E.wait_class 'Idle'

Hi,
I'm following this article :
http://www.dba-oracle.com/oracle10g_tuning/t_waits_events_signatures.htm
And there is a query : ash_user_wait_time.sql
select
   s.sid,
   s.username,
   sum(h.wait_time + h.time_waited ) "total wait time"
from
   v$active_session_history     h,
   v$session                  s,
   v$event_name                  e
where
   h.session_id = s.sid
and
   e.event_id = h.event_id
and
   e.wait_class <> 'Idle'
and
   s.username IS NOT NULL
group by
   s.sid, s.username
order by 3;But it shows the Idle sessions also even if e.wait_class is different from 'Idle'.
To show thatI modified it to :
select
   s.sid,
   s.username,
   s.wait_class,
   sum(h.wait_time + h.time_waited ) "total wait time"
from
   v$active_session_history     h,
   v$session                  s,
   v$event_name                  e
where
   h.session_id = s.sid
and
   e.event_id = h.event_id
and
   e.wait_class <> 'Idle'
and
   s.username IS NOT NULL
group by
   s.sid, s.username, s.wait_class
order by 4;The result for me is :
       SID USERNAME   WAIT_CLASS total wait time
       139 USER1      Idle                 14328
       145 USER1      Idle                 21075
       159 USER1      Idle                270668
       140 USER2      Idle               1515525
       136 USER1      Idle               2374375I do not see the interest of e.wait_class <> 'Idle' condition. And the result of query without this condition is the same.
Any idea ? Thank you.

Because you need to suppress the idle event class, as idle events are events like 'Sqlnet message from client': they are inevitable, everyone will have them, and they usually can not be tuned.
Sybrand Bakker
Senior Oracle DBA

Similar Messages

  • Open cursors problem- j2ee + oracle 10g

    Hi,
    I am using EJB on sunOne application server 8.1., Oracle 10g DB.
    EJB container connects to Oracle DB through a set of connection pools.
    BMP for all entity beans.
    I have about 160 PL/SQL functions that make up the business logic of the online application. everytime the application runs, I get an increasing number of open cursors, including some of the ones that are explicitly closed within PL/SQL (inspection with sys.v_$open_cursor).
    I made sure all CallableStatements, PreparedStatements, RecordSets within the beans are closed and set to NULL. All PL/SQL functions use explicit cursors, so every select statement is managed within a cursor which is explicitly closed when the function finishes with it.
    From v$open_cursor, I identified the sessions with the cursors still open, and issued (ALTER SYSTEM KILL SESSION �sid, #serial�) for each of the sessions (this is done via a PL/SQL function for all inactive sessions).
    These sessions have state INACTIVE, and wait_class IDLE. This has Killed all sessions, but I was not able to use the application anymore. I suspect by killing those sessions we have also caused the connections between EJB container and the Oracle DB. The only way to use the application now is to stop and restart the sunONE domain � this is very inconvenient.
    Has anyone encountered a similar problem? any suggestions to reduce or eliminate the open cursors number? Please help.
    Thank you all

    Maybe you can try to have a smaller steady-pool-size and idle-timeout-in-millis for your connection pools.
    Also, if that's at all possible, have smaller number of connection pools being shared by more apps.
    Just my 2c.
    thanks.

  • Constraint takes long time to enable, other performance issues

    Hello and thanks for any suggestions.
    I have an instance that has 13 different schemas.  All schemas are working fine, but one.  This one schema, it appears to hang. 
    I am in the process of loading all 13 schemas with identical data.  12 loaded fine.  In the 13th schema the data loaded just fine, but the enabling of the constraint takes days (yes days).  The other regions completed this successfully in about 2 minutes. 
    The users also report that their application hangs in this schema, and only this schema.
    I am running Standard Edition Oracle V11.2.0.1 on a Windows platform running Windows Server 2008.  I have checked and the tablespaces for all the schemas are on the same disk packs.
    What other information can I give to help solve this?
    Thanks

    I would start by seeing why sessions are hanging:
    select sid,event,seconds_in_wait from v$session where wait_time=0 and wait_class<>'Idle';

  • Need a group by clause for two varibles in one table

    I can't find any way to merge two variables in a group by statement with both vertical and horizobtable output.
    The statement I got for a vertical output is (in 10gR2):
    select to_char(d.begin_interval_time, 'YYYY-MM-DD') as Datenumber,
    w.wait_class as Type,
    round(sum(w.total_waits)/1024/1024, 0) as "Tot-waits"
    from dba_hist_snapshot d, dba_hist_bg_event_summary w
    where d.snap_id = w.snap_id
    group by to_char(d.begin_interval_time, 'YYYY-MM-DD'), w.wait_class
    order by to_char(d.begin_interval_time, 'YYYY-MM-DD');
    2010-02-28 System I/O 234567
    2010-02-28 User I/O 34444
    2010-03-01 System I/0 773777
    But I wanted to have the following summary output in one SQL:
    System I/O User I/O Idle Commit .... and so on
    2010-02-26 3442322 344555 335455 5
    2010-02-27 533222 2233 445455 2334
    2010-02-28 3434444 244444 345555 39
    2010-03-01 34444 55445 3444 534
    Anyone have a idea?

