Monitor defunct process

Monitoring defunct process doesn't seem to work
adding a row in the process monitoring module with
name pattern "<defunct>"
doesn't matches any thing (Process count = 0)
where in fact, there is at least 10 on the host
how is this module suppose to work?
thanks

Hi Phler2,
Monitoring defunct process doesn't seem to work
adding a row in the process monitoring module with
name pattern "<defunct>"
doesn't matches any thing (Process count = 0)
where in fact, there is at least 10 on the host
how is this module suppose to work?I'm don't think the stock Process Monitoring module will pick up a defunct/zombie state. If you're just looking for raw zombie count it's in the SystemMonitor module:
http://www.halcyoninc.com/products/PlusPack/help/SystemMonitor/HALSolarisSystemAlert-process-h.html
...part of PlusPack (free 30-day eval):
http://www.halcyoninc.com/products/PlusPack/index.php
Regards,
[email protected]
http://www.HalcyonInc.com

Similar Messages

  • Monitoring BI PROCESS CHAIN with Solution Manager BPMON

    Hello,
    I'm trying to setup the monitoring of BI process chain using the BPMON
    I setup the process chain monitoring using  BPmon application monitoring "BI Process Chain Monitoring" (BOBIPCMO)
    but than I'm not able to read the process chain names,steps,and variant from satellite system
    On other systems,using different application monitor (like PI MONITOR) is working fine
    i'm using the latest plug-in ST-A/PI 01L_BCO700 on entire landscape
    My bi system is 7.01 EHP1 sp4
    solution Manager is 7.01 EHP1 SP20
    Any suggestiions?
    Thanks
    Marco

    Hello Marco,
    the problem is that the Process Chain monitor is not yet available with ST-A/PI 01L*. Unfortunately the empty hull that you see during the setup was shipped without the proper content. The Process Chain monitor will be properly shipped with the next ST-PI and ST-A/PI combination in Q2/2010.
    Sorry for any inconvenience caused.
    If you want to have this monitor earlier on it could be implemented via Customer Exit in case that you have some MaxAttention or Safeguarding engagement.
    Best Regards
    Volker

  • How to monitor RFC processing in XI

    Hi Friends,
    In the IDoc to RFC scenario please tell me How to monitor RFC processing in XI..
    Thanks and Regards
    Sri

    Sri,
      You can monitor in SXI_MONITOR/SXMB_MONI. If its Syncronous then u can see the request and response messages in the above mentioned transaction.
      To get more details you can go for RWB and select ur message and check for audit log's.
    Best regards,
    raj.

  • Monitor Integration process log in BPM

    Hi
    I am practising the following BPM
    /people/krishna.moorthyp/blog/2005/06/09/walkthrough-with-bpm
    and I wanted to know how to monitor integration process log in BPM.
    To monitor Integration process I logged, into transaction SXMB_MONI_BPE ->Process Selection->Selected corresponding integration process (in Service Field) - and Clicked Execute as suggested in the blog
    But I get the message work item doesn't exist ??
    What do I need to do in this case ??
    Thanks,
    Kiran

    SAKHARDANDE ,
    Go to sxmb_moni-->Monitor for processed xml message --> In the view box there is standard and process. Select Process for checking the messages of BPM.
    Else as in the weblog mentioned go to SXMB_MONI_BPE ->Process Selection->Select corresponding integration process (in Service Field) ->Execute->Select work item. Select your service which you have used in ID.
    If you dont find anything then first see whether you see your message in sxmb_moni  or not? If not then check your file sender communicaiton channel in adapter monitoring.
    Regards,
    ---Satish

  • Monitor all processes EXCEPT _Total

    Hi!
    My goal is to monitor all processes on a given server, and raise an alert if memory usage (working set) exceeds 1 GB. I've tried the following: https:// sites.google.com/site/scomblogs/journal-blog/processperformancemonitoringinscom
    The problem is that the instance _Total will always contain a value higher than 1 GB. I can't seem to figure out how to override this. Is it possible at all?

