dwight
|
41ff2157a3
|
SERVER-3962 previous commit wasnt right. backport.
|
2011-10-08 23:32:17 -04:00 |
|
dwight
|
a7796c9c09
|
finish SERVER-3962
|
2011-10-08 23:32:11 -04:00 |
|
dwight
|
8814b71b26
|
SERVER-3962 very high extraneous ismaster type commands from mongos to repl set if slaveok and no pure secondary is up
|
2011-10-08 23:32:01 -04:00 |
|
Spencer T Brody
|
861bc071b8
|
Fix mongos skipping checking the health of the first node in the _nodes array
in the replica set monitor. SERVER-3605
|
2011-08-15 19:12:37 -04:00 |
|
gregs
|
b52af9b709
|
fix for sharding slaveok
|
2011-08-09 02:29:41 -04:00 |
|
Eliot Horowitz
|
5598fe195b
|
fix mongos slave ok
|
2011-08-08 23:37:44 -04:00 |
|
Eliot Horowitz
|
760f0b28d4
|
update hidden status and check secondary more accurately SERVER-3432
|
2011-08-08 19:00:59 -04:00 |
|
Aaron
|
94412807e6
|
SERVER-3249 add so_timeout support in DBClientReplicaSet
|
2011-08-07 15:09:50 -07:00 |
|
Eliot Horowitz
|
41f43db6e2
|
better slave handling for SERVER-3462
|
2011-08-06 19:50:16 -04:00 |
|
Eliot Horowitz
|
7355a811ba
|
finish SERVER-3462
|
2011-08-06 15:02:15 -04:00 |
|
gregs
|
8ac0b95fee
|
port tech could be >= 0
|
2011-07-28 13:24:59 -04:00 |
|
gregs
|
dad6351cfb
|
check host empty not port, port() always > 0
|
2011-07-28 12:56:48 -04:00 |
|
gregs
|
b776b8c9d3
|
additional logging for slave selection
|
2011-07-28 10:45:41 -04:00 |
|
Eliot Horowitz
|
20a0374718
|
expose ReplicaSetMonitor stats
SERVER-3487
|
2011-07-26 18:07:29 -04:00 |
|
Spencer T Brody
|
39032a0f97
|
Make doing slaveOk reads use all secondaries - SERVER-3462
|
2011-07-26 10:33:01 -04:00 |
|
Kristina
|
8a3a9eb501
|
fix null deref SERVER-3285
Patch by Mike Mueller
|
2011-06-23 17:51:19 -04:00 |
|
Mathias Stearn
|
3eb1de47f1
|
gcc 4.6 compile
|
2011-06-17 17:28:40 -04:00 |
|
gregs
|
884e100ec0
|
better logic for no-result in rs for parallel ops SERVER-2481
|
2011-06-16 17:34:40 -04:00 |
|
gregs
|
4d8ee4cc7c
|
parallel cursor recover gracefully from replica set and other errors SERVER-2481
|
2011-06-16 17:34:39 -04:00 |
|
Greg Studer
|
7db56fc74c
|
add return
|
2011-06-01 13:18:11 -07:00 |
|
gregs
|
b2236461fa
|
check for not master or secondary error on slave queries, re-route SERVER-2253
|
2011-06-01 15:51:10 -04:00 |
|
gregs
|
be7074a9dd
|
check if master is null CS-596
|
2011-05-27 19:42:44 -04:00 |
|
Mathias Stearn
|
b690e237fd
|
Add InsertOption_KeepGoing to keep going after error on bulk insert. SERVER-509
|
2011-05-23 18:46:27 -04:00 |
|
gregs
|
cd0dbafc47
|
Fix for potential unsafe access of address
|
2011-05-19 15:41:01 -04:00 |
|
gregs
|
1f4c905cff
|
fix for duplicate host names
|
2011-05-19 15:40:34 -04:00 |
|
Eliot Horowitz
|
225227e8e2
|
fix deadlock
|
2011-05-12 16:57:28 -04:00 |
|
Eliot Horowitz
|
5d22fbae02
|
