Author: windows7bugs

Oracle certified developer with more than ten years of experiences with Windows based domains and client operating systems.

VirtualBox | SATA vs SCSI

On By windows7bugsIn Linux, Oracle Applications(R12), Oracle related, Oracle VirtualBox5 Comments

Recently I tried to build a cloned instance of our production instance over VirtualBox for some emergency issues faced by our inventory module. As this instance was supposed to be only accessed by me, I opted to use my desktop machine for the same. Throughout the last many years I built my own machines, choosing the best available hardware at the time of building them. My current desktop configuration is like following

i7 processor, 16GB memory, 2x1TB 7200 RPM HDD, 2x2TB 5200 RPM HDD, 1x500GB HDD for the OS (Windows 8.1 64Bit)

and throughout the years I built dozens of Virtual Machines using Oracle VirtualBox, mainly for testing un-certified Oracle & other products in a sand-boxed environment, against the crippled VMplayer, VirtualBox’s unrestricted interface supported almost everything I needed from a virtual environment.

So I built my R12 instance, that is around 600GB roughly in size with almost 4.5-5 years of business data, media etc. The following resources were dedicated for the fresh VM

  • 4 processors
  • 10GB memory
  • 40GB fixed size SATA VDI for the Operating System (I used both OEL 5 & OEL 7 64bit)
  • 1.2TB fixed size SATA VDI for the instance files
  • A dedicated D-Link 10/100MB NIC

Once the instance came online, I removed, cancelled all the scheduled concurrent programs, changed the database level parameters like job_queue_processes etc, however the lag experienced throughout the access attempts remained the same. Sometimes the HTML pages took 5-6 minutes to open, forms based modules took 8-10 minutes to open and timeouts were happening, frustrating me to the most possible levels

That is when I decided to give VMPlayer a try, I converted the existing VDI for the OS as vmdk and created a fresh 850 fixed size vmdk for the instance files and attached the same as SCSI to the VM. Did the complete clone process and to my utter surprise, the login page loaded within a minutes once after the instance was started!

This lead me to do various attempts with the fresh instance, I was able to shutdown the instance much faster, forms were opening faster, though LOVs having more than thousands of items were taking more time than anticipated

Once again, I created another fresh VM with VirtualBox and attached the disks created for VMplayer with it and repeated the tests. Well, I got the same performances from the new VM and somehow I came to a conclusion that, both VirtualBox and VMPlayer provide better I/O for SCSI interfaces compared to plain SATA emulators, ironically, the disks were created over SATA drives!

This difference you may not experience with VMs those are not hosting resource hungry applications like Oracle E-Business Suite. So, if you are attempting what I had described above and notice the differences, please update me with comments section.

 

regards,

 

rajesh

 

 

VirtualBox | CentOS/Oracle 7.x Linux| Memory cannot be read/write error

On By windows7bugsIn Linux, Oracle VirtualBoxLeave a comment

Hi guys

Many times I pointed out the importance of testing technical previews, latest releases of other OS using VirtualBox as the early releases are prone to have frustrating bugs, which are many time specific to a particular hardware or distro. I was playing around with Oracle linux 7, which is a forked version of Red Hat Linux 7, named Unbrekable Linux by Oracle. Relatively I hardly had many issues with guest OS in recent times while tested over VirtualBox  and with OEL 7 guest, all of a sudden I started getting memory read errors like the one you can find with below image,

7-vb-memory error

once after the guest addons were installed. Although the change log of VirtualBox for 4.3.28 states the 3D bugs fix,  3D acceleration is the main culprit for the guest crashing once after a login attempt.

VirtualBox change log: https://www.virtualbox.org/wiki/Changelog

So, all you need to do is to disable the 3D acceleration for the guest using the settings panel. Well, unless you are going to use your VM for some “gaming” ;) I think this temporary fix should get you online with your CentOS/OEL 7.1 VM box!

Additional reading: https://www.virtualbox.org/ticket/13332

regards,

 

Oracle EBS R12 | Inventory | Receipts stuck in pending status

On By windows7bugsIn Oracle Applications(R12), Oracle Inventory (R12)Leave a comment

Sometimes the receipts against internal requests could get into trouble with the status pending. While the requester tries to receive the transaction, receipts get created with zero quantities and the internal sales order yet in open status and the quantities are seen in staging location. Please follow the below to delete such pending lines to receive the quantities once again to the targeted sub-inventory

As inventory super user

Transactions -> Receiving -> Transactions Status Summary

1

Under the tab “Supplier and Internal”

Select the “Destination” tab

Enter the name of the requester name (the name of the user who has created the internal material request to limit the lines fetched) & find the transactions

Delete the lines matching the transaction & save

2

Go back to receipts and enter the request number once again to receive the materials

regards,

for Windows7bugs

rajesh

Oracle EBS R12, Install/Clone on Oracle/Red Hat Linux 7.x

On By windows7bugsIn Linux, Oracle Applications (R12.0.4), Oracle Applications(R12), Oracle Enterprise LinuxLeave a comment

Hi guys

Last reviewed on: 10th January 2022

Made changes to the body for Oracle Linux 7.9

References: http://www.slideshare.net/tianpan/oracle-ebs-r1213installationlinux64bitpantian

(and dozens of EBS R12 related blogs and oracle community posts)

As far I could recollect, Oracle EBS R12.0.x & R12 12.1.x are not certified yet for Linux 7.x, let it be Red Hat Linux or Oracle enterprise Linux. Does the certification path really make guys like us trying to clone, install EBS R12 over yet to be certified platforms? Well, my case, I always did attempt & succeeded to a certain levels. Starting with Oracle 10g over Windows 7, 8 etc.

This time, my attempt was to clone our existing EBS R12 12.0.6 instance from Red Hat linux Enterprise 5 64Bit to Oracle Enterprise linux 7 64Bit.

I can’t say whether it is a rightful thing for you or for your environment. For me it was a fun project, that I achieved to a maximum satisfactory level.

Environment: Source System

Red Hat Linux 5.11 64Bit, 48GB physical memory, HBA interfaced external storage, single node EBS R12 12.0.6 with Oracle database 10.2.0.3 database

Environment: Target System

1. Oracle VirtualBox VM with 4 processors, 10GB memory, Oracle Enterprise Linux 7.9 with GUI on demand

VDI fixed size of 1.3T holding both application tier & database tier files copied from source system

2. A Desktop machine with i7 Processor, 4.5 TB Storage, 16GB physical RAM for real hardware performance testing

How to?

Definitely, this is not for a production instance, not for a production instance, not for a production instance & REMEMBER, even if you have valid Oracle support, Oracle will NOT support YOU as the platform is NOT certified for 12.0.x or 12.1.x

Oracle Enterprise Linux 7.x is not certified for EBS R12.0.x , hence I was unable to find the pre-requisites install for EBS R12 for the OS. The hacks started from this point onward

Update(10th January 2022): Oracle Linux 7.9 has the “Add-on” repository that allows you to install the pre-requisites package, however meant for later certified versions of Oracle EBS. Hence, stick to Oracle Linux 6 public yum repository based package installation as mentioned below.

(Please make sure that you fully updated your OEL 7.x box prior the attempts listed below)

As root switch the directory to

cd /etc/yum.repos.d

Go to

http://public-yum.oracle.com/ and install the ol6 repo following instructions provided. The instruction might get in to folds, just follow the below.

cd /etc/yum.repos.d/
wget http://yum.oracle.com/public-yum-ol6.repo

Please note public yum repositories will be soon removed, if you are reading this post much later, make sure that you will find relevant sources to install the pre-requisites.

Open the ol6 repo file using gedit/vim and enable the add-on repo. Make sure the OEL7 repo doesn’t have the add-ons enabled. If you miss this, you might end up rebuilding the Linux Server (That’s what I did!)

install pre-requisites for EBS R12 by issuing the following command

yum install oracle-ebs-server-R12-preinstall -y

Once the installation completed, you will notice that you have two new users created

oracle

applmgr

If your cloned instance was having different user names for both application and database, create same usernames and add them to relevant groups

for example applprod to oinstall group and oraprod to dba, oinstall groups (mandatory)

Special note: If you are attempting these hacks for Production instances, sorry buddy, YOU will never be supported by Oracle as they have recently confirmed to me through a private message that they don’t have any plans to certify 12.0.x versions on OEL 6 or 7

Prior you attempt to clone or install the 12.0.6 instance on OEL 6, 7 make sure that you have taken care of few things prior the attempt. There are few OS level files formatting mandatory for Oracle applications, those caused me almost 3 weeks of reading various blog posts, and NOT a single solution to address a weird error associated with “Error: Missing ormi[s]://<host>:<port>” error while shutting down the application services using adstpall.sh script file! And the reason was default formatting of /etc/hosts file by OEL 7 installation!

Few errors you should see when the /etc/hosts file is NOT formatted as per Oracle’s installation notes…

Executing service control script:
 …../admin/scripts/adoacorectl.sh stop
 Timeout specified in context file: 100 second(s) 

script returned:

  ERROR : Timed out( 100000 ): Interrupted Exception
  You are running adoacorectl.sh version 120.13

So the /etc/hosts file must be formatted like below with your OEL/RHEL 6.x & 7.x installations

127.0.0.1 localhost.localdomain localhost 
::1 localhost6.localdomain6 localhost6 
192.168.10.15 erp-prodbak.xyz.com erp-prodbak #Your box 
192.168.10.16 erp-prod.xyz.com erp-prod #Your production server, if at all required

Rest of the network related mandatory elements specified in the reference documents could be ignored, or you may format them as well. If you are going to touch your /etc/resolv.conf file at all, make sure that after changing the content of the resolv.conf, you make it read-only by issuing the following command

chattr +i resolv.conf

else after a reboot, this file be rewritten by the networking admin services

Now let us refer to the /etc/sysctl.conf & /etc/security/limits.conf files, which are setting system level parameters for the application. While I noticed that the previously configured parameters were causing performance degradation once after the clone, I just copied these elements from a 12.1.3 vision instance and replaced to find that, my cloned instance started flying, giving me much better performance than our Production instance that has the latest hardware and 48GB physical memory!

Once again, thoroughly check the below parameters, if something looks not appropriate for your environment, refrain from updating

as a precaution, do a backup for both the above mentioned files

cp /etc/sysctl.conf /etc/sysctl.conf.original 
cp /etc/security/limits.conf /etc/security/limits.conf.original

Now using your favorite editor, replace the content of you sysctl.conf file with following

# Kernel sysctl configuration file for Red Hat Linux
#
# For binary values, 0 is disabled, 1 is enabled.  See sysctl(8) and
# sysctl.conf(5) for more details.
#
# Use '/sbin/sysctl -a' to list all possible parameters.

# Controls IP packet forwarding
net.ipv4.ip_forward = 0

# Controls source route verification
net.ipv4.conf.default.rp_filter = 1

# Do not accept source routing
net.ipv4.conf.default.accept_source_route = 0

# Controls the System Request debugging functionality of the kernel
kernel.sysrq = 0

# Controls whether core dumps will append the PID to the core filename.
# Useful for debugging multi-threaded applications.
kernel.core_uses_pid = 1

# Controls the use of TCP syncookies
net.ipv4.tcp_syncookies = 1

# Controls the default maxmimum size of a mesage queue
kernel.msgmnb = 65536

# Controls the maximum size of a message, in bytes
kernel.msgmax = 65536

# Controls the maximum shared segment size, in bytes

# Controls the maximum number of shared memory segments, in pages
kernel.shmall = 4294967296

# oracle-ebs-server-R12-preinstall setting for fs.file-max is 6815744
fs.file-max = 6815744

# oracle-ebs-server-R12-preinstall setting for kernel.sem is '256 32000 100 142'
kernel.sem = 256 32000 100 142

# oracle-ebs-server-R12-preinstall setting for kernel.shmmni is 4096
kernel.shmmni = 4096

# oracle-ebs-server-R12-preinstall setting for kernel.shmall is 1073741824 on x86_64
# oracle-ebs-server-R12-preinstall setting for kernel.shmall is 2097152 on i386

# oracle-ebs-server-R12-preinstall setting for kernel.shmmax is 4398046511104 on x86_64
# oracle-ebs-server-R12-preinstall setting for kernel.shmmax is 4294967295 on i386
kernel.shmmax = 4398046511104

# oracle-ebs-server-R12-preinstall setting for kernel.msgmni is 2878
kernel.msgmni = 2878

# oracle-ebs-server-R12-preinstall setting for net.core.rmem_default is 262144
net.core.rmem_default = 262144

# oracle-ebs-server-R12-preinstall setting for net.core.rmem_max is 4194304
net.core.rmem_max = 4194304

# oracle-ebs-server-R12-preinstall setting for net.core.wmem_default is 262144
net.core.wmem_default = 262144

# oracle-ebs-server-R12-preinstall setting for net.core.wmem_max is 1048576
net.core.wmem_max = 1048576

# oracle-ebs-server-R12-preinstall setting for fs.aio-max-nr is 1048576
fs.aio-max-nr = 1048576

# oracle-ebs-server-R12-preinstall setting for net.ipv4.ip_local_port_range is 9000 65500
net.ipv4.ip_local_port_range = 9000 65500

Save the file and issue the following command

sysctl -p

Now, change the limits.conf file with following content

# /etc/security/limits.conf
#
#Each line describes a limit for a user in the form:
#
#            
#
#Where:
# can be:
#        - a user name
#        - a group name, with @group syntax
#        - the wildcard *, for default entry
#        - the wildcard %, can be also used with %group syntax,
#                 for maxlogin limit
#
# can have the two values:
#        - "soft" for enforcing the soft limits
#        - "hard" for enforcing hard limits
#
# can be one of the following:
#        - core - limits the core file size (KB)
#        - data - max data size (KB)
#        - fsize - maximum filesize (KB)
#        - memlock - max locked-in-memory address space (KB)
#        - nofile - max number of open file descriptors
#        - rss - max resident set size (KB)
#        - stack - max stack size (KB)
#        - cpu - max CPU time (MIN)
#        - nproc - max number of processes
#        - as - address space limit (KB)
#        - maxlogins - max number of logins for this user
#        - maxsyslogins - max number of logins on the system
#        - priority - the priority to run user process with
#        - locks - max number of file locks the user can hold
#        - sigpending - max number of pending signals
#        - msgqueue - max memory used by POSIX message queues (bytes)
#        - nice - max nice priority allowed to raise to values: [-20, 19]
#        - rtprio - max realtime priority
#
#                 
#

