APEX error: Found dead shared server

Hi,
I frequently encounter an error when working with APEX. If I submit any change to APEX, fi a change on some page item, APEX asks me to whether I'd like to save the wwv_flow_accept URL. I then get a file save dialog. This doesn't happen all the time, but something like 2 out of 10 times I submit something.
Looking at the database logs, I see an entry in alert.log with this information:
<pre>
Fri May 22 17:56:36 2009
Errors in file c:\oraclexe\app\oracle\admin\xe\bdump\xe_s004_1948.trc:
ORA-07445: exception encountered: core dump [ACCESS_VIOLATION] [_npierr+487] [PC:0x5F22C3] [ADDR:0x4] [UNABLE_TO_READ] []
Fri May 22 17:56:40 2009
found dead shared server 'S004', pid = (18, 1)
</pre>
Looking into the respective trace file gives me this information:
<pre>
Dump file c:\oraclexe\app\oracle\admin\xe\bdump\xe_s004_1948.trc
Fri May 22 17:56:36 2009
ORACLE V10.2.0.1.0 - Production vsnsta=0
vsnsql=14 vsnxtr=3
Oracle Database 10g Express Edition Release 10.2.0.1.0 - Production
Windows XP Version V5.1 Service Pack 3
CPU : 1 - type 586, 1 Physical Cores
Process Affinity : 0x00000000
Memory (Avail/Total): Ph:727M/1511M, Ph+PgF:1181M/2133M, VA:1428M/2047M
Instance name: xe
Redo thread mounted by this instance: 1
Oracle process number: 18
Windows thread id: 1948, image: ORACLE.EXE (S004)
*** ACTION NAME:(PAGE 4311) 2009-05-22 17:56:36.625
*** MODULE NAME:(APEX:APPLICATION 4000) 2009-05-22 17:56:36.625
*** CLIENT ID:(CKH_DEV:2151496271973569) 2009-05-22 17:56:36.625
*** SESSION ID:(29.4) 2009-05-22 17:56:36.625
*** 2009-05-22 17:56:36.625
ksedmp: internal or fatal error
ORA-07445: exception encountered: core dump [ACCESS_VIOLATION] [_npierr+487] [PC:0x5F22C3] [ADDR:0x4] [UNABLE_TO_READ] []
Current SQL statement for this session:
COMMIT
----- PL/SQL Call Stack -----
object line object
handle number name
267D174C 1275 package body APEX_030200.WWV_FLOW
267D174C 5421 package body APEX_030200.WWV_FLOW
267D174C 13366 package body APEX_030200.WWV_FLOW
26521088 30 anonymous block
----- Call Stack Trace -----
memory dump follows
</pre>
Has anybody an idea what is going wrong on my machine here?
Best regards,
Jürgen

Hi,
I took a bit of time to dig deeper into this issue. I came to the conclusion that there is a bug in Oracle XE:
If you connect to Oracle XE database using the EPG path and if you create a report that gets its data over a database link, the shared server processes used to connect to the database break.
What I did to produce this error:
I installed Oracle XE Universal Edition on a Windows XP system
I then updated APEX to the latest version 3.2
At the database I simply created a database link pointing to the same schema the APEX-application is running at. I tested the database link.
I then produced a simple test application with just one Report (I also tried a PPR-report with the same result) that displays orders with a query in line with this:
<pre>
select o.order_id, i.order_item_id, i.quantity, p.product_name, i.unit_price, i.unit_price * i.quantity amount
from demo_orders@ckh o, demo_order_items@ckh i, demo_product_info@ckh p
where o.order_id = i.order_id
and i.product_id = p.product_id
order by o.order_id, i.order_item_id
</pre>
When I ran this application, I was sometimes able to see the report without an error, sometimes a shared server process broke. You can watch the processes breaking by opening the admin/bdump directory and see how the database writes trace files for the shared server processes.
Interesting to see is that the error happens not only when you try to see the report. If you hit the edit page link while running the page or if you submit the page definition with the link query in it, the same is likely to happen. You can see this error happening when APEX offers you to download the call to wwv_flow or the f-procedure. The contents of this download then is:
<pre>IBMPC/WIN_NT-8.1.0</pre>
with some binary information in front of the message.
Another bug I'd like to add came up when I was lucky to see the report with the PPR-template. I tried pagination but this wouldn't work. Looking into this, I found the following link:
<pre>
javascript:html_PPR_Report_Page(this,'R2060428410112776','javascript:$a_report('2060428410112776','16','15','15');')
</pre>
To my understanding this can't work at all if you look a the apostrophes. So how to proceed with this???
This to me is a severe problem, as I wanted to utilize APEX as a frontend for a production database to report the contents. I was informed by Varad that a way around this is to use an Apache access to Oracle XE but as far as I'm informed, this forces me into a licenced version in order to be able to run mod_plsql.
I read about the mod_owa-approach, but the programmers state that this is not compatible with APEX. So my question is: Do you have any suggestion on how to proceed with this?
Thanks,
Jürgen
Edited by: j.sieben on May 27, 2009 5:41 PM

