bufctl
bufctl_audit
> 0000000008a6cb60::bufctl -v
ADDR BUFADDR TIMESTAMP THREAD
CACHE LASTLOG CONTENTS
8a6cb60 8a6b880 c3fb052e883da 5
48a028 0 0
libumem.so.1`umem_cache_alloc_debug+0xfd
libumem.so.1`umem_cache_alloc+0xb3
libumem.so.1`umem_alloc+0x64
libumem.so.1`malloc+0x3f
libstdc++.so.6.0.17`operator new+0x1e
ep.so`UprConsumer::snapshotMarker+0xc7
ep.so`EvpUprSnapshotMarker+0x75
bucket_engine.so`upr_snapshot_marker+0x8a
upr_snapshot_marker_executor+0x100
process_bin_packet+0xb8
complete_nread+0x182
conn_nread+0x48
event_handler+0x1ce
libevent_core-2.0.so.5.1.9`event_base_loop+0x436
worker_libevent+0x56

Change-Id: I8aeafa031b6fd9ae5d918abd9820558b311a9a72
Reviewed-on: http://review.couchbase.org/38664
Reviewed-by: Dave Rigby <daver@couchbase.com>
Tested-by: Trond Norbye <trond.norbye@gmail.com>

show more ...

33ee8aba23-Jun-2014 Trond Norbye <trond.norbye@gmail.com>

Release allocated resources by cJSON

> ::findleaks
BYTES LEAKED VMEM_SEG CALLER
8192 6 fffffd7ffc360000 MMAP
8192 1 fffffd

Release allocated resources by cJSON

> ::findleaks
BYTES LEAKED VMEM_SEG CALLER
8192 6 fffffd7ffc360000 MMAP
8192 1 fffffd7ffcf6d000 MMAP
8192 1 fffffd7ffcd6e000 MMAP
8192 1 fffffd7ffc95d000 MMAP
8192 1 fffffd7ffc75e000 MMAP
8192 1 fffffd7ffc55f000 MMAP
------------------------------------------------------------------------
Total 6 oversized leaks, 49152 bytes

CACHE LEAKED BUFCTL CALLER
0000000000485028 1 0000000008977380 ep.so`MemoryTracker::MemoryTracker+0x6b
00000000004a5028 1 000000001000b9a0 ep.so`EventuallyPersistentEngine::setClusterConfig+0x5f
000000000048a668 2046 000000000a0579a0 libcJSON.so.1.0.0`cJSON_New_Item+0x17
000000000048a668 2048 000000000a057c40 libcJSON.so.1.0.0`cJSON_New_Item+0x17
000000000048a668 2047 000000000a057b60 libcJSON.so.1.0.0`cJSON_New_Item+0x17
000000000048a668 2046 000000000a057a80 libcJSON.so.1.0.0`cJSON_New_Item+0x17
0000000000485668 2046 000000000a053540 libcJSON.so.1.0.0`parse_string+0x8a
0000000000485668 2046 000000000a053620 libcJSON.so.1.0.0`parse_string+0x8a
000000000048a028 512 000000000fb3e0e0 libstdc++.so.6.0.17`operator new+0x1e
------------------------------------------------------------------------
Total 12793 buffers, 761400 bytes
> 000000000a053620::bufctl -v
ADDR BUFADDR TIMESTAMP THREAD
CACHE LASTLOG CONTENTS
a053620 a04dca0 c3978651e3b4b 14
485668 0 0
libumem.so.1`umem_cache_alloc_debug+0xfd
libumem.so.1`umem_cache_alloc+0xb3
libumem.so.1`umem_alloc+0x64
libumem.so.1`malloc+0x3f
libcJSON.so.1.0.0`parse_string+0x8a
libcJSON.so.1.0.0`parse_object+0x8f
libcJSON.so.1.0.0`parse_value+0xbe
libcJSON.so.1.0.0`parse_array+0xa6
libcJSON.so.1.0.0`parse_value+0x9b
libcJSON.so.1.0.0`cJSON_Parse+0x41
ep.so`FailoverTable::loadFromJSON+0x24
ep.so`FailoverTable::FailoverTable+0x65
ep.so`Warmup::createVBuckets+0x152
ep.so`WarmupCreateVBuckets::run+0x29
ep.so`ExecutorThread::run+0x262

