Running out of available connections
Hi, we have been used OO4O for about 5 years in an ASP.NET application under Windows 2000, which appear to be fine.
1.5 years ago we upgraded to Windows 2003 where we noticed that connections was not being returned back to the datapool. We had Microsoft help us identity that the OO40 COM object was the root cause (a threading issue for I can remember).
I know that we have done everything in our code to close the connections, but to no still having the same problem(s). What we did was to try to obtain a connction from the datapool (50) which worked but for any reason this failed (99% of the time the connection pool was full) we would make a direct connection to the database (which we closed, destroyed etc).
Over the last 6 months we have rewritten 60% of our Web application to improve performance using ODP.NET and still using OO40 for the remainder 40%. Becauase these new changes did not impact any existing functionality, we only done performance tests on the new framework (ASP.NET V2.0/V3.0, ODP.NET).
We went live over yesturday (Monday 18/08/2008), with no problems until 1pm where we experienced errors on the old version (using OO4O).
We then perform a load test (15 users for about 5-10 minutes) on the old version on one of these page(s) and noticed when the connections to the database hit 23 then these page(s) started to fail.
As soon as these connection dropped, these page(s) started to work fine.
We have 2 version of Oracle Client 9.2 and 10 (ODP.NET), could installing V10 along side 9 cause any problems?
Any help would be great.
Thank you
Hello,
Did you ckeck the TCP/IP port of the SQL Server service? Please verify you had configured the firewall on the computer to allow this instance of SQL Server to accept connections, for example Tcp port 1433, by default.
The following thread is about same issue, you can try to the solution as Nitin post:
http://social.msdn.microsoft.com/Forums/sqlserver/en-US/27684811-1a59-4273-b7ed-3cc990b4f20a/sql-server-error-53-and-sql-server-error-17?forum=sqlgetstarted
The following KB article may related to this issue:
http://support.microsoft.com/kb/817179
Regards,
Fanny Liu
If you have any feedback on our support, please click here.
Fanny Liu
TechNet Community Support
Similar Messages
-
Running out of data connections
Problem Statement:
Running out of data connections after some time and
the following SQLException is thrown:
"failed to create a data connection with any of the
specified drivers"
Problem Description:
I have an application that is running under iPlanet
6.0 on solaris using Oracle 8 and type 2 Oracle
driver.
The application is using a number of stateless
session beans.
Normally, my servlets instantiate the remote beans in
order to make access to their interfaces. For
example, I have a servlet that fetches all of
the accounts for a given customer. First, The servlet
does the fetch by calling JNDI to lookup the "home"
interface, then does a "create" call on the home
interface to get the "remote" interface
(see snippets of code below, the ejbCreate()).
Second, the servlet calls up a specific method that
extracts the customer's accounts (see snippets of
code below, the method getAccountList()).
At first all seem to work well, however after some
random time my beans start to throw the SQLException
"failed to create a data connection with any of the
specified drivers"
At first I thought that I am not closing the
connections, but all of of my connections,
statements, and result sets, are closed
and even set to null in a finally block as you can
see in the snippets of code below.
If I restart iAS, all of the connections get released
and things look normal again till the problem
re-occurs.
Initially, we had this problem on SP2, so we upgraded
to SP3 thinking that this may solve the problem,
however the problem remained on SP3.
Another thing, this same application works perfectly
well on my laptop which has similar iPlanet
configurations except that operating system is
Windows 2000 Advanced Server, and Oracle 8.0.4.0.0.
System Configuration:
Following is my system configuration and related
iPlanet 6.0 settings.
-1- iPlanet 6.0 SP3
-2- Type 2 Oracle driver
-3- SunOs, sparc SUNW, Ultra-250, solaris 5.8
-4- Oracle 8.1.6.0.0 64-bit Production
-5- Following are the entry settings in the iPlanet
registry under
Software\iPlanet\Application Server\6.0\CCS0\DAE2\ORACLE_OCI
- CacheCleanInterval = 120
- CacheConnTimeOut = 120
- CacheDebugMsgs = 0
- CacheFreeSlots = 16
- CacheInitSlots = 64
- CacheMaxConn = 64
- CachMaxGlobalConn = 128
- ConnGiveUpTime = 60
- RMThreadMax = 32
- RMThreadMin = 0
- RSBufferInitRows = 25
- RSBufferMaxBufferSize = 6553600
- RSBufferMaxRows = 100
- RSBufferMaxSize = 32768
- SQLDebugMsgs = 0
Snippets of Code:
The following snippets of code come from a stateless
session bean and consist of a local data member, the
ejbCreate() method, and a typical rmi method that
does the database connection and extraction of
related records.
javax.sql.DataSource dataSourceObj = null;
public void ejbCreate()
throws java.rmi.RemoteException, javax.ejb.CreateException
javax.naming.Context ctx = null;
// Ensure first that the _props have been
// successfully instantiated by the constructor.
if (_props != null) {
try{
ctx = new javax.naming.InitialContext();
}catch (Exception ex){
nbUtility.logError(ex,"Error while creating Initial Context !");
try{
// DEBUG:
System.out.println("ejbCreate(): NB_DATASOURCE = " + props.getPropertyValue(props.NB_DATASOURCE));
// DEBUG:
dataSourceObj = (javax.sql.DataSource) ctx.lookup(_props.getPropertyValue(_props.NB_DATASOURCE));
catch (Exception ex){
ex.printStackTrace();
if (dataSourceObj == null)
throw new javax.ejb.CreateException("Couldn't get DataSource object from environment");
} else {
throw new javax.ejb.CreateException("Couldn't create the property manager: NB_CONFIG_FILE is null, empty, not set, or the file doesn't exist.");
* Given a customer number, this method
* returns the list of accounts that belongs to
* this customers.
