Re: FW: tomcat 8080 thread not reduced

[Christopher Schultz]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/22/17 1:37 AM, smith wrote:
> We don't just care about free, we care about active too. We know
> how many free also means how many are active

If you are using a dynamically-sized pool, then you don't really know
how many you have free if you sample active, or how many you have
active if you sample free ones. You have to collect both stats in
order to know whether or not there is a problem.

Of course, the size of the pool can chance between samples of
free/active so you are never getting a perfect picture.

> We are using HTTP

That wasn't clear from any of your previous posts. In that case
keepalive could be an issue. The solution is to use a NIO connector.

> Let's me take one example from manager status: (this is the
> manager output) Max threads: 200 Current thread count: 23 Current
> thread busy: 1 Keeped alive sockets count: 1
> 
> Current thread count: 23 I want to know if these 23 threads are
> free to use, or they are active, cannot accept new connection? Here
> are Keeped alive sockets count, it's only 1, is this 1 for the
> busy thread or a thread in 23?

The current thread count tells you how many java.lang.Thread objects
are being managed by the thread pool. It says nothing about their
state. The 200 is obviously the limit of threads that *would* be
managed by the pool.

I would have to read the code to find out if a keep-alive connection
counts as "busy" or not. But you don't want keepAlive connections to
hold-up a thread, so you should use an NIO connector.

Since you are using Tomcat 8 (albeit an old version) without
specifying a certain type of connector, you are already getting either
the NIO connector or the APR connector (if the native library is
available). Neither of those connectors will hold-up a thread for a
keepalive connection.

I suspect that "busy:1, keepalive:1" for you means that there is
currently 1 thread actually processing a request, and you have 22
threads idle. The keepalive:1 means that one socket connection is in
the keepalive state, but it's not consuming a thread.

- -chris

> -Original Message- From: Christopher Schultz 
> [mailto:ch...@christopherschultz.net] Sent: Friday, January 20,
> 2017 10:29 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080
> thread not reduced
> 
> Smith,
> 
> On 1/19/17 9:59 PM, smith wrote:
>>> "busy" is the same as "active".
> 
>> When not use , our busy thread always keep under 10
>> while the currentThreadCount keeps high (these are get from
>> tomcat manager), So we really don't know how many threads are
>> truly free.
> 
> Why do you care how many are free? Isn't it more important to know 
> how many are active? For example, I run my app servers with 
> maxThreads="150" for the most part. Then, I have Nagios notify me
> if the "active count" goes above 135 (that's 90% of my
> maxThreads).
> 
> Nagios, like many monitoring systems, won't scream the first time
> it sees something. Instead, it will check again a few times before 
> complaining. That means that as long as I don't see any
> significant duration where the active thread count exceeds 135, I
> don't get any alarms going off.
> 
> And I don't care how many "free" threads I have. I've decided that
> I want "150 and NO MORE". Nothing else matters except the active 
> count. I don't happen to allow the thread pool to re-size downward 
> (fewer threads), but if I did, I wouldn't have to change my 
> monitoring at all. What? The active count is 10 and the total pool 
> size is 20? I don't care. Wake me up when the active thread count 
> stays high for a while, indicating to me that we are getting 
> hammered.
> 
>> How many are in keepAlivedStatus
> 
> Use NIO and forget about it.
> 
> Besides, you are using *AJP*. Every thread is *always* in a
> keepalive state. So keepalive == idle as far as Tomcat is
> concerned.
> 
> If you were using HTTP, then keepalive would be an issue (and NIO 
> would be the answer), but you aren't, so it's not.
> 
> -chris
> 
>> -Original Message- From: Christopher Schultz 
>> [mailto:ch...@christopherschultz.net] Sent: Thursday, January 19,
>>  2017 4:38 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080 
>> thread not reduced
> 
>> Smith,
> 
>> On 1/18/17 8:25 PM, smith wrote:
>>> I don't care if the threads will be reduced, I just want to
>>> know why.
> 
>> Okay.
> 
>>> And we want to use the account to determine when the tomcat 
>>> capacity is not enough that we need to add max configuration
>>> or add new tomcat servers.
> 
>> Set your initial and max threads to the same v

Re: FW: tomcat 8080 thread not reduced

[tomcat]

On 22.01.2017 07:37, smith wrote:

Hi, Chris

We don't just care about free, we care about active too. We know how many free 
also means how many are active

We are using HTTP

Let's me take one example from manager status: (this is the manager output)
Max threads: 200 Current thread count: 23 Current thread busy: 1 Keeped alive 
sockets count: 1

Current thread count: 23 I want to know if these 23 threads are free to use, or 
they are active, cannot accept new connection? Here are Keeped alive sockets 
count, it's only 1, is this 1 for the busy thread or a thread in 23?



Max threads : 200
-
That is the maximum number of threads that tomcat will ever allocate, no matter how busy 
your server becomes.


Current thread count : 23
-
That is how many threads have been allocated and exist so far. If you do not use an 
Executor, then this number will never go down again (but it may go up). Included in this 
count are :

a) the threads currently busy processing a request
b) the threads which may not be actually executing a request, but are blocked on a 
keep-alive connection, waiting if there is another request coming on that same client 
connection. These threads are not available for a new client connection, but they are 
available to process a new request on the same client connection for which they have 
already processed at least one request recently.
c) the threads which are not doing anything, and which are available to process new 
connections and new requests on such new connections


Current thread busy: 1
--
This the thread (a) above.  It is in fact the thread which is processing your "manager" 
request.


Keeped alive sockets count: 1
-
This a thread in status (b) above. It is not the same thread as (a).
It is not available for a new client connection, because it is still waiting on its 
current client connection. But it is available for a new request *on the same connection*, 
for the remainder of its keep-alive timeout.
If there is no no new request on that same connection, then when the keep-alive timeout 
expires, it will close the connection and go to status (c).
If there is a new request on the same connection, it will go to status (a) and process 
that request; and when it is finished, the timeout will be reset to the keep-alive timeout 
and it will return to status (b).


So you have :
total threads :   23
- busy threads :  -1
- keepalive threads : -1

= available for new connections : 21

From time to time, tomcat will look at how many threads are in status (c) above.  It if 
it not at least minSpareThreads, additional threads will be created until there are 
minSpareThreads in status (c), with a maximum of maxThreads threads in total.



If you server was receiving now suddenly 50 new requests from 50 new clients at the same 
time, then :
- 21 requests would be processed by the 21 threads currently in status (c) (which will go 
to status (a)
- tomcat would create 29 new threads immediately, to process the other 29 requests. These 
threads would start in status (a), and when their request is finished, they would go to 
status (b) for the keepAlive timeout
- possibly, the background watchdog would run at that same moment and see that there are 
no longer minSpareThreads available in status (c). It would then create new threads in 
status (c), to bring this back to minSpareThreads.
At the end of this, you would now have (possibly) 60 threads in total, forever, which will 
slowly go back all to status (c) if no more requests are coming.



If you want another behaviour, then you should use an Executor.




-Smith

-Original Message-
From: Christopher Schultz [mailto:ch...@christopherschultz.net]
Sent: Friday, January 20, 2017 10:29 PM
To: Tomcat Users List
Subject: Re: FW: tomcat 8080 thread not reduced

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/19/17 9:59 PM, smith wrote:

"busy" is the same as "active".


When not use , our busy thread always keep under 10 while
the currentThreadCount keeps high (these are get from tomcat manager),
So we really don't know how many threads are truly free.


Why do you care how many are free? Isn't it more important to know how many are active? For 
example, I run my app servers with maxThreads="150" for the most part. Then, I have 
Nagios notify me if the "active count" goes above 135 (that's 90% of my maxThreads).

Nagios, like many monitoring systems, won't scream the first time it sees 
something. Instead, it will check again a few times before complaining. That 
means that as long as I don't see any significant duration where the active 
thread count exceeds 135, I don't get any alarms going off.

And I don't care how many "free" threads I have. I've decided that I want "150 and 
NO MORE". Nothing else matters except the active count.
I do

RE: FW: tomcat 8080 thread not reduced

[smith]

Hi, Chris

We don't just care about free, we care about active too. We know how many free 
also means how many are active

We are using HTTP

Let's me take one example from manager status: (this is the manager output)
Max threads: 200 Current thread count: 23 Current thread busy: 1 Keeped alive 
sockets count: 1

Current thread count: 23 I want to know if these 23 threads are free to use, or 
they are active, cannot accept new connection? Here are Keeped alive sockets 
count, it's only 1, is this 1 for the busy thread or a thread in 23?

-Smith

-Original Message-
From: Christopher Schultz [mailto:ch...@christopherschultz.net] 
Sent: Friday, January 20, 2017 10:29 PM
To: Tomcat Users List
Subject: Re: FW: tomcat 8080 thread not reduced

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/19/17 9:59 PM, smith wrote:
>> "busy" is the same as "active".
> 
> When not use , our busy thread always keep under 10 while 
> the currentThreadCount keeps high (these are get from tomcat manager), 
> So we really don't know how many threads are truly free.

Why do you care how many are free? Isn't it more important to know how many are 
active? For example, I run my app servers with maxThreads="150" for the most 
part. Then, I have Nagios notify me if the "active count" goes above 135 
(that's 90% of my maxThreads).

Nagios, like many monitoring systems, won't scream the first time it sees 
something. Instead, it will check again a few times before complaining. That 
means that as long as I don't see any significant duration where the active 
thread count exceeds 135, I don't get any alarms going off.

And I don't care how many "free" threads I have. I've decided that I want "150 
and NO MORE". Nothing else matters except the active count.
I don't happen to allow the thread pool to re-size downward (fewer threads), 
but if I did, I wouldn't have to change my monitoring at all. What? The active 
count is 10 and the total pool size is 20? I don't care. Wake me up when the 
active thread count stays high for a while, indicating to me that we are 
getting hammered.

> How many are in keepAlivedStatus

Use NIO and forget about it.

Besides, you are using *AJP*. Every thread is *always* in a keepalive state. So 
keepalive == idle as far as Tomcat is concerned.

If you were using HTTP, then keepalive would be an issue (and NIO would be the 
answer), but you aren't, so it's not.

- -chris

> -Original Message- From: Christopher Schultz 
> [mailto:ch...@christopherschultz.net] Sent: Thursday, January 19,
> 2017 4:38 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080 thread 
> not reduced
> 
> Smith,
> 
> On 1/18/17 8:25 PM, smith wrote:
>> I don't care if the threads will be reduced, I just want to know why.
> 
> Okay.
> 
>> And we want to use the account to determine when the tomcat capacity 
>> is not enough that we need to add max configuration or add new tomcat 
>> servers.
> 
> Set your initial and max threads to the same value (pool size =
> constant) and then monitor the "active count" with a Nagios warning at 
> e.g. 80% usage.
> 
>> Since not use the , the busy thread account also cannot 
>> tell us the correct active threads count.
> 
> "busy" is the same as "active".
> 
>> In another email thread, you said if use , it will tell us 
>> the right active thread count not just busy count, right?
> 
> I would always use an  for at least two different
> reasons:
> 
> 1. Thread management (e.g. reducing threads if necessary) 2. Shared 
> thread-pools (no need to have port 8080 and 8443 with separate
> pools)
> 
> -chris
> 
>> -Original Message- From: Christopher Schultz 
>> [mailto:ch...@christopherschultz.net] Sent: Wednesday, January 18, 
>> 2017 3:28 PM To: Tomcat Users List Subject: Re: FW: tomcat
>> 8080 thread not reduced
> 
>> Smith,
> 
>> On 1/18/17 12:47 AM, smith wrote:
>>> So the tomcat default executor will not reduce the thread count 
>>> until it reach to the max configuration?
> 
>> By default, you get a thread pool that isn't as smart as an executor.
> 
>>> Will it reduce when it reach to max?
> 
>> Not unless you use an .
> 
>>> And why the default not reduce the thread?
> 
>> Because it didn't do so in the past, before  was 
>> introduced.
> 
>> I'm curious: if you are willing to have e.g. 200 threads available at 
>> any time during the life of the JVM, why does it matter if those 
>> threads are reduced during times of inactivity?
> 
>> I think of threads as a resource like memory, where if you are going 
>> to allocate X resources, you may as well allocate X resources an

Re: FW: tomcat 8080 thread not reduced

[Christopher Schultz]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/19/17 9:59 PM, smith wrote:
>> "busy" is the same as "active".
> 
> When not use , our busy thread always keep under 10
> while the currentThreadCount keeps high (these are get from tomcat
> manager), So we really don't know how many threads are truly free.

Why do you care how many are free? Isn't it more important to know how
many are active? For example, I run my app servers with
maxThreads="150" for the most part. Then, I have Nagios notify me if
the "active count" goes above 135 (that's 90% of my maxThreads).

Nagios, like many monitoring systems, won't scream the first time it
sees something. Instead, it will check again a few times before
complaining. That means that as long as I don't see any significant
duration where the active thread count exceeds 135, I don't get any
alarms going off.

And I don't care how many "free" threads I have. I've decided that I
want "150 and NO MORE". Nothing else matters except the active count.
I don't happen to allow the thread pool to re-size downward (fewer
threads), but if I did, I wouldn't have to change my monitoring at
all. What? The active count is 10 and the total pool size is 20? I
don't care. Wake me up when the active thread count stays high for a
while, indicating to me that we are getting hammered.

> How many are in keepAlivedStatus

Use NIO and forget about it.