    Thanks a lot! The WMI link did the trick. My final script came out as follows:
    On Error Resume Next
    strComputer = "."
    ' Instantiate MOM API
    Set objMOM = CreateObject("MOM.ScriptAPI")
    Set ObjBag = objMOM.CreatePropertyBag()
    Set objWMIService = GetObject("winmgmts:" _
    & "{impersonationLevel=impersonate}!\\" _
    & strComputer & "\root\cimv2")
    Set colProcesses = objWMIService.ExecQuery _
    ("Select * from Win32_Process")
    counter = 0
    proc = ""
    For Each objProcess in colProcesses
    ' 1 GB
    wsetGb = objProcess.WorkingSetSize / 1073741824
    '100 MB
    ' wsetGb = objProcess.WorkingSetSize / 104857600
    IF (wsetGb > 1) THEN
    usr = objProcess.GetOwner(strNameOfUser,strUserDomain)
    counter = counter + 1
    proc = proc + objProcess.Name +", owner: " +strUserDomain + " " + strNameOfUser +" cmd: " +objProcess.CommandLine
    END IF
    Next
    IF (counter > 0) THEN
    Call ObjBag.AddValue("State","Error")
    Call ObjBag.AddValue("Culprit",proc)
    Call ObjBag.AddValue("WorkingSetSizeInGb",wsetGb)
    ELSE
    Call ObjBag.AddValue("State","Ok")
    Call ObjBag.AddValue("Culprit","None")
    Call ObjBag.AddValue("WorkingSetSizeInGb","Not Applicable")
    END IF
    IF Err.Number <> 0 THEN
    EH
    END IF
    Sub EH
    Call ObjBag.AddValue("State","Warning")
    Call ObjBag.AddValue("Culprit",Err.Source)
    Call ObjBag.AddValue("Warning",Err.Description)
    Err.Clear
    End Sub
    objMOM.AddItem(ObjBag)
    Call objMOM.ReturnItems

  • How to enable an monitor parallel processing in Oracle

    Hi All,
    I have 2 short questions:
    1. When we want parallel processing, we can either use a parallel hint in the query, or alter a table to be parrallel. My question is what is the difference in the following 2 syntax:
    a. ALTER TABLE myTable PARALLEL (DEGREE 3);
    b. ALTER TABLE myTable PARALLEL 3;
    Does the "DEGREE" keywor make any difference? or they both are same statements?
    2. When we enable parallel processing, how can we monitor oracle processes, to confirm that a certain table is being actually processed by multiple threads of a singe user process?
    An early response would be highly appreciated. Thanks.

    1)The parallel clause lets you change the default degree of parallelism for queries and DML on the table.
    2) PARALLEL DEGREE specifies the number of query server processes that can scan the table in parallel. Either specify a positive integer or DEFAULT which signifies to use the initialization parameter
    check further http://mywebsys.com/oracle/syntax/view_syntax.php?id=23
    Thanks

  • Getting Defunct processes while using Oracle iPlanet Web Server 7.0.9 in Solaris 10

    Hi,
    I have installed Oracle iPlanet Web Server 7.0.9 in our Solaris 10. \
    Successfully Configured iPlanet web server using the manual and the application scripts are working fine with this iPlanet webserver.
    Even if the application is working fine, the iPlanet webserver stop responding at a period of time and needs to be restart manually...
    Found that lots of defunct processes(PPID of webservd) are created and these defuncts (zombies) leads to stop the services.
    I am not good in iPlanet webserver configuration settings on the Solaris 10 OS.
    Could anyone help me to fix this issue ?
    Thanks,
    Antony.