public nbAccountList getAccountList(String CustomerNo)
throws java.rmi.RemoteException {
nbAccountList accountList = null;
if ((CustomerNo != null) && (!CustomerNo.equals(""))) {
java.sql.Connection conn = null;
java.sql.Statement stmt = null;
java.sql.ResultSet rset = null;
try {
accountList = new nbAccountList();
String sql = "SELECT * FROM " +
props.getPropertyValue(props.NB_DBTABLE_ACCOUNT) +
" WHERE " +
"(" + props.getPropertyValue(props.NB_DBFIELD_ACCOUNT_CUSTOMERNO) +
"='" + CustomerNo + "')";
// DEBUG:
nbDebug.write("accounts list SQL= " + sql);
// DEBUG:
// Let's get the connection, the statement, and the record set.
conn = dataSourceObj.getConnection();
stmt = conn.createStatement();
rset = stmt.executeQuery(sql);
// Let's loop for each single account
int index = 0;
nbAccount account = null;
while (rset.next()) {
String AccountNo = rset.getString(_props.getPropertyValue(_props.NB_DBFIELD_ACCTPERMIS_ACCOUNTNO));
// Instantiate an account object
account = new nbAccount(AccountNo);
index++;
} catch (SQLException e) {
nbUtility.logError(e, "SQLException while trying to get accounts data.");
} finally {
try {
if (rset != null) { rset.close(); rset = null; }
if (stmt != null) { stmt.close(); stmt = null; }
if (conn != null) { conn.close(); conn = null; }
} catch (SQLException e) {
nbUtility.logError(e, "SQLException while trying to close connection.");
return (accountList);
}I've experienced similar problems. Unfortunately, all efforts by iPlanet
technical support to resolve the issue have failed. (They do keep calling
and asking if they can close the ticket for some reason)
One thing that's totally anoying is the ksvradmin monitory crashes when I
try to have it report any connection pool information. They verified it's a
bug in SP3 but won't say if it's fixed in SP4 or provide an estimate.
To date here's what I've tried (by tech support's recomendation):
1) Configure for global transactions. (even thow I'm not using them)
(Failed)
2) Switch to using 3rd party driver (We were previously using native)
(Failed)
3) Ran report on oracle showing number of connections used during iplanet's
failed attempt. Report from Oracle shows 2 connections open, but iPlanet is
configured for 120.
4) Increased the connection pool size. (I didn't know why based on the
above info) (Increased to 300) (Failed)
Well there's my history. We crash after about 3 days of heavy usage. I'm
about to give up and just reset my servers each night. Will help me with
logfile rotation of the kjs files as well.
Rodger Ball
Sr. Engineer
Business Wire
"Bilal Chouman" <[email protected]> wrote in message
news:[email protected]...
Problem Statement:
Running out of data connections after some time and
the following SQLException is thrown:
"failed to create a data connection with any of the
specified drivers"
Problem Description:
I have an application that is running under iPlanet
6.0 on solaris using Oracle 8 and type 2 Oracle
driver.
The application is using a number of stateless
session beans.
Normally, my servlets instantiate the remote beans in
order to make access to their interfaces. For
example, I have a servlet that fetches all of
the accounts for a given customer. First, The servlet
does the fetch by calling JNDI to lookup the "home"
interface, then does a "create" call on the home
interface to get the "remote" interface
(see snippets of code below, the ejbCreate()).
Second, the servlet calls up a specific method that
extracts the customer's accounts (see snippets of
code below, the method getAccountList()).
At first all seem to work well, however after some
random time my beans start to throw the SQLException
"failed to create a data connection with any of the
specified drivers"
At first I thought that I am not closing the
connections, but all of of my connections,
statements, and result sets, are closed
and even set to null in a finally block as you can
see in the snippets of code below.
If I restart iAS, all of the connections get released
and things look normal again till the problem
re-occurs.
Initially, we had this problem on SP2, so we upgraded
to SP3 thinking that this may solve the problem,
however the problem remained on SP3.
Another thing, this same application works perfectly
well on my laptop which has similar iPlanet
configurations except that operating system is
Windows 2000 Advanced Server, and Oracle 8.0.4.0.0.
System Configuration:
Following is my system configuration and related
iPlanet 6.0 settings.
-1- iPlanet 6.0 SP3
-2- Type 2 Oracle driver
-3- SunOs, sparc SUNW, Ultra-250, solaris 5.8
-4- Oracle 8.1.6.0.0 64-bit Production
-5- Following are the entry settings in the iPlanet
registry under
Software\iPlanet\Application Server\6.0\CCS0\DAE2\ORACLE_OCI
- CacheCleanInterval = 120
- CacheConnTimeOut = 120
- CacheDebugMsgs = 0
- CacheFreeSlots = 16
- CacheInitSlots = 64
- CacheMaxConn = 64
- CachMaxGlobalConn = 128
- ConnGiveUpTime = 60
- RMThreadMax = 32
- RMThreadMin = 0
- RSBufferInitRows = 25
- RSBufferMaxBufferSize = 6553600
- RSBufferMaxRows = 100
- RSBufferMaxSize = 32768
- SQLDebugMsgs = 0
Snippets of Code:
The following snippets of code come from a stateless
session bean and consist of a local data member, the
ejbCreate() method, and a typical rmi method that
does the database connection and extraction of
related records.
javax.sql.DataSource dataSourceObj = null;
public void ejbCreate()
throws java.rmi.RemoteException, javax.ejb.CreateException
javax.naming.Context ctx = null;
// Ensure first that the _props have been
// successfully instantiated by the constructor.
if (_props != null) {
try{
ctx = new javax.naming.InitialContext();
}catch (Exception ex){
nbUtility.logError(ex,"Error while creating Initial Context
try{
// DEBUG:
System.out.println("ejbCreate(): NB_DATASOURCE = " +
props.getPropertyValue(props.NB_DATASOURCE));
// DEBUG:
dataSourceObj = (javax.sql.DataSource)
ctx.lookup(_props.getPropertyValue(_props.NB_DATASOURCE));
catch (Exception ex){
ex.printStackTrace();
if (dataSourceObj == null)
throw new javax.ejb.CreateException("Couldn't get DataSource
object from environment");
} else {
throw new javax.ejb.CreateException("Couldn't create the
property manager: NB_CONFIG_FILE is null, empty, not set, or thefile
doesn't exist.");
* Given a customer number, this method
* returns the list of accounts that belongs to
* this customers.
public nbAccountList getAccountList(String CustomerNo)
throws java.rmi.RemoteException {
nbAccountList accountList = null;
if ((CustomerNo != null) && (!CustomerNo.equals(""))) {
java.sql.Connection conn = null;
java.sql.Statement stmt = null;
java.sql.ResultSet rset = null;
try {
accountList = new nbAccountList();
String sql = "SELECT * FROM " +
props.getPropertyValue(props.NB_DBTABLE_ACCOUNT)
+
" WHERE " +
"(" +
props.getPropertyValue(props.NB_DBFIELD_ACCOUNT_CUSTOMERNO) +
"='" + CustomerNo + "')";
// DEBUG:
nbDebug.write("accounts list SQL= " + sql);
// DEBUG:
// Let's get the connection, the statement, and the record
set.
conn = dataSourceObj.getConnection();
stmt = conn.createStatement();
rset = stmt.executeQuery(sql);
// Let's loop for each single account
int index = 0;
nbAccount account = null;
while (rset.next()) {
String AccountNo =
rset.getString(_props.getPropertyValue(_props.NB_DBFIELD_ACCTPERMIS_ACCOUNTN
O));
>
// Instantiate an account object
account = new nbAccount(AccountNo);
index++;
} catch (SQLException e) {
nbUtility.logError(e, "SQLException while trying to get
accounts data.");
} finally {
try {
if (rset != null) { rset.close(); rset = null; }
if (stmt != null) { stmt.close(); stmt = null; }
if (conn != null) { conn.close(); conn = null; }
} catch (SQLException e) {
nbUtility.logError(e, "SQLException while trying to close
connection.");
return (accountList);
Try our New Web Based Forum at http://softwareforum.sun.com
Includes Access to our Product Knowledge Base! -
IPOD POWER RUNS OUT QUICKLY WHEN CONNECTED TO PC
Hi All,
My fully charged IPod runs out of power in half an hour when I connect to PC, whereas when it is connected to my MAC it recharges the power.
What can I do to prolong a battery time connected to PC?
Thanks
GWelcome to Apple Discussions.
You have a 3G iPod. If you are using USB to connect, then I would say it is perfectly normal since 3G iPods cannot charge via USB and therefore your battery will be depleted when you sync. If indeed you are using USB, I suggest you get an USB2/Firewire iPod cable and connect the iPod in this way: How to connect the iPod Dock Connector to FireWire and USB 2.0 Cable (You need the Firewire iPod charger which should be included with your iPod.). -
Running out of available ip addresses
I have a Windows 2012 domain controller on a /24 network. I currently have about 10 free ip addresses. No phones or portable devices on the network currently. Getting ready to automate our warehouse to expand network with 40 or more wireless
hand-held warehouse computers, running windows 8 for handhelds (mostly barcode readers and scanners) and tablet computers. 18 Access points will be installed in about 2 weeks. Need advice on how to update the ip address range of the server. Have
multiple VM's available. Also have been having some intermittent problems with the server, think it might have something to do with importing the Active Directory from SBS2003. Can I get some suggestions on pro's and con's of changing ip address
range to fix the shortage of ip addresses. Currently a 192.168.1.1/24 considering 10.0.0.1
Devices on network;
10-12 servers multiple nics
8 switches (mostly managed)
Cisco 2921 router
Palo Alto firewall
110 workstations
50 printers
40 ip cameras
100 meg fiber Internet connection
Thanks for any help or thoughts on solutions.You don't need to change to a 10. subnet. You could increase the subnet mask on the 192.168.1.0 network to a 23-bit mask. Or you could add a second 24-bit subnet and route between them, with say 4 switches in each subnet. That is the way I would go,
with two 24-bit scopes (say 192.168.1.0/24 and 192.168.2.0/24).
Bill -
Running out of pooled connections
i have web application developed with jsp/servlets and mysql. i use JNDI datasource connection pooling. i have the following element defined in my server.xml file:
<Resource name="jdbc/JSNDB" auth="Container" type="javax.sql.DataSource"
maxActive="100" maxIdle="30" maxWait="10000"
removeAbandoned="true" removeAbandonedTimeout="60"
logAbandoned="false"
username="root" password="pn2072"
driverClassName="com.mysql.jdbc.Driver"
url="jdbc:mysql://localhost:3306/globe?autoReconnect=true"/>
my site is highly dependent on the databse for content and makes extensive use of db conections. only after a few clicks, my application cannot get any more connections from the pool. when i check the mysql server, i see that after 100 connections are created, they are never released even though i close every resultset, statement and connection immediately after i use them.
can any one please help.
many thanks
akzhi
this is the method from which I get connections from the pool:
public static Connection getConnection() {
DataSource ds=null;
try {
Context ctx = new InitialContext();
if(ctx == null ) throw new Exception("Boom - No Context");
ds = (DataSource)ctx.lookup("java:comp/env/jdbc/JSNDB");
return ds.getConnection();
} catch(Exception e) {
e.printStackTrace();
return null;
and this is how i close my connections (in every method that gets a connection):
finally
try{
rst.xlose();
stmt.close();
conn.close();
catch(SQLException e){}
} -
JCo Connections in WD Model running out
Hello,
we have WD application that uses RFC model.
During a load test we have observed that the connection
pools are running out of JCo connections. The result is a an error message in the trace file
and Http 500 response code to the end user.
Number of concurrent users: 20
Max Pool Size: 20
Max Connections: 20
Exception:
[code]aused by: com.sap.mw.jco.JCO$Exception: (106) JCO_ERROR_RESOURCE: Connection pool ME_LP_MODELDATA_TSTWEB1_PT_useSSO is exhausted. The current pool size limit is 20 connections.
at com.sap.mw.jco.JCO$Pool.getClient(JCO.java:5150)
at com.sap.mw.jco.JCO$PoolManager.getClient(JCO.java:5849)
at com.sap.mw.jco.JCO$PoolManager.getClient(JCO.java:5799)
at com.sap.mw.jco.JCO.getClient(JCO.java:8076)
at com.sap.tc.webdynpro.serverimpl.core.sl.AbstractJCOClientConnection.getClient(AbstractJCOClientConnection.java:393)
at com.sap.tc.webdynpro.modelimpl.dynamicrfc.AiiModel.getCurrentlyUsedJcoClient(AiiModel.java:191)[/code]
Questions:
Is it necessary to have equal number of Max Connections as
number of potential concurrent users?
Or is there a way to manually release not used connections?
Best Regards, MaikHi,
It is not necessary to create as many connections as there are concurrent users.
We have to handle the scope of the connection properly and also close the connections once we complete the task.
Re: How to close a model object connection for Adaptive RFC?
Regards
Bharathwaj -
Connection pool running out of connections
Hi all,
I am using Oracle's connection pool implementation:
OracleConnectionPoolDataSource dbConnection;
OracleConnectionCacheImpl connectionPool;
and tomcat 3 and 4. Sometimes on Tomcat 4 (Linux) I see that the connections are not being reutilized and therefore I run out of free connections. The same application runs fine on NT and Tomcat 3 and 4. I checked all my methods and all have a close(connection) in my finally{} and besides this the problem only happens once in a while and the program is using the same methods all the time. Anybody has a clue about what could the issue be?
Thanks,
A.have u tried setCacheScheme?
-
Our company has been using BDB to store real-time traffic information (speed and incident). Traffic information, which we call a feed, comes in every three minutes for the entire US. We are creating databases for speed and incident separately. Each time we receive a feed, we creates DBs something like flow.traffic.5.db and incident.traffic.5.db. When DBs are becoming too old, we delete from the directory where we store them.
We are having trouble in using the Berkeley DB library (version 4.5.20) in our application. In our production environment (on rh5 machines), our server crashes after some time, apparently because certain memory and/or other resources are being used up. We need assistance in getting to the cause of this problem, which we weren't seeing back when we used an earlier version of BDB (4.3.27).
In my testing under Windows, I find that BDB soon runs out of available mutex locks, giving the message "unable to allocate memory for mutex; resize mutex region" from inside the __mutex_alloc_int() function. This is not under any kind of load at all, and a very simple configuration. The test is basically a repeated opening and closing of the databases in a serial fashion, so it is hard to see why the locks should be "used up" like that.
The way we are using BDB is like this; first we create a DBENV and this DBENV will never be closed in run-time. Then, we create, use and delete db files under this DBENV. In the life period of a db file, many mutex will be used by DBENV to manage access to db files.
We tried to increase the total mutex count by DBENV->mutex_set_max(int), but it just delay the crash.
Does anyone have similar problem or remedy for this type of issue? Thanks in advance,Here are requeted information when there was a crash. I hope this help you to understand what we are experiencing .
By the way, we are using BDB 4.5.20
=============================================================================
Stderr output was
“unable to allocate memory for mutex; resize mutex region
[errno: 12]: incident: Db::open: Cannot allocate memory
[errno: 12]: incident: open: unretryable error
terminate called after throwing an instance of 'DbException'
what(): Error opening incident database -- session aborted!
Aborted
=============================================================================
And,
output of db_stat -x:
3MB 792KB Mutex region size
0 The number of region locks that required waiting (0%)
4 Mutex alignment
200 Mutex test-and-set spins
34034 Mutex total count
0 Mutex free count
34034 Mutex in-use count
34034 Mutex maximum in-use count
Mutex counts
0 Unallocated
2 db handle
1 env dblist
1 env region
1 lock region
4 logical lock
1 log filename
1 log flush
2 log region
1231 mpoolfile handle
2 mpool filehandle
17 mpool file bucket
1 mpool handle
16381 mpool hash bucket
16381 mpool buffer I/O
1 mpool region
1 unknown mutex type
1 replication database
1 replication region
1 twister
1 txn active list
1 unknown mutex type
1 txn region
output of db_stat -t
1/12437971 File/offset for last checkpoint LSN
Sun Oct 26 20:19:28 2008 Checkpoint timestamp
0x80001964 Last transaction ID allocated
100 Maximum number of active transactions configured
0 Active transactions
2 Maximum active transactions
6500 Number of transactions begun
410 Number of transactions aborted
6090 Number of transactions committed
0 Snapshot transactions
0 Maximum snapshot transactions
0 Number of transactions restored
40KB Transaction region size
0 The number of region locks that required waiting (0%)
Active transactions:
output of db_stat -CA
Default locking region information:
4100 Last allocated locker ID
0x7fffffff Current maximum unused locker ID
9 Number of lock modes
1000 Maximum number of locks possible
1000 Maximum number of lockers possible
1000 Maximum number of lock objects possible
2 Number of current locks
4 Maximum number of locks at any one time
3 Number of current lockers
6 Maximum number of lockers at any one time
2 Number of current lock objects
4 Maximum number of lock objects at any one time
15458 Total number of locks requested
15456 Total number of locks released
0 Total number of locks upgraded
2049 Total number of locks downgraded
0 Lock requests not available due to conflicts, for which we waited
0 Lock requests not available due to conflicts, for which we did not wait
0 Number of deadlocks
1000000 Lock timeout value
0 Number of locks that have timed out
1000000 Transaction timeout value
0 Number of transactions that have timed out
344KB The size of the lock region
0 The number of region locks that required waiting (0%)
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Lock REGINFO information:
Lock Region type
5 Region ID
__db.005 Region name
0xad907000 Original region address
0xad907000 Region address
0xad95cf40 Region primary address
0 Region maximum allocation
0 Region allocated
REGION_JOIN_OK Region flags
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Lock region parameters:
32790 Lock region region mutex [0/64769 0% 11406/3086178512]
1031 locker table size
1031 object table size
343720 obj_off
335464 locker_off
0 need_dd
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Lock conflict matrix:
0 0 0 0 0 0 0 0 0
0 0 1 0 1 0 1 0 1
0 1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 0 0
0 1 1 0 0 0 0 1 1
0 0 1 0 0 0 0 0 1
0 1 1 0 0 0 0 1 1
0 0 1 0 1 0 1 0 0
0 1 1 0 1 1 1 0 1
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Locks grouped by lockers:
Locker Mode Count Status ----------------- Object ---------------
1000 dd= 0 locks held 1 write locks 0 pid/thread 12222/2748504976
1000 READ 1 HELD flow.navteq.410.db handle 0
1001 dd= 0 locks held 0 write locks 0 pid/thread 12222/2748504976
1002 dd= 0 locks held 1 write locks 0 pid/thread 12222/2748504976
1002 READ 1 HELD sp_key.navteq.410.db handle 0
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Locks grouped by object:
Locker Mode Count Status ----------------- Object ---------------
1000 READ 1 HELD flow.navteq.410.db handle 0
1002 READ 1 HELD sp_key.navteq.410.db handle 0 -
Sybase ASE 12.5.4 EBF 14124 ESD#4 running out of connections
I am working on a web application which works on SQL sever back end as well as Sybase ASE 12.5.4 EA server. Our project main focus was to move everything to sybase backend.
Earlier there were 2 databases, the app was connecting to - db1 on sql server, and db2 on sybase. Now we have created a separate namespace within db2 in sybase. So application only connects with db2, but with different users and in separate schemas.
Application is in java and connecting to db server using jcon3 driver library. It is using SybXADataSource driver class (provided by jcon3 jar) to connect to the server. Till last week application was able to connect to the server, everything was working fine (though I must say that code is still not optimized to use connection pools etc., I am opening a connection, and closing it for every request).
For sometime, I have been facing an issue from the server- login failed, try again. In the server error logs I found it was throwing an error- "Error 1601 Severity 17 state 3: no user connections available to run the process" I checked for available connections (using sp_configure), allowed connections are 900, and total active connection at the time this error shows up is typically between 30- 100 (using sp_monitorconfig). Seems like this is well within limits.
Also yesterday, I did one more exercise, I restarted the server and logged in as sa from isql console only. Logged out and logged in again afimeter 20 minutes (no application connection was made), and isql login again gave this error. So I am inclined to think that something could be wrong with server configuration as well. Server is on linux red hat.
I have searched all the db forums and SCN website as well, and all of these refer to one resolution of increasing the number of connections, but as from my observation regarding that, I do not think it is an issue.
Any help would be greatly appreciated!!During your sa/isql test, how did you exit from your sessions? Did you type 'exit + go', hit ^S, hit ^C, something else?
Are you running your sa/isql test from the same linux host where the ASE is runing, or from another host on the network?
Is your isql session connecting directly to the dataserver or are you going through an intermediary service?
Are there any login triggers on your sa account? (See the output from sp_displaylogin sa)
Could you run your sa/isql test again, but this time:
session #1: log in as sa using isql; stay logged in
session #2: perform your login/logout test using sa + isql
session #1: periodically run 'sp_who' to see the active connections; does the list of connections grow over time? -
Running out of connections...
Hello,
I think I'm having the same problem ..
Did you solve it ? How ?
Thanks a lot.