Besides, you are using *AJP*. Every thread is *always* in a keepalive
state. So keepalive == idle as far as Tomcat is concerned.

If you were using HTTP, then keepalive would be an issue (and NIO
would be the answer), but you aren't, so it's not.

- -chris

> -Original Message- From: Christopher Schultz
> [mailto:ch...@christopherschultz.net] Sent: Thursday, January 19,
> 2017 4:38 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080
> thread not reduced
> 
> Smith,
> 
> On 1/18/17 8:25 PM, smith wrote:
>> I don't care if the threads will be reduced, I just want to know
>> why.
> 
> Okay.
> 
>> And we want to use the account to determine when the tomcat
>> capacity is not enough that we need to add max configuration or
>> add new tomcat servers.
> 
> Set your initial and max threads to the same value (pool size = 
> constant) and then monitor the "active count" with a Nagios warning
> at e.g. 80% usage.
> 
>> Since not use the , the busy thread account also cannot
>> tell us the correct active threads count.
> 
> "busy" is the same as "active".
> 
>> In another email thread, you said if use , it will tell
>> us the right active thread count not just busy count, right?
> 
> I would always use an  for at least two different
> reasons:
> 
> 1. Thread management (e.g. reducing threads if necessary) 2. Shared
> thread-pools (no need to have port 8080 and 8443 with separate
> pools)
> 
> -chris
> 
>> -Original Message- From: Christopher Schultz 
>> [mailto:ch...@christopherschultz.net] Sent: Wednesday, January
>> 18, 2017 3:28 PM To: Tomcat Users List Subject: Re: FW: tomcat
>> 8080 thread not reduced
> 
>> Smith,
> 
>> On 1/18/17 12:47 AM, smith wrote:
>>> So the tomcat default executor will not reduce the thread count
>>> until it reach to the max configuration?
> 
>> By default, you get a thread pool that isn't as smart as an
>> executor.
> 
>>> Will it reduce when it reach to max?
> 
>> Not unless you use an .
> 
>>> And why the default not reduce the thread?
> 
>> Because it didn't do so in the past, before  was
>> introduced.
> 
>> I'm curious: if you are willing to have e.g. 200 threads
>> available at any time during the life of the JVM, why does it
>> matter if those threads are reduced during times of inactivity?
> 
>> I think of threads as a resource like memory, where if you are
>> going to allocate X resources, you may as well allocate X
>> resources and be done with it. Growing and shrinking pools of
>> things just adds complexity and reduces performance.
> 
>> Idle threads are "free" other than using a little bit of memory. 
>> So why is it so important for those threads to stop when they
>> don't have any work for a while?
> 
>> -chris
> 
>>> -Original Message- From: Christopher Schultz 
>>> [mailto:ch...@christopherschultz.net] Sent: Tuesday, January
>>> 17, 2017 7:18 PM To: Tomcat Users List Subject: Re: FW: tomcat
>>> 8080 thread not reduced
> 
>>> Smith,
> 
>>> On 1/16/17 8:22 PM, smith wrote:
>>>> Yes, I think thread count should be reduced when those

Re: FW: tomcat 8080 thread not reduced

[tomcat]

On 20.01.2017 03:59, smith wrote:

Hi, chris:


"busy" is the same as "active".

When not use , our busy thread always keep under 10 while the 
currentThreadCount keeps high (these are get from tomcat manager), So we really don't 
know how many threads are truly free. How many are in keepAlivedStatus



That may be my fault, in my contributions earlier.
I believe that it is reasonable to assume that "busy" also counts the threads which are 
currently "in keepAlive" status. That is because these threads are not "free" to handle 
new requests yet from other clients than the one which "owns" the connection on which they 
are in keepAlive.

But I do not know the code, so someone else may want to confirm/correct this.

Anyway, you do have a simple way to verify this : reduce the keepAlive to 5 seconds, 
instead of 20, and see how it impacts the counts that you are seeing.

If keepAlive counts as "busy", then you should see less "busy" threads on 
average.
If keepAlive does not count as busy, then you will see no difference.

One more proviso : the above is for the case where clients connect to tomcat directly, 
without a front-end, which may itself create a pool of permanent connections to tomcat.

(Chris already mentioned that).
If there is such a front-end, and it does something like "pings" on each connection to the 
tomcat back-end connections all the time, then you would have, on the tomcat side, an 
alsmost constant number of threads in keepAlive wait status.


All in all, considering your ultimate purpose (to determine when your tomcat is getting 
close to the limit of what it can handle, and decide if you need more tomcats or not), I 
believe that your current focus on the number of tomcat threads active or not, may not be 
the best diagnostic tool.


Maybe a better tool would be to look at the number of connections which are waiting to be 
accepted. This should always be 0, or small.  If it starts growing, then that is a clear 
signal that tomcat is getting saturated. (This is related to the "acceptCount" attribute 
of the Connectors).


Another tool may be to look at how long it takes tomcat to respond to requests (or at 
least some specific types of requests).  You can probably set a baseline, when you know 
that tomcat is really not heavily loaded (maybe on your development system e.g.).  And 
then, when you see this response time increasing seriously, is when you should start 
analysing why. (*)


This kind of approach would have the advantage of focusing on what really matters in the 
end : the ability of your server (as a whole : hardware, OS, tomcat and applications 
included) to handle a given number of requests within a given time interval, with /your/ 
mix of clients and applications.



(*) because the reason may be for example that you need more memory, not more 
threads.



-Original Message-
From: Christopher Schultz [mailto:ch...@christopherschultz.net]
Sent: Thursday, January 19, 2017 4:38 PM
To: Tomcat Users List
Subject: Re: FW: tomcat 8080 thread not reduced

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/18/17 8:25 PM, smith wrote:

I don't care if the threads will be reduced, I just want to know why.


Okay.


And we want to use the account to determine when the tomcat capacity
is not enough that we need to add max configuration or add new tomcat
servers.


Set your initial and max threads to the same value (pool size =
constant) and then monitor the "active count" with a Nagios warning at e.g. 80% 
usage.


Since not use the , the busy thread account also cannot tell
us the correct active threads count.


"busy" is the same as "active".


In another email thread, you said if use , it will tell us
the right active thread count not just busy count, right?


I would always use an  for at least two different reasons:

1. Thread management (e.g. reducing threads if necessary) 2. Shared 
thread-pools (no need to have port 8080 and 8443 with separate pools)

- -chris


-Original Message- From: Christopher Schultz
[mailto:ch...@christopherschultz.net] Sent: Wednesday, January 18,
2017 3:28 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080 thread
not reduced

Smith,

On 1/18/17 12:47 AM, smith wrote:

So the tomcat default executor will not reduce the thread count until
it reach to the max configuration?


By default, you get a thread pool that isn't as smart as an executor.


Will it reduce when it reach to max?


Not unless you use an .


And why the default not reduce the thread?


Because it didn't do so in the past, before  was introduced.

I'm curious: if you are willing to have e.g. 200 threads available at
any time during the life of the JVM, why does it matter if those
threads are reduced during times of inactivity?

I think of threads as a resource like memory, where if you are going
to allocate X resources

RE: FW: tomcat 8080 thread not reduced

[smith]

Hi, chris:

>"busy" is the same as "active".
When not use , our busy thread always keep under 10 while the 
currentThreadCount keeps high (these are get from tomcat manager), So we really 
don't know how many threads are truly free. How many are in keepAlivedStatus


-Original Message-
From: Christopher Schultz [mailto:ch...@christopherschultz.net] 
Sent: Thursday, January 19, 2017 4:38 PM
To: Tomcat Users List
Subject: Re: FW: tomcat 8080 thread not reduced

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/18/17 8:25 PM, smith wrote:
> I don't care if the threads will be reduced, I just want to know why.

Okay.

> And we want to use the account to determine when the tomcat capacity 
> is not enough that we need to add max configuration or add new tomcat 
> servers.

Set your initial and max threads to the same value (pool size =
constant) and then monitor the "active count" with a Nagios warning at e.g. 80% 
usage.

> Since not use the , the busy thread account also cannot tell 
> us the correct active threads count.

"busy" is the same as "active".

> In another email thread, you said if use , it will tell us 
> the right active thread count not just busy count, right?

I would always use an  for at least two different reasons:

1. Thread management (e.g. reducing threads if necessary) 2. Shared 
thread-pools (no need to have port 8080 and 8443 with separate pools)

- -chris

> -Original Message- From: Christopher Schultz 
> [mailto:ch...@christopherschultz.net] Sent: Wednesday, January 18,
> 2017 3:28 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080 thread 
> not reduced
> 
> Smith,
> 
> On 1/18/17 12:47 AM, smith wrote:
>> So the tomcat default executor will not reduce the thread count until 
>> it reach to the max configuration?
> 
> By default, you get a thread pool that isn't as smart as an executor.
> 
>> Will it reduce when it reach to max?
> 
> Not unless you use an .
> 
>> And why the default not reduce the thread?
> 
> Because it didn't do so in the past, before  was introduced.
> 
> I'm curious: if you are willing to have e.g. 200 threads available at 
> any time during the life of the JVM, why does it matter if those 
> threads are reduced during times of inactivity?
> 
> I think of threads as a resource like memory, where if you are going 
> to allocate X resources, you may as well allocate X resources and be 
> done with it. Growing and shrinking pools of things just adds 
> complexity and reduces performance.
> 
> Idle threads are "free" other than using a little bit of memory.
> So why is it so important for those threads to stop when they don't 
> have any work for a while?
> 
> -chris
> 
>> -Original Message- From: Christopher Schultz 
>> [mailto:ch...@christopherschultz.net] Sent: Tuesday, January 17,
>>  2017 7:18 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080 
>> thread not reduced
> 
>> Smith,
> 
>> On 1/16/17 8:22 PM, smith wrote:
>>> Yes, I think thread count should be reduced when those threads are 
>>> idle
> 
>>> Is this right? Or it will not reduced?
> 
>> Id you want Tomcat to reduce the number of idle threads, you'll need 
>> to explicitly configure an  and use that with your  
>>  .
> 
>> -chris
> 
>>> -Original Message- From: Christopher Schultz 
>>> [mailto:ch...@christopherschultz.net] Sent: Monday, January 16,
>>>  2017 2:20 PM To: Tomcat Users List Subject: Re: FW: tomcat
>>> 8080 thread not reduced
> 
>>> Smith,
> 
>>> There are your only active s:
> 
>>> On 1/14/17 1:30 AM, smith wrote:
>>>> >>> connectionTimeout="2" redirectPort="8443" />
> 
>>>> [snip]
> 
>>>> >>>  />
> 
>>> You have not changed any settings from the default. What makes you 
>>> think that your thread count should be reduced when those threads 
>>> are idle?
> 
>>> -chris
> 
>>> 
- -
>
>>>
>>> 
>>> 
> 
>> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
>>> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> 
>>> 
- -
>
>>>
>>> 
>>> 
> 
>> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
>>> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
>

Re: FW: tomcat 8080 thread not reduced

[Christopher Schultz]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/18/17 8:25 PM, smith wrote:
> I don't care if the threads will be reduced, I just want to know
> why.

Okay.

> And we want to use the account to determine when the tomcat
> capacity is not enough that we need to add max configuration or add
> new tomcat servers.

Set your initial and max threads to the same value (pool size =
constant) and then monitor the "active count" with a Nagios warning at
e.g. 80% usage.

> Since not use the , the busy thread account also cannot 
> tell us the correct active threads count.

"busy" is the same as "active".

> In another email thread, you said if use , it will tell
> us the right active thread count not just busy count, right?

I would always use an  for at least two different reasons:

1. Thread management (e.g. reducing threads if necessary)
2. Shared thread-pools (no need to have port 8080 and 8443 with
separate pools)

- -chris

> -Original Message- From: Christopher Schultz 
> [mailto:ch...@christopherschultz.net] Sent: Wednesday, January 18, 
> 2017 3:28 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080 
> thread not reduced
> 
> Smith,
> 
> On 1/18/17 12:47 AM, smith wrote:
>> So the tomcat default executor will not reduce the thread count 
>> until it reach to the max configuration?
> 
> By default, you get a thread pool that isn't as smart as an 
> executor.
> 
>> Will it reduce when it reach to max?
> 
> Not unless you use an .
> 
>> And why the default not reduce the thread?
> 
> Because it didn't do so in the past, before  was 
> introduced.
> 
> I'm curious: if you are willing to have e.g. 200 threads available
> at any time during the life of the JVM, why does it matter if
> those threads are reduced during times of inactivity?
> 
> I think of threads as a resource like memory, where if you are
> going to allocate X resources, you may as well allocate X resources
> and be done with it. Growing and shrinking pools of things just
> adds complexity and reduces performance.
> 
> Idle threads are "free" other than using a little bit of memory.
> So why is it so important for those threads to stop when they don't
> have any work for a while?
> 
> -chris
> 
>> -Original Message- From: Christopher Schultz 
>> [mailto:ch...@christopherschultz.net] Sent: Tuesday, January 17,
>>  2017 7:18 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080 
>> thread not reduced
> 
>> Smith,
> 
>> On 1/16/17 8:22 PM, smith wrote:
>>> Yes, I think thread count should be reduced when those threads 
>>> are idle
> 
>>> Is this right? Or it will not reduced?
> 
>> Id you want Tomcat to reduce the number of idle threads, you'll 
>> need to explicitly configure an  and use that with your
>>   .
> 
>> -chris
> 
>>> -Original Message- From: Christopher Schultz 
>>> [mailto:ch...@christopherschultz.net] Sent: Monday, January 16,
>>>  2017 2:20 PM To: Tomcat Users List Subject: Re: FW: tomcat
>>> 8080 thread not reduced
> 
>>> Smith,
> 
>>> There are your only active s:
> 
>>> On 1/14/17 1:30 AM, smith wrote:
>>>> >>> connectionTimeout="2" redirectPort="8443" />
> 
>>>> [snip]
> 
>>>> >>>  />
> 
>>> You have not changed any settings from the default. What makes 
>>> you think that your thread count should be reduced when those 
>>> threads are idle?
> 
>>> -chris
> 
>>> 
- -
>
>>>
>>> 
>>> 
> 
>> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
>>> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> 
>>> 
- -
>
>>>
>>> 
>>> 
> 
>> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
>>> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
>> -
>
>>
>> 
> 
> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
>> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> 
>> -
>
>>
>> 
> 
> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
>> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> --

RE: FW: tomcat 8080 thread not reduced

[smith]

Hi, Christopher

I don't care if the threads will be reduced, I just want to know why. And we 
want to use the account to determine when the tomcat capacity is not enough 
that we need to add max configuration or add new tomcat servers.

