Recommended Solaris package upgrade procedure

We would like to know the recommended Application Package upgrade procedure. We followed the procedure as defined in the Application Packaging Developer's Guide.
Currently we followed the following procedure during upgrade:
1. Check if old version is available.
2. If it is available execute the pkgrm command to remove the old package as indicated in the Application Packaging Developer's Guide. But since the package names are same and we are executing the pkgrm command from the preinstall script, in some systems it's not allowing to remove the old package since it's executed from pkgadd command scope. The following error message is displayed while package verifying step is executed
[ verifying class <none> ]
pkgadd: ERROR: missing entry in contents file for </opt/XXXXxxxx/jre>
3. We cannot manually remove the old package first since it will completely wipe out the file. We need to migrate some of the files from old release to new. So we need the old files during upgrade.
Please suggest a suitable procedure to handle this.
Some more details about the OS:
1. uname -a
SunOS rduipv6 5.10 Generic_142900-03 sun4u sparc SUNW,UltraAX-i2 Solaris

Hi.
I can't say how Oracle recomend. But I see that for update main part of Sun Product used two ways:
1) Documentation say that user should remove all previos version of product by hand.
2) Used extarnal installation tool that do it automaticaly, but not as part package installation script (pre/postinstall ).
Regards.

Similar Messages

Upgrade procedure from 4.6c to ECC 6.0

Hi
I had gone through SAP solution browser tool and observed the differences for 4.6C and 6.0
Can u please let me know How to start and end the upgrade project i.e sequence of activities.
1) should the configuration be done in ECC 6.0 same as 4.6C?
2)master data upload?
3)Z-objects movement from 4.6C to 6.0....?
4)Open PO's and stock upload......in ECC 6.0 ?
5) what are the precautions to be taken before Go-live?
Please send any links for upgrade procedure from 4.6C to 6.0 i.e sequence of activities.........
Thanking you
Deepak

Hi,
About your questions:
1) should the configuration be done in ECC 6.0 same as 4.6C?
The configuration (if you mean how the system is configured from the Basis point of view) will be almost the same as you had in 4.6C. Of course you will find new aspects you don't have in the old version (for example the integrated ITS, ICM, etc.), but you will have the system running after the upgrade.
2)master data upload?
Master data will be adapted to new version during the upgrade. Unless you need fresh data, you won't need to reload master data.
3)Z-objects movement from 4.6C to 6.0....?
Z objects are not touched by the upgrade process. It's your responsibility to keep them running after the upgrade.
4)Open PO's and stock upload......in ECC 6.0 ?
Same as question 2). Business data remains in the system after the upgrade.
5) what are the precautions to be taken before Go-live?
It is recommended to carry out functional tests, as well as connection tests with external interfaces. In any case, you shouldn't take any special precaution, at least not so different from those you would take in a normal installation.
Regards
Francisco

Solaris Live Upgrade

I didnt find a specific place to place this topic, so maybe you can replace.
I read a lot about Solaris Live Upgrade, but I'd would like to know if I can copy a BE from one computer to another over network via NFS.
Or if there is another software that can make a image of my system and replace to the others computers (they are all the same), so I will dont need to configure all of them.
Thx for any help.

Thanks for answering.
I read about JumpStart installation, but is it possible to install a customized image, but not with just the partition, packages and patches selected. I wanted to share a whole system configured. For example, the DNS, IP, User's account settings, etc.
Because, I just found a way to share the installation image with the selected packages and patches. But you will need to configure the files of the packages later in the clients.
Is it possible?

Solaris package patch creation

I am looking for documentation or procedure regarding Solaris package patch creation.
I have created Solaris packages for an ISV application. Now I'd like to create a Solaris
patch to replace one of those package's binaries - where if the patch is backed out, the
previous binary is restored.
I have read the Sun Application Packaging Developer's Guide (both 1994 and current),
and the subsection on Patching Packages does not contain sufficient information to
accomplish my goal. I have searched the Net for information, and found other's
questions and few answers.
Can anyone produce the procedure for Solaris package patch creation?
Thank you.

Here is why no info is available. Sun advises:
"We cannot make patch building utility publicly available irrespective of the customer having support contract. We suggest you to provide an updated package instead of patches to your customer and tell them to remove the old version and add the new version of the package."

Sun packaging upgrade

hi,
i would like to create a package that can be upgraded .
Meaning , will NOT have the same name as old pkg , will need to keep some files (config) from old pkg into new pkg , and can do it in one command.
the RPM has an option like that ,.
how can it be doen in the Sun Packing mechanism ?
thanks,

This upgrade is delivered as a patch set and some
additional packages. The upgrade procedure is
basically the same as a normal patch procedure.
You might consider getting the latest core Sun Cluster
patch as well, just to be sure you are at the latest rev.
The additional packages provide new features. The
most popular is the Generic Data Service which can
really save development time for simple agents.
-- richard

Issue while running a custom package:ORA-20100: File o2670336.tmp creation for FND_FILE failed.You will find more information on the cause of the error in request log. in Package PA_OPPORTUNITY_MGT_PVT Procedure modify_project_attributes ORA-20100: File o

Hi Guys,
We have created a custom package where in we are trying to call the standard API's of Oracle projects i.e PA_PROJECT_PUB.When we are trying to call these APIs we are facing the below issue.
We have tried testing in two instances ,Initially it worked in both instances.
Using the same API's multiple times we tested the same data set in these instances.
For the first few runs it works fine.But when we go on using the same API's again and again for our testing we face now and then the below issue.
Standard API's
=========
add_task
update_project
change_structure_status
create_draft_plan
Error:
ORA-20100: File o2670336.tmp creation for FND_FILE failed.You will find more information on the cause of the error in request log. in Package PA_OPPORTUNITY_MGT_PVT Procedure modify_project_attributes ORA-20100: File o2670336.tmp creation for FND_FILE failed.
You will find more information on the cause of the error in request log. in Package PA_PROJECT_PUB Procedure update_project:ORA-0000: normal, successful completionError while publishing the task
Please let us know if anyone of has come across the same situation.
Regards,
Vijay

But have issue only with validating invoice batch. I followed metalink ID's 605542.1, 167990.1,261693.1,1088553.1,749491.1, 461271.1 and few more.Was this working before? If yes, any changes have been done recently?
Please see if these docs help.
APXAPRVL: Invoice Validation Errors With MSG-00001 and REP-1419 [ID 333305.1]
Invoice Validation Errors When Others:101505:Non-Oracle Exception Rep-1419 [ID 464125.1]
Invoice Validation Fails with REP-1419 Error in Ap_approval_matched_pkg.Execute_matched_checks [ID 293425.1]
Invoice Validation Failing On Fnd_file Could Not Write To File L0202306.Tmp [ID 461520.1]
Invoice Validation (APPRVL) Errors ORA-20001 APP-SQLAP-10000 PSA_FUNDS_CONTROL_PKG.glxfck [ID 463184.1]
Validate Invoice Error With ORA-20001: APP-SQLAP-10000: AP_FUNDS_CONTROL_PKG.Calc_QV [ID 432702.1]
Invoice Validation Program Is Erroring Out [ID 382844.1]
Error On Validation Of Invoices From Previous Periods [ID 412814.1]
Thanks,
Hussein

SSIS package Upgrade from 2005 to 2012

Hi,
We have a Solution file with 8 projects for SSIS packages. All of them are in VSS. They all have been created in 2005. Now we installed VS 2012 and SSDT on our local machines. I got latest of solution and projects from VSS on new machine. I tried to open
solution file which took me SSIS package Upgrade wizard. I followed the steps and in the last step of conversion, it displayed that it had errors. The errors are:
Error 0xc001f429: Package Upgrade: The file, "C:\SSIS\compay\hello.dtsx", could not be opened for writing. The file could be read-only, or you do not have the correct permissions.
and for some packages, it is like this:
Messages
* Information 0x40019316: : The provider name for the connection manager "OLEDBConnection" has been changed from "SQLNCLI.1" to "SQLNCLI10".
* Information 0x40019316: : The provider name for the connection manager "OLEDBConnection" has been changed from "SQLNCLI10" to "SQLNCLI11".
* Information 0x40016019: : The package format was migrated from version 2 to version 6. It must be saved to retain migration changes.
* Warning: PackageName: Found SQL Server Integration Services 2005 Script Task "ScriptTask_49664f73e0f440bbb0e7d44f01036bff" that requires migration!
* Information 0x4001601a:PackageName: The Script Task "ScriptTask_49664f73e0f440bbb0e7d44f01036bff" has been migrated. The package must be saved to retain migration changes.
* Error 0xc001f429: Package Upgrade: The file, "C:\SSIS\company\Hi.dtsx", could not be opened for writing. The file could be read-only, or you do not have the correct permissions.
But solution and all projects and packages show that they are checked out. I checked properties also. Read only checkbox is not checked.
Please suggest what to do. I appreciate your response.
Thanks,