Franco
Francesco Costa
BEA Systems Italia
[email protected]
Phone +39 02 3327 3000
Mobile + 39 335 1246931
Fax +39 02 3327 3001
"Gerhard Henning" <[email protected]> ha scritto nel messaggio
news:[email protected]...
>
Hi,
we seem to have a similar problem. Can you please tell me,
which service pack solves the problem.
We are using SP8 but the problem seems to be still there!
Thanks
Gerhard
Joseph Weinstein <[email protected]> wrote:
Gene Chuang wrote:
Hi Joe,
Good news; bug ISOLATED and FIXED! First off, I was able to determinethis bug only manifests if
we are running clustered apps. under a single app server, this bugdoes not appear, and under
clustered app this bug always appears.
Furthermore, I appologize if I may have lead you on a wild goosechasewith all the subsequent
patches you sent me. As it turns out, when I received each patch Ihanded it off to our tester, who
applied the patch and probably only bounced a node in our server
cluster
before testing, hence
getting negative results. Today I applied your latest patch,
clean-bounced
all the nodes, and the
bug disappeared! Thanks Joe!
Now the 2 questions of the day:
1) Is your latest patch production worthy? Can I apply it on ourlive site?
Yes.
2) If not, when's the next service pack coming out with this fix?It will be in future service packs.
Joe
Gene
"Joseph Weinstein" <[email protected]> wrote in message
news:[email protected]...
>>>>
>>>>
Gene Chuang wrote:
Hi Joe,
We tried it and found the same exception raised, except for this
time it took a lot longer to
bug
out; the server hung for almost 2 minutes before dumping the
following.
However I was just
informed by another developer that we may have a recursive bugon our side, and that we are
doing a
non-terminating db search. Perhaps this is causing this
out-of-resource
allocation bug, i.e.
our
bug exposed your bug! I'm sure once we fix our recursion bug we
won't see this again, but maybe
something should also be done on your side that'll handle thismore gracefully...
I'll let you know what happens once we fix our bug; thanks forthe help!
DO let me know. In fact, please use the attached jar (in place ofall the others:-)
because the line ResourceAllocator.java:561 is not executable code.There may be a
JVM problem... I'm fairly sure there's no way our code can causean NPE in this
area...
Joe
javax.ejb.FinderException: Couldn't build CallableStatement from
JDBCCommandImpl:
java.sql.SQLException: Exception raised by connection pool
:java.lang.NullPointerException
atweblogic.common.internal.ResourceAllocator.reserve(ResourceAllocator.java:56
1)
atweblogic.common.internal.ResourceAllocator.reserve(ResourceAllocator.java:55
5)
atweblogic.common.internal.ResourceAllocator.reserve(ResourceAllocator.java:50
2)
atweblogic.jdbc.common.internal.ConnectionPool.reserve(ConnectionPool.java:172
atweblogic.jdbc.common.internal.ConnectionPool.reserveWaitSecs(ConnectionPool.
java:145)
atweblogic.jdbcbase.jts.Connection.openConnectionIfNecessary(Connection.java:5
87)
atweblogic.jdbcbase.jts.Connection.prepareCall(Connection.java:159)
atweblogic.jdbc20.rmi.internal.ConnectionImpl.prepareCall(ConnectionImpl.java:
89)
atweblogic.jdbc20.rmi.SerialConnection.prepareCall(SerialConnection.java:71)
atcom.kiko.db.JDBCConnection.getCallableStatement(JDBCConnection.java:158)
>>>>>
Gene Chuang
Join Kiko.com!
"Joseph Weinstein" <[email protected]> wrote in message
news:[email protected]...
>>>>>>
>>>>>>
Gene Chuang wrote:
Hi Joe,
We were finally able to consistently replicate the
ResourceAllocator.reserve()
stacktrace,
although
from a slightly different route. I tried your jar, and now
have a slightly different
exception:
Hi Gene. Thank you for your patience. Here's one last jar file.
Ditch the previous one. I appreciate your energy and willingness
to help. Let me know,
Joe
javax.ejb.EJBException: Failed to select row 0: Exception
raised
by connection pool
:java.lang.NullPointerException
at
weblogic.common.internal.ResourceAllocator.reserve(ResourceAllocator.java:56
1)
atweblogic.common.internal.ResourceAllocator.reserve(ResourceAllocator.java:55
5)
atweblogic.common.internal.ResourceAllocator.reserve(ResourceAllocator.java:50
2)
atweblogic.jdbc.common.internal.ConnectionPool.reserve(ConnectionPool.java:172
atweblogic.jdbc.common.internal.ConnectionPool.reserveWaitSecs(ConnectionPool.
java:145)
atweblogic.jdbcbase.jts.Connection.openConnectionIfNecessary(Connection.java:5
87)
atweblogic.jdbcbase.jts.Connection.prepareStatement(Connection.java:135)
atweblogic.jdbc20.rmi.internal.ConnectionImpl.prepareStatement(ConnectionImpl.
java:80)
atweblogic.jdbc20.rmi.SerialConnection.prepareStatement(SerialConnection.java:
55)
atcom.kiko.db.JDBCConnection.getPreparedStatement(JDBCConnection.java:141)
>>>>>>>
Gene Chuang
Join Kiko.com!
"Joseph Weinstein" <[email protected]> wrote in message
news:[email protected]...
Hi Gene.