Since not use the , the busy thread account also cannot tell us the 
correct active threads count. In another email thread, you said if use 
, it will tell us the right active thread count not just busy count, 
right?

-smith

-Original Message-
From: Christopher Schultz [mailto:ch...@christopherschultz.net] 
Sent: Wednesday, January 18, 2017 3:28 PM
To: Tomcat Users List
Subject: Re: FW: tomcat 8080 thread not reduced

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/18/17 12:47 AM, smith wrote:
> So the tomcat default executor will not reduce the thread count until 
> it reach to the max configuration?

By default, you get a thread pool that isn't as smart as an executor.

> Will it reduce when it reach to max?

Not unless you use an .

> And why the default not reduce the thread?

Because it didn't do so in the past, before  was introduced.

I'm curious: if you are willing to have e.g. 200 threads available at any time 
during the life of the JVM, why does it matter if those threads are reduced 
during times of inactivity?

I think of threads as a resource like memory, where if you are going to 
allocate X resources, you may as well allocate X resources and be done with it. 
Growing and shrinking pools of things just adds complexity and reduces 
performance.

Idle threads are "free" other than using a little bit of memory. So why is it 
so important for those threads to stop when they don't have any work for a 
while?

- -chris

> -Original Message- From: Christopher Schultz 
> [mailto:ch...@christopherschultz.net] Sent: Tuesday, January 17,
> 2017 7:18 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080 thread 
> not reduced
> 
> Smith,
> 
> On 1/16/17 8:22 PM, smith wrote:
>> Yes, I think thread count should be reduced when those threads are 
>> idle
> 
>> Is this right? Or it will not reduced?
> 
> Id you want Tomcat to reduce the number of idle threads, you'll need 
> to explicitly configure an  and use that with your 
>  .
> 
> -chris
> 
>> -Original Message- From: Christopher Schultz 
>> [mailto:ch...@christopherschultz.net] Sent: Monday, January 16,
>> 2017 2:20 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080 
>> thread not reduced
> 
>> Smith,
> 
>> There are your only active s:
> 
>> On 1/14/17 1:30 AM, smith wrote:
>>> >> connectionTimeout="2" redirectPort="8443" />
> 
>>> [snip]
> 
>>> >> />
> 
>> You have not changed any settings from the default. What makes you 
>> think that your thread count should be reduced when those threads are 
>> idle?
> 
>> -chris
> 
>> -
>
>> 
> 
> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
>> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> 
>> -
>
>> 
> 
> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
>> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> -
>
> 
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> 
> -
>
> 
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
-BEGIN PGP SIGNATURE-
Comment: GPGTools - http://gpgtools.org
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=ks8D
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org



-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: FW: tomcat 8080 thread not reduced

[Christopher Schultz]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/18/17 12:47 AM, smith wrote:
> So the tomcat default executor will not reduce the thread count 
> until it reach to the max configuration?

By default, you get a thread pool that isn't as smart as an executor.

> Will it reduce when it reach to max?

Not unless you use an .

> And why the default not reduce the thread?

Because it didn't do so in the past, before  was introduced.

I'm curious: if you are willing to have e.g. 200 threads available at
any time during the life of the JVM, why does it matter if those
threads are reduced during times of inactivity?

I think of threads as a resource like memory, where if you are going
to allocate X resources, you may as well allocate X resources and be
done with it. Growing and shrinking pools of things just adds
complexity and reduces performance.

Idle threads are "free" other than using a little bit of memory. So
why is it so important for those threads to stop when they don't have
any work for a while?

- -chris

> -Original Message- From: Christopher Schultz
> [mailto:ch...@christopherschultz.net] Sent: Tuesday, January 17,
> 2017 7:18 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080
> thread not reduced
> 
> Smith,
> 
> On 1/16/17 8:22 PM, smith wrote:
>> Yes, I think thread count should be reduced when those threads
>> are idle
> 
>> Is this right? Or it will not reduced?
> 
> Id you want Tomcat to reduce the number of idle threads, you'll
> need to explicitly configure an  and use that with your
>  .
> 
> -chris
> 
>> -Original Message- From: Christopher Schultz 
>> [mailto:ch...@christopherschultz.net] Sent: Monday, January 16, 
>> 2017 2:20 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080
>> thread not reduced
> 
>> Smith,
> 
>> There are your only active s:
> 
>> On 1/14/17 1:30 AM, smith wrote:
>>> >> connectionTimeout="2" redirectPort="8443" />
> 
>>> [snip]
> 
>>> >> />
> 
>> You have not changed any settings from the default. What makes
>> you think that your thread count should be reduced when those
>> threads are idle?
> 
>> -chris
> 
>> -
>
>> 
> 
> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
>> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> 
>> -
>
>> 
> 
> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
>> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> -
>
> 
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> 
> -
>
> 
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
-BEGIN PGP SIGNATURE-
Comment: GPGTools - http://gpgtools.org
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=ks8D
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: FW: tomcat 8080 thread not reduced

[Christopher Schultz]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/18/17 7:42 AM, Smith Hua wrote:
> our monitor is getting the result from the tomcat manager page
> which provides the busy threads count and totol threads count, i'm
> not sure if tomcat manager treat the keepAlived thread as free.

If you are using AJP as your protocol, then keepalive is pretty much
irrelevant, since the threads used on the Tomcat side are essentially
permanently bound to a connection coming from the reverse proxy. Those
connections aren't allocated to individual clients from the Internet;
they are aggregated and sent-over (mostly) permanent connections from
the reverse proxy to Tomcat.

> if the tomcat manager treat the keepAlived thread as free, seems we
> can't kown what's the concurrent work threads, when we need to add
> the max threads or add more tomcat servers. shall we use current
> total threads to judge this? if it reaches to max configured
> threads, need to add max threads or add tomcat servers?

The data point you actually want to look at is called "activeCount" on
the thread pool. If you aren't using an , then you should be
using one.

See
http://people.apache.org/~schultz/ApacheCon%20NA%202016/Monitoring%20Apa
che%20Tomcat%20with%20JMX.pdf
starting on slide 17.

- -chris

> -- 从myMail的Android专用app所发送 星期三, 18 一月 2017, 06:08下午 +08:00 发件人
> André Warnier (tomcat)  a...@ice-sa.com :
> 
>> On 18.01.2017 10:56, smith wrote:
>>> Hi, André
>>> 
>>> Thanks for the great explanation. So the current thread count 
>>> will grow always until it reaches to the max configuration.
>>> 
>>> I have another strange thing: We never monitored tomcat busy 
>>> thread count high (we monitored one minutes interval through 
>>> Nagios to get tomcat manager result, not high than 10). So if
>>> the threads grows, it should monitored the busy thread count
>>> reach that high(or some lower). So I'm still not understand
>>> when the current thread count will grow while current busy
>>> thread has not change a lot.
>>> 
>> 
>> I think that if you combine my explanation of the KeepAlive
>> effect, with Philippe's notes, you may get the explanation. I
>> would personally guess that your monitoring agent does not
>> consider a thread that is idle in KeepAlive state, as "busy".  So
>> it only counts as busy, the threads which are effectively in the
>> state of processing a request. (This is just a guess; I think you
>> may want to draw a little time-dependent diagram, to see what
>> really happens, with a KeepAlive of 20 seconds, and ping or
>> heartbeat requests coming in regularly.) In any case, if you want
>> to avoid this phenomenon altogether, try configuring an Executor.
>> You need to configure the  itself, and then in your
>> , refer to that Executor).
>> 
>> 
>> 
>> 
>>> 
>>> 
>>> -Original Message- From: André Warnier (tomcat) 
>>> [mailto:a...@ice-sa.com] Sent: Wednesday, January 18, 2017 9:41
>>> AM To:  users@tomcat.apache.org Subject: Re: FW: tomcat 8080
>>> thread not reduced
>>> 
>>> Hi.
>>> 
>>> I believe that what Philippe mentions below is somewhat
>>> different : in his configuration, there is apparently a
>>> front-end httpd server, which communicates with Tomcat via AJP
>>> and the tomcat AJP Connector. In such a case, the "mod_jk" or
>>> "mod_proxy_ajp" connector module inside of httpd, creates a
>>> series of connections to the back-end tomcat, and keeps these
>>> connections alive in a pool of connections, according to rules
>>> which are different, than what applies to your case (which
>>> consists of direct HTTP connections from browsers, to the
>>> tomcat HTTP connector). So Philippe's case introduces probably
>>> an additional level of complexity into the equation.
>>> semi-graphically :
>>> 
>>> smith case :
>>> 
>>> client <- HTTP -> tomcat HTTP connector - tomcat thread
>>> 
>>> Philippe's case :
>>> 
>>> client <- HTTP -> apache httpd + ajp module <- AJP -> tomcat
>>> AJP connector - tomcat thread
>>> 
>>> The real question is :
>>> 
>>> The tomcat 8.0 HTTP Connector documentation ( 
>>> http://tomcat.apache.org/tomcat-8.0-doc/config/http.html ) 
>>> mentions a series of attributes/parameters (maxThreads, 
>>> minSpareThreads, connectionTimeout, keepAliveTimeout,..)
>>> which, together, would lead one to believe that tomcat is
>>> managing the numbe

Re[2]: FW: tomcat 8080 thread not reduced

[Smith Hua]

thanks, 
our monitor is getting the result from the tomcat manager page which provides 
the busy threads count and totol threads count, i'm not sure if tomcat manager 
treat the keepAlived thread as free.
if the tomcat manager treat the keepAlived thread as free, seems we can't kown 
what's the concurrent work threads, when we need to add the max threads or add 
more tomcat servers.
shall we use current total threads to judge this? if it reaches to max 
configured threads, need to add max threads or add tomcat servers?
--
从myMail的Android专用app所发送 星期三, 18 一月 2017, 06:08下午 +08:00 发件人 André Warnier 
(tomcat)  a...@ice-sa.com :

>On 18.01.2017 10:56, smith wrote:
>> Hi, André
>>
>> Thanks for the great explanation. So the current thread count will grow 
>> always until it reaches to the max configuration.
>>
>> I have another strange thing:
>> We never monitored tomcat busy thread count high (we monitored one minutes 
>> interval through Nagios to get tomcat manager result, not high than 10). So 
>> if the threads grows, it should monitored the busy thread count reach that 
>> high(or some lower). So I'm still not understand when the current thread 
>> count will grow while current busy thread has not change a lot.
>>
>
>I think that if you combine my explanation of the KeepAlive effect, with 
>Philippe's notes, 
>you may get the explanation.
>I would personally guess that your monitoring agent does not consider a thread 
>that is 
>idle in KeepAlive state, as "busy".  So it only counts as busy, the threads 
>which are 
>effectively in the state of processing a request.
>(This is just a guess; I think you may want to draw a little time-dependent 
>diagram, to 
>see what really happens, with a KeepAlive of 20 seconds, and ping or heartbeat 
>requests 
>coming in regularly.)
>In any case, if you want to avoid this phenomenon altogether, try configuring 
>an Executor. 
>You need to configure the  itself, and then in your , 
>refer to that 
>Executor).
>
>
>
>
>>
>>
>> -Original Message-
>> From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
>> Sent: Wednesday, January 18, 2017 9:41 AM
>> To:  users@tomcat.apache.org
>> Subject: Re: FW: tomcat 8080 thread not reduced
>>
>> Hi.
>>
>> I believe that what Philippe mentions below is somewhat different : in his 
>> configuration, there is apparently a front-end httpd server, which 
>> communicates with Tomcat via AJP and the tomcat AJP Connector.
>> In such a case, the "mod_jk" or "mod_proxy_ajp" connector module inside of 
>> httpd, creates a series of connections to the back-end tomcat, and keeps 
>> these connections alive in a pool of connections, according to rules which 
>> are different, than what applies to your case (which consists of direct HTTP 
>> connections from browsers, to the tomcat HTTP connector).
>> So Philippe's case introduces probably an additional level of complexity 
>> into the equation.
>> semi-graphically :
>>
>> smith case :
>>
>> client <- HTTP -> tomcat HTTP connector - tomcat thread
>>
>> Philippe's case :
>>
>> client <- HTTP -> apache httpd + ajp module <- AJP -> tomcat AJP connector - 
>> tomcat thread
>>
>> The real question is :
>>
>> The tomcat 8.0 HTTP Connector documentation
>> ( http://tomcat.apache.org/tomcat-8.0-doc/config/http.html ) mentions a 
>> series of attributes/parameters (maxThreads, minSpareThreads, 
>> connectionTimeout,
>> keepAliveTimeout,..) which, together, would lead one to believe that tomcat 
>> is managing the number of threads dynamically (increasing the number of 
>> threads (up to maxThreads) when needed to process simultaneous requests, and 
>> decreasing the number of threads again (down to minSpareThreads) when less 
>> requests are being processed).
>>
>> According to observations however (and to Chris' comments), that does not 
>> seem to be the case : once the number of threads has grown to a certain 
>> number, it never goes down back down again.
>> It has to be said that the above interpretation was mine, and the current 
>> documentation of the HTTP Connector never says explicitly that, on its own, 
>> it manages the number of threads dynamically. But it *does* mention a 
>> minSpareThreads parameter.
>>
>> On the other hand, when you use an Executor 
>> (http://tomcat.apache.org/tomcat-8.0-doc/config/executor.html), then the 
>> documentation states explicitly that the number of threads *is* dynamically 
>> managed, up and down.
>>
>> I found something else 