Similar Messages

  • What found dead shared server error mean?

    Please, I found the error found dead shared server in my alert.log file and I need to know what cause this error?
    If it is critical?
    What I should look up?

    There is a known issue with using heterogenous services routine and MTS. Kuljeet is correct. Most probably you are hitting bug 3243584. You may have to check with oracle support on this!
    -Ramesh

  • "found dead shared server ... " errors occured in oracle alert log

    Dear all,
    I have found a error message in my oracle alert log which is "found dead shared server 's000, pid=(10,1)'". What's this mean? and this error will casue a core dump in ../cdump directory.
    btw. Thanks the answers of antti.koskinen and yingkuan.

    one more thing, my server is not configured as shared server mode, it is dedicated server mode. but I don't know why system inform me "found dead shared server..".

  • Found dead shared server errors keep appearing

    I'm on Windows Server 2003 64-bit SP2
    Running Oracle 11.1.0.6.0 64-big Enterprise
    Configured as shared server running ASM
    In my alert log I keep getting:
    Thu May 01 18:36:05 2008
    Exception [type: ACCESS_VIOLATION, UNABLE_TO_READ] [ADDR:0x3951B64C] [PC:0x72A7ABB, kghfrf()+161]
    Errors in file g:\oracle\diag\rdbms\cma\cma\trace\cma_s004_3648.trc (incident=30201):
    ORA-07445: exception encountered: core dump [kghfrf()+161] [ACCESS_VIOLATION] [ADDR:0x3951B64C] [PC:0x72A7ABB] [UNABLE_TO_READ] []
    Thu May 01 18:36:05 2008
    Errors in file g:\oracle\diag\rdbms\cma\cma\cdump\cmacore.log
    ORA-07445: caught exception [ACCESS_VIOLATION] at [kghfrf()+161] [0x00000000072A7ABB]
    Thu May 01 18:36:06 2008
    Sweep Incident[30201]: completed
    Thu May 01 18:36:07 2008
    found dead shared server 'S004', pid = (24, 10)
    Any idea what is causing this? I can find no references on metalink about this.
    Thanks!

    Thank you for your reply.
    Unfortunately I'm on 11.1.0.6.0.
    Note 310177.1 lists 4 bugs that have been fixed prior to 11.1.0.6.0
    Note 4739114.8 list is a bug fixed in 11.1.0.6.0
    And the ORA-600/ORA-7445 Troubleshooter on Metalink unfortunately does not bring up anything other than the bug docs fixed prior to 11.1.0.6.0 (nor has 11g support yet).
    Has anyone else experienced this?

  • Found dead shared server

    every day i found this message in the db log :
    "Tue Mar 29 22:21:08 2005
    found dead shared server 'S000', pid = (10, 5)
    Wed Mar 30 13:51:58 2005
    found dead shared server 'S000', pid = (10, 6)"
    what do it means ??
    it is danger for the databse ???

    mrashid07 wrote:
    In /bdump oracle only provide this information---
    found dead shared server 'S000', pid = (17, 49)
    How can I find out these processes (17, 49) which made the dead lock event??? There's no evidence from this message that the process was terminated due to a dead lock event.
    Oracle also does not terminate a dead lock process - it raises a deadlock exception, allowing the process to catch this exception and re-act on it.
    When Oracle does terminate a process, it will be due to some kind of fatal error. In which case there will be an ORA-0600 or similar error message in the alert log, together with a trace file reference (or cdmp ref) that contains the technical details as to what happened.
    Dead shared servers are usually caused by an external process (or user) explicitly issuing a command to terminate the shared server process. E.g. killing the shared process at o/s level (using the o/s CLI), or via Oracle using the "+alter system kill session+" command.
    You have not shown any evidence yet to the contrary - so the only conclusion that can be made thus far is that Oracle is not an issue here, but application code and/or users are.

