Tablespace Replication Problem - high disk I/O

Hi.
I'm doing some R&D on Oracle Streams. Have setup Tablespace replication between 2 10g R2 instances. data seems to replicate between the 2. These 2 instances have no applications running off of them apart from OEM and queries I run using SQL Developer and SQL*PLUS.
The problem i'm seeing is that since setting up and switching on this replication config disk I/O is high.I'm using windows Performance Monitor to look at
- % Disk time = 100%
- Avg Disk Writes/sec = 20
- Avg Disk Reads/sec = 30
- CPU % = 1
- % Commited Mem = 40%
To me this just looks/sounds wrong.
This has been like this for about 24hrs.
OEM ADDM report says "Investigate the cause for high "Streams capture: waiting for archive log" waits. Refer to Oracle's "Database Reference" for the description of this wait event. Use given SQL for further investigation." I haven't found any reference to this anywhere.
Anybody got any ideas on how to track this one down? Where in my db's can I look for more info?
Platform details:
(P4, 1GB RAM, IDE disk) x 2
Windows Server 2003 x64 SP1
Oracle 10.2.0.1 Enterprise x64
Script used to setup replication:
set echo on;
connect streamadmin/xxxx;
BEGIN
  DBMS_STREAMS_ADM.SET_UP_QUEUE(
    queue_table => '"STM_QT"',
    queue_name  => '"STM_Q"',
    queue_user  => '"STREAMADMIN"');
END;
--connect streamadmin/xxxx@(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=oratest2.xxxx.co.uk)(PORT=1521)))(CONNECT_DATA=(SID=oratest2.xxxx.co.uk)(server=DEDICATED)));
connect streamadmin/xxxx@oratest2;
BEGIN
  DBMS_STREAMS_ADM.SET_UP_QUEUE(
    queue_table => '"STM_QT"',
    queue_name  => '"STM_Q"',
    queue_user  => '"STREAMADMIN"');
END;
connect streamadmin/xxxx;
create or replace directory "EMSTRMTBLESPCEDIR_0" AS 'D:\ORACLE\DATA\ORATEST1';
DECLARE 
    t_names DBMS_STREAMS_TABLESPACE_ADM.TABLESPACE_SET;
BEGIN   
t_names(1) := '"ADFERO_TS"';
DBMS_STREAMS_ADM.MAINTAIN_TABLESPACES(
   tablespace_names             => t_names,
   source_directory_object       => '"DPUMP_DIR"',
   destination_directory_object  => '"DPUMP_DIR"',
   destination_database          => 'ORATEST2.xxxx.CO.UK' ,
   setup_streams                 => true,
   script_name                    => 'Strm_100407_1172767271909.sql',
   script_directory_object        => '"DPUMP_DIR"',
   dump_file_name                 => 'Strm_100407_1172767271909.dmp',
   capture_name                  => '"STM_CAP"',
   propagation_name              => '"STM_PROP"',
   apply_name                    => '"STM_APLY"',
   source_queue_name             => '"STREAMADMIN"."STM_Q"',
   destination_queue_name        => '"STREAMADMIN"."STM_Q"',
   log_file                      => 'Strm_100407_1172767271909.log',
   bi_directional                => true);
END;
/

ok dont know why this didn't work before but here are the results.
select segment_name, bytes from dba_segments where owner='SYSTEM' and segment_name like 'LOGMNR%' ORDER BY bytes desc
SEGMENT_NAME                                                                      BYTES                 
LOGMNR_RESTART_CKPT$                                                              14680064              
LOGMNR_OBJ$                                                                       5242880               
LOGMNR_COL$                                                                       4194304               
LOGMNR_I2COL$                                                                     3145728               
LOGMNR_I1OBJ$                                                                     2097152               
LOGMNR_I1COL$                                                                     2097152               
LOGMNR_RESTART_CKPT$_PK                                                           2097152               
LOGMNR_ATTRIBUTE$                                                                 655360                
LOGMNRC_GTCS                                                                      262144                
LOGMNR_I1CCOL$                                                                    262144                
LOGMNR_CCOL$                                                                      262144                
LOGMNR_CDEF$                                                                      262144  
LOGMNR_USER$                                                                      65536  
160 rows selected
select segment_name, bytes from dba_extents where segment_name=upper( 'logmnr_restart_ckpt$' );
SEGMENT_NAME                                                                      BYTES                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              65536                 
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
LOGMNR_RESTART_CKPT$                                                              1048576               
29 rows selectedMessage was edited by:
JA
Message was edited by:
JA

Similar Messages

  • High Disk I/O Problem

    Hello all
    why every server have a capture Process have this problem "High Disk I/O" ???
    in all configurations , transactions load (high or low) , number of connections (high or low)
    i try to keep server without any transactions or any connections for 6 day and i still have this Problem
    it seems a serious problem without fixing
    release 10.2.0.1

    hi rijesh
    SEGMENT_NAME Size in MB SUM(BLOCKS)
    LOGMNR_RESTART_CKPT$ .5 64
    in Oracle Enterprise manager wait event is :
    Wait event "Streams capture: waiting for archive log" in wait class "Other" was consuming significant database time.
    i traced session that cause this wait event and generate tkprof file but i think there is nothing useful :
    {color:#0000ff}TKPROF: Release 10.2.0.1.0 - Production on Mon Nov 17 16:08:35 2008
    Copyright (c) 1982, 2005, Oracle. All rights reserved.
    {color}
    {color:#0000ff}Trace file: C:\oracle\admin\test\bdump\test_p000_3740.trc
    Sort options: prsdsk exedsk fchdsk
    count = number of times OCI procedure was executed
    cpu = cpu time in seconds executing
    elapsed = elapsed time in seconds executing
    disk = number of physical reads of buffers from disk
    query = number of buffers gotten for consistent read
    current = number of buffers gotten in current mode (usually for update)
    rows = number of rows processed by the fetch or execute call
    Trace file: C:\oracle\admin\test\bdump\test_p000_3740.trc
    Trace file compatibility: 10.01.00
    Sort options: prsdsk exedsk fchdsk
    1 session in tracefile.
    0 user SQL statements in trace file.
    0 internal SQL statements in trace file.
    0 SQL statements in trace file.
    0 unique SQL statements in trace file.
    10268 lines in trace file.
    0 elapsed seconds in trace file.
    {color}
    i tried to open trace file "test_p000_3740.trc" and i found this :
    {color:#0000ff}Dump file c:\oracle\admin\test\bdump\test_p000_3740.trc
    Mon Nov 17 14:52:21 2008
    ORACLE V10.2.0.1.0 - Production vsnsta=0
    vsnsql=14 vsnxtr=3
    Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - Production
    With the Partitioning, Oracle Label Security, OLAP and Data Mining options
    Windows Server 2003 Version V5.2 Service Pack 1
    CPU : 4 - type 586, 1 Physical Cores
    Process Affinity : 0x00000000
    Memory (Avail/Total): Ph:2789M/4094M, Ph+PgF:4519M/5972M, VA:1627M/2047M
    Instance name: test
    {color}
    {color:#0000ff}Redo thread mounted by this instance: 1
    Oracle process number: 29
    {color}
    {color:#0000ff}Windows thread id: 3740, image: ORACLE.EXE (P000)
    *** 2008-11-17 14:52:21.652
    *** SERVICE NAME:(SYS$USERS) 2008-11-17 14:52:21.652
    *** SESSION ID:(144.2) 2008-11-17 14:52:21.652
    WAIT #0: nam='control file sequential read' ela= 234 file#=0 block#=1 blocks=1 obj#=-1 tim=3793171610
    WAIT #0: nam='control file sequential read' ela= 294 file#=1 block#=1 blocks=1 obj#=-1 tim=3793186744
    WAIT #0: nam='control file sequential read' ela= 208 file#=2 block#=1 blocks=1 obj#=-1 tim=3793187894
    WAIT #0: nam='control file sequential read' ela= 182 file#=0 block#=16 blocks=1 obj#=-1 tim=3793188866
    WAIT #0: nam='control file sequential read' ela= 233 file#=0 block#=18 blocks=1 obj#=-1 tim=3793189902
    WAIT #0: nam='log file sequential read' ela= 12608 log#=0 block#=43745 blocks=8192 obj#=-1 tim=3793255308
    WAIT #0: nam='log file sequential read' ela= 68218 log#=0 block#=43745 blocks=8192 obj#=-1 tim=3793363640
    WAIT #0: nam='log file sequential read' ela= 100 log#=0 block#=1 blocks=1 obj#=-1 tim=3793518549
    WAIT #0: nam='log file sequential read' ela= 89 log#=0 block#=1 blocks=1 obj#=-1 tim=3793541091
    WAIT #0: nam='control file sequential read' ela= 14717 file#=0 block#=1 blocks=1 obj#=-1 tim=3793556425
    WAIT #0: nam='control file sequential read' ela= 6630 file#=1 block#=1 blocks=1 obj#=-1 tim=3793563619
    WAIT #0: nam='control file sequential read' ela= 9355 file#=2 block#=1 blocks=1 obj#=-1 tim=3793573635
    WAIT #0: nam='control file sequential read' ela= 11426 file#=0 block#=16 blocks=1 obj#=-1 tim=3793585608
    WAIT #0: nam='control file sequential read' ela= 201 file#=0 block#=18 blocks=1 obj#=-1 tim=3793586348
    WAIT #0: nam='control file sequential read' ela= 190 file#=0 block#=1 blocks=1 obj#=-1 tim=3793587134
    WAIT #0: nam='control file sequential read' ela= 186 file#=1 block#=1 blocks=1 obj#=-1 tim=3793587842
    WAIT #0: nam='control file sequential read' ela= 185 file#=2 block#=1 blocks=1 obj#=-1 tim=3793588548
    WAIT #0: nam='control file sequential read' ela= 187 file#=0 block#=16 blocks=1 obj#=-1 tim=3793589241
    WAIT #0: nam='control file sequential read' ela= 185 file#=0 block#=18 blocks=1 obj#=-1 tim=3793589944
    WAIT #0: nam='log file sequential read' ela= 72452 log#=0 block#=43745 blocks=8192 obj#=-1 tim=3793663620
    WAIT #0: nam='log file sequential read' ela= 106739 log#=0 block#=43745 blocks=8192 obj#=-1 tim=3793771943
    WAIT #0: nam='log file sequential read' ela= 95 log#=0 block#=1 blocks=1 obj#=-1 tim=3793908609
    WAIT #0: nam='log file sequential read' ela= 111 log#=0 block#=1 blocks=1 obj#=-1 tim=3793925054
    WAIT #0: nam='control file sequential read' ela= 22382 file#=0 block#=1 blocks=1 obj#=-1 tim=3793948070
    WAIT #0: nam='control file sequential read' ela= 6591 file#=1 block#=1 blocks=1 obj#=-1 tim=3793955243
    WAIT #0: nam='control file sequential read' ela= 9478 file#=2 block#=1 blocks=1 obj#=-1 tim=3793965270
    WAIT #0: nam='control file sequential read' ela= 11449 file#=0 block#=16 blocks=1 obj#=-1 tim=3793977257
    WAIT #0: nam='control file sequential read' ela= 203 file#=0 block#=18 blocks=1 obj#=-1 tim=3793977980
    {color}
    any suggestions my friend
    Best regards
    Edited by: wassim haddad on Nov 17, 2008 5:47 AM

  • Problem  High running Temperature  Specification iMac 27 inch  Mac OS X  Includes Software Snow Leopard upgraded to Mountain Lion  OS X 10.8.03 (1278) 251 GB Flash Drive 2 TB  Sata disk Processor 3.4 Ghz Intel Core I7 Memory 16 GB 1333 MHz DDR3  Graphics