Re[2]: FW: tomcat 8080 thread not reduced

[Smith Hua]

thanks, our monitor is getting the result from the tomcat manager page which 
provides the busy threads count and totol threads count
--
从myMail的Android专用app所发送 星期三, 18 一月 2017, 06:08下午 +08:00 发件人 André Warnier 
(tomcat)  a...@ice-sa.com :

>On 18.01.2017 10:56, smith wrote:
>> Hi, André
>>
>> Thanks for the great explanation. So the current thread count will grow 
>> always until it reaches to the max configuration.
>>
>> I have another strange thing:
>> We never monitored tomcat busy thread count high (we monitored one minutes 
>> interval through Nagios to get tomcat manager result, not high than 10). So 
>> if the threads grows, it should monitored the busy thread count reach that 
>> high(or some lower). So I'm still not understand when the current thread 
>> count will grow while current busy thread has not change a lot.
>>
>
>I think that if you combine my explanation of the KeepAlive effect, with 
>Philippe's notes, 
>you may get the explanation.
>I would personally guess that your monitoring agent does not consider a thread 
>that is 
>idle in KeepAlive state, as "busy".  So it only counts as busy, the threads 
>which are 
>effectively in the state of processing a request.
>(This is just a guess; I think you may want to draw a little time-dependent 
>diagram, to 
>see what really happens, with a KeepAlive of 20 seconds, and ping or heartbeat 
>requests 
>coming in regularly.)
>In any case, if you want to avoid this phenomenon altogether, try configuring 
>an Executor. 
>You need to configure the  itself, and then in your , 
>refer to that 
>Executor).
>
>
>
>
>>
>>
>> -Original Message-
>> From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
>> Sent: Wednesday, January 18, 2017 9:41 AM
>> To:  users@tomcat.apache.org
>> Subject: Re: FW: tomcat 8080 thread not reduced
>>
>> Hi.
>>
>> I believe that what Philippe mentions below is somewhat different : in his 
>> configuration, there is apparently a front-end httpd server, which 
>> communicates with Tomcat via AJP and the tomcat AJP Connector.
>> In such a case, the "mod_jk" or "mod_proxy_ajp" connector module inside of 
>> httpd, creates a series of connections to the back-end tomcat, and keeps 
>> these connections alive in a pool of connections, according to rules which 
>> are different, than what applies to your case (which consists of direct HTTP 
>> connections from browsers, to the tomcat HTTP connector).
>> So Philippe's case introduces probably an additional level of complexity 
>> into the equation.
>> semi-graphically :
>>
>> smith case :
>>
>> client <- HTTP -> tomcat HTTP connector - tomcat thread
>>
>> Philippe's case :
>>
>> client <- HTTP -> apache httpd + ajp module <- AJP -> tomcat AJP connector - 
>> tomcat thread
>>
>> The real question is :
>>
>> The tomcat 8.0 HTTP Connector documentation
>> ( http://tomcat.apache.org/tomcat-8.0-doc/config/http.html ) mentions a 
>> series of attributes/parameters (maxThreads, minSpareThreads, 
>> connectionTimeout,
>> keepAliveTimeout,..) which, together, would lead one to believe that tomcat 
>> is managing the number of threads dynamically (increasing the number of 
>> threads (up to maxThreads) when needed to process simultaneous requests, and 
>> decreasing the number of threads again (down to minSpareThreads) when less 
>> requests are being processed).
>>
>> According to observations however (and to Chris' comments), that does not 
>> seem to be the case : once the number of threads has grown to a certain 
>> number, it never goes down back down again.
>> It has to be said that the above interpretation was mine, and the current 
>> documentation of the HTTP Connector never says explicitly that, on its own, 
>> it manages the number of threads dynamically. But it *does* mention a 
>> minSpareThreads parameter.
>>
>> On the other hand, when you use an Executor 
>> (http://tomcat.apache.org/tomcat-8.0-doc/config/executor.html), then the 
>> documentation states explicitly that the number of threads *is* dynamically 
>> managed, up and down.
>>
>> I found something else which seems to explain the riddle :
>>
>>  http://tomcat.apache.org/migration-6.html#Connector_thread_pools
>> That section explicitly says that since tomcat 6.0, for a Connector without 
>> an Executor, the number of threads always grows, and never decreases. And it 
>> also explicitly says that the Connector's "minSpareThreads" attribute will 
>&

Re: FW: tomcat 8080 thread not reduced

[tomcat]

On 18.01.2017 10:56, smith wrote:

Hi, André

Thanks for the great explanation. So the current thread count will grow always 
until it reaches to the max configuration.

I have another strange thing:
We never monitored tomcat busy thread count high (we monitored one minutes 
interval through Nagios to get tomcat manager result, not high than 10). So if 
the threads grows, it should monitored the busy thread count reach that high(or 
some lower). So I'm still not understand when the current thread count will 
grow while current busy thread has not change a lot.



I think that if you combine my explanation of the KeepAlive effect, with Philippe's notes, 
you may get the explanation.
I would personally guess that your monitoring agent does not consider a thread that is 
idle in KeepAlive state, as "busy".  So it only counts as busy, the threads which are 
effectively in the state of processing a request.
(This is just a guess; I think you may want to draw a little time-dependent diagram, to 
see what really happens, with a KeepAlive of 20 seconds, and ping or heartbeat requests 
coming in regularly.)
In any case, if you want to avoid this phenomenon altogether, try configuring an Executor. 
You need to configure the  itself, and then in your , refer to that 
Executor).








-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
Sent: Wednesday, January 18, 2017 9:41 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

Hi.

I believe that what Philippe mentions below is somewhat different : in his 
configuration, there is apparently a front-end httpd server, which communicates 
with Tomcat via AJP and the tomcat AJP Connector.
In such a case, the "mod_jk" or "mod_proxy_ajp" connector module inside of 
httpd, creates a series of connections to the back-end tomcat, and keeps these connections alive in 
a pool of connections, according to rules which are different, than what applies to your case 
(which consists of direct HTTP connections from browsers, to the tomcat HTTP connector).
So Philippe's case introduces probably an additional level of complexity into 
the equation.
semi-graphically :

smith case :

client <- HTTP -> tomcat HTTP connector - tomcat thread

Philippe's case :

client <- HTTP -> apache httpd + ajp module <- AJP -> tomcat AJP connector - 
tomcat thread

The real question is :

The tomcat 8.0 HTTP Connector documentation
(http://tomcat.apache.org/tomcat-8.0-doc/config/http.html) mentions a series of 
attributes/parameters (maxThreads, minSpareThreads, connectionTimeout,
keepAliveTimeout,..) which, together, would lead one to believe that tomcat is 
managing the number of threads dynamically (increasing the number of threads 
(up to maxThreads) when needed to process simultaneous requests, and decreasing 
the number of threads again (down to minSpareThreads) when less requests are 
being processed).

According to observations however (and to Chris' comments), that does not seem 
to be the case : once the number of threads has grown to a certain number, it 
never goes down back down again.
It has to be said that the above interpretation was mine, and the current 
documentation of the HTTP Connector never says explicitly that, on its own, it 
manages the number of threads dynamically. But it *does* mention a 
minSpareThreads parameter.

On the other hand, when you use an Executor 
(http://tomcat.apache.org/tomcat-8.0-doc/config/executor.html), then the 
documentation states explicitly that the number of threads *is* dynamically 
managed, up and down.

I found something else which seems to explain the riddle :

http://tomcat.apache.org/migration-6.html#Connector_thread_pools
That section explicitly says that since tomcat 6.0, for a Connector without an Executor, 
the number of threads always grows, and never decreases. And it also explicitly says that 
the Connector's "minSpareThreads" attribute will be ignored.

So in fact the only thing wrong, is the online documentation for the Connectors 
: the minSpareThreads attribute should be removed (since it is anyway ignored).
That seems to have been an oversight ever since tomcat 6.0.

As far as I am concerned thus, the mystery is solved.

One question which is a bit left open is :

What - if any - is the real advantage/disadvantage of having perhaps maxThreads 
idle threads, as opposed to using an Executor to manage the number of threads 
dynamically ?
But that is probably the kind of question to which the appropriate answer is : "it 
depends.."




On 18.01.2017 07:33, smith wrote:

Thanks, Philippe

But we never monitored tomcat busy thread count high (we monitored one minutes 
interval through nagios to get tomcat manager result, not high than 10). This 
is strange

-Original Message-
From: Philippe Busque [mailto:pbus...@mediagrif.com]
Sent: Monday, January 16, 2017 8:09 PM
To: users@tomcat.apache.org
Subject

RE: FW: tomcat 8080 thread not reduced

[smith]

Hi, André 

Thanks for the great explanation. So the current thread count will grow always 
until it reaches to the max configuration.

I have another strange thing:
We never monitored tomcat busy thread count high (we monitored one minutes 
interval through Nagios to get tomcat manager result, not high than 10). So if 
the threads grows, it should monitored the busy thread count reach that high(or 
some lower). So I'm still not understand when the current thread count will 
grow while current busy thread has not change a lot.



-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com] 
Sent: Wednesday, January 18, 2017 9:41 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

Hi.

I believe that what Philippe mentions below is somewhat different : in his 
configuration, there is apparently a front-end httpd server, which communicates 
with Tomcat via AJP and the tomcat AJP Connector.
In such a case, the "mod_jk" or "mod_proxy_ajp" connector module inside of 
httpd, creates a series of connections to the back-end tomcat, and keeps these 
connections alive in a pool of connections, according to rules which are 
different, than what applies to your case (which consists of direct HTTP 
connections from browsers, to the tomcat HTTP connector).
So Philippe's case introduces probably an additional level of complexity into 
the equation.
semi-graphically :

smith case :

client <- HTTP -> tomcat HTTP connector - tomcat thread

Philippe's case :

client <- HTTP -> apache httpd + ajp module <- AJP -> tomcat AJP connector - 
tomcat thread

The real question is :

The tomcat 8.0 HTTP Connector documentation
(http://tomcat.apache.org/tomcat-8.0-doc/config/http.html) mentions a series of 
attributes/parameters (maxThreads, minSpareThreads, connectionTimeout,
keepAliveTimeout,..) which, together, would lead one to believe that tomcat is 
managing the number of threads dynamically (increasing the number of threads 
(up to maxThreads) when needed to process simultaneous requests, and decreasing 
the number of threads again (down to minSpareThreads) when less requests are 
being processed).

According to observations however (and to Chris' comments), that does not seem 
to be the case : once the number of threads has grown to a certain number, it 
never goes down back down again.
It has to be said that the above interpretation was mine, and the current 
documentation of the HTTP Connector never says explicitly that, on its own, it 
manages the number of threads dynamically. But it *does* mention a 
minSpareThreads parameter.

On the other hand, when you use an Executor 
(http://tomcat.apache.org/tomcat-8.0-doc/config/executor.html), then the 
documentation states explicitly that the number of threads *is* dynamically 
managed, up and down.

I found something else which seems to explain the riddle :

http://tomcat.apache.org/migration-6.html#Connector_thread_pools
That section explicitly says that since tomcat 6.0, for a Connector without an 
Executor, the number of threads always grows, and never decreases. And it also 
explicitly says that the Connector's "minSpareThreads" attribute will be 
ignored.

So in fact the only thing wrong, is the online documentation for the Connectors 
: the minSpareThreads attribute should be removed (since it is anyway ignored).
That seems to have been an oversight ever since tomcat 6.0.

As far as I am concerned thus, the mystery is solved.

One question which is a bit left open is :

What - if any - is the real advantage/disadvantage of having perhaps maxThreads 
idle threads, as opposed to using an Executor to manage the number of threads 
dynamically ?
But that is probably the kind of question to which the appropriate answer is : 
"it depends.."