  • Found dead shared server in aler file

    hi
    some times i got these entries in my alert log file.
    found dead shared server 'S001', pid = (12, 2)
    Tue Apr 01 15:47:28 2008
    found dead shared server 'S001', pid = (12, 3)
    Tue Apr 01 15:51:31 2008
    found dead shared server 'S001', pid = (12, 4)
    Tue Apr 01 15:55:37 2008
    found dead shared server 'S001', pid = (12, 5)
    Tue Apr 01 15:57:34 2008
    found dead shared server 'S001', pid = (12, 6)
    Tue Apr 01 16:49:41 2008
    found dead shared server 'S000', pid = (11, 3)
    Tue Apr 01 18:00:21 2008
    Private_strands 7 at log switchi want to know is it harmful for oracle because its on my oracle live server.
    so can any body tell me what are the negative effects of this error means is it cause
    some serious effects.
    thanks

    What is your database version (complete) and O/S? Have the users been facing -ORA-3113 and ORA-3114 errors?
    Oracle Metalink Document: Dead Shared Servers And Connections Fail with ORA-3113 Calling External Procedures: Doc ID: Note:333026.1
    https://metalink.oracle.com/metalink/plsql/f?p=130:14:12268887976412175026::::p14_database_id,p14_docid,p14_show_header,p14_show_help,p14_black_frame,p14_font:NOT,333026.1,1,1,1,helvetica
    Adith

  • "found dead shared server" in alert.log

    Hi.
    I am getting the following message in the alert log of 10.2.0.2 database.
    Fri Feb 2 11:13:54 2007
    found dead shared server 'S005', pid = (32, 3)
    Fri Feb 2 11:14:24 2007
    found dead shared server 'S005', pid = (32, 4)
    Fri Feb 2 11:59:48 2007
    found dead shared server 'S001', pid = (11, 1)
    found dead shared server 'S005', pid = (32, 5)
    Fri Feb 2 12:00:04 2007
    found dead shared server 'S001', pid = (80, 165)
    found dead shared server 'S001', pid = (80, 166)
    What does it mean actaully...
    will it affect(what impact) my database.
    Thanks
    JD

    Hi,
    Are you using MTS feature? If yes,
    Then as given in the site:
    http://searchoracle.techtarget.com/tip/0,289483,sid41_gci1018471,00.html
    major issue I have seen with MTS is the dispatcher process getting abruptly killed. For some unknown reason the dispatcher process all of a sudden dies, killing the connected sessions. The following is an extract from the alert.log file for one such error:
    Wed Sep 8 09:07:35 2004
    Errors in file /u01/bdump/abc_d050_521.trc:
    ORA-07445: exception encountered: core dump [00000001026BA2D4] [SIGSEGV] [Address
    not mapped to object] [0x000000018] [] []
    Wed Sep 8 09:07:45 2004
    found dead dispatcher 'D050', pid = (667, 207)
    The dead dispatcher process is restarted once pmon cleans the in-doubt sessions. The workaround for this issue is to turn off the DCD (dead connection detection) feature. Set SQLNET.EXPIRE_TIME = 0 in sqlnet.ora file. This issue occurred in v9.2.0.4 and v9.2.0.5 and Oracle is currently working to resolve the same.
    I think you are not using MTS feature, otherwise you would have received errors
    like above & trace file would have been generated & error.ORA-07445
    If you are not using MTS feature,
    set the initialization parameter shared_servers = 0. also make sure that the parameter dispatcher is not set (it has no default) to anything.
    Message was edited by:
    Seema