Change-Id: I20f4d46d39996df04b60545e72e5250e8ef2e533
Reviewed-on: http://review.couchbase.org/38658
Reviewed-by: Dave Rigby <daver@couchbase.com>
Tested-by: Trond Norbye <trond.norbye@gmail.com>

show more ...

4a5d415e23-Jun-2014 Trond Norbye <trond.norbye@gmail.com>

Release cluster configuration during shutdown

> ::findleaks -v
BYTES LEAKED VMEM_SEG CALLER
12288 6 fffffd7ffc35f000 MMAP
12288

Release cluster configuration during shutdown

> ::findleaks -v
BYTES LEAKED VMEM_SEG CALLER
12288 6 fffffd7ffc35f000 MMAP
12288 1 fffffd7ffcf6c000 MMAP
12288 1 fffffd7ffcd6d000 MMAP
12288 1 fffffd7ffc95c000 MMAP
12288 1 fffffd7ffc75d000 MMAP
12288 1 fffffd7ffc55e000 MMAP
------------------------------------------------------------------------
Total 6 oversized leaks, 73728 bytes

CACHE LEAKED BUFCTL CALLER
0000000000485028 1 00000000089d0380 ep.so`MemoryTracker::MemoryTracker+0x6b
00000000004a5028 1 0000000009b92ee0 ep.so`EventuallyPersistentEngine::setClusterConfig+0x5f
000000000048a028 512 00000000098300e0 libstdc++.so.6.0.17`operator new+0x1e
------------------------------------------------------------------------
Total 514 buffers, 40968 bytes
> 0000000009b92ee0::bufctl -v
ADDR BUFADDR TIMESTAMP THREAD
CACHE LASTLOG CONTENTS
9b92ee0 df8f000 c39fa1a13ad83 4
4a5028 0 0
libumem.so.1`umem_cache_alloc_debug+0xfd
libumem.so.1`umem_cache_alloc+0xb3
libumem.so.1`umem_alloc+0x64
libumem.so.1`malloc+0x3f
ep.so`EventuallyPersistentEngine::setClusterConfig+0x5f
ep.so`processUnknownCommand+0x73a
ep.so`EvpUnknownCommand+0x38
bucket_engine.so`bucket_unknown_command+0x23d
default_unknown_command+0xb8
process_bin_unknown_packet+0xa9
process_bin_packet+0xc6
complete_nread+0x182
conn_nread+0x48
event_handler+0x1ce
libevent_core-2.0.so.5.1.9`event_base_loop+0x436

Change-Id: I093d7d7112701e7c23bb9522246489d532668aa6
Reviewed-on: http://review.couchbase.org/38659
Reviewed-by: Dave Rigby <daver@couchbase.com>
Tested-by: Trond Norbye <trond.norbye@gmail.com>

show more ...

197f9c9620-Jun-2014 Sundar Sridharan <sundar.sridharan@gmail.com>

MB-11405 Reduce CPU use in active system

Instead of having all the threads snooze for the minimum
snooze time across all task sets, have only 1 thread per
task set snooze while other

MB-11405 Reduce CPU use in active system

Instead of having all the threads snooze for the minimum
snooze time across all task sets, have only 1 thread per
task set snooze while others sleep for longer duration.
This not only uses lower cpu in an active system with 5+ buckets
but also reduces lock contention in smaller configurations.

Change-Id: I99a203c3d0fb36612fb884c44684f747d2f09e42
Reviewed-on: http://review.couchbase.org/38620
Reviewed-by: Michael Wiederhold <mike@couchbase.com>
Tested-by: Sundararaman Sridharan <sundar@couchbase.com>

show more ...