On 18.01.2017 07:33, smith wrote:
> Thanks, Philippe
>
> But we never monitored tomcat busy thread count high (we monitored one 
> minutes interval through nagios to get tomcat manager result, not high than 
> 10). This is strange
>
> -Original Message-
> From: Philippe Busque [mailto:pbus...@mediagrif.com]
> Sent: Monday, January 16, 2017 8:09 PM
> To: users@tomcat.apache.org
> Subject: Re: Re: FW: tomcat 8080 thread not reduced
>
> We're having a similar issues with our numberous Tomcat instances.
>
> Our connector config look like this.
>
>enableLookups="false" redirectPort="8443" acceptCount="100"
> connectionTimeout="2"  URIEncoding="UTF-8" />
>
>
> Sometime, the number of active connection would jump very high (up to 190), 
> due to some external issues (database lock, etc) and threads would accumulate.
>
> Even though a connectionTimeout is set, and therefor set keepAliveTimeout as 
> the same value,  threads are never released once the problem is  resolved 
> un

RE: FW: tomcat 8080 thread not reduced

[smith]

Hi, André
My case is same, my tomcat also has load balancer in front and it has health 
check to tomcat

-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com] 
Sent: Wednesday, January 18, 2017 9:41 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

Hi.

I believe that what Philippe mentions below is somewhat different : in his 
configuration, there is apparently a front-end httpd server, which communicates 
with Tomcat via AJP and the tomcat AJP Connector.
In such a case, the "mod_jk" or "mod_proxy_ajp" connector module inside of 
httpd, creates a series of connections to the back-end tomcat, and keeps these 
connections alive in a pool of connections, according to rules which are 
different, than what applies to your case (which consists of direct HTTP 
connections from browsers, to the tomcat HTTP connector).
So Philippe's case introduces probably an additional level of complexity into 
the equation.
semi-graphically :

smith case :

client <- HTTP -> tomcat HTTP connector - tomcat thread

Philippe's case :

client <- HTTP -> apache httpd + ajp module <- AJP -> tomcat AJP connector - 
tomcat thread

The real question is :

The tomcat 8.0 HTTP Connector documentation
(http://tomcat.apache.org/tomcat-8.0-doc/config/http.html) mentions a series of 
attributes/parameters (maxThreads, minSpareThreads, connectionTimeout,
keepAliveTimeout,..) which, together, would lead one to believe that tomcat is 
managing the number of threads dynamically (increasing the number of threads 
(up to maxThreads) when needed to process simultaneous requests, and decreasing 
the number of threads again (down to minSpareThreads) when less requests are 
being processed).

According to observations however (and to Chris' comments), that does not seem 
to be the case : once the number of threads has grown to a certain number, it 
never goes down back down again.
It has to be said that the above interpretation was mine, and the current 
documentation of the HTTP Connector never says explicitly that, on its own, it 
manages the number of threads dynamically. But it *does* mention a 
minSpareThreads parameter.

On the other hand, when you use an Executor 
(http://tomcat.apache.org/tomcat-8.0-doc/config/executor.html), then the 
documentation states explicitly that the number of threads *is* dynamically 
managed, up and down.

I found something else which seems to explain the riddle :

http://tomcat.apache.org/migration-6.html#Connector_thread_pools
That section explicitly says that since tomcat 6.0, for a Connector without an 
Executor, the number of threads always grows, and never decreases. And it also 
explicitly says that the Connector's "minSpareThreads" attribute will be 
ignored.

So in fact the only thing wrong, is the online documentation for the Connectors 
: the minSpareThreads attribute should be removed (since it is anyway ignored).
That seems to have been an oversight ever since tomcat 6.0.

As far as I am concerned thus, the mystery is solved.

One question which is a bit left open is :

What - if any - is the real advantage/disadvantage of having perhaps maxThreads 
idle threads, as opposed to using an Executor to manage the number of threads 
dynamically ?
But that is probably the kind of question to which the appropriate answer is : 
"it depends.."




On 18.01.2017 07:33, smith wrote:
> Thanks, Philippe
>
> But we never monitored tomcat busy thread count high (we monitored one 
> minutes interval through nagios to get tomcat manager result, not high than 
> 10). This is strange
>
> -Original Message-
> From: Philippe Busque [mailto:pbus...@mediagrif.com]
> Sent: Monday, January 16, 2017 8:09 PM
> To: users@tomcat.apache.org
> Subject: Re: Re: FW: tomcat 8080 thread not reduced
>
> We're having a similar issues with our numberous Tomcat instances.
>
> Our connector config look like this.
>
>enableLookups="false" redirectPort="8443" acceptCount="100"
> connectionTimeout="2"  URIEncoding="UTF-8" />
>
>
> Sometime, the number of active connection would jump very high (up to 190), 
> due to some external issues (database lock, etc) and threads would accumulate.
>
> Even though a connectionTimeout is set, and therefor set keepAliveTimeout as 
> the same value,  threads are never released once the problem is  resolved 
> until Tomcat is restarted.  We would end up with maybe 5-10 busy workers, but 
> 190 idle workers/ threads
>
> I think the issue is related to how the StandardThreadExecutor is 
> implemented. The StandardThreadExecutor is a front for the default Java 
> ThreadPoolExecutor.  If I'm not mistaken, ThreadPoolExecutor is distributing 
> work in round robin fashion among all defined workers, rather than sticking 
> to

Re: FW: tomcat 8080 thread not reduced

[tomcat]

Hi.

I believe that what Philippe mentions below is somewhat different : in his configuration, 
there is apparently a front-end httpd server, which communicates with Tomcat via AJP and 
the tomcat AJP Connector.
In such a case, the "mod_jk" or "mod_proxy_ajp" connector module inside of httpd, creates 
a series of connections to the back-end tomcat, and keeps these connections alive in a 
pool of connections, according to rules which are different, than what applies to your 
case (which consists of direct HTTP connections from browsers, to the tomcat HTTP connector).

So Philippe's case introduces probably an additional level of complexity into 
the equation.
semi-graphically :

smith case :

client <- HTTP -> tomcat HTTP connector - tomcat thread

Philippe's case :

client <- HTTP -> apache httpd + ajp module <- AJP -> tomcat AJP connector - 
tomcat thread

The real question is :

The tomcat 8.0 HTTP Connector documentation 
(http://tomcat.apache.org/tomcat-8.0-doc/config/http.html) mentions a series of 
attributes/parameters (maxThreads, minSpareThreads, connectionTimeout, 
keepAliveTimeout,..) which, together, would lead one to believe that tomcat is managing 
the number of threads dynamically (increasing the number of threads (up to maxThreads) 
when needed to process simultaneous requests, and decreasing the number of threads again 
(down to minSpareThreads) when less requests are being processed).


According to observations however (and to Chris' comments), that does not seem to be the 
case : once the number of threads has grown to a certain number, it never goes down back 
down again.
It has to be said that the above interpretation was mine, and the current documentation of 
the HTTP Connector never says explicitly that, on its own, it manages the number of 
threads dynamically. But it *does* mention a minSpareThreads parameter.


On the other hand, when you use an Executor 
(http://tomcat.apache.org/tomcat-8.0-doc/config/executor.html), then the documentation 
states explicitly that the number of threads *is* dynamically managed, up and down.


I found something else which seems to explain the riddle :

http://tomcat.apache.org/migration-6.html#Connector_thread_pools
That section explicitly says that since tomcat 6.0, for a Connector without an Executor, 
the number of threads always grows, and never decreases. And it also explicitly says that 
the Connector's "minSpareThreads" attribute will be ignored.


So in fact the only thing wrong, is the online documentation for the Connectors : the 
minSpareThreads attribute should be removed (since it is anyway ignored).

That seems to have been an oversight ever since tomcat 6.0.

As far as I am concerned thus, the mystery is solved.

One question which is a bit left open is :

What - if any - is the real advantage/disadvantage of having perhaps maxThreads idle 
threads, as opposed to using an Executor to manage the number of threads dynamically ?

But that is probably the kind of question to which the appropriate answer is : "it 
depends.."




On 18.01.2017 07:33, smith wrote:

Thanks, Philippe

But we never monitored tomcat busy thread count high (we monitored one minutes 
interval through nagios to get tomcat manager result, not high than 10). This 
is strange

-Original Message-
From: Philippe Busque [mailto:pbus...@mediagrif.com]
Sent: Monday, January 16, 2017 8:09 PM
To: users@tomcat.apache.org
Subject: Re: Re: FW: tomcat 8080 thread not reduced

We're having a similar issues with our numberous Tomcat instances.

Our connector config look like this.

  


Sometime, the number of active connection would jump very high (up to 190), due 
to some external issues (database lock, etc) and threads would accumulate.

Even though a connectionTimeout is set, and therefor set keepAliveTimeout as 
the same value,  threads are never released once the problem is  resolved until 
Tomcat is restarted.  We would end up with maybe 5-10 busy workers, but 190 
idle workers/ threads

I think the issue is related to how the StandardThreadExecutor is implemented. 
The StandardThreadExecutor is a front for the default Java ThreadPoolExecutor.  
If I'm not mistaken, ThreadPoolExecutor is distributing work in round robin 
fashion among all defined workers, rather than sticking to the core threads.


As a result, should a website has any constant traffic (Apache AJP ping, load 
balancer monitoring, normal traffic, etc), all thread will be hit at least once 
within the configured  keepAliveTimeout, reseting it. So unless the 
keepAliveTimeout is set to a very low value, which defeat the purpose, thread 
will never be released .


This is what I've come to suspect from looking at the StandardThreadExecutor, 
but never really had the opportunity to do deeper test with load.
But from Tomcat 6 to tomcat 8, we were never able to  decrease the number of 
'idle' workers back from the highest value it h

RE: Re: FW: tomcat 8080 thread not reduced

[smith]

Thanks, Philippe

But we never monitored tomcat busy thread count high (we monitored one minutes 
interval through nagios to get tomcat manager result, not high than 10). This 
is strange

-Original Message-
From: Philippe Busque [mailto:pbus...@mediagrif.com] 
Sent: Monday, January 16, 2017 8:09 PM
To: users@tomcat.apache.org
Subject: Re: Re: FW: tomcat 8080 thread not reduced

We're having a similar issues with our numberous Tomcat instances.

Our connector config look like this.

 


Sometime, the number of active connection would jump very high (up to 190), due 
to some external issues (database lock, etc) and threads would accumulate.

Even though a connectionTimeout is set, and therefor set keepAliveTimeout as 
the same value,  threads are never released once the problem is  resolved until 
Tomcat is restarted.  We would end up with maybe 5-10 busy workers, but 190 
idle workers/ threads

I think the issue is related to how the StandardThreadExecutor is implemented. 
The StandardThreadExecutor is a front for the default Java ThreadPoolExecutor.  
If I'm not mistaken, ThreadPoolExecutor is distributing work in round robin 
fashion among all defined workers, rather than sticking to the core threads.


As a result, should a website has any constant traffic (Apache AJP ping, load 
balancer monitoring, normal traffic, etc), all thread will be hit at least once 
within the configured  keepAliveTimeout, reseting it. So unless the 
keepAliveTimeout is set to a very low value, which defeat the purpose, thread 
will never be released .


This is what I've come to suspect from looking at the StandardThreadExecutor, 
but never really had the opportunity to do deeper test with load.
But from Tomcat 6 to tomcat 8, we were never able to  decrease the number of 
'idle' workers back from the highest value it had reached.


Le 2017-01-16 à 05:24, André Warnier (tomcat) a écrit :
> On 16.01.2017 11:10, smith wrote:
>> We has same problem on dev env that no any traffic to the serive,
>
> Ah. That is /new/ information, which may change the suggestions below.
> It looks like you should really find out what these threads are doing, 
> probably by doing a few thread dumps.
> See here e.g. :
> http://stackoverflow.com/questions/18573411/tomcat-thread-dump
>
> Again : we do not know your application, so we can only make guesses 
> based on the information that you provide.
>
>  will try on dev first
>>
>> -Original Message-
>> From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
>> Sent: Monday, January 16, 2017 10:08 AM
>> To: users@tomcat.apache.org
>> Subject: Re: FW: tomcat 8080 thread not reduced
>>
>> On 16.01.2017 09:50, smith wrote:
>>> Busy one is process customer request, do not know what non-busy one 
>>> is doing, always keep 120 for many days. I don't think 20s timeout 
>>> will not cause so long connection
>>>
>>> -smith
>>
>> And did you actually try it ?
>>
>> We do not know your website or your application, so we cannot tell 
>> how many clients there are, what these clients are really requesting, 
>> how many requests each client is sending before going away, etc.
>>
>> KeepAlive means that when a client has sent its /last/ request and 
>> received the response, one thread is going to remain "not free" (but 
>> doing nothing) for the duration of the KeepAlive timeout. This thread 
>> will keep waiting, for KeepAliveTimeout seconds, just in case the 
>> client would still send another request (which it may never do, 
>> depending on the application).
>>
>> Imagine that your application is so that the average client
>> - connects to your site
>> - sends a single HTTP request, which gets processed in 0.1 s
>> - receives the response
>> - and then goes away
>> and that the above sequence happens once every second, from different 
>> clients.
>> After one second, there will be one thread waiting for another 19 
>> seconds before becoming free (and potentially destroyed or re-used).
>> After 2 seconds, there will be 2 such threads. After 3 seconds, 3 
>> threads. And so on. After 20 seconds, the first thread will be freed, 
>> but there will be 19 other threads still waiting, and one new thread 
>> just created.
>> If everything stays perfectly regular like that, your will have 
>> /permanently/ 20 threads in existence, even if the minimum is 10.
>> If you change the above so that there is a new client every 0.5 s, 
>> you will have permanently 40 threads (of which only 2 maximum are 
>> really doing something).
>>
>> The point is : KeepAlive is not "bad", and in some cases having a 
>> relatively long KeepAliveTimeout is the ri

RE: FW: tomcat 8080 thread not reduced

[smith]

Thanks chris

So the tomcat default executor will not reduce the thread count until it reach 
to the max configuration? Will it reduce when it reach to max?
And why the default not reduce the thread?

