Max concurrent RT engines is ignored

Dear experts,
my use case is to execute initial load jobs for a large number of backend systems in an automated fashion without any resource contention.
I've configured an IdM 7.2 SP7 system with one dispatcher to start at most two concurrent runtime engines. This is configured both as a global setting and on the single dispatcher (see screenshots).
Then I execute a normal job with one toGeneric pass whose entry script uses uRunJobNow() to start the ~50 initial load jobs. My expectation was that these jobs would now be executed in sequence, with at most two jobs running in parallel.
Unfortunately, that's not the case. In fact, up to 16 runtimes are started in parallel, so resource contention occurs with the usual problems (OutOfMemory, timeout etc.). The dispatcher log emits warning messages:
Not allowed to start ANY runtimes (right now running runtimes = 16, allowed total runtimes = 2
Question: is there anything I need to change in the dispatcher configuration to achieve the desired behavior?
Best regards,
Lambert

I think that in SP7 it simply counts running jobs (mc_jobs.state=2), so if a job takes a long time to start and the check frequency is low (1 or 2 seconds for instance) then the dispatcher has time to start more runtimes before the state of 2 jobs are changed to 2. I believe that later versions of the dispatcher checks the pids that were started to see how many are still active.
To test, change the max engines to start to 1, increase the dispatchers policy/check interval to 30 seconds. This should result in 1 runtime being started every 30 seconds and make sure that the jobs have had time to start properly and set their state to running. Setting the state to 2 should be almost instantaneous though so that 16 have time to start almost sounds like some setting is being ignored. You might also want to check that there isn't an excessive amount of javaclasses getting loaded that bloats the startup times.
Br,
Per Christian
Message was edited by: Per Krabsetsve

Similar Messages

  • Does Max threads in Viusal Admin overide Max concurrency in Channels...

    Hi All,
    Here are the settings for no of JDBC threads in our Java Visual Admin:
    AFWCall 10
    AFWRecv 5
    AFWRqst 10
    AFWSend 5
    JDBC_http://sap.com/xi/XI/SystemCall 5
    JDBC_http://sap.com/xi/XI/SystemRecv 5
    JDBC_http://sap.com/xi/XI/SystemRqst 5
    JDBC_http://sap.com/xi/XI/SystemSend 5
    Now the issue is that even if I put Max concurrency as 1 or 5 in JDBC adapter, I see 5 messages with status 'Delivering' at one time.
    So does that mean Max threads in Viusal Admin overide Max concurrency in Channels
    Pls advice...on this...
    Regards,
    XIer
    Edited by: XIer on May 14, 2008 2:16 AM

    Hi All,
    Here are the settings for no of JDBC threads in our Java Visual Admin:
    AFWCall 10
    AFWRecv 5
    AFWRqst 10
    AFWSend 5
    JDBC_http://sap.com/xi/XI/SystemCall 5
    JDBC_http://sap.com/xi/XI/SystemRecv 5
    JDBC_http://sap.com/xi/XI/SystemRqst 5
    JDBC_http://sap.com/xi/XI/SystemSend 5
    Now the issue is that even if I put Max concurrency as 1 or 5 in JDBC adapter, I see 5 messages with status 'Delivering' at one time.
    So does that mean Max threads in Viusal Admin overide Max concurrency in Channels
    Pls advice...on this...
    Regards,
    XIer
    Edited by: XIer on May 14, 2008 2:16 AM

  • What is CR returned error code / status when hit max concurrent access ?

    Hi,
    Do anyone knows what is the returned error code or status return from crystal report object when it hits its max concurrent access license?
    I'm developing on VS 2005 With crystal report, which comes along with it. If i'm not wrong its CR 9. Correct me if I got e versioning wrong. =)
    Thanks in advance!!

    Hi,
    Do anyone knows what is the returned error code or status return from crystal report object when it hits its max concurrent access license?
    I'm developing on VS 2005 With crystal report, which comes along with it. If i'm not wrong its CR 9. Correct me if I got e versioning wrong. =)
    Thanks in advance!!