Would you put the attached jar to the head of your
weblogic.classpath,
and let me know if this fixes the problem? thanks,
Joe
Gene Chuang wrote:
WL 5.1 sp 6, running on Solaris 2.7 over Oracle 8i
Gene Chuang
Join Kiko.com!
"Joseph Weinstein" <[email protected]> wrote in message
news:[email protected]...
Gene Chuang wrote:
Hi Joe,
Now I'm getting this slightly different stack trace...
do u think it's the same
"out of
connection"
error, or something else?It's something else. What version of the product is this?
Joe
Couldn't build PreparedStatement from JDBCCommandImpl:
java.sql.SQLException:
Exception
raised
by
connection pool :java.lang.NullPointerException
at
weblogic.common.internal.ResourceAllocator.reserve(ResourceAllocator.java:56
1)
atweblogic.common.internal.ResourceAllocator.reserve(ResourceAllocator.java:55
5)
atweblogic.common.internal.ResourceAllocator.reserve(ResourceAllocator.java:50
2)
atweblogic.jdbc.common.internal.ConnectionPool.reserve(ConnectionPool.java:172
atweblogic.jdbc.common.internal.ConnectionPool.reserveWaitSecs(ConnectionPool.
java:145)
atweblogic.jdbcbase.jts.Connection.openConnectionIfNecessary(Connection.java:5
87)
atweblogic.jdbcbase.jts.Connection.prepareStatement(Connection.java:135)
atweblogic.jdbc20.rmi.internal.ConnectionImpl.prepareStatement(ConnectionImpl.
java:80)
atweblogic.jdbc20.rmi.SerialConnection.prepareStatement(SerialConnection.java:
55)
>>>>>>>>>>>
Gene Chuang
Join Kiko.com!
"Joseph Weinstein" <[email protected]> wrote in message
news:[email protected]...
Hi.
The server won't let an EJB wait indefinitely for
a connection to become
available in a pool. All the while it waits, it iscommandeering one
of the fixed number of execute-threads, doing nothingwhile the thread
could possibly run another task which already hasit's pool connection,
and could release it if it ran. The default waitallowed is 5 seconds
the server decides to devote it's resources to thework it can do, and
not let overflow requests cause an unintentionalor malicious denial-
of-service attack. The fundamental solution is tosupply enough DBMS
connections to serve the load you have.
Joe
Gene Chuang wrote:
Hi,
Here's our pool init info:
weblogic.jdbc.connectionPool.oraclePool=\
driver=oracle.jdbc.driver.OracleDriver,\
url=jdbc:oracle:oci8:@xxx,\
loginDelaySecs=0,\
initialCapacity=4,\
maxCapacity=10,\
capacityIncrement=2,\
allowShrinking=true,\
shrinkPeriodMins=15,\
refreshMinutes=10,\
testTable=dual,\
props=user=xxx;password=xxxx;server=xxx
I run a db-intensive data migration program on
my ejbs (hence JDBC connection
requests
are
made
on
the serverside). After a while, I run out of
connections,
and the exception
pasted
below is
thrown.
Why is this exception thrown? If the pool is at
maxCapacity and all connections
are
taken,
shouldn't weblogic place the connection-requestor
on wait until a connection is
available?
And
I'm
pretty sure I return the connections because I
wrap connection.close() on a
finally
block...
JDBCPoolConnection.connect: Problem connecting
Exception:
java.sql.SQLException: Pool connect failed: Noneavailable
java.sql.SQLException: Pool connect failed: Noneavailable
at
weblogic.jdbcbase.pool.Driver.connect(Driver.java:184)
atweblogic.jdbcbase.jts.Driver.connect(Driver.java:233)
at
weblogic.jdbc20.common.internal.RmiDataSource.getConnection(RmiDataSourc
e.java:55)
at
weblogic.jdbc20.common.internal.RmiDataSource_ServiceStub.getConnection(
RmiDataSource_ServiceStub.java:179)
at
com.kiko.db.JDBCPoolConnection.connect(JDBCPoolConnection.java:24)ABMEB.
ejbActivate: Getting bean out of pooled state.