-Smith

-Original Message-
From: Christopher Schultz [mailto:ch...@christopherschultz.net] 
Sent: Tuesday, January 17, 2017 7:18 PM
To: Tomcat Users List
Subject: Re: FW: tomcat 8080 thread not reduced

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/16/17 8:22 PM, smith wrote:
> Yes, I think thread count should be reduced when those threads are 
> idle
> 
> Is this right? Or it will not reduced?

Id you want Tomcat to reduce the number of idle threads, you'll need to 
explicitly configure an  and use that with your  .

- -chris

> -Original Message- From: Christopher Schultz 
> [mailto:ch...@christopherschultz.net] Sent: Monday, January 16,
> 2017 2:20 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080 thread 
> not reduced
> 
> Smith,
> 
> There are your only active s:
> 
> On 1/14/17 1:30 AM, smith wrote:
>> > connectionTimeout="2" redirectPort="8443" />
> 
>> [snip]
> 
>> 
> 
> You have not changed any settings from the default. What makes you 
> think that your thread count should be reduced when those threads are 
> idle?
> 
> -chris
> 
> -
>
> 
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> 
> -
>
> 
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
-BEGIN PGP SIGNATURE-
Comment: GPGTools - http://gpgtools.org
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=W1gG
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org



-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: FW: tomcat 8080 thread not reduced

[Christopher Schultz]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

On 1/16/17 8:22 PM, smith wrote:
> Yes, I think thread count should be reduced when those threads are
> idle
> 
> Is this right? Or it will not reduced?

Id you want Tomcat to reduce the number of idle threads, you'll need
to explicitly configure an  and use that with your 
.

- -chris

> -Original Message- From: Christopher Schultz
> [mailto:ch...@christopherschultz.net] Sent: Monday, January 16,
> 2017 2:20 PM To: Tomcat Users List Subject: Re: FW: tomcat 8080
> thread not reduced
> 
> Smith,
> 
> There are your only active s:
> 
> On 1/14/17 1:30 AM, smith wrote:
>> > connectionTimeout="2" redirectPort="8443" />
> 
>> [snip]
> 
>> 
> 
> You have not changed any settings from the default. What makes you
> think that your thread count should be reduced when those threads
> are idle?
> 
> -chris
> 
> -
>
> 
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
> 
> 
> -
>
> 
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
-BEGIN PGP SIGNATURE-
Comment: GPGTools - http://gpgtools.org
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=W1gG
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

RE: FW: tomcat 8080 thread not reduced

[smith]

Yes, I also think it should act like this, but it did not.

-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com] 
Sent: Monday, January 16, 2017 2:33 PM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 16.01.2017 15:19, Christopher Schultz wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA256
>
> Smith,
>
> There are your only active s:
>
> On 1/14/17 1:30 AM, smith wrote:
>> > connectionTimeout="2" redirectPort="8443" />
>>
>> [snip]
>>
>> 
>
> You have not changed any settings from the default. What makes you 
> think that your thread count should be reduced when those threads are 
> idle?

Hi.  (André responding for "smith")

Maybe this :
http://tomcat.apache.org/tomcat-8.0-doc/config/http.html#Standard_Implementation
--> minSpareThreads
?
(That's also the way I understand this : inactive threads disappear after a 
while, down to the number mentioned here - which is 10 by default. Not so ?)

>
> - -chris
> -BEGIN PGP SIGNATURE-
> Comment: GPGTools - http://gpgtools.org
> Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
>
> iQIcBAEBCAAGBQJYfNaBAAoJEBzwKT+lPKRY3ZsP+wdrzjUHAVPoF8KXIAj+vz0O
> lZ0zXBC2XID9Bh6dQRKvW+9k/RQRCuOrnZ+Z4qVdeo9jdtQMPGkQcpr5mw3m7rCZ
> XefMU8KBsSl/31lBhl8ajxAUqpTx37dSUV6mUFVtVVEZdp2/+3Jy1slmEuO7+COJ
> wUYtD7y4yk9oitxj0S3Omeglo+g00yJmTVgEJClCF19wYIn0uSfwZFgOpkftj2HO
> ct1ndMa5WIUQYtQ6ueiOgIoZb4aEhhS69JilN0GZ1WZpdTUCj04bosnOJ94w8fPi
> 1+z/5GCAPVhLBYN+LTr/03EoC9u+gm7163+eIbGK8g/1GiNqY+m6plI7LMGaNy0j
> 3pfgitcK2Mte4w2T0yJpe87lwNOCrkVL4oXCmOGx8b/P5f9M2MukKTIfmRsT7Q/G
> u4skc+LYCxVwIqsND/4DuBatAx8FSzi3UebHIJoBVOe22b6zPYKGaXUBo530dAvK
> iVnBgzxIOvQuQr1pD6ylRjCmrmooIH6P8iknHX2DrKyaqEcdnR0g/iJnfuGeW+JO
> VrbTWEcb29IrKsn4WGuglALijl+I1UCEvru5aTlvkXyWhePK1wlHYaupy69sy3a/
> 5z4LJdOebtN3QFr2I6hM54L0+bjjwCznlVa8b/6ky84ya/7tzWGoHqpqYod46HZZ
> 7E1OkTnDk67xhmh7iXbB
> =e2NW
> -END PGP SIGNATURE-
>
> -
> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: users-h...@tomcat.apache.org
>


-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org



-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

RE: FW: tomcat 8080 thread not reduced

[smith]

Yes, I think thread count should be reduced when those threads are idle

Is this right? Or it will not reduced?

-Original Message-
From: Christopher Schultz [mailto:ch...@christopherschultz.net] 
Sent: Monday, January 16, 2017 2:20 PM
To: Tomcat Users List
Subject: Re: FW: tomcat 8080 thread not reduced

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

There are your only active s:

On 1/14/17 1:30 AM, smith wrote:
>  connectionTimeout="2" redirectPort="8443" />
> 
> [snip]
> 
> 

You have not changed any settings from the default. What makes you think that 
your thread count should be reduced when those threads are idle?

- -chris
-BEGIN PGP SIGNATURE-
Comment: GPGTools - http://gpgtools.org
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=e2NW
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org



-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: Re: FW: tomcat 8080 thread not reduced

[Philippe Busque]

We're having a similar issues with our numberous Tomcat instances.

Our connector config look like this.




Sometime, the number of active connection would jump very high (up to
190), due to some external issues (database lock, etc) and threads would
accumulate.

Even though a connectionTimeout is set, and therefor set
keepAliveTimeout as the same value,  threads are never released once the
problem is  resolved until Tomcat is restarted.  We would end up with
maybe 5-10 busy workers, but 190 idle workers/ threads

I think the issue is related to how the StandardThreadExecutor is
implemented. The StandardThreadExecutor is a front for the default Java
ThreadPoolExecutor.  If I'm not mistaken, ThreadPoolExecutor is
distributing work in round robin fashion among all defined workers,
rather than sticking to the core threads.


As a result, should a website has any constant traffic (Apache AJP ping,
load balancer monitoring, normal traffic, etc), all thread will be hit
at least once within the configured  keepAliveTimeout, reseting it. So
unless the keepAliveTimeout is set to a very low value, which defeat the
purpose, thread will never be released .


This is what I've come to suspect from looking at the
StandardThreadExecutor, but never really had the opportunity to do
deeper test with load.
But from Tomcat 6 to tomcat 8, we were never able to  decrease the
number of 'idle' workers back from the highest value it had reached.


Le 2017-01-16 à 05:24, André Warnier (tomcat) a écrit :

On 16.01.2017 11:10, smith wrote:

We has same problem on dev env that no any traffic to the serive,


Ah. That is /new/ information, which may change the suggestions below.
It looks like you should really find out what these threads are doing,
probably by doing a few thread dumps.
See here e.g. :
http://stackoverflow.com/questions/18573411/tomcat-thread-dump

Again : we do not know your application, so we can only make guesses
based on the information that you provide.

 will try on dev first


-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
Sent: Monday, January 16, 2017 10:08 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 16.01.2017 09:50, smith wrote:

Busy one is process customer request, do not know what non-busy one is
doing, always keep 120 for many days. I don't think 20s timeout will
not cause so long connection

-smith


And did you actually try it ?

We do not know your website or your application, so we cannot tell
how many clients there are, what these clients are really requesting,
how many requests each client is sending before going away, etc.

KeepAlive means that when a client has sent its /last/ request and
received the response, one thread is going to remain "not free" (but
doing nothing) for the duration of the KeepAlive timeout. This thread
will keep waiting, for KeepAliveTimeout seconds, just in case the
client would still send another request (which it may never do,
depending on the application).

Imagine that your application is so that the average client
- connects to your site
- sends a single HTTP request, which gets processed in 0.1 s
- receives the response
- and then goes away
and that the above sequence happens once every second, from different
clients.
After one second, there will be one thread waiting for another 19
seconds before becoming free (and potentially destroyed or re-used).
After 2 seconds, there will be 2 such threads. After 3 seconds, 3
threads. And so on. After 20 seconds, the first thread will be freed,
but there will be 19 other threads still waiting, and one new thread
just created.
If everything stays perfectly regular like that, your will have
/permanently/ 20 threads in existence, even if the minimum is 10.
If you change the above so that there is a new client every 0.5 s,
you will have permanently 40 threads (of which only 2 maximum are
really doing something).

The point is : KeepAlive is not "bad", and in some cases having a
relatively long KeepAliveTimeout is the right thing to do. Also,
having a high number of threads sitting idle is not necessarily a
problem.
Your own scenario is probably not like the above perfectly regular
and irrealistic one above. But there may be a perfectly logical
reason why you have so many threads on average, and I am just trying
to give you ideas for finding out the reason.




-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
Sent: Monday, January 16, 2017 8:33 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 16.01.2017 03:41, Smith Hua wrote:


actually there is not much busy threads, less tahn 10,so i think this
parameter may has nothing to do with this


It depends on what you call "busy". What are the busy ones doing ?
and what are the non-busy ones doing ?



--
从myMail的Android专用app所发送 星期一, 16 一月 2017, 03:01上午 +08:00
发件人 André Warnier (tomcat)  a...@ice-sa.com :


Hi.

I

Re: FW: tomcat 8080 thread not reduced

[tomcat]

On 16.01.2017 15:19, Christopher Schultz wrote:

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

There are your only active s:

On 1/14/17 1:30 AM, smith wrote:



[snip]




You have not changed any settings from the default. What makes you
think that your thread count should be reduced when those threads are
idle?


Hi.  (André responding for "smith")

Maybe this :
http://tomcat.apache.org/tomcat-8.0-doc/config/http.html#Standard_Implementation
--> minSpareThreads
?
(That's also the way I understand this : inactive threads disappear after a while, down to 
the number mentioned here - which is 10 by default. Not so ?)




- -chris
-BEGIN PGP SIGNATURE-
Comment: GPGTools - http://gpgtools.org
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=e2NW
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org




-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: FW: tomcat 8080 thread not reduced

[Christopher Schultz]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Smith,

There are your only active s:

On 1/14/17 1:30 AM, smith wrote:
>  connectionTimeout="2" redirectPort="8443" />
> 
> [snip]
> 
> 

You have not changed any settings from the default. What makes you
think that your thread count should be reduced when those threads are
idle?

- -chris
-BEGIN PGP SIGNATURE-
Comment: GPGTools - http://gpgtools.org
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQIcBAEBCAAGBQJYfNaBAAoJEBzwKT+lPKRY3ZsP+wdrzjUHAVPoF8KXIAj+vz0O
lZ0zXBC2XID9Bh6dQRKvW+9k/RQRCuOrnZ+Z4qVdeo9jdtQMPGkQcpr5mw3m7rCZ
XefMU8KBsSl/31lBhl8ajxAUqpTx37dSUV6mUFVtVVEZdp2/+3Jy1slmEuO7+COJ
wUYtD7y4yk9oitxj0S3Omeglo+g00yJmTVgEJClCF19wYIn0uSfwZFgOpkftj2HO
ct1ndMa5WIUQYtQ6ueiOgIoZb4aEhhS69JilN0GZ1WZpdTUCj04bosnOJ94w8fPi
1+z/5GCAPVhLBYN+LTr/03EoC9u+gm7163+eIbGK8g/1GiNqY+m6plI7LMGaNy0j
3pfgitcK2Mte4w2T0yJpe87lwNOCrkVL4oXCmOGx8b/P5f9M2MukKTIfmRsT7Q/G
u4skc+LYCxVwIqsND/4DuBatAx8FSzi3UebHIJoBVOe22b6zPYKGaXUBo530dAvK
iVnBgzxIOvQuQr1pD6ylRjCmrmooIH6P8iknHX2DrKyaqEcdnR0g/iJnfuGeW+JO
VrbTWEcb29IrKsn4WGuglALijl+I1UCEvru5aTlvkXyWhePK1wlHYaupy69sy3a/
5z4LJdOebtN3QFr2I6hM54L0+bjjwCznlVa8b/6ky84ya/7tzWGoHqpqYod46HZZ
7E1OkTnDk67xhmh7iXbB
=e2NW
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: FW: tomcat 8080 thread not reduced

[tomcat]

On 16.01.2017 11:10, smith wrote:

We has same problem on dev env that no any traffic to the serive,


Ah. That is /new/ information, which may change the suggestions below.
It looks like you should really find out what these threads are doing, probably by doing a 
few thread dumps.