    Here is the truss output (last few lines before the application hungs...
    /14:    2516.1065        0.0000 time()                                          = 1428330724
    /49:    2516.1382        0.0001 pollsys(0xFFFFFFFF71E3FB40, 1, 0xFFFFFFFF71E3FA80, 0x00000000) = 0
    /49:    2516.1386        0.0000 lwp_sigmask(SIG_SETMASK, 0xFFBFFEFF, 0x0000FFF7) = 0xFFBFFEFF [0x0000FFFF]
    /49:    2516.1389        0.0000 lwp_kill(66, SIG#0)                             Err#3 ESRCH
    /49:    2516.1412        0.0022 munmap(0xFFFFFFFF70D00000, 262144)              = 0
    /49:    2516.1413        0.0000 lwp_exit()
    /39:    2516.1553        0.0000 pollsys(0xFFFFFFFF7283FB40, 1, 0xFFFFFFFF7283FA80, 0x00000000) = 0
    /39:    2516.1557        0.0000 lwp_sigmask(SIG_SETMASK, 0xFFBFFEFF, 0x0000FFF7) = 0xFFBFFEFF [0x0000FFFF]
    /39:    2516.1559        0.0000 lwp_kill(75, SIG#0)                             Err#3 ESRCH
    /39:    2516.1561        0.0001 munmap(0xFFFFFFFF70400000, 262144)              = 0
    /39:    2516.1562        0.0000 lwp_exit()
    /25:    2516.1614        0.0000 lwp_park(0xFFFFFFFF73CFBA90, 0)                 Err#62 ETIME
    /16:    2516.1671        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /35:    2516.2558        0.0000 lwp_park(0xFFFFFFFF72EFBA90, 0)                 Err#62 ETIME
    /16:    2516.2673        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /16:    2516.3675        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /36:    2516.4317        0.0000 lwp_park(0xFFFFFFFF72CFBE60, 0)                 Err#62 ETIME
    /15:    2516.4505        0.0000 lwp_park(0xFFFFFFFF7613F9D0, 0)                 Err#62 ETIME
    /16:    2516.4677        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /16:    2516.5679        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /27:    2516.6533        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF7383FD20, 0x00000000) = 0
    /16:    2516.6681        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /1:     2516.7105        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF7FFFA6A0, 0x00000000) = 0
    /16:    2516.7684        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /34:    2516.8137        0.0000 lwp_park(0xFFFFFFFF730FBA70, 0)                 Err#62 ETIME
    /16:    2516.8685        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /16:    2516.9686        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /16:    2517.0688        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /14:    2517.1057        0.0001 pollsys(0x00000000, 0, 0xFFFFFFFF744FBDA0, 0x00000000) = 0
    /14:    2517.1061        0.0000 time()                                          = 1428330725
    /25:    2517.1618        0.0000 lwp_park(0xFFFFFFFF73CFBA90, 0)                 Err#62 ETIME
    /16:    2517.1691        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /35:    2517.2561        0.0000 lwp_park(0xFFFFFFFF72EFBA90, 0)                 Err#62 ETIME
    /16:    2517.2693        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /16:    2517.3695        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /36:    lwp_park(0xFFFFFFFF72CFBE60, 0) (sleeping...)
    /15:    lwp_park(0xFFFFFFFF7613F9D0, 0) (sleeping...)
    /16:    2517.4696        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /54:    2517.5362        0.0000 pollsys(0xFFFFFFFF7193FB40, 1, 0xFFFFFFFF7193FA80, 0x00000000) = 0
    /54:    2517.5364        0.0000 lwp_sigmask(SIG_SETMASK, 0xFFBFFEFF, 0x0000FFF7) = 0xFFBFFEFF [0x0000FFFF]
    /54:    2517.5367        0.0000 lwp_kill(53, SIG#0)                             Err#3 ESRCH
    /54:    2517.5368        0.0001 munmap(0xFFFFFFFF71A00000, 262144)              = 0
    /54:    2517.5369        0.0000 lwp_exit()
    /16:    2517.5697        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /27:    pollsys(0x00000000, 0, 0xFFFFFFFF7383FD20, 0x00000000) (sleeping...)
    /16:    2517.6699        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /1:     2517.7106        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF7FFFA6A0, 0x00000000) = 0
    /1:     2517.7111        0.0001 close(26)                                       = 0
    /1:     2517.7113        0.0000 write(24, " 8", 1)                              = 1
    /1:     2517.7115        0.0000 close(23)                                       = 0
    /1:     2517.7116        0.0000 close(24)                                       = 0
    /1:     2517.7118        0.0000 lwp_unpark(25)                                  = 0
    /25:    2517.7118        0.0000 lwp_park(0xFFFFFFFF73CFBA90, 0)                 = 0
    /1:     2517.7120        0.0000 lwp_unpark(35)                                  = 0
    /35:    2517.7120        0.0000 lwp_park(0xFFFFFFFF72EFBA90, 0)                 = 0
    /25:    2517.7121        0.0000 lwp_sigmask(SIG_SETMASK, 0xFFBFFEFF, 0x0000FFF7) = 0xFFBFFEFF [0x0000FFFF]
    /25:    2517.7122        0.0000 lwp_kill(61, SIG#0)                             Err#3 ESRCH
    /35:    2517.7123        0.0000 lwp_sigmask(SIG_SETMASK, 0xFFBFFEFF, 0x0000FFF7) = 0xFFBFFEFF [0x0000FFFF]
    /25:    2517.7124        0.0001 munmap(0xFFFFFFFF71200000, 262144)              = 0
    /25:    2517.7125        0.0000 lwp_exit()
    /1:     2517.7126        0.0000 lwp_wait(25, 0xFFFFFFFF7FFFA47C)                = 0
    /35:    2517.7127        0.0000 lwp_kill(52, SIG#0)                             Err#3 ESRCH
    /35:    2517.7128        0.0001 munmap(0xFFFFFFFF71B00000, 262144)              = 0
    /35:    2517.7129        0.0000 lwp_unpark(1)                                   = 0
    /1:     2517.7129        0.0000 lwp_park(0x00000000, 0)                         = 0
    /35:    2517.7130        0.0000 lwp_exit()
    /1:     2517.7132        0.0000 lwp_wait(35, 0xFFFFFFFF7FFFA47C)                = 0
    /1:     2517.7134        0.0000 close(13)                                       = 0
    /16:    2517.7700        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /34:    2517.8141        0.0000 lwp_park(0xFFFFFFFF730FBA70, 0)                 Err#62 ETIME
    /16:    2517.8702        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /16:    2517.9704        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /22:    2517.9743        0.0001 pollsys(0xFFFFFFFF73F3FA98, 1, 0xFFFFFFFF73F3F9D0, 0x00000000) = 0
    /16:    2518.0706        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF742FBDD0, 0x00000000) = 0
    /14:    2518.1063        0.0000 pollsys(0x00000000, 0, 0xFFFFFFFF744FBDA0, 0x00000000) = 0
    /14:    2518.1073        0.0000 time()                                          = 1428330726
    /1:     2518.1082        0.0000 lseek(1, 0, SEEK_CUR)                           = 0
    /1:     2518.1083        0.0000 lseek(2, 0, SEEK_CUR)                           = 4128
    /1:     2518.1085        0.0000 lseek(2, 0, SEEK_CUR)                           = 4128
    /1:     2518.1093        0.0000 lseek(1, 0, SEEK_CUR)                           = 0
    /1:     2518.1094        0.0000 lseek(2, 0, SEEK_CUR)                           = 4128
    /1:     2518.1095        0.0000 lseek(2, 0, SEEK_CUR)                           = 4128
    /1:     2518.1102        0.0000 _exit(0)

