cache miss when number of initial concurrent connections > poolSize #50

Open
elarcent opened this Issue May 14, 2012 · 3 comments

Comments

Projects
None yet
2 participants
@elarcent

Playing with ab, I discovered the following problem:
with poolSzie set, for example, as 10, and no pool operating(initial start of application), 'ab -c 11' will yield cache miss on 11th request, with no error message returned. Just 'false' in result. If you start 20 concurrent connection, 10 will get the miss, and so on, basically any number over poolSize of connections will be in trouble. That also will happen if all pooled connections close during normal application run, and start over again.

@3rd-Eden

This comment has been minimized.

Show comment Hide comment
@3rd-Eden

3rd-Eden Sep 14, 2012

Owner

What would you think would be the best way to solve this issue for you. Just wait until a connection is available again?

Owner

3rd-Eden commented Sep 14, 2012

What would you think would be the best way to solve this issue for you. Just wait until a connection is available again?

@elarcent

This comment has been minimized.

Show comment Hide comment
@elarcent

elarcent Dec 6, 2012

Yeah, that should work just fine, there are no many options left how to do it. And then developer will have to find proper poolSize for his environment.

elarcent commented Dec 6, 2012

Yeah, that should work just fine, there are no many options left how to do it. And then developer will have to find proper poolSize for his environment.

@3rd-Eden

This comment has been minimized.

Show comment Hide comment
@3rd-Eden

3rd-Eden Dec 6, 2012

Owner

Is this still occurring? We recently moved over to a different connection pool backend which should have solved this issue.

Owner

3rd-Eden commented Dec 6, 2012

Is this still occurring? We recently moved over to a different connection pool backend which should have solved this issue.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment