I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed what appears
to be a handle leak. When I execute a DB2 connect statement in a command
window, followed by a DB2 connect reset statement, The number of handles
displayed in Windows Task Manager for the db2syscs process does not return
to its original value, but to 1 handle higher.
While this is not a serious issue most of the time, we have applications
connecting every 30 seconds or so and they are expected to have up-times of
months. We have seen the handles listed at ~20,000, when they start at
~300.
Can anyone confirm whether this is a real issue or is this just a display
oddity in Task Manager?
12  3881 
Ken Brubaker wrote: I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed what
appears
to be a handle leak. When I execute a DB2 connect statement in a command
window, followed by a DB2 connect reset statement, The number of handles
displayed in Windows Task Manager for the db2syscs process does not return
to its original value, but to 1 handle higher.
While this is not a serious issue most of the time, we have applications
connecting every 30 seconds or so and they are expected to have up-times
of
months. We have seen the handles listed at ~20,000, when they start at
~300.
Can anyone confirm whether this is a real issue or is this just a display
oddity in Task Manager?
Maybe you are using agent pooling? What's the setting of the dbm cfg
parameter NUM_POOLAGENTS?
--
Knut Stolze
Information Integration
IBM Germany / University of Jena
The NUM_POOLAGENTS has been left at the default value of 100.
"Knut Stolze" <st****@de.ibm.com> wrote in message
news:c5**********@fsuj29.rz.uni-jena.de... Ken Brubaker wrote:
I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed what
appears
to be a handle leak. When I execute a DB2 connect statement in a
command window, followed by a DB2 connect reset statement, The number of handles
displayed in Windows Task Manager for the db2syscs process does not
return to its original value, but to 1 handle higher.
While this is not a serious issue most of the time, we have applications
connecting every 30 seconds or so and they are expected to have up-times
of
months. We have seen the handles listed at ~20,000, when they start at
~300.
Can anyone confirm whether this is a real issue or is this just a
display oddity in Task Manager?
Maybe you are using agent pooling? What's the setting of the dbm cfg
parameter NUM_POOLAGENTS?
--
Knut Stolze
Information Integration
IBM Germany / University of Jena
Ken Brubaker wrote: The NUM_POOLAGENTS has been left at the default value of 100.
What is NUM_INITAGENTS set to? I believe the default is 0.
NUM_POOLAGENTS defines how many agents DB2 will keep alive (after
an application disconnects from the database) in order to avoid
the overhead of starting new agents.
NUM_INITAGENTS specifies the number of agents DB2 starts when the
instance is started.
If NUM_INITAGENTS < NUM_POOLAGENTS, then you could see the behavior
that you describe, up to the point where you have > NUM_POOLAGENTS
active connections to the database.
Good luck,
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =----- http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----
Anyone can tell if what's displayed in the task manager as handles are the
file/object handles or some other type?
sysinternals has a tool called handle.exe that shows them / the windows
handles.
I think we need clarification on the term handle here because i don't think
we're talking about this :
SQLAllocHandle() is a generic function that allocates environment,
connection, statement, or descriptor handles.
I would read on
Maximum Total Files Open configuration parameter - maxtotfilop
and get snapshots of 'files closed' to see if you're in a bad situation.
PM
"Ken Brubaker" <ke**********@interconsecurity.com> a écrit dans le message
de news:fu********************@news20.bellglobal.com. .. I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed what
appears to be a handle leak. When I execute a DB2 connect statement in a command
window, followed by a DB2 connect reset statement, The number of handles
displayed in Windows Task Manager for the db2syscs process does not return
to its original value, but to 1 handle higher.
While this is not a serious issue most of the time, we have applications
connecting every 30 seconds or so and they are expected to have up-times
of months. We have seen the handles listed at ~20,000, when they start at
~300.
Can anyone confirm whether this is a real issue or is this just a display
oddity in Task Manager?
Yes, I am talking about Windows handles, not DB2 handles allocated through
an SQLAllocHandle() call.
Using that application from sysinternals, I have the following information:
When I start DB2, I took a handle list from the db2syscs process. On the
command line, I connected to my database and disconnected. I took a handle
list again and diff'ed the two captures. I had two new 'thread' handles and
6 'event' handles listed. Since connection caching is occurring, this is
not surprising.
After that, every time I issue a DB2 Connect to db followed by a DB2 Connect
reset, I show one more 'event' handle in the output. These handles do not
seem to ever get reclaimed by the system until I issue a DB2stop command.
This sounds to me like it is a resource leak in DB2, specifically the
db2syscs process.
MAXTOTFILOP is set to the default of 16000.
"PM (pm3iinc-nospam) CGO" <PM (pm3iinc-nospam)@cgocable.ca> wrote in message
news:0W*****************@charlie.risq.qc.ca... Anyone can tell if what's displayed in the task manager as handles are the
file/object handles or some other type?
sysinternals has a tool called handle.exe that shows them / the windows
handles.
I think we need clarification on the term handle here because i don't
think we're talking about this :
SQLAllocHandle() is a generic function that allocates environment,
connection, statement, or descriptor handles.
I would read on
Maximum Total Files Open configuration parameter - maxtotfilop
and get snapshots of 'files closed' to see if you're in a bad situation.
PM
"Ken Brubaker" <ke**********@interconsecurity.com> a écrit dans le message
de news:fu********************@news20.bellglobal.com. .. I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed what
appears to be a handle leak. When I execute a DB2 connect statement in a
command window, followed by a DB2 connect reset statement, The number of handles
displayed in Windows Task Manager for the db2syscs process does not
return to its original value, but to 1 handle higher.
While this is not a serious issue most of the time, we have applications
connecting every 30 seconds or so and they are expected to have up-times
of months. We have seen the handles listed at ~20,000, when they start at
~300.
Can anyone confirm whether this is a real issue or is this just a
display oddity in Task Manager?
Write a small script to loop over connect / resets and see whether db2syscs
tops out at some number of handles or not. If it doesn't, open a PMR against
the handle leak (failure to call the Win32 CloseHandle API).
"Ken Brubaker" <ke**********@interconsecurity.com> wrote in message
news:AR*********************@news20.bellglobal.com ... Yes, I am talking about Windows handles, not DB2 handles allocated through
an SQLAllocHandle() call.
Using that application from sysinternals, I have the following
information:
When I start DB2, I took a handle list from the db2syscs process. On the
command line, I connected to my database and disconnected. I took a
handle list again and diff'ed the two captures. I had two new 'thread' handles
and 6 'event' handles listed. Since connection caching is occurring, this is
not surprising.
After that, every time I issue a DB2 Connect to db followed by a DB2
Connect reset, I show one more 'event' handle in the output. These handles do not
seem to ever get reclaimed by the system until I issue a DB2stop command.
This sounds to me like it is a resource leak in DB2, specifically the
db2syscs process.
MAXTOTFILOP is set to the default of 16000.
"PM (pm3iinc-nospam) CGO" <PM (pm3iinc-nospam)@cgocable.ca> wrote in
message news:0W*****************@charlie.risq.qc.ca... Anyone can tell if what's displayed in the task manager as handles are
the file/object handles or some other type?
sysinternals has a tool called handle.exe that shows them / the windows
handles.
I think we need clarification on the term handle here because i don't
think we're talking about this :
SQLAllocHandle() is a generic function that allocates environment,
connection, statement, or descriptor handles.
I would read on
Maximum Total Files Open configuration parameter - maxtotfilop
and get snapshots of 'files closed' to see if you're in a bad situation.
PM
"Ken Brubaker" <ke**********@interconsecurity.com> a écrit dans le
message de news:fu********************@news20.bellglobal.com. .. I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed what
appears to be a handle leak. When I execute a DB2 connect statement in a command window, followed by a DB2 connect reset statement, The number of
handles displayed in Windows Task Manager for the db2syscs process does not return to its original value, but to 1 handle higher.
While this is not a serious issue most of the time, we have
applications connecting every 30 seconds or so and they are expected to have
up-times of months. We have seen the handles listed at ~20,000, when they start
at ~300.
Can anyone confirm whether this is a real issue or is this just a
display oddity in Task Manager?
Ran a small script over the weekend:
:start
db2 connect to dbname
sleep 1
db2 connect reset
sleep 1
goto :start
This morning when I arrived, Task Manager shows 102,121 handles allocated to
the db2syscs.exe service. When you recommend I submit a PMR, can you tell
me where to do this? Also, what does PMR mean?
"Mark Yudkin" <my***********************@nospam.org> wrote in message
news:c5**********@ngspool-d02.news.aol.com... Write a small script to loop over connect / resets and see whether
db2syscs tops out at some number of handles or not. If it doesn't, open a PMR
against the handle leak (failure to call the Win32 CloseHandle API).
"Ken Brubaker" <ke**********@interconsecurity.com> wrote in message
news:AR*********************@news20.bellglobal.com ... Yes, I am talking about Windows handles, not DB2 handles allocated
through an SQLAllocHandle() call.
Using that application from sysinternals, I have the following
information:
When I start DB2, I took a handle list from the db2syscs process. On
the command line, I connected to my database and disconnected. I took a
handle list again and diff'ed the two captures. I had two new 'thread' handles
and 6 'event' handles listed. Since connection caching is occurring, this
is not surprising.
After that, every time I issue a DB2 Connect to db followed by a DB2
Connect reset, I show one more 'event' handle in the output. These handles do
not seem to ever get reclaimed by the system until I issue a DB2stop
command. This sounds to me like it is a resource leak in DB2, specifically the
db2syscs process.
MAXTOTFILOP is set to the default of 16000.
"PM (pm3iinc-nospam) CGO" <PM (pm3iinc-nospam)@cgocable.ca> wrote in
message news:0W*****************@charlie.risq.qc.ca... Anyone can tell if what's displayed in the task manager as handles are
the file/object handles or some other type?
sysinternals has a tool called handle.exe that shows them / the
windows handles.
I think we need clarification on the term handle here because i don't
think we're talking about this :
SQLAllocHandle() is a generic function that allocates environment,
connection, statement, or descriptor handles.
I would read on
Maximum Total Files Open configuration parameter - maxtotfilop
and get snapshots of 'files closed' to see if you're in a bad
situation.
PM
"Ken Brubaker" <ke**********@interconsecurity.com> a écrit dans le message de news:fu********************@news20.bellglobal.com. ..
> I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed what
appears
> to be a handle leak. When I execute a DB2 connect statement in a
command > window, followed by a DB2 connect reset statement, The number of handles > displayed in Windows Task Manager for the db2syscs process does not
return > to its original value, but to 1 handle higher.
>
> While this is not a serious issue most of the time, we have applications > connecting every 30 seconds or so and they are expected to have up-times of
> months. We have seen the handles listed at ~20,000, when they start at > ~300.
>
> Can anyone confirm whether this is a real issue or is this just a
display > oddity in Task Manager?
>
>
Can you inform me of how or where to open a PMR on this issue? I ran a
script over the weekend as follows:
:start
db2 connect to dbname
sleep 1
db2 connect reset
sleep 1
goto start
This morning, db2syscs.exe showed 102,121 handles allocated according to
Windows Task Manager.
"Mark Yudkin" <my***********************@nospam.org> wrote in message
news:c5**********@ngspool-d02.news.aol.com... Write a small script to loop over connect / resets and see whether
db2syscs tops out at some number of handles or not. If it doesn't, open a PMR
against the handle leak (failure to call the Win32 CloseHandle API).
"Ken Brubaker" <ke**********@interconsecurity.com> wrote in message
news:AR*********************@news20.bellglobal.com ... Yes, I am talking about Windows handles, not DB2 handles allocated
through an SQLAllocHandle() call.
Using that application from sysinternals, I have the following
information:
When I start DB2, I took a handle list from the db2syscs process. On
the command line, I connected to my database and disconnected. I took a
handle list again and diff'ed the two captures. I had two new 'thread' handles
and 6 'event' handles listed. Since connection caching is occurring, this
is not surprising.
After that, every time I issue a DB2 Connect to db followed by a DB2
Connect reset, I show one more 'event' handle in the output. These handles do
not seem to ever get reclaimed by the system until I issue a DB2stop
command. This sounds to me like it is a resource leak in DB2, specifically the
db2syscs process.
MAXTOTFILOP is set to the default of 16000.
"PM (pm3iinc-nospam) CGO" <PM (pm3iinc-nospam)@cgocable.ca> wrote in
message news:0W*****************@charlie.risq.qc.ca... Anyone can tell if what's displayed in the task manager as handles are
the file/object handles or some other type?
sysinternals has a tool called handle.exe that shows them / the
windows handles.
I think we need clarification on the term handle here because i don't
think we're talking about this :
SQLAllocHandle() is a generic function that allocates environment,
connection, statement, or descriptor handles.
I would read on
Maximum Total Files Open configuration parameter - maxtotfilop
and get snapshots of 'files closed' to see if you're in a bad
situation.
PM
"Ken Brubaker" <ke**********@interconsecurity.com> a écrit dans le message de news:fu********************@news20.bellglobal.com. ..
> I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed what
appears
> to be a handle leak. When I execute a DB2 connect statement in a
command > window, followed by a DB2 connect reset statement, The number of handles > displayed in Windows Task Manager for the db2syscs process does not
return > to its original value, but to 1 handle higher.
>
> While this is not a serious issue most of the time, we have applications > connecting every 30 seconds or so and they are expected to have up-times of
> months. We have seen the handles listed at ~20,000, when they start at > ~300.
>
> Can anyone confirm whether this is a real issue or is this just a
display > oddity in Task Manager?
>
>
PMR = Problem Management Report
You can open a PMR by creating a support request with IBM. http://www-306.ibm.com/software/support/probsub.html
--
Matt Emmerton
IBM Toronto Lab
DB2 OLTP Performance
"Ken Brubaker" <ke**********@interconsecurity.com> wrote in message
news:XO********************@news20.bellglobal.com. .. Can you inform me of how or where to open a PMR on this issue? I ran a
script over the weekend as follows:
:start
db2 connect to dbname
sleep 1
db2 connect reset
sleep 1
goto start
This morning, db2syscs.exe showed 102,121 handles allocated according to
Windows Task Manager.
"Mark Yudkin" <my***********************@nospam.org> wrote in message
news:c5**********@ngspool-d02.news.aol.com... Write a small script to loop over connect / resets and see whether
db2syscs tops out at some number of handles or not. If it doesn't, open a PMR
against the handle leak (failure to call the Win32 CloseHandle API).
"Ken Brubaker" <ke**********@interconsecurity.com> wrote in message
news:AR*********************@news20.bellglobal.com ... Yes, I am talking about Windows handles, not DB2 handles allocated through an SQLAllocHandle() call.
Using that application from sysinternals, I have the following
information:
When I start DB2, I took a handle list from the db2syscs process. On the command line, I connected to my database and disconnected. I took a
handle list again and diff'ed the two captures. I had two new 'thread'
handles and 6 'event' handles listed. Since connection caching is occurring, this is not surprising.
After that, every time I issue a DB2 Connect to db followed by a DB2
Connect reset, I show one more 'event' handle in the output. These handles do not seem to ever get reclaimed by the system until I issue a DB2stop command. This sounds to me like it is a resource leak in DB2, specifically the
db2syscs process.
MAXTOTFILOP is set to the default of 16000.
"PM (pm3iinc-nospam) CGO" <PM (pm3iinc-nospam)@cgocable.ca> wrote in
message news:0W*****************@charlie.risq.qc.ca...
> Anyone can tell if what's displayed in the task manager as handles
are the > file/object handles or some other type?
> sysinternals has a tool called handle.exe that shows them / the windows > handles.
>
> I think we need clarification on the term handle here because i
don't think
> we're talking about this :
> SQLAllocHandle() is a generic function that allocates environment,
> connection, statement, or descriptor handles.
>
> I would read on
> Maximum Total Files Open configuration parameter - maxtotfilop
> and get snapshots of 'files closed' to see if you're in a bad
situation. >
> PM
>
> "Ken Brubaker" <ke**********@interconsecurity.com> a écrit dans le
message > de news:fu********************@news20.bellglobal.com. ..
> > I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed
what > appears
> > to be a handle leak. When I execute a DB2 connect statement in a
command
> > window, followed by a DB2 connect reset statement, The number of
handles > > displayed in Windows Task Manager for the db2syscs process does
not return
> > to its original value, but to 1 handle higher.
> >
> > While this is not a serious issue most of the time, we have
applications > > connecting every 30 seconds or so and they are expected to have
up-times > of
> > months. We have seen the handles listed at ~20,000, when they
start at > > ~300.
> >
> > Can anyone confirm whether this is a real issue or is this just a
display
> > oddity in Task Manager?
> >
> >
>
>
PMR = Problem Management Report
You can open a PMR by creating a support request with IBM. http://www-306.ibm.com/software/support/probsub.html
--
Matt Emmerton
IBM Toronto Lab
DB2 OLTP Performance
"Ken Brubaker" <ke**********@interconsecurity.com> wrote in message
news:XO********************@news20.bellglobal.com. .. Can you inform me of how or where to open a PMR on this issue? I ran a
script over the weekend as follows:
:start
db2 connect to dbname
sleep 1
db2 connect reset
sleep 1
goto start
This morning, db2syscs.exe showed 102,121 handles allocated according to
Windows Task Manager.
"Mark Yudkin" <my***********************@nospam.org> wrote in message
news:c5**********@ngspool-d02.news.aol.com... Write a small script to loop over connect / resets and see whether
db2syscs tops out at some number of handles or not. If it doesn't, open a PMR
against the handle leak (failure to call the Win32 CloseHandle API).
"Ken Brubaker" <ke**********@interconsecurity.com> wrote in message
news:AR*********************@news20.bellglobal.com ... Yes, I am talking about Windows handles, not DB2 handles allocated through an SQLAllocHandle() call.
Using that application from sysinternals, I have the following
information:
When I start DB2, I took a handle list from the db2syscs process. On the command line, I connected to my database and disconnected. I took a
handle list again and diff'ed the two captures. I had two new 'thread'
handles and 6 'event' handles listed. Since connection caching is occurring, this is not surprising.
After that, every time I issue a DB2 Connect to db followed by a DB2
Connect reset, I show one more 'event' handle in the output. These handles do not seem to ever get reclaimed by the system until I issue a DB2stop command. This sounds to me like it is a resource leak in DB2, specifically the
db2syscs process.
MAXTOTFILOP is set to the default of 16000.
"PM (pm3iinc-nospam) CGO" <PM (pm3iinc-nospam)@cgocable.ca> wrote in
message news:0W*****************@charlie.risq.qc.ca...
> Anyone can tell if what's displayed in the task manager as handles
are the > file/object handles or some other type?
> sysinternals has a tool called handle.exe that shows them / the windows > handles.
>
> I think we need clarification on the term handle here because i
don't think
> we're talking about this :
> SQLAllocHandle() is a generic function that allocates environment,
> connection, statement, or descriptor handles.
>
> I would read on
> Maximum Total Files Open configuration parameter - maxtotfilop
> and get snapshots of 'files closed' to see if you're in a bad
situation. >
> PM
>
> "Ken Brubaker" <ke**********@interconsecurity.com> a écrit dans le
message > de news:fu********************@news20.bellglobal.com. ..
> > I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed
what > appears
> > to be a handle leak. When I execute a DB2 connect statement in a
command
> > window, followed by a DB2 connect reset statement, The number of
handles > > displayed in Windows Task Manager for the db2syscs process does
not return
> > to its original value, but to 1 handle higher.
> >
> > While this is not a serious issue most of the time, we have
applications > > connecting every 30 seconds or so and they are expected to have
up-times > of
> > months. We have seen the handles listed at ~20,000, when they
start at > > ~300.
> >
> > Can anyone confirm whether this is a real issue or is this just a
display
> > oddity in Task Manager?
> >
> >
>
>
A PMR (Problem Management Record) is the way you get IBM to investigate and
track problems (known as an APARs) and their resolutions, and coordinate
within IBM and between you and IBM. Your "local" support centre will handle
all communication.
Info on opening PMRs in your country should be in your IBM support contract,
also Matt gave a URL. If you don't have a support contract, you'll need to
open a PMR for your specific case, in which case you may be asked to sign
some form of case-specific contract. Contact information for your country
was in the box your DB2 came in.
"Ken Brubaker" <ke**********@interconsecurity.com> wrote in message
news:XO********************@news20.bellglobal.com. .. Ran a small script over the weekend:
:start
db2 connect to dbname
sleep 1
db2 connect reset
sleep 1
goto :start
This morning when I arrived, Task Manager shows 102,121 handles allocated
to the db2syscs.exe service. When you recommend I submit a PMR, can you tell
me where to do this? Also, what does PMR mean?
"Mark Yudkin" <my***********************@nospam.org> wrote in message
news:c5**********@ngspool-d02.news.aol.com... Write a small script to loop over connect / resets and see whether
db2syscs tops out at some number of handles or not. If it doesn't, open a PMR
against the handle leak (failure to call the Win32 CloseHandle API).
"Ken Brubaker" <ke**********@interconsecurity.com> wrote in message
news:AR*********************@news20.bellglobal.com ... Yes, I am talking about Windows handles, not DB2 handles allocated through an SQLAllocHandle() call.
Using that application from sysinternals, I have the following
information:
When I start DB2, I took a handle list from the db2syscs process. On the command line, I connected to my database and disconnected. I took a
handle list again and diff'ed the two captures. I had two new 'thread'
handles and 6 'event' handles listed. Since connection caching is occurring, this is not surprising.
After that, every time I issue a DB2 Connect to db followed by a DB2
Connect reset, I show one more 'event' handle in the output. These handles do not seem to ever get reclaimed by the system until I issue a DB2stop command. This sounds to me like it is a resource leak in DB2, specifically the
db2syscs process.
MAXTOTFILOP is set to the default of 16000.
"PM (pm3iinc-nospam) CGO" <PM (pm3iinc-nospam)@cgocable.ca> wrote in
message news:0W*****************@charlie.risq.qc.ca...
> Anyone can tell if what's displayed in the task manager as handles
are the > file/object handles or some other type?
> sysinternals has a tool called handle.exe that shows them / the windows > handles.
>
> I think we need clarification on the term handle here because i
don't think
> we're talking about this :
> SQLAllocHandle() is a generic function that allocates environment,
> connection, statement, or descriptor handles.
>
> I would read on
> Maximum Total Files Open configuration parameter - maxtotfilop
> and get snapshots of 'files closed' to see if you're in a bad
situation. >
> PM
>
> "Ken Brubaker" <ke**********@interconsecurity.com> a écrit dans le
message > de news:fu********************@news20.bellglobal.com. ..
> > I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed
what > appears
> > to be a handle leak. When I execute a DB2 connect statement in a
command
> > window, followed by a DB2 connect reset statement, The number of
handles > > displayed in Windows Task Manager for the db2syscs process does
not return
> > to its original value, but to 1 handle higher.
> >
> > While this is not a serious issue most of the time, we have
applications > > connecting every 30 seconds or so and they are expected to have
up-times > of
> > months. We have seen the handles listed at ~20,000, when they
start at > > ~300.
> >
> > Can anyone confirm whether this is a real issue or is this just a
display
> > oddity in Task Manager?
> >
> >
>
>
A PMR (Problem Management Record) is the way you get IBM to investigate and
track problems (known as an APARs) and their resolutions, and coordinate
within IBM and between you and IBM. Your "local" support centre will handle
all communication.
Info on opening PMRs in your country should be in your IBM support contract,
also Matt gave a URL. If you don't have a support contract, you'll need to
open a PMR for your specific case, in which case you may be asked to sign
some form of case-specific contract. Contact information for your country
was in the box your DB2 came in.
"Ken Brubaker" <ke**********@interconsecurity.com> wrote in message
news:XO********************@news20.bellglobal.com. .. Ran a small script over the weekend:
:start
db2 connect to dbname
sleep 1
db2 connect reset
sleep 1
goto :start
This morning when I arrived, Task Manager shows 102,121 handles allocated
to the db2syscs.exe service. When you recommend I submit a PMR, can you tell
me where to do this? Also, what does PMR mean?
"Mark Yudkin" <my***********************@nospam.org> wrote in message
news:c5**********@ngspool-d02.news.aol.com... Write a small script to loop over connect / resets and see whether
db2syscs tops out at some number of handles or not. If it doesn't, open a PMR
against the handle leak (failure to call the Win32 CloseHandle API).
"Ken Brubaker" <ke**********@interconsecurity.com> wrote in message
news:AR*********************@news20.bellglobal.com ... Yes, I am talking about Windows handles, not DB2 handles allocated through an SQLAllocHandle() call.
Using that application from sysinternals, I have the following
information:
When I start DB2, I took a handle list from the db2syscs process. On the command line, I connected to my database and disconnected. I took a
handle list again and diff'ed the two captures. I had two new 'thread'
handles and 6 'event' handles listed. Since connection caching is occurring, this is not surprising.
After that, every time I issue a DB2 Connect to db followed by a DB2
Connect reset, I show one more 'event' handle in the output. These handles do not seem to ever get reclaimed by the system until I issue a DB2stop command. This sounds to me like it is a resource leak in DB2, specifically the
db2syscs process.
MAXTOTFILOP is set to the default of 16000.
"PM (pm3iinc-nospam) CGO" <PM (pm3iinc-nospam)@cgocable.ca> wrote in
message news:0W*****************@charlie.risq.qc.ca...
> Anyone can tell if what's displayed in the task manager as handles
are the > file/object handles or some other type?
> sysinternals has a tool called handle.exe that shows them / the windows > handles.
>
> I think we need clarification on the term handle here because i
don't think
> we're talking about this :
> SQLAllocHandle() is a generic function that allocates environment,
> connection, statement, or descriptor handles.
>
> I would read on
> Maximum Total Files Open configuration parameter - maxtotfilop
> and get snapshots of 'files closed' to see if you're in a bad
situation. >
> PM
>
> "Ken Brubaker" <ke**********@interconsecurity.com> a écrit dans le
message > de news:fu********************@news20.bellglobal.com. ..
> > I am using DB2 8.5 FP5 on Windows 2000 Server and have noticed
what > appears
> > to be a handle leak. When I execute a DB2 connect statement in a
command
> > window, followed by a DB2 connect reset statement, The number of
handles > > displayed in Windows Task Manager for the db2syscs process does
not return
> > to its original value, but to 1 handle higher.
> >
> > While this is not a serious issue most of the time, we have
applications > > connecting every 30 seconds or so and they are expected to have
up-times > of
> > months. We have seen the handles listed at ~20,000, when they
start at > > ~300.
> >
> > Can anyone confirm whether this is a real issue or is this just a
display
> > oddity in Task Manager?
> >
> >
>
>
This thread has been closed and replies have been disabled. Please start a new discussion. Similar topics
by: Yair |
last post by:
Hello,
I've encounter a leak of 50 handles, when trying to query performance
counters through the registry using HKEY_PERFORMANCE_DATA registry...
|
by: Bryan Mayland |
last post by:
I've got a problem with a System.Net.Socket that is failing on
connect. Each time I attempt a connection, 3 handles are allocated,
and when the...
|
by: Manfred Braun |
last post by:
Hi All,
I am writing a proggi, which should monitor some processes. While doing
this, I needed a test-program and wrote one, which does nothing...
|
by: Jayme Pechan |
last post by:
I was wondering if someone could help me understand why it seems that
creating threads using the code below causes handles to seemingly leak.
...
|
by: NickP |
last post by:
Hi there,
This is really crazy!
1. Make a folder
2. Put a text file in the folder
3. Run the following code...
Dim pop As New...
|
by: Pranil Kanderi |
last post by:
I have a huge application with lots of custom controls. The app had a
memory leak but fixed that by calling the dispose method where needed
and now...
|
by: PlayQ |
last post by:
Hello,
I am using the C# example code in the WMEncoder Series 9 SDK to encode a series of audio files but after a while I notice a handle leak in...
|
by: =?Utf-8?B?QU1lcmNlcg==?= |
last post by:
I have a vb.net program that leaks handles as reported by task manager.
After running for a couple of hours, task manager reports over 1000 handles,...
|
by: not_a_commie |
last post by:
My C# app (using managed DirectInput) leaks handles like mad on one of
my test boxes. (The other test boxes are fine.) The handle being
leaked (as...
|
by: concettolabs |
last post by:
In today's business world, businesses are increasingly turning to PowerApps to develop custom business applications. PowerApps is a powerful tool...
|
by: better678 |
last post by:
Question:
Discuss your understanding of the Java platform. Is the statement "Java is interpreted" correct?
Answer:
Java is an object-oriented...
|
by: teenabhardwaj |
last post by:
How would one discover a valid source for learning news, comfort, and help for engineering designs? Covering through piles of books takes a lot of...
|
by: Kemmylinns12 |
last post by:
Blockchain technology has emerged as a transformative force in the business world, offering unprecedented opportunities for innovation and...
|
by: CD Tom |
last post by:
This only shows up in access runtime. When a user select a report from my report menu when they close the report they get a menu I've called Add-ins...
|
by: Naresh1 |
last post by:
What is WebLogic Admin Training?
WebLogic Admin Training is a specialized program designed to equip individuals with the skills and knowledge...
|
by: Matthew3360 |
last post by:
Hi, I have a python app that i want to be able to get variables from a php page on my webserver. My python app is on my computer. How would I make it...
|
by: AndyPSV |
last post by:
HOW CAN I CREATE AN AI with an .executable file that would suck all files in the folder and on my computerHOW CAN I CREATE AN AI with an .executable...
|
by: Matthew3360 |
last post by:
Hi,
I have been trying to connect to a local host using php curl. But I am finding it hard to do this. I am doing the curl get request from my web...
| |