Gene Chuang
Join Kiko.com!--
PS: Folks: BEA WebLogic is in S.F. with both entryand advanced positions for
people who want to work with Java and E-Commerceinfrastructure products. Send
resumes to [email protected]
The Weblogic Application Serverfrom BEA
JavaWorld Editor's Choice Award: Best WebApplication Server
Java Developer's Journal Editor's Choice Award:Best Web Application Server
Crossroads A-List Award: Rapid Application
Development
Tools for Java
Intelligent Enterprise RealWare: Best ApplicationUsing a Component Architecture
>
http://www.bea.com/press/awards_weblogic.html
>>>>>>>>>>
PS: Folks: BEA WebLogic is in S.F. with both entry andadvanced positions for
people who want to work with Java and E-Commerce
infrastructure
products. Send
resumes to [email protected]
The Weblogic Application Server fromBEA
JavaWorld Editor's Choice Award: Best Web
Application
Server
Java Developer's Journal Editor's Choice Award: BestWeb Application Server
Crossroads A-List Award: Rapid Application
Development
Tools for Java
Intelligent Enterprise RealWare: Best Application Usinga Component Architecture
>
http://www.bea.com/press/awards_weblogic.html
>>>>>>>>
PS: Folks: BEA WebLogic is in S.F. with both entry andadvanced
positions for
people who want to work with Java and E-Commerce
infrastructure
products. Send
resumes to [email protected]
The Weblogic Application Server fromBEA
JavaWorld Editor's Choice Award: Best Web
Application
Server
Java Developer's Journal Editor's Choice Award: Best WebApplication Server
Crossroads A-List Award: Rapid Application DevelopmentTools for Java
Intelligent Enterprise RealWare: Best Application Using aComponent Architecture
>
http://www.bea.com/press/awards_weblogic.html
>>>>>>
PS: Folks: BEA WebLogic is in S.F. with both entry and advancedpositions for
people who want to work with Java, XML, SOAP and E-Commerce
infrastructure
products. Send resumes to [email protected]
The Weblogic Application Server from BEA
JavaWorld Editor's Choice Award: Best Web ApplicationServer
Java Developer's Journal Editor's Choice Award: Best Web
Application
Server
Crossroads A-List Award: Rapid Application Development Toolsfor Java
Intelligent Enterprise RealWare: Best Application Using a
Component
Architecture
http://www.bea.com/press/awards_weblogic.html--
PS: Folks: BEA WebLogic is in S.F. with both entry and advanced
positions
for
people who want to work with Java, XML, SOAP and E-Commerce
infrastructure
products. Send resumes to [email protected]
The Weblogic Application Server from BEA
JavaWorld Editor's Choice Award: Best Web Application Server
Java Developer's Journal Editor's Choice Award: Best WebApplication
Server
Crossroads A-List Award: Rapid Application Development Toolsfor Java
Intelligent Enterprise RealWare: Best Application Using a ComponentArchitecture
http://www.bea.com/press/awards_weblogic.html--
PS: Folks: BEA WebLogic is in S.F. with both entry and advanced positions
for
people who want to work with Java, XML, SOAP and E-Commerceinfrastructure
products. Send resumes to [email protected]
The Weblogic Application Server from BEA
JavaWorld Editor's Choice Award: Best Web Application Server
Java Developer's Journal Editor's Choice Award: Best Web Application
Server
Crossroads A-List Award: Rapid Application Development Tools for
Java
Intelligent Enterprise RealWare: Best Application Using a ComponentArchitecture
http://www.bea.com/press/awards_weblogic.html -
Connection leaks and application is running out of connection
Hi All,
We have configured the SQL Database external resource for OBPM specific connection pool. All the business processes are using the Fuego.Sql package for the data base transaction calls. I have no clue how this package is managing the database connections. If more than 25 users perform concurrent testing, the application is running out of connections. Connection pool configuration details as below.
Maximum Pool Size : 500
Maximum connections per user : 50
Minimun Pool Size : 0
Connection Idle Time (mins). : 5
Maximum Opened Cursors : 1000
Please share your thoughts on how I can track and fix this issue. Also please let me know the answers below.
1. Is there any way to find out the stats about the connection pool
2. If I configure the remote JDBC that points to J2EE datasource, would that fix this issue.
I appreciate your help.
Thanks,Can anyone please share your ideas?
Thanks, -
How do I add to my data plan on the iPad once I have run out and no longer have an internet connection? I am working out of the country and got a message that I was running low,followed immediatly by a you're out message.
Should you not be asking your carrier? They are the people supplying your data plan.
-
Hey my iPod is not showing up on iTunes and my computer doesn't recognize it. It runs out of battery very fast and only shows the 'Connect to iTunes" screen. What do I do?
http://support.apple.com/kb/ht1808
http://support.apple.com/kb/ts1567
Basic troubleshooting steps
17" 2.2GHz i7 Quad-Core MacBook Pro 8G RAM 750G HD + OCZ Vertex 3 SSD Boot HD
Got problems with your Apple iDevice-like iPhone, iPad or iPod touch? Try Troubleshooting 101 -
I am trying to download photos off my IPhone as the "Cloud" tells me I have run out of storage space. I have connected my phone to IPhoto on my Mac but when I clik on Import nothing happens. What am I doing wrong?
Help for iCloud storage space >> Managing your iCloud storage
-
i am having problems with my new ipad that i got for Christmas today, i don't get free wi-fi at my home and it said if the wi-fi is not available, connect to itunes. i have done that and it didnt help me. mind helping me out?
Further info - I changed my security settings to allow apps from anywhere, and again tried to install. I again got an error message
that the update failed to install, and to contact Customer Support.
Maybe you are looking for
-
I have recently changed my e-mail address which was used as the ID for my updating the apps. on my itouch 4g.Now when I try to update the apps.the ID comes up as the old e-mail that I used in the past and that I cant change to the new e-mail that I
-
Printing parameter values in xml report, and displaying it in header of the
Hi All, Can anybody give me an idea to print the parameter values in the xml report. My requirement is like, I have 10 parameters in report builder, I have generated the xml file and created the .rtf template and finally registerted in oracle apps. B
-
Data flow not visible while creating remote cube.
Hi SDN, I am working on a remote cube. I need to link the infosource to remote cube. I have selected the source system and assigned. I have done, like--Remote cube --> Context Menu --> Show Data flow , and I wanted to see the source system , Info s
-
BEX: Changing date format from 'mm/dd/yyyy' to 'mon/yyyy'
Happy Friday! I need to change the date format in my BEX query from a regular date format (dd/mm/yyyy) to display as 'JUN/2008'. Please note that I dont want to modify the date field but just need to display it month and year format in the bex query.
-
Skype desktop crashes on windows 8.1
Interesting, Skype for Desktop has been working for me for the past week since I did the Vista / Service Pack compatability thing. However, I just installed Chrome and I can go to a few simple website (like my own) but it crashes on most websites.Wha