See here e.g. : http://stackoverflow.com/questions/18573411/tomcat-thread-dump

Again : we do not know your application, so we can only make guesses based on the 
information that you provide.


 will try on dev first


-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
Sent: Monday, January 16, 2017 10:08 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 16.01.2017 09:50, smith wrote:

Busy one is process customer request, do not know what non-busy one is
doing, always keep 120 for many days. I don't think 20s timeout will
not cause so long connection

-smith


And did you actually try it ?

We do not know your website or your application, so we cannot tell how many 
clients there are, what these clients are really requesting, how many requests 
each client is sending before going away, etc.

KeepAlive means that when a client has sent its /last/ request and received the response, 
one thread is going to remain "not free" (but doing nothing) for the duration 
of the KeepAlive timeout. This thread will keep waiting, for KeepAliveTimeout seconds, 
just in case the client would still send another request (which it may never do, 
depending on the application).

Imagine that your application is so that the average client
- connects to your site
- sends a single HTTP request, which gets processed in 0.1 s
- receives the response
- and then goes away
and that the above sequence happens once every second, from different clients.
After one second, there will be one thread waiting for another 19 seconds 
before becoming free (and potentially destroyed or re-used). After 2 seconds, 
there will be 2 such threads. After 3 seconds, 3 threads. And so on. After 20 
seconds, the first thread will be freed, but there will be 19 other threads 
still waiting, and one new thread just created.
If everything stays perfectly regular like that, your will have /permanently/ 
20 threads in existence, even if the minimum is 10.
If you change the above so that there is a new client every 0.5 s, you will 
have permanently 40 threads (of which only 2 maximum are really doing 
something).

The point is : KeepAlive is not "bad", and in some cases having a relatively 
long KeepAliveTimeout is the right thing to do.  Also, having a high number of threads 
sitting idle is not necessarily a problem.
Your own scenario is probably not like the above perfectly regular and 
irrealistic one above. But there may be a perfectly logical reason why you have 
so many threads on average, and I am just trying to give you ideas for finding 
out the reason.




-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
Sent: Monday, January 16, 2017 8:33 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 16.01.2017 03:41, Smith Hua wrote:


actually there is not much busy threads, less tahn 10,so i think this
parameter may has nothing to do with this


It depends on what you call "busy". What are the busy ones doing ? and what are 
the non-busy ones doing ?



--
从myMail的Android专用app所发送 星期一, 16 一月 2017, 03:01上午 +08:00 发件人 André Warnier 
(tomcat)  a...@ice-sa.com :


Hi.

I can find nothing really wrong in your configuration below.
But, what happens if in this section :

>   maxThreads="300" connectionTimeout="2"
> redirectPort="8443" />

you change the connectionTimeout to 3000 (= 3 seconds, instead of the above 20 
seconds) ?

Do you still see the number of threads remaining at the maximum ?

See :
http://tomcat.apache.org/tomcat-8.0-doc/config/http.html#Standard_Im
plementation
--> connectionTimeout
and the fact that it is also the default for keepAliveTimeout


On 14.01.2017 07:30, smith wrote:

The server.xml:





  
  
  
  
  
  
  
  

  
  


  

  
  





















  
  

  
  


  

  









  

  


-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
Sent: Friday, January 13, 2017 10:42 AM
To:  users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 13.01.2017 09:38, smith wrote:





From: smith [mailto:smith@zoom.us]
Sent: Tuesday, January 10, 2017 9:57 AM
To: 'users'
Subject: tomcat 8080 thread not reduced



Hi,



We have installed Apache Tomcat/8.0

RE: FW: tomcat 8080 thread not reduced

[smith]

We has same problem on dev env that no any traffic to the serive, will try on 
dev first

-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com] 
Sent: Monday, January 16, 2017 10:08 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 16.01.2017 09:50, smith wrote:
> Busy one is process customer request, do not know what non-busy one is 
> doing, always keep 120 for many days. I don't think 20s timeout will 
> not cause so long connection
>
> -smith

And did you actually try it ?

We do not know your website or your application, so we cannot tell how many 
clients there are, what these clients are really requesting, how many requests 
each client is sending before going away, etc.

KeepAlive means that when a client has sent its /last/ request and received the 
response, one thread is going to remain "not free" (but doing nothing) for the 
duration of the KeepAlive timeout. This thread will keep waiting, for 
KeepAliveTimeout seconds, just in case the client would still send another 
request (which it may never do, depending on the application).

Imagine that your application is so that the average client
- connects to your site
- sends a single HTTP request, which gets processed in 0.1 s
- receives the response
- and then goes away
and that the above sequence happens once every second, from different clients.
After one second, there will be one thread waiting for another 19 seconds 
before becoming free (and potentially destroyed or re-used). After 2 seconds, 
there will be 2 such threads. After 3 seconds, 3 threads. And so on. After 20 
seconds, the first thread will be freed, but there will be 19 other threads 
still waiting, and one new thread just created.
If everything stays perfectly regular like that, your will have /permanently/ 
20 threads in existence, even if the minimum is 10.
If you change the above so that there is a new client every 0.5 s, you will 
have permanently 40 threads (of which only 2 maximum are really doing 
something).

The point is : KeepAlive is not "bad", and in some cases having a relatively 
long KeepAliveTimeout is the right thing to do.  Also, having a high number of 
threads sitting idle is not necessarily a problem.
Your own scenario is probably not like the above perfectly regular and 
irrealistic one above. But there may be a perfectly logical reason why you have 
so many threads on average, and I am just trying to give you ideas for finding 
out the reason.


>
> -Original Message-
> From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
> Sent: Monday, January 16, 2017 8:33 AM
> To: users@tomcat.apache.org
> Subject: Re: FW: tomcat 8080 thread not reduced
>
> On 16.01.2017 03:41, Smith Hua wrote:
>>
>> actually there is not much busy threads, less tahn 10,so i think this 
>> parameter may has nothing to do with this
>
> It depends on what you call "busy". What are the busy ones doing ? and what 
> are the non-busy ones doing ?
>
>
>> --
>> 从myMail的Android专用app所发送 星期一, 16 一月 2017, 03:01上午 +08:00 发件人 André Warnier 
>> (tomcat)  a...@ice-sa.com :
>>
>>> Hi.
>>>
>>> I can find nothing really wrong in your configuration below.
>>> But, what happens if in this section :
>>>
>>>>  >>> maxThreads="300" connectionTimeout="2"
>>>> redirectPort="8443" />
>>>
>>> you change the connectionTimeout to 3000 (= 3 seconds, instead of the above 
>>> 20 seconds) ?
>>>
>>> Do you still see the number of threads remaining at the maximum ?
>>>
>>> See :  
>>> http://tomcat.apache.org/tomcat-8.0-doc/config/http.html#Standard_Im
>>> plementation
>>> --> connectionTimeout
>>> and the fact that it is also the default for keepAliveTimeout
>>>
>>>
>>> On 14.01.2017 07:30, smith wrote:
>>>> The server.xml:
>>>>
>>>> 
>>>> 
>>>> 
>>>> 
>>>>  >>> className="org.apache.catalina.startup.VersionLoggerListener" />
>>>>  
>>>>  
>>>>  >>> SSLEngine="on" />
>>>>  
>>>>  >>> className="org.apache.catalina.core.JreMemoryLeakPreventionListener" />
>>>>  >>> className="org.apache.catalina.mbeans.GlobalResourcesLifecycleListener" />
>>>>  >>

Re: FW: tomcat 8080 thread not reduced

[tomcat]

On 16.01.2017 09:50, smith wrote:

Busy one is process customer request, do not know what non-busy one is doing, 
always keep 120 for many days. I don't think 20s timeout will not cause so long 
connection

-smith


And did you actually try it ?

We do not know your website or your application, so we cannot tell how many clients there 
are, what these clients are really requesting, how many requests each client is sending 
before going away, etc.


KeepAlive means that when a client has sent its /last/ request and received the response, 
one thread is going to remain "not free" (but doing nothing) for the duration of the 
KeepAlive timeout. This thread will keep waiting, for KeepAliveTimeout seconds, just in 
case the client would still send another request (which it may never do, depending on the 
application).


Imagine that your application is so that the average client
- connects to your site
- sends a single HTTP request, which gets processed in 0.1 s
- receives the response
- and then goes away
and that the above sequence happens once every second, from different clients.
After one second, there will be one thread waiting for another 19 seconds before becoming 
free (and potentially destroyed or re-used). After 2 seconds, there will be 2 such 
threads. After 3 seconds, 3 threads. And so on. After 20 seconds, the first thread will be 
freed, but there will be 19 other threads still waiting, and one new thread just created.
If everything stays perfectly regular like that, your will have /permanently/ 20 threads 
in existence, even if the minimum is 10.
If you change the above so that there is a new client every 0.5 s, you will have 
permanently 40 threads (of which only 2 maximum are really doing something).


The point is : KeepAlive is not "bad", and in some cases having a relatively long 
KeepAliveTimeout is the right thing to do.  Also, having a high number of threads sitting 
idle is not necessarily a problem.
Your own scenario is probably not like the above perfectly regular and irrealistic one 
above. But there may be a perfectly logical reason why you have so many threads on 
average, and I am just trying to give you ideas for finding out the reason.





-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
Sent: Monday, January 16, 2017 8:33 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 16.01.2017 03:41, Smith Hua wrote:


actually there is not much busy threads, less tahn 10,so i think this
parameter may has nothing to do with this


It depends on what you call "busy". What are the busy ones doing ? and what are 
the non-busy ones doing ?



--
从myMail的Android专用app所发送 星期一, 16 一月 2017, 03:01上午 +08:00 发件人 André Warnier 
(tomcat)  a...@ice-sa.com :


Hi.

I can find nothing really wrong in your configuration below.
But, what happens if in this section :

   >   maxThreads="300" connectionTimeout="2"
   > redirectPort="8443" />

you change the connectionTimeout to 3000 (= 3 seconds, instead of the above 20 
seconds) ?

Do you still see the number of threads remaining at the maximum ?

See :  
http://tomcat.apache.org/tomcat-8.0-doc/config/http.html#Standard_Implementation
--> connectionTimeout
and the fact that it is also the default for
keepAliveTimeout


On 14.01.2017 07:30, smith wrote:

The server.xml:





 
 
 
 
 
 
 
 

 
 
   
   
 

 
 

   
   


   
   
   
   
   
   

   
   


   

   
   

 
 

 
 
   
   
 

 

   
   

   



   
 
   
 


-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
Sent: Friday, January 13, 2017 10:42 AM
To:  users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 13.01.2017 09:38, smith wrote:





From: smith [mailto:smith@zoom.us]
Sent: Tuesday, January 10, 2017 9:57 AM
To: 'users'
Subject: tomcat 8080 thread not reduced



Hi,



We have installed Apache Tomcat/8.0.14, and found that after one period of 
time, the thread count for 8080(our port published) goes to 120 and never 
reduced even the busy count is only 3-4.

Why? Tomcat8 not reduced the thread pool even the thread is idle, and the 
minSpareThreads for tomcat8 default is only 10.

When will the thread reduce?





Best regards

Smith




Hi.

Please copy/paste your complete server.xml configuration file (confidential 
things removed), so that we could have a useful look at it.
Please edit *only* the confidential things, not entire sections. Often, the 
issue is in the details.



-
To unsubscribe, e-mail:  users-unsubscr...@tomcat.a

RE: FW: tomcat 8080 thread not reduced

[smith]

Busy one is process customer request, do not know what non-busy one is doing, 
always keep 120 for many days. I don't think 20s timeout will not cause so long 
connection

-smith

-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com] 
Sent: Monday, January 16, 2017 8:33 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 16.01.2017 03:41, Smith Hua wrote:
>
> actually there is not much busy threads, less tahn 10,so i think this 
> parameter may has nothing to do with this

It depends on what you call "busy". What are the busy ones doing ? and what are 
the non-busy ones doing ?