  • Found dead shared server - ORA-07445:

    Hi,
    Apologies if this has been posted before but can someone help with getting a resolution to the above error that I noticed in my Alert log. Here's an except from the log. i can post the error file but it is 3.4mb in size.
    Thanks in Advance,
    Alan
    Current log# 3 seq# 20641 mem# 0: D:\ORADATA\LIVE\REDO003.ORA
    Sun May 02 20:47:05 2010
    Thread 1 advanced to log sequence 20642 (LGWR switch)
    Current log# 4 seq# 20642 mem# 0: D:\ORADATA\LIVE\REDO004.ORA
    Sun May 02 20:49:47 2010
    Errors in file d:\oracle\product\10.2.0\admin\live\bdump\live_s005_4612.trc:
    ORA-07445: exception encountered: core dump [ACCESS_VIOLATION] [kghfrf+157] [PC:0x51F4AE9] [ADDR:0x21B1D6CC] [UNABLE_TO_READ] []
    Sun May 02 20:50:16 2010
    found dead shared server 'S005', pid = (16, 229)
    Sun May 02 20:59:27 2010
    ALTER SYSTEM ARCHIVE LOG
    Sun May 02 20:59:27 2010
    Thread 1 cannot allocate new log, sequence 20643
    Private strand flush not complete
    Current log# 4 seq# 20642 mem# 0: D:\ORADATA\LIVE\REDO004.ORA
    Sun May 02 20:59:30 2010

    please check metalink..
    ORA-07445 [ACCESS_VIOLATION] [unable_to_trans_pc] on Windows Platforms [ID 456801.1]
    ORA-07445 [ACCESS_VIOLATION] [_kkslmtl+524] [PC:0xF41EC0] [ADDR:0x0] [UNABLE_TO_READ] [] [ID 755278.1]
    Regards
    Rajesh

  • In FORMS 10G found dead shared server 'S000', pid = (10, 9)

    during a test phase that we are doing before convert to FORMS 10G sometime the form crash (random) and in the same moment we foun the follow message in the log :
    found dead shared server 'S000', pid = (10, 9)
    DO YOU SOMEThING ???

    what is MTS ??
    we have doa standard installation with forms and report as application server and oracle 9i as db.
    we leve the standard configuration for the db.
    I don't Know what is MTS?

  • Found dead shared server 'S002', pid = (12, 37)

    Whats up with that ? I got a client running a utility (Tora) against a 9.2.0.1 database, and every 30 seconds he get's disconnected, and this appears in the alert log
    "found dead shared server 'S002', pid = (12, 37)"
    .. Mind that this exact setup has worked flawlessly for at least 2 weeks, just begun doing it today... Only other thing going on is a massive insert operation, for wich i've had to add tablespaces to make room for... Should i restart the db ? shouldnt be neccesary, right ?

    I recommend you to read completely this chapter:
    Configuring Shared Servers
    Proper configuration of shared servers can result in significant performance improvement.
    This chapter contains the following topic:
    Introduction to Shared Server Performance
    Configuring the Number of Shared Servers
    Introduction to Shared Server Performance
    Using shared servers enables you to reduce the number of processes and the amount of memory consumed on the server machine. Shared servers are beneficial for systems where there are many OLTP users performing intermittent transactions.
    Using shared servers rather than dedicated servers is also generally better for systems that have a high connection rate to the database. With shared servers, when a connect request is received, a dispatcher is already available to handle concurrent connection requests. With dedicated servers, on the other hand, a connection-specific dedicated server is sequentially initialized for each connection request.
    Performance of certain database features can improve when a shared server architecture is used, and performance of certain database features can degrade slightly when a shared server architecture is used. For example, a session can be prevented from migrating to another shared server while parallel execution is active.
    A session can remain nonmigratable even after a request from the client has been processed, because not all the user information has been stored in the UGA. If a server were to process the request from the client, then the part of the user state that was not stored in the UGA would be inaccessible. To avoid this, individual shared servers often need to remain bound to a user session.
    Configuring the Number of Shared Servers
    When using some features, you may need to configure more shared servers, because some servers might be bound to sessions for an excessive amount of time.
    This section discusses how to reduce contention for processes used by Oracle's architecture:
    Identifying Contention Using the Dispatcher-Specific Views
    Reducing Contention for Dispatcher Processes
    Reducing Contention for Shared Servers
    Determining the Optimal Number of Dispatchers and Shared Servers
    http://download-west.oracle.com/docs/cd/B10501_01/server.920/a96533/features.htm#22666
    Joel Pérez
    http://otn.oracle.com/experts