  • FNDLIBR defunct process on RAC/PCP

    Hi,
    EBS - 12.1.3
    DB - 11.2.0.1
    OS - Redhat Linux
    In one of our client we have RAC with PCP setup.
    We are facing issue of FNDLIBR defunct process on one of our application node.
    [applprod@XXXXXXX ~]$ ps -ef|grep defunct
    applprod 16541 3970  0 00:02 ?        00:00:00 [FNDLIBR] <defunct>
    applprod 16542 3970  0 00:02 ?        00:00:00 [FNDLIBR] <defunct>
    applprod 16544 3970  0 00:02 ?        00:00:00 [FNDLIBR] <defunct>
    applprod 16545 3970  0 00:02 ?        00:00:00 [FNDLIBR] <defunct>
    applprod 23207 23077 0 09:43 pts/1    00:00:00 grepdefunct
    applprod 27286 3970  0 Apr11 ?        00:00:00 [FNDLIBR] <defunct>
    applprod 29654 3970  0 Apr12 ?        00:00:00 [FNDLIBR] <defunct>
    This defunct process get cleared by itself after some time.
    Could you please advise, what causing this process to go into defunct state and get cleared by itself.

    A defunct process, just like a regular process, may be able to keep a file open on the filesystem. If you stop the environment to take a backup, and processes are still left running when the backup starts, the backup is not clean. Doesn't matter if the process is defunct or not. It's the system telling you there is a problem with a process. The bigger picture is that they often go away by themselves, but after what time ? Is it 1 second, 1 minute, 1 hour ... ? The longer, the worse the problem. Some may not go away. In most cases, the explication is that it is a known feature when you stop component X, and you can determine an average time frame in which these processes continue to live, before they are gone.
    FNDLIBR and defunct are know to be a common combination spotted, but they should disappear after some minutes, assuming no requests are still running.