#*               soft    core            0
#*               hard    rss             10000
#@student        hard    nproc           20
#@faculty        soft    nproc           20
#@faculty        hard    nproc           50
#ftp             hard    nproc           0
#@student        -       maxlogins       4

# * hard nofile 65535
# * soft nofile 4096
# * hard nproc 16384
# * soft nproc 2047


# oracle-ebs-server-R12-preinstall setting for nofile soft limit is 4096
oratest   soft   nofile    4096
appltest   soft   nofile    4096

# oracle-ebs-server-R12-preinstall setting for nofile hard limit is 65536
oratest   hard   nofile    65536
appltest   hard   nofile    65536

# oracle-ebs-server-R12-preinstall setting for nproc soft limit is 16384
# refer orabug15971421 for more info.
oratest   soft   nproc    16384
appltest   soft   nproc    16384

# oracle-ebs-server-R12-preinstall setting for nproc hard limit is 16384
oratest   hard   nproc    16384
appltest   hard   nproc    16384

# oracle-ebs-server-R12-preinstall setting for stack soft limit is 10240KB
oratest   soft   stack    10240
appltest   soft   stack    10240

# oracle-ebs-server-R12-preinstall setting for stack hard limit is 32768KB
oratest   hard   stack    32768
appltest   hard   stack    32768


# End of file

Install unzip version 5 from https://oss.oracle.com/el4/unzip/unzip.html

(This is mandatory as 12.0.x doesn’t support unzip versions above 5, and your cloning will fail with error notifying unzip is not supported)

Source the libdb.so.2 library file to /usr/lib from your source system, as the libgdbm.so.4.0.0 soft link hack will not work. If the libdb.so.2 is already a soft link in your source system, move the source file “libgdbm.so.2.0.0” from source to target box and create the soft link.

Now go ahead and enjoy the cloning/installation of your R12 12.0.x, or R12.1.x releases over OEL/RHEL 7.x (CENTOS is GNU release of RHEL, so you may experiment with that distro as well, nothing guaranteed though)

Please be warned, I have read many places that, few modules do not work on Linux 6,7 releases. Hence, the entire exercises you painfully executed may turn futile at later stages, and once again, if you have a live Oracle support, do not attempt such with your Production instance!

Enjoy guys :)

for Windows7bugs

rajesh

Windows 8 | Wan miniport | device disconnected

On By windows7bugsIn Windows 7, Windows 8Leave a comment

For unknown reasons, the Wan miniport(s) available with Windows 7/8 could start showing errors & giving the user no apparent solutions to fix them, including uninstalling them.

I had the same situation when all of a sudden one of my PPTP connections started giving errors like “Device disconnected”, properties for the PPTP connection not shown and not allowing me to delete the connection in the corrupt state.

After few attempts, I came across a workable solution, posted with Microsoft support site. Unfortunately, this time I forgot to bookmark the page, however, still remember the entire steps. Please find them listed below

View – > Show hidden devices

1

Right click the Wan Miniport device with yellow exclamation mark

(Enlarge the images for more details)

2

3

4

5

Now force install the driver. Once the installation is over, you can safely uninstall this device. Follow the same for all the Miniport devices those are showing the exclamation mark. Once you are done with the uninstall processes, don’t do anything else. Just restart the computer and start creating the VPN connections once again.

The above exercises must have fixed the Wan miniport issues for you.

Regards,

rajesh

Windows 10 on Virtual Box | Unable to allocate and lock memory | Virtualbox Guru Meditation error

On By windows7bugsIn Windows 10Leave a comment

As Microsoft is continuously updating their technical preview for Windows 10, we are sure many of you out there are trying out the latest release 10074, that adds(?) many features. As a precaution, we always use Oracle VirtualBox for testing such releases and never had more issues than with the pre-releases of Windows 10!

One of the regular errors we came across were about “Unable to allocate and lock memory” while the VM started, and gradually pops up another window titled “Guru Meditation”, asking us to view the log files (/me chuckles, when Google can do that for us). We came across a wonderful post, that explains how to resolve this particular dreadful situation with VMs!

and here we are linking the same POST for you! Enjoy and Enjoy your Windows 10 VM

regards,

rajesh