> --
> 从myMail的Android专用app所发送 星期一, 16 一月 2017, 03:01上午 +08:00 发件人 André Warnier 
> (tomcat)  a...@ice-sa.com :
>
>> Hi.
>>
>> I can find nothing really wrong in your configuration below.
>> But, what happens if in this section :
>>
>>   >  >   > maxThreads="300" connectionTimeout="2"
>>   > redirectPort="8443" />
>>
>> you change the connectionTimeout to 3000 (= 3 seconds, instead of the above 
>> 20 seconds) ?
>>
>> Do you still see the number of threads remaining at the maximum ?
>>
>> See :  
>> http://tomcat.apache.org/tomcat-8.0-doc/config/http.html#Standard_Implementation
>> --> connectionTimeout
>> and the fact that it is also the default for
>> keepAliveTimeout
>>
>>
>> On 14.01.2017 07:30, smith wrote:
>>> The server.xml:
>>>
>>> 
>>> 
>>> 
>>> 
>>> >> />
>>> 
>>> 
>>> >> SSLEngine="on" />
>>> 
>>> >> className="org.apache.catalina.core.JreMemoryLeakPreventionListener" />
>>> >> className="org.apache.catalina.mbeans.GlobalResourcesLifecycleListener" />
>>> >> className="org.apache.catalina.core.ThreadLocalLeakPreventionListener" />
>>>
>>> 
>>> 
>>>   
>>>   >> type="org.apache.catalina.UserDatabase"
>>> description="User database that can be updated and saved"
>>> 
>>> factory="org.apache.catalina.users.MemoryUserDatabaseFactory"
>>> pathname="conf/tomcat-users.xml" />
>>> 
>>>
>>> 
>>> 
>>>
>>>   
>>>   
>>>
>>>
>>>   
>>>   >>  maxThreads="300" connectionTimeout="2"
>>>  redirectPort="8443" />
>>>   
>>>   
>>>   
>>>   
>>>
>>>   
>>>   
>>>
>>>
>>>       
>>>
>>>   
>>>   
>>>
>>> 
>>> 
>>>
>>> 
>>> 
>>>   
>>>   >>  resourceName="UserDatabase"/>
>>> 
>>>
>>> >>   unpackWARs="true" autoDeploy="true">
>>>
>>>   
>>>   
>>>
>>>   
>>>
>>> >> directory="logs"
>>>  prefix="localhost_access_log" suffix=".txt"
>>>  pattern="%h,%t,%m,%U,%H,%s,%B,%D,%{User-Agent}i" />
>>>
>>>   >> docBase="//t" sessionCookieName="" />
>>> 
>>>   
>>> 
>>> 
>>>
>>> -Original Message-
>>> From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
>>> Sent: Friday, January 13, 2017 10:42 AM
>>> To:  users@tomcat.apache.org
>>> Subject: Re: FW: tomcat 8080 thread not reduced
>>>
>>> On 13.01.2017 09:38, smith wrote:
>>>>
>>>>
>>>>
>>>>
>>>> From: smith [mailto:smith@zoom.us]
>>>> Sent: Tuesday, January 10, 2017 9:57 AM
>>>> To: 'users'
>>>> Subject: tomcat 8080 thread not reduced
>>>>
>>>>
>>>>
>>>> Hi,
>>>>
>>>>
>>>>
>>>> We have installed Apache Tomcat/8.0.14, and found that after one period of 
>>>> time, the thread

Re: FW: tomcat 8080 thread not reduced

[tomcat]

On 16.01.2017 03:41, Smith Hua wrote:


actually there is not much busy threads, less tahn 10,so i think this parameter 
may has nothing to do with this


It depends on what you call "busy". What are the busy ones doing ? and what are the 
non-busy ones doing ?




--
从myMail的Android专用app所发送 星期一, 16 一月 2017, 03:01上午 +08:00 发件人 André Warnier 
(tomcat)  a...@ice-sa.com :


Hi.

I can find nothing really wrong in your configuration below.
But, what happens if in this section :

  >   maxThreads="300" connectionTimeout="2"
  > redirectPort="8443" />

you change the connectionTimeout to 3000 (= 3 seconds, instead of the above 20 
seconds) ?

Do you still see the number of threads remaining at the maximum ?

See :  
http://tomcat.apache.org/tomcat-8.0-doc/config/http.html#Standard_Implementation
--> connectionTimeout
and the fact that it is also the default for
keepAliveTimeout


On 14.01.2017 07:30, smith wrote:

The server.xml:
















  
  





  
  


  
  
  
  
  
  

  
  


  

  
  






  
  




  
  

  



  

  



-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
Sent: Friday, January 13, 2017 10:42 AM
To:  users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 13.01.2017 09:38, smith wrote:





From: smith [mailto:smith@zoom.us]
Sent: Tuesday, January 10, 2017 9:57 AM
To: 'users'
Subject: tomcat 8080 thread not reduced



Hi,



We have installed Apache Tomcat/8.0.14, and found that after one period of 
time, the thread count for 8080(our port published) goes to 120 and never 
reduced even the busy count is only 3-4.

Why? Tomcat8 not reduced the thread pool even the thread is idle, and the 
minSpareThreads for tomcat8 default is only 10.

When will the thread reduce?





Best regards

Smith




Hi.

Please copy/paste your complete server.xml configuration file (confidential 
things removed), so that we could have a useful look at it.
Please edit *only* the confidential things, not entire sections. Often, the 
issue is in the details.



-
To unsubscribe, e-mail:  users-unsubscr...@tomcat.apache.org
For additional commands, e-mail:  users-h...@tomcat.apache.org



-
To unsubscribe, e-mail:  users-unsubscr...@tomcat.apache.org
For additional commands, e-mail:  users-h...@tomcat.apache.org




-
To unsubscribe, e-mail:  users-unsubscr...@tomcat.apache.org
For additional commands, e-mail:  users-h...@tomcat.apache.org




-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re[2]: FW: tomcat 8080 thread not reduced

[Smith Hua]

actually there is not much busy threads, less tahn 10,so i think this parameter 
may has nothing to do with this
--
从myMail的Android专用app所发送 星期一, 16 一月 2017, 03:01上午 +08:00 发件人 André Warnier 
(tomcat)  a...@ice-sa.com :

>Hi.
>
>I can find nothing really wrong in your configuration below.
>But, what happens if in this section :
>
> >   > maxThreads="300" connectionTimeout="2"
> > redirectPort="8443" />
>
>you change the connectionTimeout to 3000 (= 3 seconds, instead of the above 20 
>seconds) ?
>
>Do you still see the number of threads remaining at the maximum ?
>
>See :  
>http://tomcat.apache.org/tomcat-8.0-doc/config/http.html#Standard_Implementation
>--> connectionTimeout
>and the fact that it is also the default for
>keepAliveTimeout
>
>
>On 14.01.2017 07:30, smith wrote:
>> The server.xml:
>>
>> 
>> 
>> 
>> 
>>
>>
>>
>>> SSLEngine="on" />
>>
>>> className="org.apache.catalina.core.JreMemoryLeakPreventionListener" />
>>> className="org.apache.catalina.mbeans.GlobalResourcesLifecycleListener" />
>>> className="org.apache.catalina.core.ThreadLocalLeakPreventionListener" />
>>
>>
>>
>>  
>>  >type="org.apache.catalina.UserDatabase"
>>description="User database that can be updated and saved"
>>factory="org.apache.catalina.users.MemoryUserDatabaseFactory"
>>pathname="conf/tomcat-users.xml" />
>>
>>
>>
>>
>>
>>  
>>  
>>
>>
>>  
>>  > maxThreads="300" connectionTimeout="2"
>> redirectPort="8443" />
>>  
>>  
>>  
>>  
>>
>>  
>>  
>>
>>
>>  
>>
>>  
>>  
>>
>>
>>
>>
>>
>>
>>  
>>  > resourceName="UserDatabase"/>
>>
>>
>>>  unpackWARs="true" autoDeploy="true">
>>
>>  
>>  
>>
>>  
>>
>>  > directory="logs"
>> prefix="localhost_access_log" suffix=".txt"
>> pattern="%h,%t,%m,%U,%H,%s,%B,%D,%{User-Agent}i" />
>>
>>  > docBase="//t" sessionCookieName="" />
>>
>>  
>>
>> 
>>
>> -Original Message-
>> From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
>> Sent: Friday, January 13, 2017 10:42 AM
>> To:  users@tomcat.apache.org
>> Subject: Re: FW: tomcat 8080 thread not reduced
>>
>> On 13.01.2017 09:38, smith wrote:
>>>
>>>
>>>
>>>
>>> From: smith [mailto:smith@zoom.us]
>>> Sent: Tuesday, January 10, 2017 9:57 AM
>>> To: 'users'
>>> Subject: tomcat 8080 thread not reduced
>>>
>>>
>>>
>>> Hi,
>>>
>>>
>>>
>>> We have installed Apache Tomcat/8.0.14, and found that after one period of 
>>> time, the thread count for 8080(our port published) goes to 120 and never 
>>> reduced even the busy count is only 3-4.
>>>
>>> Why? Tomcat8 not reduced the thread pool even the thread is idle, and the 
>>> minSpareThreads for tomcat8 default is only 10.
>>>
>>> When will the thread reduce?
>>>
>>>
>>>
>>>
>>>
>>> Best regards
>>>
>>> Smith
>>>
>>>
>>
>> Hi.
>>
>> Please copy/paste your complete server.xml configuration file (confidential 
>> things removed), so that we could have a useful look at it.
>> Please edit *only* the confidential things, not entire sections. Often, the 
>> issue is in the details.
>>
>>
>>
>> -
>> To unsubscribe, e-mail:  users-unsubscr...@tomcat.apache.org
>> For additional commands, e-mail:  users-h...@tomcat.apache.org
>>
>>
>>
>> -
>> To unsubscribe, e-mail:  users-unsubscr...@tomcat.apache.org
>> For additional commands, e-mail:  users-h...@tomcat.apache.org
>>
>
>
>-
>To unsubscribe, e-mail:  users-unsubscr...@tomcat.apache.org
>For additional commands, e-mail:  users-h...@tomcat.apache.org
>

Re: FW: tomcat 8080 thread not reduced

[tomcat]

Hi.

I can find nothing really wrong in your configuration below.
But, what happens if in this section :

>   maxThreads="300" connectionTimeout="2"
> redirectPort="8443" />

you change the connectionTimeout to 3000 (= 3 seconds, instead of the above 20 
seconds) ?

Do you still see the number of threads remaining at the maximum ?

See : 
http://tomcat.apache.org/tomcat-8.0-doc/config/http.html#Standard_Implementation
--> connectionTimeout
and the fact that it is also the default for
keepAliveTimeout


On 14.01.2017 07:30, smith wrote:

The server.xml:





   
   
   
   
   
   
   
   

   
   
 
 
   

   
   

 
 


 
 
 
 
 
 

 
 


 

 
 

   
   

   
   
 
 
   

   

 
 

 



 
   
 
   


-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com]
Sent: Friday, January 13, 2017 10:42 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 13.01.2017 09:38, smith wrote:





From: smith [mailto:smith@zoom.us]
Sent: Tuesday, January 10, 2017 9:57 AM
To: 'users'
Subject: tomcat 8080 thread not reduced



Hi,



We have installed Apache Tomcat/8.0.14, and found that after one period of 
time, the thread count for 8080(our port published) goes to 120 and never 
reduced even the busy count is only 3-4.

Why? Tomcat8 not reduced the thread pool even the thread is idle, and the 
minSpareThreads for tomcat8 default is only 10.

When will the thread reduce?





Best regards

Smith




Hi.

Please copy/paste your complete server.xml configuration file (confidential 
things removed), so that we could have a useful look at it.
Please edit *only* the confidential things, not entire sections. Often, the 
issue is in the details.



-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org



-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org




-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

RE: FW: tomcat 8080 thread not reduced

[smith]

The server.xml:





  
  
  
  
  
  
  
  

  
  


  

  
  





















  
  

  
  


  

  









  

  


-Original Message-
From: André Warnier (tomcat) [mailto:a...@ice-sa.com] 
Sent: Friday, January 13, 2017 10:42 AM
To: users@tomcat.apache.org
Subject: Re: FW: tomcat 8080 thread not reduced

On 13.01.2017 09:38, smith wrote:
>
>
>
>
> From: smith [mailto:smith@zoom.us]
> Sent: Tuesday, January 10, 2017 9:57 AM
> To: 'users'
> Subject: tomcat 8080 thread not reduced
>
>
>
> Hi,
>
>
>
> We have installed Apache Tomcat/8.0.14, and found that after one period of 
> time, the thread count for 8080(our port published) goes to 120 and never 
> reduced even the busy count is only 3-4.
>
> Why? Tomcat8 not reduced the thread pool even the thread is idle, and the 
> minSpareThreads for tomcat8 default is only 10.
>
> When will the thread reduce?
>
>
>
>
>
> Best regards
>
> Smith
>
>

Hi.

Please copy/paste your complete server.xml configuration file (confidential 
things removed), so that we could have a useful look at it.
Please edit *only* the confidential things, not entire sections. Often, the 
issue is in the details.



-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org



-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: FW: tomcat 8080 thread not reduced

[tomcat]

On 13.01.2017 09:38, smith wrote:





From: smith [mailto:smith@zoom.us]
Sent: Tuesday, January 10, 2017 9:57 AM
To: 'users'
Subject: tomcat 8080 thread not reduced



Hi,



We have installed Apache Tomcat/8.0.14, and found that after one period of 
time, the thread count for 8080(our port published) goes to 120 and never 
reduced even the busy count is only 3-4.

Why? Tomcat8 not reduced the thread pool even the thread is idle, and the 
minSpareThreads for tomcat8 default is only 10.

When will the thread reduce?





Best regards

Smith




Hi.

Please copy/paste your complete server.xml configuration file (confidential things 
removed), so that we could have a useful look at it.
Please edit *only* the confidential things, not entire sections. Often, the issue is in 
the details.




-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

FW: tomcat 8080 thread not reduced

[smith]

 

 

From: smith [mailto:smith@zoom.us] 
Sent: Tuesday, January 10, 2017 9:57 AM
To: 'users'
Subject: tomcat 8080 thread not reduced

 

Hi,

 

We have installed Apache Tomcat/8.0.14, and found that after one period of 
time, the thread count for 8080(our port published) goes to 120 and never 
reduced even the busy count is only 3-4. 

Why? Tomcat8 not reduced the thread pool even the thread is idle, and the 
minSpareThreads for tomcat8 default is only 10.

When will the thread reduce?

 

 

Best regards

Smith