Why didnt you check it out directly from VSS rather than going this long route? How would VSS now that file is in checked out status if you go and manually change readwrite property? I think it may be that its getting confused when you tried to save
it into its working folder as it still has status as checked in and not locked by anyone.
Can you try saving packages copies to different folder and then do the upgrade by including them onto a IS project?
Please Mark This As Answer if it helps to solve the issue Visakh ---------------------------- http://visakhm.blogspot.com/ https://www.facebook.com/VmBlogs

What is the recommended way to upgrade system ruby and python binaries and their libraries?

I'd like to upgrade my system ruby to 1.9.2 instead of old 1.8.7. Also was wondering what's the recommended way of upgrading any scripting languages installed on system?

Hi Sogaard,
see below:
How does SAP recommend to integrate Webi-report in the SAP Portal? Is it through an URL iview, the iview templates (thumbnail, folder and alert) or through the Master Iview?
The sample iViews are just samples for a particular case. The iView template allows you to use any application on the BusinessObjects system - including the viewing of content. What you are looking for is the creation process for WebI. So you have 2 options: you can use the KM integration and you can build a java application that offers the workflow you looking for and integrate it with the iView template
Depending on what method should be used, I'd like to know which settings to be focused on. For instance, if the Master Iview is to be used, should opendocument be used instead of reports? And what other customizations will have to be done in order to integrate a Webi-report instead of a Crystal-Report?
OpenDocument is for viewing reports. The Installation Guide outlines the integration of the iView template and the KM part.
What is the intended use for the BOBJ repositoy manager? Would that be the way to integrate the Info-view?
This is the integration into the KM repository which provides a richer functionality then just an iView.
Ingo

Sql developer doesn't show any packages and procedure.

Hi All,
I am perty new here and I am facing a strange problem in SQL DEVELOPER. I have one schema and it has packages and functions created in it, but when I click
packages or functions in sql developer it says table or views does not exist. however, I can see those packages and procedures from sql plus and oracle enterprise manager. you swift reply would be highly appriciated
hare krishna
Alok

Which version of sqldeveloper?
Which version of Oracle?
This sounds like an old bug where sqldeveloper checked for 'select any table', but you actually need 'select any dictionary'

Strange download size of on calling packaged stored procedure in VB6

Hi,
I tried to use VB6 to call oracle stored procedure with ref cursor as output parameter under a package in 10g database (10.1.03). Oracle client uses windows 9.2 / 10.1.02 (Oracle OLEDB Provider) . I found that when the package size increases(> 1000 lines codes), the data size transfer from database to VB client also increases. I try to trap the binary data flow between VB6 and database when calling packaged stored procedure. I found that VB6(or oracle client ??) will first download package spec then execute the requested stored procedure. This may explain the cause of increasing size.
However, this large download size for calling only one of the procedure will make my application run slowly when my application uses the package much. How can I solve this problem ? Does this relate to oracle client(p.s. this problem disappear when using Microsoft OLEDB Provider)?
Thanks
Steven

You can use a wrapper to get compiled code and if it's 10g you can do even more by putting it in a library.
OLEDB from VB6 still using Oracle 'Client' as you call Oracle NET.

Why apps don't need to be granted for synonyms of packages and procedures?