    Do you mean this?
    SQL> select to_char(d.begin_interval_time, 'YYYY-MM-DD') as Datenumber,
      2  round(sum(decode(w.wait_class,'Concurrency', w.total_waits/1024/1024, 0)),0) as "Concurrency",
      3  round(sum(decode(w.wait_class,'System I/O', w.total_waits/1024/1024, 0)),0) as "System I/O",
      4  round(sum(decode(w.wait_class,'User I/O', w.total_waits/1024/1024, 0)),0) as "User I/O",
      5  round(sum(decode(w.wait_class,'Configuration', w.total_waits/1024/1024, 0)),0) as "Configuration",
      6  round(sum(decode(w.wait_class,'Other', w.total_waits/1024/1024, 0)),0) as "Other",
      7  round(sum(decode(w.wait_class,'Commit', w.total_waits/1024/1024, 0)),0) as "Commit",
      8  round(sum(decode(w.wait_class,'Idle', w.total_waits/1024/1024, 0)),0) as "Idle"
      9  from dba_hist_snapshot d, dba_hist_bg_event_summary w
    10  where d.snap_id = w.snap_id
    11  group by to_char(d.begin_interval_time, 'YYYY-MM-DD')
    12  order by to_char(d.begin_interval_time, 'YYYY-MM-DD');
    DATENUMBER Concurrency System I/O   User I/O Configuration      Other     Commit       Idle
    2010-02-22           1        298          0             0          2          0       2107
    2010-02-23           1        299          0             0          2          0       2114
    2010-02-24           1        300          0             0          2          0       2121
    2010-02-25           1        301          0             0          2          0       2129
    2010-02-26           1        303          0             0          2          0       2136
    2010-02-27           1        304          0             0          2          0       2143
    2010-02-28           1        305          0             0          2          0       2150
    2010-03-01           0        140          0             0          1          0        988Max
    http://oracleitalia.wordpress.com

