2f02bfd0cd
Use volatile field instead of AtomicReference
...
it's not necessary to use `AtomicReference` here since the result of atomic method `updateAndGet()` is unused, and not necessary to use `AtomicReferenceFieldUpdater` here since only `set()` is using, volatile field access is more appropriate.
5 months ago
d54418061f
"merged" #2126 Added Support For beginRequest and endRequest (new version)
6 months ago
a671b6e24d
Close connections marked as evicted instead of returning them to the pool ( #2157 )
6 months ago
b3e6d23dc5
Consistent handling of errorCode and sqlState in timeout exception ( #2226 )
...
Co-authored-by: the-vj <one.vallu@gmail.com>
6 months ago
7ec98c3ee2
Minor debug logging change.
6 months ago
f601340304
Minor debug logging change.
6 months ago
0a6ccdb334
Only log 'Connection not added' when connection not added
1 year ago
61ac830709
Remove error log for pool initialization exception ( #1660 )
...
Co-authored-by: Jeff Fang <jeff.fang@activenetwork.com>
2 years ago
69394742a6
Add pool stats to connection acquisition timeout exception
3 years ago
3104292f68
Minor code cleanups.
3 years ago
fbd2a18910
Add log when connection not added.
3 years ago
c2148006ce
Simplify connection creation.
3 years ago
ed2da5f1f4
Fix idle timeout tests, sleep too tight to idle threshold. (+1 squashed commit)
...
Squashed commits:
[4844f9e1] Fix idle timeout tests, sleep too tight to idle threshold. (+8 squashed commits)
Squashed commits:
[e0ec6b7f] Fix idle timeout tests, sleep too tight to idle threshold.
[c40a690d] Fix idle timeout tests, sleep too tight to idle threshold.
[7a0a1560] Synchronize idleTimeout manipulation methods
[bab9f648] Fix idle timeout tests, sleep too tight to idle threshold.
[4ee5259b] Add missing ) character in message
[e9d49fc2] Enable debug logging for tests.
[89c1ab45] Fix idle timeout tests, sleep too tight to idle threshold.
[aa06bb85] Simplify pool entry last-accessed time handling.
4 years ago
1991355e6b
Convert to Java 11
4 years ago
d0c7d54bc9
Fix regression cause by #1605 . When returning the addConnectionExecutor pool sizes
...
back to defaults, the core pool size must be changed before the maximum pool size.
4 years ago
04919ad540
lint cleanup and Visual Studio Code files
4 years ago
ade83d14cd
Code cleanup.
4 years ago
0561309e32
features: add heartbeat to keepalive connection if configure it ( #1699 )
...
* features: add heartbeat to keepalive connection if configure it
* features: add heartbeat to keepalive connection if configure it
* features: add heartbeat to keepalive connection if configure it
* features: keepalive unit test
* optimize code
* optimize code
4 years ago
00f5f202a4
Due to change in JDK-11 and later in ThreadPoolExecutor.java, setMaximumPoolSize needs to be done before setCorePoolSize, the current implementation throws IllegalArgumentException and hence prefilling of connections by setting com.zaxxer.hikari.blockUntilFilled is broken. ( #1605 )
...
Co-authored-by: rampr_000 <spachanady@gmail.com>
4 years ago
8df41bf748
Additional logging.
5 years ago
fe29962233
When blocking initialization is specified, use a minimum of 16 connection creation threads.
5 years ago
93c7178012
Fixes #1528 regression in fail fast handling.
5 years ago
e8e9cdd00a
fixes #1287 set core pool size before max pool size
6 years ago
1ac94f16a2
Fixes #1287 when system property blockUntilFilled is set, use multiple threads to fill the pool.
6 years ago
f005a4769f
Various clean-ups pointed out by sonarcloud.io
6 years ago
471e27ece0
Use orElseGet to delay call until necessary ( #1210 )
7 years ago
74fa2a84e8
Log uncaught error in acquiring connection (+1 squashed commit)
...
Squashed commits:
[7a00d8bd
7 years ago
b5c340eee3
Permit changing the catalog via the HikariConfigMXBean
7 years ago
00cd48a7c8
Fixes #1117 Add (currently undocumented) system property to block pool startup until minimumIdle connections exist in the pool.
7 years ago
3313174891
Add pool metrics ( #1110 )
7 years ago
7bff6723c1
Fixes #1095 fix breakage caused by sealed configuration with respect to special handling
...
for the metricsRegistry and metricsTrackerFactory properties, which are allowed to be
altered *once* after the pool has started.
7 years ago
91bd7dfb33
Allowing anonymous subclasses of MetricRegistry. Also changing checks for metrics libraries from classname contains to assignableFrom the metrics class ( #1089 )
7 years ago
4a4c139654
Fix Java file permissions ( #1015 )
...
Some files were inadvertently marked executable.
7 years ago
4b796b5209
Clean up and better JavaDoc.
7 years ago
d6a9cd435c
Fixes #983 fix logic that determines how many idle connections can be removed, without violating the minimumIdle contract.
8 years ago
6735fcfe3a
Closes #989 Merged the parts I agree with.
8 years ago
2a9e0185e4
Fix NPE caused by merge.
8 years ago
43b423cc5a
Log better message when connection was evicted or dead. ( #977 )
8 years ago
63f6ff371f
Fixed ClassNotFoundException when com.codahale.metrics.MetricRegistry is not in the classpath ( #984 )
...
* Fixed ClassNotFoundException when com.codahale.metrics.MetricRegistry is not in the classpath
* Changed equals to contains to handle CompositeMeterRegistry
8 years ago
051f78c276
Alpha-level support for micrometer metrics
8 years ago
47b827d11e
Fixes #929 length the shutdown timeouts to 10 seconds. Beyond this,
...
user's need to poll the pool themselves if they want to wait for all the
connections to disappear.
8 years ago
2a0a065b24
Introduce factory for leak detection tasks ( #943 )
...
This change splits off a ProxyLeakTaskFactory from the ProxyLeakTask class
so that scheduling the task is done after its constructor completes. This
prevents publishing the tasks "this" reference during construction.
See "Safe construction techniques" by Brian Goetz.
8 years ago
61be9b923d
Fixes #880 Fix race condition caused by sorting collection
...
while the condition of sort can change.
8 years ago
e7a3e39f9b
Better place to capture pool.Wait metric ( #831 )
...
@wvuong Thanks. Yes, that is better.
8 years ago
d875e838c3
pool.Wait metric doesn't include timed out requests
...
We noticed that the pool.Wait metric only captured the timings for successful connection requests and doesn't capture when a connection request times out.
I didn't want to mess with IMetricsTrackerDelegate.recordBorrowStats() so I created a recordBorrowTimeoutStats() and implementation in MetricsTrackerDelegate instead. Feel free to tweak to your liking.
8 years ago
0ab35e7445
Minor cleanup.
8 years ago
2414ebdaa4
Queue immediate replacement of retired maxLifetime connection.
8 years ago
ef7da7dad4
Minor cleanup.
8 years ago
ea3043dea2
Fixes #817 restore former failfast=false behavior by completely
...
bypassing the connection validation if initializationFailTimeout is
negative.
8 years ago
1587198e6a
Change the internal handling of the initializationFailTimeout.
8 years ago
Yanming Zhou
Brett Wooldridge
Thomas Aregger
VJ
James Christian
Jeff Fang
nothing
sw-pachanady
Matt Nelson
Jungtaek Lim
Manabu Matsuzaki
NersesAM
Stephan Schroevers
Johno Crawford
Arthur Gavlyukovskiy
Andreas Brenk
will vuong
Will Vuong