3adce30420-Jun-2014 Mike Wiederhold <mike@couchbase.com>

MB-11459: Add debug logging for high seqno incorrectly set

Change-Id: I44e80cf415f8b7c4bba70a9b684301534e026c03
Reviewed-on: http://review.couchbase.org/38606
Reviewed-by: abhinav da

MB-11459: Add debug logging for high seqno incorrectly set

Change-Id: I44e80cf415f8b7c4bba70a9b684301534e026c03
Reviewed-on: http://review.couchbase.org/38606
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

d69977b120-Jun-2014 abhinavdangeti <abhinav@couchbase.com>

MB-11344: Convert a temp item during an insert

- If a temp item is already found to exist during
warmup, then convert the temp item into a regular
item.

Change-Id: I12bc1ead

MB-11344: Convert a temp item during an insert

- If a temp item is already found to exist during
warmup, then convert the temp item into a regular
item.

Change-Id: I12bc1ead075cb19f9ef397397f84abe6086f50e4
Reviewed-on: http://review.couchbase.org/38531
Reviewed-by: Michael Wiederhold <mike@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

c7d941cc17-Jun-2014 Dave Rigby <daver@couchbase.com>

CMake: Search project include dirs before /usr/local/include

This ensures that if /usr/local/include has outdated header files
(e.g. you previously compiled and installed CB 2.5.1), then

CMake: Search project include dirs before /usr/local/include

This ensures that if /usr/local/include has outdated header files
(e.g. you previously compiled and installed CB 2.5.1), then the build
of master doesn't fail due to it picking up outdated header files.

Change-Id: I9c4d7987adebb0780d3a0b08f27b666af128708d
Reviewed-on: http://review.couchbase.org/38361
Reviewed-by: Trond Norbye <trond.norbye@gmail.com>
Tested-by: Dave Rigby <daver@couchbase.com>

show more ...

e1e06be218-Jun-2014 Sriram Ganesan <sriram@couchbase.com>

MB-11407: Add logs to the UPR consumer rollback code

On the UPR consumer side, there aren't enough logs to indicate
if a rollback has been processed or not.

Change-Id: I5df76097

MB-11407: Add logs to the UPR consumer rollback code

On the UPR consumer side, there aren't enough logs to indicate
if a rollback has been processed or not.

Change-Id: I5df76097db803e4ed75915a92058f109f526dc16
Reviewed-on: http://review.couchbase.org/38456
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Reviewed-by: Michael Wiederhold <mike@couchbase.com>
Tested-by: Sriram Ganesan <sriram@couchbase.com>

show more ...

694f5a6118-Jun-2014 Mike Wiederhold <mike@couchbase.com>

MB-11477: Fix crash in process items due to no default argument

Change-Id: I4932ce2094273ce1e877bea70c8e279e232d35f2
Reviewed-on: http://review.couchbase.org/38451
Reviewed-by: abhin

MB-11477: Fix crash in process items due to no default argument

Change-Id: I4932ce2094273ce1e877bea70c8e279e232d35f2
Reviewed-on: http://review.couchbase.org/38451
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Reviewed-by: Sriram Ganesan <sriram@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

f43d865d18-Jun-2014 Mike Wiederhold <mike@couchbase.com>

Revert "MB-11361: Implement upr noop command"

We will add this back once some of the other components fix
issues that were caused by this patch.

This reverts commit e2e75cb16b1e

Revert "MB-11361: Implement upr noop command"

We will add this back once some of the other components fix
issues that were caused by this patch.

This reverts commit e2e75cb16b1e5bf715e9437a55580df46a72a011.