  • Lmnop

    ===========================================
    DB-Maintenance checks
    ===========================================
    What is the Library Cache Hit Ratio, it should be >90%
    SELECT SUM (PINHITS) / SUM (PINS) * 100 FROM V$LIBRARYCACHE;
    What is Library Cache Reloads Ratio, It should be <1%
    SELECT SUM (PINS), SUM (RELOADS), SUM (RELOADS) / SUM (PINS) FROM V$LIBRARYCACHE;
    Dictionary Cache Miss Ratio should be <15%
    SELECT (SUM (GETMISSES) / SUM (GETS)) * 100 FROM V$ROWCACHE;
    Hit Ratio For DB Buffer Cache should be >90%
    Select (sum(GETS-GETMISSES)) / SUM(GETS)*100 "Dictionary Cache Hit Ratio" From v$rowcache;
    Full Table Scans Ratio should be <5%
    SELECT D.VALUE "disk", M.VALUE "mem", (D.VALUE / M.VALUE) * 100 "Ratio" FROM V$SYSSTAT M, V$SYSSTAT D WHERE M.NAME = 'sorts (memory)' AND D.NAME = 'sorts (disk)';
    If Full Table Scan is more than 5% run below query to find full details:-
    SELECT * FROM V$SYSSTAT WHERE NAME LIKE '%table scan%';
    #echo 'ALERT - Oracle Access (HRMI - 192.168.68.212) on:' `date` `who` | mail -s "Alert: Oracle Access from `who -m | cut -d"(" -f2 | cut -d")" -f1`" [email protected]
    =======================================================
    RMAN Backup job details:
    select to_char(START_TIME,'mm/dd/yy hh24:mi') start_time,INPUT_TYPE,STATUS,to_char(END_TIME,'mm/dd/yy hh24:mi') end_time,
    elapsed_seconds/3600 hrs,OUTPUT_BYTES_DISPLAY from V$RMAN_BACKUP_JOB_DETAILS order by session_key ;
    RMAN Backup details:
    select to_char(START_TIME,'mm/dd/yy hh24:mi') start_time,INPUT_TYPE,STATUS,to_char(END_TIME,'mm/dd/yy hh24:mi') end_time,
    elapsed_seconds/3600 hrs,OUTPUT_BYTES_DISPLAY from V$RMAN_BACKUP_JOB_DETAILS order by session_key ;
    Waits by class:
    Select wait_class, sum(time_waited), sum(time_waited)/sum(total_waits) Sum_Waits From v\$system_wait_class Group by wait_class Order by 3 desc;
    waits by instance:
    select event, time_waited from v\$system_event where ROWNUM <= 10 order by 1;
    waits datafile level:
    select f.file_name "Data File",count(*) "Wait Number",sum(h.time_waited) "Total Time Waited" from v$active_session_history h,dba_data_files f where h.sample_time between sysdate - 1/24 and sysdate
    and h.current_file#=f.file_id
    group by f.file_name
    order by 3 desc;
    waiting sessions sql:
    select h.user_id,u.username,s.sql_text,sum ( h.wait_time + h.time_waited ) "Total wait time" from v$active_session_history h, v$sqlarea s,dba_users u,v$event_name e where h.sample_time between sysdate - 1/24 and sysdate and h.sql_id=s.sql_id and h.user_id = u.user_id and e.event_id = h.event_id and e.wait_class <> 'idle'
    group by h.user_id,s.sql_text,u.username order by 4 desc;
    what are users currently waiting on:
    select s.sid,s.username,sum(h.wait_time+h.time_waited) " Total Waited Time " from v$active_session_history h,v$session s,v$event_name e where h.sample_time between sysdate - 1/24 and sysdate
    and h.session_id = s.sid
    and e.event_id = h.event_id
    and e.wait_class <> 'idle'
    and s.username is not null
    group by s.sid,s.username
    order by 1;
    buffer busy waits:
    select owner, segment_name, segment_type from dba_extents a, v$session_wait b
    where b.event='buffer busy waits' and a.file_id=b.p1;
    blocked sessions:
    SELECT b.session_id AS sid,
    NVL(b.oracle_username, '(oracle)') AS username,
    a.owner AS object_owner,
    a.object_name,
    Decode(b.locked_mode, 0, 'None',
    1, 'Null (NULL)',
    2, 'Row-S (SS)',
    3, 'Row-X (SX)',
    4, 'Share (S)',
    5, 'S/Row-X (SSX)',
    6, 'Exclusive (X)',
    b.locked_mode) locked_mode,
    b.os_user_name
    FROM dba_objects a,
    v$locked_object b
    WHERE a.object_id = b.object_id
    ORDER BY 1, 2, 3, 4;
    large objects in shared pool:
    select OWNER,NAME||' - '||TYPE object,SHARABLE_MEM
    from v\$db_object_cache
    where SHARABLE_MEM > 10000
    and type in ('PACKAGE','PACKAGE BODY','FUNCTION','PROCEDURE')
    order by SHARABLE_MEM desc;
    blocked sessions:
    SELECT b.session_id AS sid,
    NVL(b.oracle_username, '(oracle)') AS username,
    a.owner AS object_owner,
    a.object_name,
    Decode(b.locked_mode, 0, 'None',
    1, 'Null (NULL)',
    2, 'Row-S (SS)',
    3, 'Row-X (SX)',
    4, 'Share (S)',
    5, 'S/Row-X (SSX)',
    6, 'Exclusive (X)',
    b.locked_mode) locked_mode,
    b.os_user_name
    FROM dba_objects a,
    v\$locked_object b
    WHERE a.object_id = b.object_id
    ORDER BY 1, 2, 3, 4;
    Find out SGA usage:
    select round(used.bytes /1024/1024 ,2) used_mb,
    round(free.bytes /1024/1024 ,2) free_mb,
    round(tot.bytes /1024/1024 ,2) total_mb
    from
    (select sum(bytes) bytes from v$sgastat where name != 'free memory') used ,
    (select sum(bytes) bytes from v$sgastat where name = 'free memory') free ,
    (select sum(bytes) bytes from v$sgastat) tot
    ==============================================
    disk capacity: fdisk -l
    Total memory: grep MemTotal /proc/meminfo
    CPU Deatils: cat /proc/cpuinfo
    OS BIt: uname -a
    OS Version: cat /etc/redhat-release
    Check CPU is 32/64 Bit: getconf LONG_BIT
    =====================================================
    Sun OS:
    ==========
    check Total physical memory:
    # prtdiag -v | grep Memory
    # prtconf | grep Memory
    check Free physical Memory:
    # top (if available)
    # sar -r 5 10
    Free Memory=freemen*8 (pagesize=8k)
    # vmstat 5 10
    Free Memory = free
    For swap:
    # swap -s
    # swap -l
    OS BIt: isalist (sparcv9,amd64 then 64bit)
    OS Version: cat /etc/release
    CPU Deatils: psrinfo -v
    ================================================
    Established sessions for specific port:
    netstat -an|grep :1800|sort|wc -l
    netstat -an|grep :1800|sort|grep 'ESTABLISHED'|wc -l
    =====================================================

    answered

  • Virtual circuit wait

    Hii All..
    I am seeing the following wait in the result of the query.Have you any idea what is the reason of the wait ?
    SELECT event, total_waits, time_waited_micro
    FROM v$system_event
    WHERE wait_class <> 'Idle' order by 3 desc
    result is
         EVENT     TOTAL_WAITS     TIME_WAITED_MICRO
    1     virtual circuit wait      588966     241791462220
    2     SQL*Net more data from client     433132     78999049842
    3     direct path read     7570504     22167783009
    4     db file sequential read     10187059     18516379084
    BANNER
    Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - 64bit Production on AIX 5.3 ML 10

    Hii..
    In the metalink note ID 6653834.8
    This is a performance monitoring enhancement to split the
    'virtual circuit status' wait event into two new
    wait events:
    "shared server idle wait" - for when the shared server is
    idle waiting for something to do
    "virtual circuit wait" - for when the shared server is
    blocked waiting on a specific
    circuit / message
    The wait "virtual circuit status" no longer exist with this fix.

  • V$rman_status query slow on exadata

    I am trying to retrieve the latest information of rman backups for different object types and using following query:
    ALTER SESSION SET nls_date_format = 'mm.dd.yyyy hh24:mi:ss';
    SELECT (SELECT host_name||',' FROM v$instance) "HOST", (SELECT db_unique_name||',' FROM v$database) "DB_UNQ_NAME", start_time||',' "START_TIME", object_type||',' "OBJ_TYPE", status "STATUS" FROM v$rman_status WHERE object_type='DB FULL' AND start_time = (SELECT max(start_time) FROM v$rman_status WHERE object_type='DB FULL')
    UNION
    SELECT (SELECT host_name||',' FROM v$instance) "HOST", (SELECT db_unique_name||',' FROM v$database) "DB_UNQ_NAME", start_time||',' "START_TIME", object_type||',' "OBJ_TYPE", status "STATUS" FROM v$rman_status WHERE object_type='DB INCR' AND start_time = (SELECT max(start_time) FROM v$rman_status WHERE object_type='DB INCR')
    UNION
    SELECT (SELECT host_name||',' FROM v$instance) "HOST", (SELECT db_unique_name||',' FROM v$database) "DB_UNQ_NAME", start_time||',' "START_TIME", object_type||',' "OBJ_TYPE", status "STATUS" FROM v$rman_status WHERE object_type='ARCHIVELOG' AND start_time = (SELECT max(start_time) FROM v$rman_status WHERE object_type='ARCHIVELOG')
    UNION
    SELECT (SELECT host_name||',' FROM v$instance) "HOST", (SELECT db_unique_name||',' FROM v$database) "DB_UNQ_NAME", start_time||',' "START_TIME", object_type||',' "OBJ_TYPE", status "STATUS" FROM v$rman_status WHERE object_type='CONTROLFILE' AND start_time = (SELECT max(start_time) FROM v$rman_status WHERE object_type='CONTROLFILE')
    UNION
    SELECT (SELECT host_name||',' FROM v$instance) "HOST", (SELECT db_unique_name||',' FROM v$database) "DB_UNQ_NAME", start_time||',' "START_TIME", object_type||',' "OBJ_TYPE", status "STATUS" FROM v$rman_status WHERE object_type='DATAFILE FULL' AND start_time = (SELECT max(start_time) FROM v$rman_status WHERE object_type='DATAFILE FULL') ORDER BY 4;
    This query is running very fast on all non-exadata database but slow (infact don't give result for 15 minutes after which I have to cancel it) on some of the exadata databases.
    The waitevents are as follows:
    control file sequential read 
    cell single block physical read
    library cache lock           
    library cache pin            
    Disk file operations I/O     
    gc current block 2-way       
    SQL*Net message to client    
    events in waitclass Other    
    Any idea, what could be the possible cause and/or a better query to retrieve backup information?
    Regards,
    Abhinav

    Thanks for your help Hemant,
    The V$SESSION view just informs the following for the "waiting" session:
    EVENT SQL*Net message from client
    STATE WAITED KNOWN TIME
    WAIT_CLASS Idle
    As this doesn't seem very useful to me, i used the oradebug utility to do a hang analysis (should show the latch contention, isn't it ?)
    Here is (part of) the result
    is not in a wait:
    last wait: 40 min 45 sec ago
    blocking: 0 sessions
    current sql: select * from v$rman_status
    wait history:
    1. event: 'SQL*Net message from client'
    time waited: 38.355161 sec
    wait id: 13 p1: 'driver id'=0x62657100
    p2: '#bytes'=0x1
    * time between wait #1 and #2: 0.000001 sec
    2. event: 'SQL*Net message to client'
    time waited: 0.000001 sec
    wait id: 12 p1: 'driver id'=0x62657100
    p2: '#bytes'=0x1
    * time between wait #2 and #3: 0.000012 sec
    3. event: 'SQL*Net message from client'
    time waited: 0.000017 sec
    wait id: 11 p1: 'driver id'=0x62657100
    p2: '#bytes'=0x1
    To complement this, i looked at the corresponding process at the os level using simple top -p. This showed a constant processor consumption and an increasing memory usage
    .Seems that the process is not blocked, but is lost in its processing. Like and infinite loop...

  • RMAN hangs on v$rman_status select

    RMAN hangs after performing a backup or even simple commands.
    I found that the "guilty" sql is:
    select /*+ rule */ round(sum(MBYTES_PROCESSED)), round(sum(INPUT_BYTES)), round(sum(OUTPUT_BYTES)) from V$RMAN_STATUS START WITH RECID = :row_id and STAMP = :row_stamp CONNECT BY PRIOR RECID = parent_recid
    I tried to do a simple select on v$rman_status, which also hangs.
    I couldn't even desc v$rman_status.
    I tried to gather statistics on the underlying tables (X$KRBMRST, X$KSFQP, X$KCCRSR) at no avail (the problem has been described in Oracle 10.1, see DocID 375386.1)
    Another possible solution found would be to recreate the control file. Not the best solution on a production database.
    Environment: Oracle SE 11.2 on Suse SLES 10.2
    Any idea ?
    Edited by: user10770996 on Feb 28, 2010 5:38 PM

    Thanks for your help Hemant,
    The V$SESSION view just informs the following for the "waiting" session:
    EVENT SQL*Net message from client
    STATE WAITED KNOWN TIME
    WAIT_CLASS Idle
    As this doesn't seem very useful to me, i used the oradebug utility to do a hang analysis (should show the latch contention, isn't it ?)
    Here is (part of) the result
    is not in a wait:
    last wait: 40 min 45 sec ago
    blocking: 0 sessions
    current sql: select * from v$rman_status
    wait history:
    1. event: 'SQL*Net message from client'
    time waited: 38.355161 sec
    wait id: 13 p1: 'driver id'=0x62657100
    p2: '#bytes'=0x1
    * time between wait #1 and #2: 0.000001 sec
    2. event: 'SQL*Net message to client'
    time waited: 0.000001 sec
    wait id: 12 p1: 'driver id'=0x62657100
    p2: '#bytes'=0x1
    * time between wait #2 and #3: 0.000012 sec
    3. event: 'SQL*Net message from client'
    time waited: 0.000017 sec
    wait id: 11 p1: 'driver id'=0x62657100
    p2: '#bytes'=0x1
    To complement this, i looked at the corresponding process at the os level using simple top -p. This showed a constant processor consumption and an increasing memory usage
    .Seems that the process is not blocked, but is lost in its processing. Like and infinite loop...

  • Unable to capture the Idle time for BSP page

    Hi Experts,
    I want to capture the Idle time of my BSP page. If that is 5 mins then i have to display the pop up displaying the remaining time.
    Please let me know how to capture the IDLE TIME. not the time after the page is loaded.
    Any suggestion will be helpful.
    Aready checked in SDN but unable to get the solution so posting it.
    Thanks in advance.
    Sravanthi.V

    hi,
    After capturing the idle time iam giving the warning popup to user before 5mins of expiry.Now my requirement is if the user clicks on OK button of popup the page should get refresh. i.e.Idle time should of system should break and we have to get one more hour for expiry.
    Thanks in advance,
    Sravanthi.V

  • Windows always shows lock screen when idle, ignores setting

    On Windows 8, I had my computer set up so that the screen would turn off after a period of inactivity. When I would move the mouse or press a key, it would resume without asking me to log in. After upgrading to Windows 8.1, it is now always showing me the
    lock screen when resuming from idle. In the lock screen settings, I found a new option named "When my PC is inactive, show the locks screen instead of turning off the screen". I have disabled this setting, but my computer still locks every time the
    screen turns off.
    How can I get my computer to stop locking when it is idle?

    Hi,
    What about the settings in Screen Saver? Please check the Option "On resume, display logon screen" is unchecked.
    We
    are trying to better understand customer views on social support experience, so your participation in this
    interview project would be greatly appreciated if you have time.
    Thanks for helping make community forums a great place.

  • CDAQ lock-up on Windows 7 when app idle.

    I scarse few details to help fix an issue I witnessed while on-site yesterday.
    A LV app running under Win 7 using a cDAQ chassis will lock-up LV if the app is left idle for a couple of hours.
    WHile Idle it just grabs samples from the AI chnnels and displays same.
    Attempts to open the FP of the sub-VI grabbing the samples locks-up so I was not able to check if there where hardware errors. memory was pegged out at the time.
    While monitoring the app for over an hour I saw no memory leaks and did not have time to wait for three hours or more so my facts are limitied.
    For the sake of this post I am only interested in finding out if others have seen cDAQ locking up when the app is idle.
    I did check the power save options and disabled the power saving feature for the USB but the isue did happena again after siting idle over-night.
    SO...
    Has anyone seen lock-ups of cDAQ ?
    Ben
    PS I added some logging to catch low-level errors when it happens again.
    Ben Rayner
    I am currently active on.. MainStream Preppers
    Rayner's Ridge is under construction

    tst wrote:
    Not locking up, but I have seen a cDAQ constantly throwing errors after plugging in (or, more likely, unplugging) a USB flash drive (Win XP in that case) and it only stopped after the cDAQ was turned off and back on. It seems to me that there are things happening in the USB stack/driver which the cDAQ doesn't always like.
    One thing you could try is looking the event log for Windows or search to see if there's a utility which would let you log USB events in the system. Assuming that's really the source of the issue (and it wouldn't shock me - your app might be stuck in the middle of a VISA call), then that might help to pinpoint it.
    Your sense that it is USB related was confirmed by the log file.
    I still have to follow-up with the customer to see if the issue is still re-occuring but the logged error indicated the buffer on the CDAQ backplne was over-flowing... NI Support found a KB article that explains that when doing highspeed I/O, and Windows going in to out of screen saver will stall the USB I/O long enough for the buffer to over-flow.
    Customer was going to work with NI Support to fiddle with Windows and shut-down all of the power-saving etc options that could get in the way.
    I THINK the error code was "-200361".
    Take care,
    Ben
    Ben Rayner
    I am currently active on.. MainStream Preppers
    Rayner's Ridge is under construction

  • RCA Connection Pool idle Time-Out takes no effect !

    My question description goes here.
    According to JCA specification, I developed my 'ManagedConnectionImpl' class from the interface 'ManagedConnection'. I realize the 'destroy()' function to send out logout request to the EIS.
    Then I deployed the connector in Sun Java System Application Server, I noticed there are two parameters in Connection Pool part, they are:
    1. Idle Timeout. It said it's the maximum time that a connection can remain idle in the pool. I assume the connection will be removed after the specific time expired and before it's removed it will call the recallable function, 'destroy()', in my concrete class, 'ManagedConnectionImpl'.
    2. Pool Resize Quantity. it said it's number of connections to be removed when pool idle time expired.
    I am weird about it. I think EIS had itself session control strategy, if the specific session time-out expired, it will invalidate this session. So I think we will set 'Idle Timeout' in application server to be shorter than the EIS session time-out. If the 'Idle Timeout' in application server expired, it should remove all connections inside otherwise maybe the connection with invalid session will exist! (the background knowledge is our system will return back soap fault when it meets invalid session, so the invalid connections will not be removed by switching on your configuration item, 'On Any Failure')
    So I set "inital and minimum pool size" to 8, "maximum pool size" to 32 and "Pool Resize Quantity" to 32. (I expect AS to remove all when pool idle time expired)
    After deploying, I send out requests at the first round and wait for the time expired but I can't see the desired logout requests from pool automatically even one. Firstly I guess if my recallable function definition is wrong but when I shut down the Application Server, the desired logout requests are sent out from pool automatically. So I think my recallable function definition is workable.
    What's your comments on it?
    P.S.
    I am using Sun Java System Application Server 8.1.
    Thanks in advance!
    BRs
    /Leo

    I have had following test to ensure I sent out notification to listener.
    [#|2005-08-23T16:14:25.061+0800|INFO|sun-appserver-pe8.1_01|javax.enterprise.system.stream.out|_ThreadID=12;|
    It's in managed env!|#]
    [#|2005-08-23T16:14:25.062+0800|INFO|sun-appserver-pe8.1_01|javax.enterprise.system.stream.out|_ThreadID=12;|
    enter matchManagedConnections() !|#]
    [#|2005-08-23T16:14:25.062+0800|INFO|sun-appserver-pe8.1_01|javax.enterprise.system.stream.out|_ThreadID=12;|
    try to find a matching and existing one, reuse it!|#]
    [#|2005-08-23T16:14:25.062+0800|INFO|sun-appserver-pe8.1_01|javax.enterprise.system.stream.out|_ThreadID=12;|
    Found a matching and existing one, reuse it!|#]
    [#|2005-08-23T16:14:25.495+0800|INFO|sun-appserver-pe8.1_01|javax.enterprise.system.stream.out|_ThreadID=12;|
    tearDown() is called, the application-level connection is released!|#]
    [#|2005-08-23T16:14:25.496+0800|INFO|sun-appserver-pe8.1_01|javax.enterprise.system.stream.out|_ThreadID=12;|
    enter calling close() of managed connection.|#]
    [#|2005-08-23T16:14:25.496+0800|INFO|sun-appserver-pe8.1_01|javax.enterprise.system.stream.out|_ThreadID=12;|
    Start calling close() of managed connection.|#]
    [#|2005-08-23T16:14:25.496+0800|INFO|sun-appserver-pe8.1_01|javax.enterprise.system.stream.out|_ThreadID=12;|
    start to notify the listener the completeness of connection.|#]
    [#|2005-08-23T16:14:25.496+0800|INFO|sun-appserver-pe8.1_01|javax.enterprise.system.stream.out|_ThreadID=12;|
    notify the listener the completeness of connection successfully.|#]
    Whether it's related with the initial and minimum number parameter? Although I found it will not be created during AS start-up and will be created at the first request.
    Whether it shall be kept to meet the minimum requirement? I have failed to set it to 0.

  • After update 3.1 - 3G does not work after phone sits idle

    Updated from 3.01 to 3.1 - went smoothly, no problems. Restored my latest backup without issue. All is good. A few minutes later I tried to make a call and it failed. I tried to access the web - failed. Rebooted phone comes up as expected. Made a call, access to web working. 10 minutes later tried to make another call - failed; access to web failed again too. Switched over to Edge, rebooted. Calls and web access work. Waited 10-15 minutes, tried again; Edge still working. Switched back to 3G, rebooted. Same issue - calls/web work initially but after a short time, all calls and web access fail. I went through the 3.1 restore process again -- started over like it was a new phone. Again, the update worked without any issues/errors. I was able to restore my backup and all was good. But, same 3G problems occurred. So just for fun, I ran the Iphone config utility and captured the console log during a phone reboot sequence. After the phone booted up calls and web access were working with 3G. Then, let the phone sit idle for a while and saw these event occur in the log; the first 5 entries were made when I successfully sent a text message. After that message was sent I let the phone sit for a while. The next couple of entries were interesting. After the 'deactivated PDP context' messages appeared I could no longer make calls or access the web with 3G. Anyone else see this?
    I never ever had this problem with 3.01 software.
    Sat Sep 12 09:04:58 unknown com.apple.SpringBoard[25] <Notice>: told to send sms 867
    Sat Sep 12 09:04:58 unknown CommCenter[29] <Notice>: queuing sms message with id 867
    Sat Sep 12 09:04:58 unknown SpringBoard[25] <Error>: mms: queued messageId 867
    Sat Sep 12 09:05:00 unknown com.apple.SpringBoard[25] <Notice>: internalID: [867]
    Sat Sep 12 09:05:00 unknown com.apple.SpringBoard[25] <Notice>: notifying clients of event: 2 (recordID: 867)
    Sat Sep 12 09:07:02 unknown CommCenter[29] <Notice>: Deactivating PDP context 1, because it has gone idle.
    Sat Sep 12 09:07:43 unknown CommCenter[29] <Notice>: Deactivated PDP context 1 that supports connection types 0x2
    Sat Sep 12 09:07:43 unknown configd[23] <Notice>: network configuration changed.
    Sat Sep 12 09:07:43 unknown configd[23] <Notice>: CaptiveNetworkSupport:CaptiveHandlePrimaryChanged:2034 new primary: 67DCB7A6-30F5-4ECE-8FBB-5FC1DA71276D on pdp_ip0

    I thought I'd post you a quick response because you described exactly the issues I was having. After my phone was idle for a while, my iPhone would refuse to get data over 3G.
    I've tested this thoroughly and I'm able to "trip up" the phone into a state where it can't download data from the 3G network. But turning airline mode on for about 30 seconds then off again resets forces a renegotiate with the cell tower/3g and everything seems to work again about a minute later.
    Nothing untoward listed in the console at this point either - The PDP context that you mentioned that was deactivated was more than likely the Internet Tethering. (It's usually PDP context 1, by the looks of things).
    I think I might switch off 3G and see if that helps until Apple fix this issue. It is related to the wifi issues that others have reported - As I've seen my iPhone unable to download information via 3G if wifi is turned on, but you don't connect to a wifi network. But I can't seem to reproduce this reliably.
    -Soen

  • Dell Venue 11 Pro 7140 WiFi becomes disconnected during idle or waking from sleep

    I have a Dell Venue 11 Pro 7140 in which the wi-fi consistently becomes disconnected when either the tablet is idle for a period of time, or if it has woken from sleep. Period of time varies -- more than 5 minutes, but less than an hour.
    When wi-fi connection is lost, the only way to connect again is to reboot. Not much of a solution. USB 3.0-to-Ethernet adapter is rock-solid, but there isn't much point in having a tablet if you can't be wireless at some point.
    Things I've tried:
    - Optimized the intel driver settings roaming agressivenes: medium-low, Throughput booster: enabled, Transmit Power: highest.
    - toggling airplane mode, disabling/enabling via Network Connections
    - creating a custom wireless profile
    Really, not sure what I can try next. I've used other Windows 8.1 Pro laptops, and none of them have been this spotty.
    If anyone has a solution out there, I'm game to try. Otherwise, I'll be returning this back to Dell as a defective product. There seems to be issues about earlier versions of the Venue 11 Pro, but I haven't been able to find much on the 7140 version.
    Fully updated Windows 8.1 Pro. Using Intel dual-band Wireless AC 7265 adapter with v17.13.11.5 of the driver. BIOS version A03.

    Hi Jennifer,                                                                                                        Saturday 4th April 2015
    This whole process from order date 14th March 2015 to now has been frustrating, my mistake was to get one of the sales consultants on chat to check the saved cart, had compatible accessories for this tablet. From then on, after multiple quotes in the end he still managed to get the address wrong and missed putting screen protector on the order, multiple emails and phone calls the following week finally got address changed but couldn't get screen protector added to order in the end I ordered it over the internet. I should have cancelled order when I wanted to.
    What should have been a simple process is now a nightmare the Tablet with screen protector needs to be replaced has not worked properly since unboxing and first switch on , spent many hours of installing updates, contacted technical assistance last Thursday they did a remote hookup changed some things worked for a little time.
    Totally disappointed
    Leigh
    PS update 7th April 2015 started factory reinstall last Friday reinstalled all recomended updates more wasted time tested over weekend WiFi still not working after sleep mode. When I check for new updates keeps asking to reinstal theses 2 updates already done, reinstal them over and over still tells me to update them, second file fails to instal have to manually instal O2 Micro file searches wrong place for missing component.
    "Automatic updates
    These updates can be installed automatically.
    Intel(R) Management Engine Components Installer View full driver details
    Chipset_Driver_3TY1R_WN_10.0.31.1000_A02.EXE | Chipset (107MB) | Recommended
    Download File
    Add to download list
    O2 Micro OZ777xxx/OZ621XX memory card reader Driver View full driver details
    Chipset_Driver_VPNKY_WN_3.0.8.45_A06.EXE | Chipset (16MB) | Recommended
    This file will automatically self-install after downloading.Restart required "
    There is something wrong with this tablet.
    Download File
    Add to download list

  • Will Not Copy After Being Idle

    THIS IS A COPYING ISSUE!!!  After installing the upgraded version of Comcast's latest modem/router, our HP 6550 would not print.  I contacted Comcast and reported the difficulty and they informed me it was a printer problem - not their new modem.  Consequently, we purchased a new HP 5740 printer  The printing function worked fine and it would copy immediatley after the printer was turned on, however, after the printer sat idle for a couple of hours, the copy function would not work.  If you then turned the printer off and immediately turned it back on, it would copy.  Again, Comcast reiterated it was a printer problem.   I returned the original 5740 back to the store from which it was purchased and exchanged it for another HP 5740 printer.  After hooking up the second HP 5740, the same problems with copying occurred.  I returned the second HP 5740 to the dealer for a refund and left town for three months.  We recently returned and purchased a HP ENVY 7645 printer from another dealer.  We are having the same issues with this printer as we had with the two previous HP 5740 printers.  I discussed the situation with a level II supervisor at Comcast and he arranged for a new modem/router to be brought to the house and installed by a technician.  The installation occurred today and we are still experiencing exactly the same problem.  The Comcast technician who installed the new modem/router (who couldn't wait around while the printer sat idle) said that if the problems occurs after the installation, we should purchase a printer/copier from another manufacturer - or turn our printer "off and then on" when we want to copy.   He also asid that if this problem does continue after the new modem/router is installed, it definitely won't be a Comcast problem.  DOES ANYONE HAVE ANY IDEA WHERE I SHOULD GO FROM HERE?

    Hi @lamoso,
    Welcome to the HP Forums!
    I understand that your HP Envy 7645 will not copy after being idle. I am happy to look into this for you!
    I believe this is happening due to the printer going into sleep mode. As the options are for 5 minutes, 10 minutes, and 15 minutes for sleep mode to activate. I would recommend trying a hard reset, by going to this post,How to perform a Hard Reset, by @Rich1. It is important that the printer's power cable is plugged directly into the wall outlet, and not a surge protector. See this article, Issues when Connected to an Uninterruptible Power Supply/Power Strip/Surge Protector for more information. This applies to Inkjet printers as well.
    After the hard reset, I would then make sure the printer's firmware is up-to-date. Getting the Latest Firmware and Product Updates.
    Let's see what happens after giving these two steps a try!
    Thank you for posting!
    “Please click the Thumbs up icon below to give me a virtual high-five for responding.”
    RnRMusicMan
    I work on behalf of HP
    Please click “Accept as Solution ” if you feel my post solved your issue, it will help others find the solution.
    Click the “Kudos Thumbs Up" to say “Thanks” for helping!

Maybe you are looking for