  • Max Concurrent API Alert with Windows MP 6.0.7061

    Several of our servers are reporting the Max Concurrent API Reached alert. Most are Windows 2012. We are running the latest version of the MP - 6.0.7061. I have collected the performance data and analyzed it as described in KB 2688798 which indicates
    there is no issue. Most notable is that the Average Hold Time is consistently 0, which makes the formula always come out to 0, indicating no issue. The output from the diagnostics in SCOM is:
    Date and Time:
    6/17/2014 11:14:45 AM
    Property Name
    Property Value
    Detection Time
    06/17/2014 11:14:45
    Secure Channel
    _Total
    Semaphore Timeouts
    5
    Semaphore Waiters
    0
    Semaphore Acquires
    1441011
    Semaphore Holders
    0
    Average Semaphore Hold Time
    0
    State
    0
    ErrorNumber
    0
    Was wondering if anyone else has seen this behavior with this MP version. I know there was a bug in previous versions but according to the release notes it has been fixed. Not sure it matters but most of the servers affected are SharePoint.

    Hi
    There is an bug in the monitor. please disable this monitor with override till issue is fixed in next version.
    Regards 
    sridhar v

  • Max Concurrent Audio Conferences my Lync Server can handle

    Hello Experts, 
    Can somebody help with this calculations:
    I was asked by my manager to to find out the "Max concurrent audio conferences" that our Lync Front End server can handle. I have tried used Capacity calculator, but I sometimes get confused with it.
    My information:
    Conferencing codec is G.722 codec,
    we have 10,000 users in Lync Front End server
    Server runs: dual processor, hexa-core 2.3 ghz
    Assuming 30 % concurrent conferences
    Thanks Guys

    Given the user model that 5% of users in a pool may be in a conference at any one time, a pool of 80,000 users could have about 4,000 users in conferences at one time. These conferences are expected
    to be a mix of media (some IM-only, some IM with audio, some audio/video, for example) and number of participants. There is no hard limit for the actual number of conferences allowed, and actual usage determines the actual performance. Then you could have
    500 users or less in conferences at one time.
    You can refer to the link below to calculate.
    http://technet.microsoft.com/en-us/library/gg615015.aspx
    Lisa Zheng
    TechNet Community Support

  • Max Concurrent connections supported by UTL_HTTP

    Hi,
    Can anyone help me out to find how many concurrent connections can UTL_HTTP.REQUEST_PIECES support. Is there any parameter that governs max no.of connections? If so, pls post the parameter name.

    Assaf:  as always, thanks so much for the info!
    When you say double-sided, do you mean a memory stick which has chips on both sides; as in 8 chips on each side?
    I bought the Kingston memory one stick of 128MB PC100/PC133 which has chips on one side.  I also bought the 256MB PC100/PC133 which has chips on both sides.
    I currently have both these memory sticks in the MS 6119 and they appear to be working well.  I am running Prime95 torture test just to make sure everything is working as it should be.
    Currently I have the following memory in the MS 6119:
    Bank 0 - the original PC100 128MB stick which has 8 chips on each side.
    Bank 1 - the Kingston (KVR133Q/128R) PC100/PC133 128MB with 8 chips on one side.
    Bank 2 - the kingston (KVR 133Q/256R) PC100/PC133 256MB with 8 chips on both sides.
    The total is, of course is 512MB - and appears to be working just fine.  Windows LOVES RAM  :D