  • How to monitor parallel processing

    Hi All,
    I have 2 short questions:
    1. When we want parallel processing, we can either use a parallel hing in the query, or alter a table to be parrallel. My question is what is the difference in the following 2 syntax:
    a. ALTER TABLE myTable PARALLEL (DEGREE 3);
    b. ALTER TABLE myTable PARALLEL 3;
    Does the "DEGREE" keywor make any difference? or they both are same statements?
    2. When we enable parallel processing, how can we monitor oracle processes, to confirm that a certain table is being actually processed by multiple threads of a singe user process?
    An early response would be highly appreciated. Thanks.

    user566817 wrote:
    2. When we enable parallel processing, how can we monitor oracle processes, to confirm that a certain table is being actually processed by multiple threads of a singe user process?There are a number of virtual performance views that can be used. Please refer to the Oracle® Database Reference guide for details on these.
    Had a look though my scripts and I have this one.. cannot recall if I "borrowed" it from somewhere and customised it and how old it is.. but it should (hopefully) still be mostly correct. It uses the virtual view v$px_process to determine the list of current PQ slaves in the pool and if they are used, map them to the Oracle session using them.
    select  distinct
            x.server_name           as "PQ",
            x.status                as "Status",
            x.sid                   as "OraPID",
            w2.sid                  as "Parent OraPID",
            v.osuser                as "O/S User",
            v.schemaname            as "User",
            w1.event                as "Child Wait",
            w2.event                as "Parent Wait"
    from    v$px_process    x,
            v$lock          l,
            v$session       v,
            v$session_wait w1,
            v$session_wait w2
    where   x.sid =! l.sid(+)
    and     to_number (substr(x.server_name,3)) = l.id2(+)
    and     x.sid = w1.sid(+)
    and     l.sid = w2.sid(+)
    and     x.sid = v.sid(+)
    and     nvl(l.type,'PS') = 'PS'Use at own risk - best would be to verify that this is still valid using the Reference Guide. Or create similar ones using the available V$ views for the details you want to see (e.g. SQL statement executed per PQ, etc).

  • How to decrease the spamming defunct processes by run_service.sh script

    Hi
    I am running OWB applications.
    I got a alert showing defunct process are incresed to 40 by the script run_services.sh and now they are increased to 68
    -rwxrwxr-x 1 oracle dba 1473 Mar 28 2005 run_service.sh
    /opt/oracle/products/OWB/owb/bin/unix/run_service.sh -a
    is spamming the defunct process
    How can i terminate those defunct processes
    Please do needful
    Thank you
    Tokala

    Thanks and
    I need to kill all 58 defunct process
    Is there any awk program to kill all defunct process
    Tokala

  • How BAM can use to monitor B2B process?

    We have a BPEL process that is using B2B piece for connecting Trading partner and want to monitor whole process from Start of the BPEL to end of the B2B.
    Here are the questions.
    Q1. We know by using BAM sensor on BPEL process we can monitor the BPEL prcocess but how about B2B piece?
    Q2. What are the best practices for capturing and reading transactional data in various phases and functional acknowledgement messages from a B2B server?
    Thanks in advance for your reply.

    Please send a note to [email protected] for a document on BAM-B2B integration.

  • Process Monitoring: Monitor BI Process Chains

    Hello,
    does someone know how to monitor Bi process chains in SolMan Process Monitoring? I tried to add the chain CCMS node, but for some reason it does not accept the chain (I guess because it is a dynamic created CCMS context).
    Regards,
    Frank

    Hi,
    I will monitor one process with the name
    pmd, but 4 processes match on these chracters. How
    can i tell SMC that the Process Monitoring shows only
    the process who exactly Match the entered value ?Have you tried the extra optional user and/or argument filtering?
    If all 4 of your processes run as the same user, with the same arguments (or none at all) then you could also symlink a different file name to the binary you want so it appears in your ps output as something unique. That only works if you have control over what launches that process to begin with (so you can change it to call that new name as well).
    If you still can't make that process unique you could also write a small script to monitor it and dump back the interesting bits to stdout to be read by ScriptRunner (http://www.halcyoninc.com/products/ScriptRunner/index.php)
    Regards,
    [email protected]
    http://www.HalcyonInc.com

  • Monitoring inbound processing and Sync/ Async communication on PI7.11

    Hi All,
    I am new to PI. We have following two requirements to be done on PI 7.11
    1)     Monitoring inbound processing (Txn Code SWF_INB_MON)
    2)     Monitoring Sync/ Async communication (Txn Code SXMS_SAMON)
    Kindly let me know how should I proceed or what document should I follow or any scenario is to be configured.