Normally, When a user create a synonym of other schemas' package or procedure, it requires to be authorized to execute via 'grant' by owner before the user can call it.
But on one of our environment of EBS 12.1.1(vis), apps user don't need to be granted. Just need to make synonym and then can call the synonym. while on another environment of EBS 12.1.1(product), it still need to be granted. I notice that apps user in the two diffierent environment has many different database roles. Is any role associated with this? Does anybody know the reason?
Thanks in advance!
Jing
Edited by: user12954958 on Aug 25, 2010 11:44 PM

Hi Jing;
Please check below thread which is posted Hussein Sawwan
Re: apps & applsys password
aaps,applsyspub,applsys
Re: apps & applsys password
Regard
Helios

SubVersion in SQL Developer - how to link to packages and procedures

My question is this - I am starting out using SubVersion in Oracle and it seems pretty straightforward to use as a source control for files that are stored in a Windows directory, that is the only way I have used it so far.
Am I able to use it to control packages and procedures that I have saved as part of my database, ie the procedures are part of a package and the package appears within the Packages section of the appropriate connection in SQL Developer.
I hope this makes sense.
Cheers,
JabJam

Hi JabJam,
So far, everyone is correct. SQL Developer just uses a file-based source control system (e.g., SVN), which means updating database object definitions via scripts under version control, or somehow exporting the new definition to a script so it can be version controlled after the fact. And wouldn't it be nice if the database objects could be version controlled directly. With the database developer having control over when any changes are committed for shared use in the database. Similar to using the SVN plug-in for an IDE like Eclipse for a given programming language in a shared development environment.
But what would your DBA have to say about that? Don't most DBAs require scripts for deploying code or structural changes?
In Oracle, definitions (DDL) are committed automatically. Anyone granted access sees them. Since you probably don't do development directly against a production database, eventually scripts (either manual or generated) will be needed for deployment to QA or production. Few mechanisms currently exist to work against an active database without other authorized users also seeing these changes.
For DDL, the closest thing would be the fairly recent edition-based redefinition feature in 11g (R2, I think). For DML, the OWM (Oracle Workspace Management) feature would be loosely analogous.
This is not a simple issue. Version control via an extension for Subversion provides essential basic support. Going beyond that might mean watching the evolution of edition-based redefinition, but DBAs would probably require scripts for that too.
Regards,
Gary
SQL Developer Team