  • Max concurrent connections

    Coming back to a topic that was posted a few months ago with
    an
    additional question:
    >I believe you mean to ask, "What is the maximum number of
    >concurrent connections such as those used in loadMovie or
    XML.load?"
    >If this is what you're asking then there are no current
    limits. However,
    >the sending server may limit concurrent connections for a
    single client
    >to any number so the loads may fail if they exceed this
    limit. The
    >limitations are therefore server-based, not Flash based,
    and they will
    >differ from server to server.
    My question:
    While Flash doesn't enforce a limit, do any/all browsers
    enforce
    limits? If I wanted to load up 50, 100, 200 movies at once,
    where if
    anywhere would I be limited?
    Also, it seems that, in a test program loading a bunch of
    images from a server concurrently, I am not able to get more than
    two at once. Is this purely a server issue?
    -andy

    Hi,
    I imagine it does.
    Also I guess if we are talking about just some random UDP traffic it would also mean that the default timeout for a connection would be 2min. The most usual UDP traffic would probably be DNS querys. In those cases I presume though that the UDP connections dont stay on firewall for long as long as the firewall sees the DNS reply.
    But as I said if we are talking about some random UDP traffic that is allowed through the firewall I would guess it stays in the connection table of the firewall for a couple of minutes. So you might be looking at 1000 concurrent connections or even more?
    I have once witnessed a single server sending so much UDP traffic that it reached the connection limit of an ASA5540 which is 400 000 concurrent connections.
    - Jouni

  • ASA max concurrent connections

    If I have a thousand nodes from the public each perform a UDP ping to a server behind the ASA, does each count as a concurrent connection?

    Hi,
    I imagine it does.
    Also I guess if we are talking about just some random UDP traffic it would also mean that the default timeout for a connection would be 2min. The most usual UDP traffic would probably be DNS querys. In those cases I presume though that the UDP connections dont stay on firewall for long as long as the firewall sees the DNS reply.
    But as I said if we are talking about some random UDP traffic that is allowed through the firewall I would guess it stays in the connection table of the firewall for a couple of minutes. So you might be looking at 1000 concurrent connections or even more?
    I have once witnessed a single server sending so much UDP traffic that it reached the connection limit of an ASA5540 which is 400 000 concurrent connections.
    - Jouni

  • Max Concurrent Users for Oracle 8i

    I am trying to find out information (white paper) that states how many concurrent users 8i can handle. Does anyone know of somewhere to get this data?
    Thanks in advance,
    JS

    Can you post the entire dump from the console?
    Where are you getting the term "Concurrent Users"? Concurrent Users is a term Oracle used to use (or maybe still do) when licensing products. I'm just interested where you are seeing this from.

  • Max Concurrent users

    Has any one idea how many concurrent users can Sun One 6.0 support ?
    We are doing Load testing of web application on Sun One 6.0 using LoadRunner. The maximum concurrent users that it is supporting is around 25. If Users are more than 25 then the application context of web application gets refreshed and few user transactions gets failed.
    Is there any way to exceed this limit ? and plz tell if anyone has done load testing on Sun One 6.0

    well by concurrent users i mean 25 users logging at that very instant.
    And application context (servletcontext) gets refreshed means - like i am maintaing connection pool object and user hashmap object at application level.
    code in JSP ::application.setAttribute ("connpool",ObjConnPool)
    when no of concurrent users get more, all logging at same time, then these objects r removed from application (servletcontext) and i get null pointer exception as objects r no more binded to application(servletcontext). This definately means application(servletcontext) has refreshed

  • Application Engine hangs

    Hi
    I find that for some reason only programs of type Application Engine are going into queued status.
    All the other programs are running fine.
    The Process scheduler processes application engine programs only after cancelling all application engine programs in queued status
    Can you please let me know what possible causes could be attributed to this behaviour

    Cyril
    As Nick told, find for earlier AE's that are in error/delete and clean them up from Prcs monitor. then,
    *Option1
    Reconfigure the prcs scheduler, clear cache and restart back. Check the last one that is in Queued. If it picks UP then let the backlog clears until u trigger a new one.
    *Option2
    Reconfigure the Prcs Scheduler by selecting 'n' to disable the PSAESRV's, then clear cache and start back.
    IN either case you can try increase the logfence in psprcs.cfg so that it gives detailed info in the logs. btw Wat is the Max concurrency set on AE's? is the issue on both NT/UNX?? (or) Just one of them?
    vKing.