    Hi,
        1. Start inbound processing monitoring with transaction SWF_INB_MON or SWF_INB_ADM.
               Note that you require administrator authorization for transaction SWF_INB_ADM
        2. You can monitor the throughput of the inbound processing either on the basis of individual process system or system-wide. In this way you can display the throughput figures in any discrete time range, right down to seconds. A similar display for message packages allows you to monitor inbound processing with the same detail." 
         3. You can display runtime information about individual messages, for example, the total runtime of a message in inbound processing or just the runtime of a specific section of the inbound processing. For monitoring individual messages click on "Display Message Packages".
    regards,
    ganesh.

  • How to monitor BPEL process..? i want to see HTTP request message..

    Hi all,
    Is there any way to monitor BPEL process. In Jdev there is tool call
    HTTP analyzer but it only monitors process which are invoked from Jdev.
    Like Java webservices can monitored via that tool. But i want to monitor BPEL
    process. I want to see Http request message. It that possible.?
    thanks a lot.
    /mishit

    Any requests you want to monitor should be routed through obtunnel. So the webservice that calls your bpel process should use the url of the machine where obtunnel is located, and obtunnel should be configured with the details of the machine where the bpel process is located.
    also, you may find this thread of interest How to capture SOAP message using TCP Monitor in JDev?

  • defunct process when stop and start web server

    Dear all,
    I am using Sun One Web Server 6.1 SP6. When i try to stop and then start the web server. The <defunct> process is generated (his parent PID points to webservd).
    The following is the result:
    1st stop and start the web server (defunct process is generated)
    2nd stop and start the web server (defunct process disappears)
    3rd stop and start the web server(defunct process is generated)
    4th stop and start the web server (defunct process disappears)
    Here is my magnus.conf
    ServerID www
    ServerName www.abc.edu.hk
    ErrorLog /i03_01/SUNWwbsvr616/https-www/logs/errors
    PidLog /i03_01/SUNWwbsvr616/https-www/logs/pid
    User nobody
    DNS off
    Security off
    ClientLanguage en
    AdminLanguage en
    DefaultLanguage en
    RqThrottle 512
    StackSize 131072
    Init fn="load-modules" funcs="caspreq,caspinit,casptrans" shlib="/i03_01/casp/module/sunos5_optimized/netscape_6.x/nes_casp2.so"
    Init fn="caspinit" casplib="/i03_01/casp/asp-server-3078"
    Init fn="load-modules" funcs="htaccess-init,htaccess-find" shlib="/i03_01/SUNWwbsvr616/plugins/htaccess/htaccess.so"
    Init fn="htaccess-init"
    Init access="/i03_01/SUNWwbsvr616/https-www/logs/access" format.access="%Ses->client.ip% - %Req->vars.auth-user% [%SYSDATE%] \"%Req->reqpb.clf-request%\" %Req->srvhdrs.clf-status% %Req->srvhdrs.content-length% \"%Req->headers.referer%\" \"%Req->headers.user-agent%\"" fn="flex-init"
    Init fn="load-modules" shlib="/i03_01/SUNWwbsvr616/bin/https/lib/libj2eeplugin.so" shlib_flags="(global|now)"
    TempDir /tmp/https-www-61c30017
    # Load modules for PHP
    Init fn="load-modules" funcs="php4_init,php4_execute,php4_auth_trans" shlib="/i03_01/SUNWwbsvr616/bin/libphp4.so"
    Init fn="php4_init" LateInit="yes" errorString="Failed to initialize PHP!" [php_ini="/i03_01/SUNWwbsvr616/https-www/config/php.ini"]
    # Enable statistics
    Init fn="stats-init" update-interval="5" virtual-servers="100" profiling="yes"
    any ideas? Thank

    - which OS & version are you running into this issue ?
    - By definition, defunct processes (or zombie processes) are processes that have exited prior to it's parent calling a wait() on it to get the return status .
    - couple of things I can think of to narrow down the cause of this issue
    1. disable ASP & php plugins and see if this still causes this issue
    2. try starting the web server with log level to finest and see fi you get any useful information from the server error logs
    hope this helps

Maybe you are looking for