fix race condition when adding hosts to the list of relica set nodes
|
2011-05-12 15:44:33 -04:00 |
|
gregs
|
0017f70d11
|
don't shadow return value for maybePrimary
|
2011-05-12 15:25:30 -04:00 |
|
gregs
|
2b922d9097
|
deterministic slaveok failover
|
2011-04-27 17:54:49 -04:00 |
|
Eliot Horowitz
|
2a4fc317fc
|
a little more aggressive trying slaves
|
2011-04-27 00:00:32 -04:00 |
|
gregs
|
627048913d
|
log output fix and slightly more verbose slave failover
|
2011-04-26 16:33:31 -04:00 |
|
Eliot Horowitz
|
f939069509
|
DBClientBase::callLazy - does the send part, and doesn't wait for response
you have to call recv later
|
2011-04-21 14:57:44 -04:00 |
|
Eliot Horowitz
|
1707fa2a7e
|
fix callback for not master errors on replica set connections and make background thread more reliable SERVER-2868
|
2011-03-30 13:52:51 -04:00 |
|
Eliot Horowitz
|
68cbfd1e48
|
fix rs client error on failover SERVER-2619
|
2011-03-03 12:20:49 -05:00 |
|
Eliot Horowitz
|
763c49f692
|
fix startup race condition
|
2011-02-13 23:29:11 -05:00 |
|
Eliot Horowitz
|
7e5a60ba66
|
protect ReplicaSet watcher with a lock
|
2011-02-13 19:13:21 -05:00 |
|
Eliot Horowitz
|
7a4399841f
|
more background job debugging
|
2011-02-13 16:04:06 -05:00 |
|
Eliot Horowitz
|
f31d82cad9
|
some better debugging and an assert
|
2011-02-13 10:07:13 -05:00 |
|
Eliot Horowitz
|
b1a9d70ed7
|
safety check for SERVER-2539
|
2011-02-12 14:54:14 -05:00 |
|
Eliot Horowitz
|
f4f7ceccf8
|
some public accessors
|
2011-02-12 14:48:20 -05:00 |
|
Eliot Horowitz
|
37b11e4dde
|
fix replica set cursor cleaning through mongos SERVER-2505
fix cursor cleaning in mongos generally by not doing lazy cleaning
|
2011-02-09 23:03:03 -05:00 |
|
Eliot Horowitz
|
1ef1e7f2b8
|
less chatty ReplicaSetMonitor for c++ client SERVER-2490
|
2011-02-07 16:42:35 -05:00 |
|
Eliot Horowitz
|
c3125e1877
|
DBClientReplicaSet::killCursor should never be called SERVER-2479
|
2011-02-05 17:56:14 -05:00 |
|
Eliot Horowitz
|
471fd7da06
|
added option paral to call to specify server address where was sent if different
|
2011-02-05 17:52:36 -05:00 |
|
Eliot Horowitz
|
1227dfe359
|
fix concurrency issues with DBClientReplicaSet SERVER-2466
|
2011-02-02 10:47:02 -05:00 |
|
Kristina
|
8867e60e88
|
more aggressive about checking for slave errors SERVER-2439
|
2011-01-31 11:34:47 -05:00 |
|
Eliot Horowitz
|
9c1381ece0
|
ReplicaSetMonitorWatcher needs to run all the time SERVER-2399
|
2011-01-24 10:03:23 -05:00 |
|
Kristina
|
f0eb24f224
|
rs throws out master on connection exception SERVER-2377
|
2011-01-21 13:57:35 -05:00 |
|
Kristina
|
a756cf2375
|
make c++ client check that rs master is ok before returning it SERVER-2377
|
2011-01-20 19:35:12 -05:00 |
|
Kristina
|
a380c189fc
|
make checkMaster more aggressive about finding new master SERVER-2261
|
2011-01-14 15:45:25 -05:00 |
|