  • Limit concurrent user logins

    Hi all,
    One of our customers is trying to limit the number of concurrent user logins to 1.
    He has deployed a 2500 WLC (v7.4) with a Dot1x SSID. Authentication against external radius server (IAS).
    Configured the following:
    Max Concurrent Logins for a user name: 1
    But doesn't work despite of the value configured in "Max-Login Ignore Identity Response" option (enable|disable).
    My doubt here is if that these parameters just work when we are using local authentication or if it could be a bug with this particular software train.
    I've found contradictory info regarding this particular topic.
    Thanks in advance.
    Best regards,
    Alberto

    Hi Saurav,
    Thanks for the info provided.
    Our problem here is that despite the value of max-login-ignore-identity-response (enable|disable) we always can establish multiple simultaneous connections with the same username credentials. 
    Does Max Concurrent Logins for a user name work with external radius authentication? Are we missing something else?
    Thanks in advance.
    Best regards,
    Alberto

  • Amount of max possible simultaneous INSERTs depend on what ??

    Hi all,
    can somebody comment on following statement , whether its is absolutely true, only a part of the truth or completely wrong :
    "The maximum possible amount of transactions that are simultaneously performing inserts into an Oracle- table is MAINLY constrained by the available ITL-slots of that table." (Assumption : Hardware performance identical)
    Background : We have a mission critical table where every user performs inserts (some kind of an audit table). These inserts are part of the original business transaction and are -normally- very fast.
    Problem is that the commit of the original business txn is only issued after populating this audit table.
    We were asked how many hanging user transactions would be required in order to completely lock the table for further inserts and as a consequence completely hangup the application.
    Cheers,
    Mig

    mbobak wrote:
    2.) You didn't mention version, or whether you're using ASSM, but, if you're still using MSSM (free list managment), then you'll need to consider freelists as well. The freelists parameter should be set to the max concurrent number of inserts on a segment.
    Mark,
    I wasn't planning to go into details about how to configure the number of blocks available for insertion, but you're right: freelists is a significant setting, as is freelist groups. ASSM is roughly equivalent to just freelists 16 freelist groups 1 - which is why systems with very high rates of concurrent DML should seriously consider avoiding it.
    P.S. Come to think of it, you're more likely to have trouble from buffer busy waits before you run into problems with ITL waits if you try increasing insertion concurrency by setting a high INITRANS.
    Regards
    Jonathan Lewis
    http://jonathanlewis.wordpress.com
    http://www.jlcomp.demon.co.uk
    "Science is more than a body of knowledge; it is a way of thinking" Carl Sagan

  • XML Publisher 5.6.2 Document Processor Engine

    Would I be correct in assuming that sending the xapi: tags through the regular concurrent request engine would be a fultile exercise in order to burst an invoice batch into multiple PDFs?
    Could somebody please confirm that I must use a JAVA concurrent program wrapper around the request in order to use the document processor engine to burst and save on the file system?
    Thanks in advance,
    G.

    XML Publisher 5.6.2 API may be used in JDeveloper 11g by importing the Oracle JDBC library definition from JDeveloper 10.1.3. Instead of the Oracle JDBC library in JDeveloper 11g create a Oracle JDBC library from the JDeveloper 10.1.3 Oracle JDBC library JAR files.

  • Max no of calls that can be queued in ICM

    Where can I find for a given installation the max no of calls that can queued for given ICM system.

    I looked at that but I think the answer is still not well defined. It talks about max call per second.That means that it can receive 300 max calls per second in ICM/IVR system.  It does not say anything about how many concurrent calls can sit in the system and be queued.
    For example on IVR side, we clearly know that it can support X no of  max concurrent calls or whatever the figure may be based on ports.  However for queued calls in ICM, it does not seem to very obvious. Unless I am missing something here.

Maybe you are looking for