    Problem
    High running Temperature
    Specification
    iMac 27 inch
    Mac OS X  Includes Software Snow Leopard upgraded to Mountain Lion  OS X 10.8.03 (1278)
    251 GB Flash Drive
    2 TB  Sata disk
    Processor 3.4 Ghz Intel Core I7
    Memory 16 GB 1333 MHz DDR3
    Graphics AMD Radeon HD 6970 2048 MB
    Video: Software Final Cut Pro X  10.0.8
    I use the imac mainly for editing Video which when doing so it has reached temperatures of near to 60° C which causes the program to become unstable and needs to be cooled down to operate correctly.
    An Apple technician (by phone) advised me to download an smcfancontrol.  This allows each individual fan to increase speed and also advises the current temperature at °C.  The smcfancontrol has been downloaded but this has not resolved the problem.  By increasing the speed of the fan more heat is generated and creates additional noise. 
    I have since taken the iMac to an Apple Store for examination as suggested by APPLECARE when the fans were found to be working correctly.  An Apple Store technician carried out a test, but not whilst working under the pressure of FINAL CUT PRO when the iMac generates a high temperature.
    The Temperature Specification  for this model is
    Operating Temperature 20 °c to 35°C
    Storage Temp.47°c
    Clearly the temperature my iMac  is operating at is excessive and could cause the computer to break down prematurely.  Has anyone experienced this problem.
    Frank Howell

    When was the last time you cleaned/vacuumed out all of your iMac's vents, slots and ports.
    You need to do this with the iMac powered down, disconnected from power and everything disconnected.
    Lay your iMac, screen down, on a soft surface, like a heavy towel. Use a small electronics vac, or portable or full size vac with hose and a crevice tool attachment and carefully and slowly vacuum every vent, slot and port on your iMac.
    Then reconnect everything up, power up and download and install iStat Pro. Start working, then report the temps of your iMac after this has been done.
    When using SMCFan Control, it is important not to run the fans at too high an RPM. This doesn't generate heat on its own, but will suck in more dust, lint and dirt into your iMac, and that will be the cause of additonal increases in heat.
    I run my fans in the range of 1400-1600 RPMs. You can go as high as 1800 RPMs but it is important to keep the RPM range below either 1800 or below 2000 RPMs.
    I, also, run a very small desktop fan pointed toward the upper back of my iMac to keep temps down even further.

  • High disk usage on DB server

    Hi All,
    I have 4 four disks connected to my db server and on disk2, I found very high usage (above 90%) every time. I have only one tablespace BME_DATA on this disk and it has only one datafile.
    There are only few tables (less than 50) that are stored in this tablespace but most of them are very huge in size (~ 500MB), and yes most of them have high DML activity also.
    I am planning to create another datafile for BME_DATA on DISK3 so that load can be balanced. Please suggest if there is any better way to do the same.
    Thanks for your interest.
    Bhupinder

    Hi Alan,
    What version database? have you analyzed the tables
    and indexes associated with the datafile on this
    disk? Has this problem recently started?I am using Oracle 9.2.0, and all the table including indexes were analyzed today. Recently I have loaded 1G+ of data into multiple tables and yes out of those tables are only causing the problem.
    I have only table datafile (BME_DATA_01.dbf) to Diks2, index datafile is located on Disk1.
    Now shall I go and add one more data file to Disk3 or is there any memory changes I can do to reduce the bottleneck on disk2.
    Thanks again
    Bhupinder

  • High disk usage/high memory usage.

    i keep getting high memory usage and high disk usage in ps cs6.
    its slow when i open a file. how can i solve this problem.
    thank you.
    philip.

    Not to put too fine a point on it, but more than what?  More RAM is almost always better, but you haven't answered most of the questions posed here.  The answer might be different if you have 2 GB of RAM or 32 GB of RAM already.  A 9 megapixel document could get rather large with a lot of layers, but again you've neglected to give any hard info about its size.
    Regarding your video card, that's a bit of a judgment call you need to make.  For Photoshop CS6 I currently advise people to try to get a video card that gets a score of 1000 or better on this benchmark:  http://www.videocardbenchmark.net/high_end_gpus.html
    Video card speed will matter when you're editing those 3D components and video card VRAM will also matter, but possibly even more importantly your GeForce 320 is a bit old, and nVidia may just not be doing their best driver work for it any more. 
    Everything in general will run more smoothly with a powerful modern video card.  I like to recommend a VisionTek ATI Radeon HD 7750 - or if the computer power supply and your budget allow, the even more powerful 7850.  VisionTek does a very good job with their hardware, IMO, and it comes with a lifetime guarantee. 
    Keep in mind that if you switch from nVidia to ATI or vice versa, you should carefully sweep your system to ensure all files from the old card are removed, then visit the web site of the new card's maker and download/install their latest display driver release.
    Best of luck.
    -Noel

  • RAM Issues + High disk usage on MSI GE60

    Recently I've noticed that my system slowly eats up RAM. It starts off at 21% and can go as high up as 80% or more. When I'm playing games on it the RAM does not clear up properly either, it stays at 50% or so. This was an issue before that I somehow managed to solve, but it is back again.
    As for high disk usage, that's been a problem since I've had the laptop.
    OS: Windows 8
    Processor: Intel Core i7 4700MQ (2.40GHz)
    GPU: NVIDIA GeForce GTX 765M
    RAM: 8GB Memory
    HDD: 750GB
    If you need more info I'll provide it.

    If you open task manager-> performance -> open resource manager / you will see whats eating up your resources
    if its windows see if indexing is off
    did you installed 3rd party software diffrent than stock ?
    how old is your laptop?
    its not normal to have such high usage numbers !! i think its a program causing the problems !
    if you check your resource manager you will be able to see whats causing it
    ps if possible please post a screen shot and ill look into it

  • [SOLVED] High disk activity after kernel boot

    Hi,
    after kernel loading (just before getty starts) i notice extremely high disk activity for ~20 seconds for no reason. I think that it is something related to systemd and mount services.
    When i set rootfs to read-only in syslinux kernel parameters then the system starts normally without delay. So, i think that something is constantly writing to (or checking?) my hard disk right after boot sequence.
    I'm using  kernel 3.7.9-1-ARCH.
    $ systemd --version
    systemd 197
    +PAM -LIBWRAP -AUDIT -SELINUX -IMA -SYSVINIT +LIBCRYPTSETUP +GCRYPT +ACL +XZ
    $ systemd-analyze
    Startup finished in 3121ms (kernel) + 6238ms (userspace) = 9360ms
    $ systemd-analyze blame
    1289ms NetworkManager.service
    1173ms colord.service
    812ms systemd-logind.service
    722ms systemd-binfmt.service
    660ms systemd-udevd.service
    639ms systemd-sysctl.service
    314ms tmp.mount
    305ms systemd-remount-fs.service
    300ms upower.service
    219ms systemd-tmpfiles-setup.service
    168ms polkit.service
    160ms dev-sda2.swap
    156ms systemd-vconsole-setup.service
    129ms wpa_supplicant.service
    125ms dev-mqueue.mount
    121ms dev-hugepages.mount
    108ms systemd-udev-trigger.service
    101ms verynice.service
    59ms systemd-user-sessions.service
    48ms boot.mount
    47ms hpfall.service
    46ms home.mount
    43ms cpupower.service
    27ms sys-kernel-config.mount
    4ms proc-sys-fs-binfmt_misc.mount
    Syslinux kernel parameters (either ro or rw has the same result)
    APPEND root=/dev/sda3 rw nomodeset vga=current clocksource=acpi_pm libahci.ignore_sss=1
    Tried also with fsck hook and removing sda3 from fstab
    APPEND root=/dev/sda3 rw nomodeset vga=current clocksource=acpi_pm libahci.ignore_sss=1 rootflags=noatime,barrier=0
    my fstab:
    # UUID=925aba97-6805-442b-bb51-3ea6c16a5a86
    /dev/sda3 / ext4 rw,noatime,barrier=0 0 0
    # UUID=dd60df02-7992-43c1-a0e0-79441e23268e
    /dev/sda1 /boot ext4 rw,noatime 0 0
    # UUID=d2621666-f2c5-4f36-8da3-c0f21a23246c
    /dev/sda4 /home ext4 rw,noatime 0 0
    # UUID=3d70bc7d-5b0c-4adc-ba84-f32c9d132bcc
    /dev/sda2 none swap defaults 0 0
    Thanks for your time.
    Last edited by anelehto (2013-02-23 14:27:41)

    It was a problem with preload and its configuration.. I removed it and everything went back to normal!

  • KDE 4.7: very slow and high disk usage

    Since upgrade to KDE 4.7 on my i686 laptop, my system behaves weirdly. I noticed a slow down caused by a very high disk usage: when the HDD spins the cpu usage reported by the CPU monitor plasmoid is very high (~ 50%), while top reports a very small CPU load. Another strange symptom is that, when I try to close the session, the system freezes for about 15 seconds before the logout/shutdown/restart dialog window appears, again with HDD spinning furiously. If I exit the dialog window and then try to close session again it works smoothly, so I think some caching is at work at second try.
    This is very painful since it slows down my workflow. Since now I tried to:
    - disable nepomuk, strigi and all that using the nepomukcontroller plasmoid
    - disable compositing
    - delete .kde folder
    - uninstall and reinstall all kde packages
    without any benefit. On the same laptop I have dual boot with a 64 bit arch installation, which does not have the same problem, nor has it my 64 bit desktop pc.
    Does anyone have similar issues? I don't understand what is it that makes the HDD spin so much, nor how to eventually track it down (are there som HDD usage profiling utilities?). Moreover, why top and CPU usage plasmoid report so different CPU usage stats?
    I'm very confused, so any help will be greatly appreciated.

    Although I'm not having any of the issues as Snack, I have found KDE 4.7.0 to be a very horrible release. In fact, I'm sick and tired of the KDE release notes bragging about killing thousands of bugs each release, when in actuality the first few revisions of each stable release are horrible. When I first upgraded, I ran into at least four really nasty bugs. Although I've worked around some of them, it left a really sour taste in my mouth. Here is what I've experienced:
    1.) With a new user or fresh install, there is no wallpaper by default, even if I install kde-wallpapers before I log in, it makes no difference.
    2.) Sometimes kwin will crash during log out.
    3.) Worse yet, if I exit a full screen game, it will more than likely crash Kwin. This is a problem on my laptop (which has intel video) and not my desktop (which has nvidia video). If I don't remember to disable desktop effects before opening a game, I may run into a problem.
    4.) Although my performance is good, it's not great. I would think a laptop with a Core i5 processor and 8GB of RAM would run faster than it is.
    I have tested XFCE for a bit, and I do like it alot. However, with XFCE I am unable to browse Samba shares, and that is a huge problem. In addition, I find KDE's applications to be much better than what ships with XFCE, but I question whether they're worth the performance hit. I really love Kate, Gwenview, Digikam, and Amarok. I know you can install those in XFCE, but they feel out of place and then my desktop pretty much becomes a mutt. Not only that, but KDE is full featured, whereas XFCE is not. It's like I have to choose between a stable desktop or a desktop with full of bugs.
    I hope that KDE 4.7.1 fixes everything, but historically it's usually the 4.x.2 release that really does the trick.

  • High disk utilization every 24 hours

    Hi all,
    While troubleshooting an user's complain about slow access to his mailbox on a Windows 2012/Exchange 2013 environment, I came across this interesting behavior:
    Disk utilization on the server will increase from 100KBps to 3000KBps (on average) and stay high for a good 10 hours or so. Then disk activity will go down for exactly 24 hours from the time it was up and increase again for another cycle.
    Basically, no matter how long it takes to go down, it will increase again 24 hours later. Looking at Perfmon and Resource Manager all I could see during both periods is that store.worker.exe is always on top, and READ has higher values. The issue does not
    appear to be related to business hours / backup window / client access to exchange, as it occurs at different times of the day and changes everyday.
    I believe it has something to do with an internal maintenance/reporting/monitoring feature of Exchange 2013, as it happens on 3 Exchange environments we manage, completely apart from each other.
    Microsoft Exchange Health Manager service has already been disabled as well as scheduled tasks under Microsoft/Windows/PLA and they all proved not to make a difference.
    Any ideas out there?
    Diego de Azevedo IT Analyst - MCSE, MCITP

    Good Call. I just spent the last few hours going over 550,000 events. 
    I reviewed the application log for a 24 hour period and here is what I noticed:
    During the high disk queue period there are lots of entries for MsExchangeMailboxAssistants. For each of the 15 mailbox databases. 
    These only run during the High Disk Queues window.
    TopN Words
    Store DS Maintenance
    Sharing Policy
    SharePoint Signal Store
    Search Index Repair Time Based 
    People-centric triage 
    People Relevance
    Managed Folder Mailbox 
    Mailbox Processor
    Mailbox Association Replication
    Calendar Repair
    People Relevance Feeder
    Orphaned Mailbox Purge
    Here are the examples of the log entries for Event ID 9017
    Service MSExchangeMailboxAssistants. TopN Words Assistant for database *** is entering a work cycle. There are *** mailboxes on this database.
    Service MSExchangeMailboxAssistants. Store DS Maintenance Assistant for database *** is entering a work cycle. There are *** mailboxes on this database.
    Service MSExchangeMailboxAssistants. Sharing Policy Assistant. for database *** is entering a work cycle. There are *** mailboxes on this database.
    Service MSExchangeMailboxAssistants. SharePoint Signal Store Assistant for database *** is entering a work cycle. There are *** mailboxes on this
    database.
    Service MSExchangeMailboxAssistants. Search Index Repair Time Based Assistant for database *** is entering a work cycle. There are *** mailboxes on this database.
    Service MSExchangeMailboxAssistants. People-centric triage for database *** is entering a work cycle. There are *** mailboxes on this database.
    Service MSExchangeMailboxAssistants. People Relevance Assistant for database *** is entering a work cycle. There are *** mailboxes on this database.
    Service MSExchangeMailboxAssistants. Managed Folder Mailbox Assistant for database *** is entering a work cycle. There are *** mailboxes on this database.
    Service MSExchangeMailboxAssistants. Mailbox Processor Assistant for database *** is entering a work cycle. There are *** mailboxes on this database.
    Service MSExchangeMailboxAssistants. Mailbox Association Replication Assistant for database *** is entering a work cycle. There are *** mailboxes
    on this database.
    Service MSExchangeMailboxAssistants. Calendar Repair Assistant for database *** is entering a work cycle. There are *** mailboxes on this database.
    There are 3039 of these messages as well:
    The people relevance feeder for mailbox *** picked 16 sent items to process. Number of sent items successfully processed 16 and number of sent items failed to be processed 0. Total time taken to process this
    mailbox 4539.8548.
    It appears the purge of orphaned mailboxes happens only at this time as well .
    These run throughout the day and not just 24 hours after the last restart/switchover.
    Calendar Sync 
    Junk Email Options 
    Probe Time Based 
    Store Maintenance 
    Examples of the Event ID 9017
    Service MSExchangeMailboxAssistants. Calendar Sync Assistant for database *** is entering a work cycle. There are *** mailboxes on this database.
    Service MSExchangeMailboxAssistants. Junk Email Options Assistant for database *** is entering a work cycle. There are *** mailboxes on this database.
    Service MSExchangeMailboxAssistants. Probe Time Based Assistant for database *** is entering a work cycle. There are *** mailboxes on this database.
    Service MSExchangeMailboxAssistants. Store Maintenance Assistant for database *** is entering a work cycle. There are *** mailboxes on this database.
    So how do I get the first group of maintenance tasks to run throughout the day, or at different times, or at least not all at the same time for all 15 databases?
    Thanks!! 
    Well, its all workload based, so I don't think you can readily adjust that.
    I'm more inclined to look at the specs of the mailbox servers. Are they up to the task and have they been vetted with the Mailbox Calculator?
    Twitter!: Please Note: My Posts are provided “AS IS” without warranty of any kind, either expressed or implied.

  • Low number users, high disk IO, 11g 106 Win32

    I can see for a high number of users in a critical high availability environment,
    all the things Oracle does for maintenance, monitoring, etc. is vital.
    But for an intranet office with 5 users and less than 4 gigs of data, using the database in the daytime only, and a couple backup jobs running at night, is there a way to reduce the frequency of oracle disk IO events for monitoring, etc. The hard drives are just running themselves to death.
    Oracle CPU use is low after fixing a known problem with Windows 11g 106 and one maintenance job problem noted on this forum previously related to sysman.mgmt_job where job_name = 'PROVISIONING DAEMON'.
    It is possible part of the high disk IO is Windows 2008 adding many behind the scenes monitoring tasks compared to Windows 2003.
    Any ideas to reduce disk thrashing in low user environment would be welcome.
    Thanks

    Process Explorer utility points to various oracle
    processes,
    oracle.exe, emagent, nmupm, etc.Take a look at the Windows event log and the listener.log file. Do you see repeating patterns such as the following, with an extremely short delay between each event?
    From the Windows Event Viewer - Application Log:
    Error  Faulting application nmefwmi.exe, version 0.0.0.0, time stamp 0x46e8bfeb, faulting module RPCRT4.dll, version 6.0.6000.16525, time stamp 0x469c344c, exception code 0xc0000005, fault offset 0x00078edf, process id 0x139c, application start time 0x01c858a1d991a940.
    Error  Faulting application nmefwmi.exe, version 0.0.0.0, time stamp 0x46e8bfeb, faulting module RPCRT4.dll, version 6.0.6000.16525, time stamp 0x469c344c, exception code 0xc0000005, fault offset 0x00079b50, process id 0x44c, application start time 0x01c863adc2f27220.
    Error  Faulting application nmefwmi.exe, version 0.0.0.0, time stamp 0x46e8bfeb, faulting module RPCRT4.dll, version 6.0.6000.16525, time stamp 0x469c344c, exception code 0xc0000005, fault offset 0x00079b50, process id 0x1324, application start time 0x01c863adec6a2350.
    From the Listener.log:
    08-AUG-2008 19:31:50 * service_update * MYDB * 0
    08-AUG-2008 19:32:02 * service_update * MYDB * 0
    08-AUG-2008 19:32:05 * service_update * MYDB * 0
    08-AUG-2008 19:32:08 * service_update * MYDB * 0
    08-AUG-2008 19:32:11 * service_update * MYDB * 0
    08-AUG-2008 19:32:15 * service_update * MYDB * 0
    08-AUG-2008 19:32:36 * (CONNECT_DATA=(SID=MYDB)(CID=(PROGRAM=D:\Oracle\Product\product\11.1.0\db_1\bin\emagent.exe)(HOST=MYSERVER)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50027)) * establish * MYDB * 0
    08-AUG-2008 19:32:36 * (CONNECT_DATA=(SID=MYDB)(CID=(PROGRAM=D:\Oracle\Product\product\11.1.0\db_1\bin\emagent.exe)(HOST=MYSERVER)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50028)) * establish * MYDB * 0
    08-AUG-2008 19:32:56 * (CONNECT_DATA=(SID=MYDB)(CID=(PROGRAM=D:\Oracle\Product\product\11.1.0\db_1\bin\emagent.exe)(HOST=MYSERVER)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50030)) * establish * MYDB * 0
    08-AUG-2008 19:32:56 * (CONNECT_DATA=(SID=MYDB)(CID=(PROGRAM=D:\Oracle\Product\product\11.1.0\db_1\bin\emagent.exe)(HOST=MYSERVER)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50029)) * establish * MYDB * 0
    08-AUG-2008 19:32:56 * service_update * MYDB * 0
    08-AUG-2008 19:33:00 * service_update * MYDB * 0
    08-AUG-2008 19:33:03 * service_update * MYDB * 0
    08-AUG-2008 19:33:06 * service_update * MYDB * 0
    08-AUG-2008 19:33:09 * service_update * MYDB * 0
    08-AUG-2008 19:33:12 * service_update * MYDB * 0
    08-AUG-2008 19:33:15 * service_update * MYDB * 0
    08-AUG-2008 19:33:42 * service_update * MYDB * 0
    08-AUG-2008 19:33:51 * service_update * MYDB * 0
    08-AUG-2008 19:33:54 * service_update * MYDB * 0
    08-AUG-2008 19:33:57 * ping * 0
    08-AUG-2008 19:33:58 * (CONNECT_DATA=(SID=MYDB)(CID=(PROGRAM=D:\Oracle\Product\product\11.1.0\db_1\perl\5.8.3\bin\MSWin32-x86-multi-thread\perl.exe)(HOST=MYSERVER)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50034)) * establish * MYDB * 0
    08-AUG-2008 19:33:58 * (CONNECT_DATA=(SID=MYDB)(CID=(PROGRAM=D:\Oracle\Product\product\11.1.0\db_1\perl\5.8.3\bin\MSWin32-x86-multi-thread\perl.exe)(HOST=MYSERVER)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50035)) * establish * MYDB * 0
    08-AUG-2008 19:33:58 * (CONNECT_DATA=(SID=MYDB)(CID=(PROGRAM=D:\Oracle\Product\product\11.1.0\db_1\perl\5.8.3\bin\MSWin32-x86-multi-thread\perl.exe)(HOST=MYSERVER)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50036)) * establish * MYDB * 0
    08-AUG-2008 19:33:58 * (CONNECT_DATA=(SID=MYDB)(CID=(PROGRAM=D:\Oracle\Product\product\11.1.0\db_1\perl\5.8.3\bin\MSWin32-x86-multi-thread\perl.exe)(HOST=MYSERVER)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50038)) * establish * MYDB * 0
    08-AUG-2008 19:33:59 * (CONNECT_DATA=(CID=(PROGRAM=)(HOST=)(USER=SYSTEM))(COMMAND=status)(ARGUMENTS=64)(SERVICE=(ADDRESS=(PROTOCOL=TCP)(HOST=bin)(PORT=1521)))(VERSION=185599488)) * status * 0
    08-AUG-2008 19:34:00 * (CONNECT_DATA=(SID=MYDB)(CID=(PROGRAM=D:\Oracle\Product\product\11.1.0\db_1\perl\5.8.3\bin\MSWin32-x86-multi-thread\perl.exe)(HOST=MYSERVER)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50041)) * establish * MYDB * 0
    08-AUG-2008 19:34:01 * (CONNECT_DATA=(SID=MYDB)(CID=(PROGRAM=D:\Oracle\Product\product\11.1.0\db_1\perl\5.8.3\bin\MSWin32-x86-multi-thread\perl.exe)(HOST=MYSERVER)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50042)) * establish * MYDB * 0
    08-AUG-2008 19:34:09 * service_update * MYDB * 0
    08-AUG-2008 19:34:12 * service_update * MYDB * 0
    08-AUG-2008 19:34:18 * service_update * MYDB * 0
    08-AUG-2008 19:34:21 * service_update * MYDB * 0
    08-AUG-2008 19:34:27 * service_update * MYDB * 0
    08-AUG-2008 19:34:42 * (CONNECT_DATA=(CID=(PROGRAM=)(HOST=__jdbc__)(USER=))(SERVICE_NAME=MYDB)) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50049)) * establish * MYDB * 0
    08-AUG-2008 19:34:45 * service_update * MYDB * 0
    08-AUG-2008 19:34:56 * (CONNECT_DATA=(CID=(PROGRAM=)(HOST=__jdbc__)(USER=))(SERVICE_NAME=MYDB)) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50052)) * establish * MYDB * 0
    08-AUG-2008 19:34:56 * (CONNECT_DATA=(CID=(PROGRAM=)(HOST=__jdbc__)(USER=))(SERVICE_NAME=MYDB)) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50053)) * establish * MYDB * 0
    08-AUG-2008 19:34:56 * (CONNECT_DATA=(CID=(PROGRAM=)(HOST=__jdbc__)(USER=))(SERVICE_NAME=MYDB)) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50054)) * establish * MYDB * 0
    08-AUG-2008 19:34:57 * service_update * MYDB * 0
    08-AUG-2008 19:34:57 * (CONNECT_DATA=(CID=(PROGRAM=)(HOST=__jdbc__)(USER=))(SERVICE_NAME=MYDB)) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50055)) * establish * MYDB * 0
    08-AUG-2008 19:34:57 * (CONNECT_DATA=(CID=(PROGRAM=)(HOST=__jdbc__)(USER=))(SERVICE_NAME=MYDB)) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50056)) * establish * MYDB * 0
    08-AUG-2008 19:34:58 * (CONNECT_DATA=(CID=(PROGRAM=)(HOST=__jdbc__)(USER=))(SERVICE_NAME=MYDB)) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50057)) * establish * MYDB * 0
    08-AUG-2008 19:35:00 * service_update * MYDB * 0
    08-AUG-2008 19:35:00 * (CONNECT_DATA=(CID=(PROGRAM=)(HOST=__jdbc__)(USER=))(SERVICE_NAME=MYDB)) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50059)) * establish * MYDB * 0
    08-AUG-2008 19:35:03 * service_update * MYDB * 0
    08-AUG-2008 19:35:15 * service_update * MYDB * 0
    08-AUG-2008 19:36:54 * service_update * MYDB * 0
    08-AUG-2008 19:36:55 * ping * 0
    08-AUG-2008 19:36:55 * (CONNECT_DATA=(CID=(PROGRAM=)(HOST=)(USER=SYSTEM))(COMMAND=status)(ARGUMENTS=64)(SERVICE=(ADDRESS=(PROTOCOL=TCP)(HOST=bin)(PORT=1521)))(VERSION=185599488)) * status * 0
    08-AUG-2008 19:37:02 * (CONNECT_DATA=(SID=MYDB)(CID=(PROGRAM=D:\Oracle\Product\product\11.1.0\db_1\perl\5.8.3\bin\MSWin32-x86-multi-thread\perl.exe)(HOST=MYSERVER)(USER=SYSTEM))) * (ADDRESS=(PROTOCOL=tcp)(HOST=192.185.10.51)(PORT=50080)) * establish * MYDB * 0
    08-AUG-2008 19:37:15 * service_update * MYDB * 0
    08-AUG-2008 19:38:30 * service_update * MYDB * 0
    ...If you do see patterns like the above, temporarily stop the OracleDBConsole service for the database instance to see if the disk activity slows considerably.
    Charles Hooper
    IT Manager/Oracle DBA
    K&M Machine-Fabricating, Inc.

  • Check process generating high disk I/O read

    Hi Gurus,
    In Solaris 8, which command can I use to check the process that is generating high disk IO read ?
    Mav

    I do not know which file is generating high disk IO read.
    Is there a command to check which process is generating more Disk IO read ?
    I'm using iostat to check Disk IO read every 15 minutes and output to a log file. But I'm unable to determine
    which process is causing high Disk IO read at times.

  • How can i solve this replication problem between TT and ORACLE

    Hi
    I have an application that using AWT cashgroup implement the replication between TT (7.0.6.7) and ORACLE(10g);
    but i encounter this problem:
    16:16:50.01 Err : REP: 2302682: ABM_BAL_WH:meta.c(4588): TT5259: Failed to store Awt runtime information for datastore /abm_wh/abm_bal_ttdata/abm_bal_wh on Oracle.
    16:16:50.02 Err : REP: 2302682: ABM_BAL_WH:meta.c(4588): TT5107: TT5107: Oracle(OCI) error in OCIStmtExecute(): ORA-08177: can't serialize access for this transaction rc = -1 -- file "bdbStmt.c", lineno 3726, procedure "ttBDbStmtExecute()"
    16:16:50.02 Err : REP: 2302682: ABM_BAL_WH:receiver.c(5612): TT16187: Transaction 1316077016/357692526; Error: transient 0, permanent 1
    the isolation level of my date store is read-committed ,and the sys.ODBC.INI file is also set Isolation=1(readcommitted mode)
    so ,I still wonder how the error ORA-08177!
    how can i solve this replication problem?
    thank you.

    I suspect this is failing on an UPDATE to the tt_03_reppeers table on Oracle. I would guess the TT repagent has to temporarily use serializable isolation when updating this table. Do you have any other datastores with AWT cachegroups propagating into the same Oracle database? Or can you identify if some other process is preventing the repagent from using serializable isolation? If you google ORA-08177 there seem to be ways out there to narrow down what's causing the contention.

  • XML Query Have High Disk Reads.

    Question regarding XML data. When I query the following sql statement, the performance is fast. In addition, the explain plan shows that this query is efficient and utliizing all the indexes. However, when I query the v$sql table, I see that this query has very high physical read. Any idea as to why this query has a high physical read?
    SELECT lpcs.TEXT_LOAN_PCH_CONT_CNTNT_XML.extract('/').getClobVal() as TEXT_CNTNT_XML
    FROM REFTDBR1.LOAN_PCH_CONT_CNTNT_SET lpcs
    WHERE lpcs.ID_LOAN_PCH_CONT = '9c24ede6-2bf0-4a9c-8345-fa986ec7516c'
    AND lpcs.ind_deltn = 'N'
    AND lpcs.ind_curr = 'Y'
    DISK_READS
    73488233

    We're on Oracle 11Rg1 RAC with ASM architecture on Liinux platform. Below is the explain plan
    SQL> explain plan for SELECT lpcs.TEXT_LOAN_PCH_CONT_CNTNT_XML.extract('/').getClobVal() as TEXT_CNTNT_XML
    FROM REFTDBR1.LOAN_PCH_CONT_CNTNT_SET lpcs
    WHERE lpcs.ID_LOAN_PCH_CONT = :1
    AND lpcs.ind_deltn = 'N'
    AND lpcs.ind_curr = 'Y' ;
    Explained.
    SQL> select * from table(dbms_xplan.display);
    PLAN_TABLE_OUTPUT
    Plan hash value: 3814252188
    Id Operation Name Rows Bytes Cost (%CPU) Time
    0 SELECT STATEMENT 1 128 5 (0) 00:00:01
    1 TABLE ACCESS BY INDEX ROWID LOAN_PCH_CONT_CNTNT_SET 1 128 5 (0) 00:00:01
    * 2 INDEX RANGE SCAN LOAN_PCH_CONT_CNTNT_SET_X2 1 3 (0) 00:00:01
    Predicate Information (identified by operation id):
    PLAN_TABLE_OUTPUT
    2 - access("LPCS"."ID_LOAN_PCH_CONT"=:1 AND "LPCS"."IND_CURR"='Y' AND "LPCS"."IND_DELTN"='N')
    14 rows selected
    This query was executed 1623 times so I guess that could explain the high disk reads? Does this mean that this query is not being stored in the buffer cache since the disk reads are high?
    DISK_READS 73488233
    FETCHES 1623
    EXECUTIONS 1623

  • Performance issue High disk I/O

    Hi All,
    we have Oracle9i Enterprise Edition Release 9.2.0.1.0 - Production running on REd Hat EE 4
    We are having some high disk I/O issues after doing some schema changes last week.
    I checked some wait events but found pmon timer and smon timer at the top all the time. but not getting any clue of reason behind it.
    If anyone has faced such issue then please advice as what could be reason of it.
    SELECT a.SID, a.event, a.wait_time, a.seconds_in_wait, a.state
    FROM v$session_wait a, v$session b
    WHERE a.SID = b.SID AND event <> 'SQL*Net message from client'
    ORDER BY seconds_in_wait DESC
    SID     EVENT     WAIT_TIME     SECONDS_IN_WAIT     STATE
    1     pmon timer     0     4090     WAITING
    8     smon timer     0     1209     WAITING
    9     rdbms ipc message     0     799     WAITING
    regards
    vinay

    To add some more information to this
    The disk activity before doing schema changes used to be somewhere around 30-45% but now it has suddenly went to 70-80%.
    The schema changes includes addition of some complex views, tables and indexes.
    but i haven't seen any considarable chnages in size of newly added objects they are all together less than 1.5 gb and our total database size is 700GB.
    i took snapshot for over period of time i found top event as direct path write
    Top 5 Wait Events
    ~~~~~~~~~~~~~~~~~ Wait % Total
    Event Waits Time (cs) Wt Time
    direct path write 1,444,566 495,020 38.06
    direct path read 677,912 139,936 10.76
    PX Deq: Execution Msg 1,252 129,432 9.95
    db file scattered read 1,235,215 117,431 9.03
    db file sequential read 912,288 98,714 7.59
    Top 5 Wait Events
    ~~~~~~~~~~~~~~~~~ Wait % Total
    Event Waits Time (cs) Wt Time
    direct path write 3,923,608 1,218,268 33.98
    db file scattered read 1,770,687 533,507 14.88
    direct path read 3,676,295 468,525 13.07
    db file sequential read 1,911,892 331,635 9.25
    SQL*Net more data to client 7,126,163 188,606 5.26
    there is considerable slow response has been reported on application but but still usable.
    Regards
    Vinay

  • Session in-memory replication problem

    Hi,
              I am running into some cluster HttpSession replication problems. Here is
              the scenario where replication fails (all servers mentioned here are a
              part of a cluster).
              1a - 2 Weblogic servers (A&B) are running - no users logged in,
              2a - user logs in and a new session in server A is created.
              3a - after several interactions, server A is killed.
              4a - after user makes susequent request, Weblogic correctly fails over
              to server B
              Problem: Not entire session data is replicated. The authentication info
              seems to
              be replicated correctly but there are some collections in the session of
              server A
              that did not make it to the session in server B.
              The interesting part is this: If there is only one server A running to
              begin with and a user
              interacts with it for a while and only then server B is started, when
              after server B starts up
              server A dies - the entire session (which is exactly the same as in the
              failing scenario) is
              corretly replicated in B, including collections that were missing in the
              failing scenario.
              How can this be possible ????
              Thanks for any info on this one - it really puzzles me.
              Andrew
              

    Yes, you are on the right track. Everytime you modify the object you should call
              putValue. We will make it more clear in the docs.
              - Prasad
              Andrzej Porebski wrote:
              > Everything is Serilizable. I get no exceptions. I did however read some old
              > posts regarding
              > session replication and I hope I found an answer. It basically seems to boil
              > down to what
              > triggers session sync-up between servers. In my case , I store an object into
              > session and
              > later on manipulate that object directly wihotu session involvment and the
              > results of those manipulations
              > are not replicated - no wonder if HttpSession's putValue method is the only
              > trigger.
              > Am i on the right track here?
              >
              > -Andrew
              >
              > Prasad Peddada wrote:
              >
              > > Do you have non serializable data by any chance?
              > >
              > > - Prasad
              > >
              > > Andrzej Porebski wrote:
              > >
              > > > Hi,
              > > > I am running into some cluster HttpSession replication problems. Here is
              > > > the scenario where replication fails (all servers mentioned here are a
              > > > part of a cluster).
              > > > 1a - 2 Weblogic servers (A&B) are running - no users logged in,
              > > > 2a - user logs in and a new session in server A is created.
              > > > 3a - after several interactions, server A is killed.
              > > > 4a - after user makes susequent request, Weblogic correctly fails over
              > > > to server B
              > > >
              > > > Problem: Not entire session data is replicated. The authentication info
              > > > seems to
              > > > be replicated correctly but there are some collections in the session of
              > > > server A
              > > > that did not make it to the session in server B.
              > > >
              > > > The interesting part is this: If there is only one server A running to
              > > > begin with and a user
              > > > interacts with it for a while and only then server B is started, when
              > > > after server B starts up
              > > > server A dies - the entire session (which is exactly the same as in the
              > > > failing scenario) is
              > > > corretly replicated in B, including collections that were missing in the
              > > > failing scenario.
              > > >
              > > > How can this be possible ????
              > > >
              > > > Thanks for any info on this one - it really puzzles me.
              > > >
              > > > Andrew
              > >
              > > --
              > > Cheers
              > >
              > > - Prasad
              >
              > --
              > -------------------------------------------------------------
              > Andrzej Porebski
              > Sailfish Systems, Ltd. Phone 1 + (212) 607-3061
              > 44 Wall Street, 17th floor Fax: 1 + (212) 607-3075
              > New York, NY 10005
              > -------------------------------------------------------------
              

Maybe you are looking for