Change-Id: I46c1d14ac60044f6a6b34f2fcd08f615eefe219b
Reviewed-on: http://review.couchbase.org/38438
Tested-by: Aliaksey Kandratsenka <alkondratenko@gmail.com>
Reviewed-by: Michael Wiederhold <mike@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

e2e75cb113-Jun-2014 Mike Wiederhold <mike@couchbase.com>

MB-11361: Implement upr noop command

The upr noop command is used to detect dead connections. It is sent
once every 3 minutes and waits 3 minutes for a response. If no
response comes

MB-11361: Implement upr noop command

The upr noop command is used to detect dead connections. It is sent
once every 3 minutes and waits 3 minutes for a response. If no
response comes in that time the connection is deemed to be dead. The
noop command is only sent when flow control is turned on by the
consumer.

Change-Id: Ib6f1eb51810d9c641ca52d1e3cda8ae9ffa3684c
Reviewed-on: http://review.couchbase.org/38355
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

3f63f2c717-Jun-2014 Mike Wiederhold <mike@couchbase.com>

MB-11447: Don't close db handle if rollback header fails

This happens automatically in couchstore and will cause a segfault
if we try to close the handle again.

Change-Id: I4e0f

MB-11447: Don't close db handle if rollback header fails

This happens automatically in couchstore and will cause a segfault
if we try to close the handle again.

Change-Id: I4e0f756749951ac869861fc150303b9df594ca65
Reviewed-on: http://review.couchbase.org/38377
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

2beda27417-Jun-2014 Mike Wiederhold <mike@couchbase.com>

Enable flow control for intra-cluster replication by default

We have fixed all know issues and run a bunch of tests to ensure
that this is now working. As a result we will turn flow cont

Enable flow control for intra-cluster replication by default

We have fixed all know issues and run a bunch of tests to ensure
that this is now working. As a result we will turn flow control
back on by default.

Change-Id: Ib3a498b9798eb7eccf03722464a220dd3f7da828
Reviewed-on: http://review.couchbase.org/38367
Reviewed-by: Sriram Ganesan <sriram@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

cd32661017-Jun-2014 Sundar Sridharan <sundar.sridharan@gmail.com>

Add support for custom unit test directory

/tmp/test being hard-coded into ep_testsuite.cc makes it
hard to have unit tests run on ramdisks. This change allows
the unit test suite to

Add support for custom unit test directory

/tmp/test being hard-coded into ep_testsuite.cc makes it
hard to have unit tests run on ramdisks. This change allows
the unit test suite to be pointed to a dir in ramdisk.

Change-Id: I2e428ed6fe1a38513db27a70aa7a1c36bbe90c6f
Reviewed-on: http://review.couchbase.org/38379
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Tested-by: Sundararaman Sridharan <sundar@couchbase.com>

show more ...

63ab9e2516-Jun-2014 Trond Norbye <trond.norbye@gmail.com>

Add memory-related stats to the memory-stat cmd

Change-Id: I95e419d6dc3429a63fa7f50fa5b744a7b2d7f34e
Reviewed-on: http://review.couchbase.org/38319
Reviewed-by: Michael Wiederhold <m

Add memory-related stats to the memory-stat cmd

Change-Id: I95e419d6dc3429a63fa7f50fa5b744a7b2d7f34e
Reviewed-on: http://review.couchbase.org/38319
Reviewed-by: Michael Wiederhold <mike@couchbase.com>
Tested-by: Trond Norbye <trond.norbye@gmail.com>

show more ...

7353d90a06-Jun-2014 Sriram Ganesan <sriram@couchbase.com>

MB-11263: Add retry logic for temporarily failed mutations

Processing a mutation on the consumer side can fail due to temporary out of memory
conditions. In that case, we just need to re

MB-11263: Add retry logic for temporarily failed mutations

Processing a mutation on the consumer side can fail due to temporary out of memory
conditions. In that case, we just need to return from the failure and re-process
the failed message in the next iteration.

Change-Id: Idd50fa66d7288db9e25db837a19d2bdd0057437f
Reviewed-on: http://review.couchbase.org/37999
Reviewed-by: Michael Wiederhold <mike@couchbase.com>
Tested-by: Sriram Ganesan <sriram@couchbase.com>

show more ...