  • "found dead shared server" while making a hot backup

    This is part of the alert, maybe it helps:
    Wed Aug 15 11:27:16 2007
    alter tablespace ERKSNEW begin backup
    Completed: alter tablespace ERKSNEW begin backup
    Wed Aug 15 11:28:23 2007
    found dead shared server 'S000', pid = (32, 127)
    Wed Aug 15 11:28:47 2007
    alter tablespace ERKSNEW end backup
    Wed Aug 15 11:28:48 2007
    Completed: alter tablespace ERKSNEW end backup
    Any help? Thanks a lot !!! :-)

    Did the backup fail? It may be a temporary condition. If it happens more often,
    1) change to dedicated server mode temporarily
    2) call Oracle support
    Werner

  • Dead shared server S000  ??

    Hello all,
    When one of the user was trying to extract oracle data to a tab separated file
    he got the error ORA-03113
    In the alert log i found
    found dead shared server 'S000', pid = (10, 1)
    SQL> select * from v$shared_server;
    NAME PADDR STATUS MESSAGES BYTES BREAKS CIRCUIT
    IDLE BUSY REQUESTS
    S000 212556FC WAIT(COMMON) 0 0 0 00
    749848 0 0
    Listener.log
    =========
    28-MAY-2007 13:38:35 * http * handoff * 12518
    TNS-12518: TNS:listener could not hand off client connection
    TNS-12571: TNS:packet writer failure
    TNS-12560: TNS:protocol adapter error
    TNS-00533: Connection dissolved or not yet made
    32-bit Windows Error: 57: Unknown error
    28-MAY-2007 13:38:35 * http * handoff * 12518
    TNS-12518: TNS:listener could not hand off client connection
    TNS-12571: TNS:packet writer failure
    TNS-12560: TNS:protocol adapter error
    TNS-00533: Connection dissolved or not yet made
    32-bit Windows Error: 57: Unknown error
    D:\oracle\ora92\admin\biprod\bdump>lsnrctl status
    LSNRCTL for 32-bit Windows: Version 9.2.0.7.0 - Production on 28-MAY-2007 15:26:02
    Copyright (c) 1991, 2002, Oracle Corporation. All rights reserved.
    Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=host1)(PORT=1521)))
    TNS-12535: TNS:operation timed out
    TNS-12560: TNS:protocol adapter error
    TNS-00505: Operation timed out
    32-bit Windows Error: 60: Unknown error
    Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=IPC)(KEY=EXTPROC0)))
    STATUS of the LISTENER
    Alias LISTENER
    Version TNSLSNR for 32-bit Windows: Version 9.2.0.7.0 - Production
    Start Date 22-MAY-2007 16:37:37
    Uptime 5 days 22 hr. 48 min. 49 sec
    Trace Level off
    Security OFF
    SNMP OFF
    Listener Parameter File D:\oracle\ora92\network\admin\listener.ora
    Listener Log File D:\oracle\ora92\network\log\listener.log
    Listening Endpoints Summary...
    (DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=host1)(PORT=1521)))
    (DESCRIPTION=(ADDRESS=(PROTOCOL=ipc)(PIPENAME=\\.\pipe\EXTPROC0ipc)))
    Services Summary...
    Service "PLSExtProc" has 1 instance(s).
    Instance "PLSExtProc", status UNKNOWN, has 1 handler(s) for this service...
    Service "bod" has 1 instance(s).
    Instance "biprod", status UNKNOWN, has 1 handler(s) for this service...
    The command completed successfully
    What could be the reason??
    TIA,
    JJ

    How large is your listener.log? This error stack occurs, when listener.log has grown too large on the Windows platform. You have to delete and recreate the file, do this in regular intervals.
    Werner

  • Configaration error during the shared servies configuration.

