Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
Scripting HTL Engine 1.4.22-1.4.0
-
None
Description
I see threaddumps which show lock contention in the ScriptDependencyResolver like this:
at jdk.internal.misc.Unsafe.park(java.base@11.0.20/Native Method) - parking to wait for <0x0000000496e16af0> (a java.util.concurrent.locks.ReentrantReadWriteLock$NonfairSync) at java.util.concurrent.locks.LockSupport.park(java.base@11.0.20/LockSupport.java:194) at java.util.concurrent.locks.AbstractQueuedSynchronizer.parkAndCheckInterrupt(java.base@11.0.20/AbstractQueuedSynchronizer.java:885) at java.util.concurrent.locks.AbstractQueuedSynchronizer.acquireQueued(java.base@11.0.20/AbstractQueuedSynchronizer.java:917) at java.util.concurrent.locks.AbstractQueuedSynchronizer.acquire(java.base@11.0.20/AbstractQueuedSynchronizer.java:1240) at java.util.concurrent.locks.ReentrantReadWriteLock$WriteLock.lock(java.base@11.0.20/ReentrantReadWriteLock.java:959) at org.apache.sling.scripting.sightly.impl.utils.ScriptDependencyResolver.resolveScript(ScriptDependencyResolver.java:100) at org.apache.sling.scripting.sightly.impl.engine.extension.use.RenderUnitProvider.provide(RenderUnitProvider.java:95) at org.apache.sling.scripting.sightly.impl.engine.extension.use.UseRuntimeExtension.call(UseRuntimeExtension.java:71) at org.apache.sling.scripting.sightly.impl.engine.runtime.RenderContextImpl.call(RenderContextImpl.java:72) at apps.components.x.y.y__002e__html.render(y__002e__html.java:39)
but also:
at jdk.internal.misc.Unsafe.park(java.base@11.0.20/Native Method) - parking to wait for <0x0000000496e16af0> (a java.util.concurrent.locks.ReentrantReadWriteLock$NonfairSync) at java.util.concurrent.locks.LockSupport.park(java.base@11.0.20/LockSupport.java:194) at java.util.concurrent.locks.AbstractQueuedSynchronizer.parkAndCheckInterrupt(java.base@11.0.20/AbstractQueuedSynchronizer.java:885) at java.util.concurrent.locks.AbstractQueuedSynchronizer.doAcquireShared(java.base@11.0.20/AbstractQueuedSynchronizer.java:1009) at java.util.concurrent.locks.AbstractQueuedSynchronizer.acquireShared(java.base@11.0.20/AbstractQueuedSynchronizer.java:1324) at java.util.concurrent.locks.ReentrantReadWriteLock$ReadLock.lock(java.base@11.0.20/ReentrantReadWriteLock.java:738) at org.apache.sling.scripting.sightly.impl.utils.ScriptDependencyResolver.resolveScript(ScriptDependencyResolver.java:93) at org.apache.sling.scripting.sightly.impl.engine.extension.use.RenderUnitProvider.provide(RenderUnitProvider.java:95) at org.apache.sling.scripting.sightly.impl.engine.extension.use.UseRuntimeExtension.call(UseRuntimeExtension.java:71) at org.apache.sling.scripting.sightly.impl.engine.runtime.RenderContextImpl.call(RenderContextImpl.java:72) at apps.components.a.b.b__002e__html$3.render(b__002e__html.java:218)
I see that the ScriptDependencyResolver holds cache, which just saves positive results where the result has been found; if the result is null, it's not cached, and it will be attempted over and over again.
Of course this situation mostly happens if a lot of these requests with invalid scriptIdentifiers are done, which points to problems in the content. But it would be great if the code behaves a bit better here, because I have a situation here where requests holding the write lock block all other requests which would have a cache hit.
(Thinking further, I don't really understand why a ReadWrite lock is used here at all. As far as I can see a ConcurrentHashMap as a cache should be sufficient.)
Attachments
Issue Links
- relates to
-
SLING-11608 Cache script dependency resolution
- Closed
- links to