8d1a921a17-Jun-2014 Mike Wiederhold <mike@couchbase.com>

Add brackets to if statement in ActiveStream

Change-Id: I1b251767fe35c74c265fd742033da4c7b1c97866
Reviewed-on: http://review.couchbase.org/38354
Reviewed-by: abhinav dangeti <abhinav

Add brackets to if statement in ActiveStream

Change-Id: I1b251767fe35c74c265fd742033da4c7b1c97866
Reviewed-on: http://review.couchbase.org/38354
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

64f8b29517-Jun-2014 Sundar Sridharan <sundar.sridharan@gmail.com>

MB-11297: Ensure lastSentSeqno < endSeqno

As we send the complete snap shot, it can so
happen that lastSentSeqno be greater than endSeqno
for an active stream, causing an underflow i

MB-11297: Ensure lastSentSeqno < endSeqno

As we send the complete snap shot, it can so
happen that lastSentSeqno be greater than endSeqno
for an active stream, causing an underflow in
stats collection.
Change-Id: I8a8f0755f988c8c1ef9046020a4702775d2ddd4c
Reviewed-on: http://review.couchbase.org/38353
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Tested-by: Sundararaman Sridharan <sundar@couchbase.com>

show more ...

2a30f37116-Jun-2014 Mike Wiederhold <mike@couchbase.com>

MB-11421: Supports ack should be false for upr connections

Upr doesn't support ack/nack logic.

Change-Id: I5ba3d1cf323f527ce59ed2d8dd901566f3bcd1f5
Reviewed-on: http://review.co

MB-11421: Supports ack should be false for upr connections

Upr doesn't support ack/nack logic.

Change-Id: I5ba3d1cf323f527ce59ed2d8dd901566f3bcd1f5
Reviewed-on: http://review.couchbase.org/38347
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Reviewed-by: Sundararaman Sridharan <sundar@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

420a088e16-Jun-2014 Mike Wiederhold <mike@couchbase.com>

MB-11376: Don't rollback if snap_start == start == upper

We shouldn't tell a client to rollback to the same seqno that it
request.

Change-Id: If6c9bd438e2a70f637ff8bcb28bc46258d

MB-11376: Don't rollback if snap_start == start == upper

We shouldn't tell a client to rollback to the same seqno that it
request.

Change-Id: If6c9bd438e2a70f637ff8bcb28bc46258d28b10c
Reviewed-on: http://review.couchbase.org/38344
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

e6e2550316-Jun-2014 abhinavdangeti <abhinav@couchbase.com>

Account for UPR_EXPIRATION in upr-producer's getNextItem()

Change-Id: Ib31c01617398c35ebbce56784835af8a49b7e260
Reviewed-on: http://review.couchbase.org/38345
Reviewed-by: Michael Wi

Account for UPR_EXPIRATION in upr-producer's getNextItem()

Change-Id: Ib31c01617398c35ebbce56784835af8a49b7e260
Reviewed-on: http://review.couchbase.org/38345
Reviewed-by: Michael Wiederhold <mike@couchbase.com>
Tested-by: abhinav dangeti <abhinav@couchbase.com>

show more ...

463abeaa16-Jun-2014 Mike Wiederhold <mike@couchbase.com>

Implement upr expiration command

We don't support doing anything different than a deletion within
the enigne yet so whenever we see and expiration we should just
treat it as a deleti

Implement upr expiration command

We don't support doing anything different than a deletion within
the enigne yet so whenever we see and expiration we should just
treat it as a deletion.

Change-Id: I709ba05e7cdfa1800a62d0a54143c991e8787f96
Reviewed-on: http://review.couchbase.org/38338
Reviewed-by: abhinav dangeti <abhinav@couchbase.com>
Tested-by: Michael Wiederhold <mike@couchbase.com>

show more ...

1...<<11121314151617181920>>...169