    Hi All,
    I have configured the shared services, after that i want to test that. I went to
    http://localhost:28080/interop/
    I got an error:
    Error: An error was encountered while performing this operation.
    I tried lot, like reboot the system, deleted cookies. none works for me.
    Please help me , I have been trying from so many days.
    thanks in advance,
    Raj.

    Could you please check with <http://localhost:58080/interop>? if that doesn't work, then check the log file. (Hyperion_Home\logs\config\configtool.log or configtool_err.log). Also please paste the log details if possible.
    Thanks,
    KK

  • Found dead server (ORA-07445)

    hello guys.
    i have this odd situation in one of my database. Sometimes i have this in my alert log "found dead shared server 'S000', pid = (19,191)" with an ORA-07445 afterwards and which sometimes cause the application to return an error "Io exception: End of TNS data channel". I wonder why this appears in my db. by the way, the machine is our Materialized view db which is part of the master-mv replication setup. Im confused why in my master db, that error doesnt appear, so i guess this is not an application or configuration issue. I wonder if this has something to do with the current setup!!??? to understand further on my problem, here is my setup
    MASTER DB (no problem on this machine)
    - multi-threaded dual processor
    - 4GB memory
    - RAID 5 on SATA disk
    - Net8 is in Dedicated server connection
    - this machine is dedicated only for the database. no other applications running
    - OS is redhat 9
    - located in our main server farm in US
    MATERIALIZED VIEW DB (this is where the errors appears)
    - 1 processor
    - 1GB memory
    - IDE disk
    - Net8 is in shared server connection
    - this machine is not only a db server but also the web application and the middle tier so jboss resides on this machine.
    - OS is in fedora core 3
    - located in one of our office somewhere in ASIA
    now you know my setup. so, any comments on this!?? if you'll ask me, my best findings why this error occurs is that maybe this has something to do with the resource!?? as you can see, jboss and oracle resides in one machine (currently, this is our setup due to budget constraints) so maybe this is the reason why i had a dead shared connection and the db has memory dumps?? and also, is there a way to test if the problem is due to lack of resources on the machine!?? i need to know so that i can justify to the management that we need to add machine dedicated for jboss and not with this kind of setup. its like our whole architecture was put into 1 machine (i know, this is a bad architecture). currently, we dont see any problem with the application as far as we know (since theres no complaints from the customers), though there is a proof the the application returns end of TNS channel error. I need some inputs from you guys so that i will have an idea on how to justify this to the management... i hope i can get reply from you..
    Thanks
    Erwin

    please check metalink..
    ORA-07445 [ACCESS_VIOLATION] [unable_to_trans_pc] on Windows Platforms [ID 456801.1]
    ORA-07445 [ACCESS_VIOLATION] [_kkslmtl+524] [PC:0xF41EC0] [ADDR:0x0] [UNABLE_TO_READ] [] [ID 755278.1]
    Regards
    Rajesh

  • Shared server problem.

    Has anyone every had this issue occur?
    I had a system slow down on one of our database servers, running Linux AS3 Oracle 10g (10.1.0.2.0). During the slow down (CPU spikes), sqlplus querries were running very slow on our client/server applications connected. The alert log shows this:
    found dead shared server 'S002', pid = (13, 1)
    I have never seen this before, what do you recommend. We normally run shared servers set at 50, I found that this was a bit high considering we normally see s001-s005 running. Shortly after the dead server error I dropped the shared servers from 50 to 20, that seemed to curb the high CPU spikes. We currently run on dispatcher as well, should I have more dispatchers running to handle the 20 shared servers? I have done the math, and it looks like we need only 2 dispatchers, but I was curious to know if increasing dispatchers and lowering shared servers is a true benefit to performance.
    Thanks again for any advice, I do not know what impact a dead shared server means, does Oracle simply move to s003 and disregard s002 if it is dead?

    We changed our database from dedicated server mode to shared server (since users were not able to connect to the database...listener log showed tns-12518 and related errors....previously it used to show ora-04030 in alert log).
    You have switched to shared server just because users were unable to connect? I think you need to fix the problem before thinking of switching to shared server.

Maybe you are looking for