Solaris 2.7 to Solaris 8 Upgrade

Hi, I'm attemting to upgrade Solaris 2.7 (11/99) to Solaris 8 (1/01) in a Sun E450 and I get an error. Now I describe the steps that I follow:
- I insert the CD 1 of 2 of Solaris Software 8 (1/01) and I do an Stop+A with "boot cdrom"
- We insert geographical region, the network configuration...
- When it asks for an Actual or Initial installation I select Actual and it shows something like this:
" The version of Solaris (solaris 2.7) of the segment c0t0d0s0 can't upgrade.
Fsck can't verify a system archives in the archives system table (vfstab)"
I verified from console that c0t0d0s0 is mounted in "a/" and then I've done an fsck and the partition not contents any errors
This is my "vfstab":
#device          device          mount          FS     fsck     mount     mount
#to mount     to fsck          point          type     pass     at boot     options
#/dev/dsk/c1d0s2 /dev/rdsk/c1d0s2 /usr          ufs     1     yes     -
fd     -     /dev/fd     fd     -     no     -
/proc     -     /proc     proc     -     no     -
/dev/dsk/c0t0d0s1     -     -     swap     -     no     -
/dev/dsk/c0t0d0s0     /dev/rdsk/c0t0d0s0     /     ufs     1     no     logging,noatime
/dev/dsk/c0t0d0s3     /dev/rdsk/c0t0d0s3     /var     ufs     1     no     logging,noatime
/dev/dsk/c0t0d0s4     /dev/rdsk/c0t0d0s4     /opt     ufs     2     yes     logging,noatime
/dev/dsk/c0t0d0s7     /dev/rdsk/c0t0d0s7     /home     vxfs     2     yes     -
/dev/dsk/c6t0d1s0     /dev/rdsk/c6t0d1s0     /programas     vxfs     3     yes     -
swap     -     /tmp     tmpfs     -     yes     -
/dev/dsk/c7t1d0s0 /dev/rdsk/c7t1d0s0 /imagen vxfs 3 yes -
/dev/dsk/c7t1d2s0 /dev/rdsk/c7t1d2s0 /imagen2 vxfs 3 yes -
/dev/dsk/c6t0d3s0 /dev/rdsk/c6t0d3s0 /imagen3 vxfs 3 yes -
and this is the output of an mount command:
proc on /proc read/write/setuid on Wed Jan 28 17:50:21 2004
/ on /dev/dsk/c0t0d0s0 read/write/setuid/logging/noatime/largefiles on Wed Jan 4
/dev/fd on fd read/write/setuid on Wed Jan 28 17:50:21 2004
/var on /dev/dsk/c0t0d0s3 read/write/setuid/logging/noatime/largefiles on Wed J4
/opt on /dev/dsk/c0t0d0s4 read/write/setuid/largefiles/noatime/logging on Wed J4
/home on /dev/dsk/c0t0d0s7 read/write/setuid/log on Wed Jan 28 17:50:24 2004
/programas on /dev/dsk/c6t0d1s0 read/write/setuid/log on Wed Jan 28 17:50:24 204
/tmp on swap read/write/setuid on Wed Jan 28 17:50:24 2004
/imagen on /dev/dsk/c7t1d0s0 read/write/setuid/log on Wed Jan 28 17:50:24 2004
/imagen2 on /dev/dsk/c7t1d2s0 read/write/setuid/log on Wed Jan 28 17:50:24 2004
/imagen3 on /dev/dsk/c6t0d3s0 read/write/setuid/log on Wed Jan 28 17:50:25 2004
and the "mnttab" file:
/proc     /proc     proc     rw,suid,dev=2f80000     1075308621
/dev/dsk/c0t0d0s0     /     ufs     rw,suid,dev=800000,logging,noatime,largefiles     1075308621
fd     /dev/fd     fd     rw,suid,dev=3040000     1075308621
/dev/dsk/c0t0d0s3     /var     ufs     rw,suid,dev=800003,logging,noatime,largefiles     1075308621
/dev/dsk/c0t0d0s4     /opt     ufs     largefiles,noatime,logging,dev=800004     1075308624
/dev/dsk/c0t0d0s7     /home     vxfs     rw,log,suid     1075308624
/dev/dsk/c6t0d1s0     /programas     vxfs     rw,log,suid     1075308624
swap     /tmp     tmpfs     rw,dev=1     1075308624
/dev/dsk/c7t1d0s0     /imagen     vxfs     rw,log,suid     1075308624
/dev/dsk/c7t1d2s0     /imagen2     vxfs     rw,log,suid     1075308624
/dev/dsk/c6t0d3s0     /imagen3     vxfs     rw,log,suid     1075308625
-hosts     /net     autofs     ignore,indirect,nosuid,nobrowse,dev=31c0001     1075308629
-xfn     /xfn     autofs     ignore,indirect,dev=31c0002     1075308629
eusleg01:vold(pid345)     /vol     nfs     ignore,noquota,dev=3180001     1075308639
I've done the same steps with a Sun E250 and the result is Ok , Solaris is upgraded ok to version 8
I need urgent help :(
Thanks

http://supportforum.sun.com/sunos/index.php?t=tree&th=673&start=0&rid=27&S=801555a93d24e3464fa33b5a32e88894

Solaris 10 upgrade - Major problems - PLEASE HELP

I have a v880 running Solaris 10 11/06 (U3) which I am trying to upgrade to the latest Solaris 8/11 (U10) release without much success. I am simply using the Solaris Installation Program and booting from DVD and running the upgrade option. However the process fails with the following error:
The selected Software exceeds the available disk space.
The Solaris operating environment selected to upgrade contains
a metadevice. Disk Space Re-allocation is not allowed.
Please do one of the following:
Deselect some software to be installed.
Choose another version of Solaris to upgrade. (If available)
Go back and select Initial Install.
Exit the installation.<Press ENTER to continue>
A df -h shows that / has 22Gb free, swap space is 3.1Gb free, /var has 7.1Gb free and /usr 3.1Gb free. All the documentation seems to show that that is ample space. I'm relatively new to Unix and Solaris so be gentle with me if I'm doing something stupid. Many thanks to all. An initial install is not an option as I need to keep the setup as is.

It's probably /usr that it's whining about. It's not very exact when calculating space and 3.1G is not ample space these days. This is why I put everything in the / partition anymore. Don't have to worry about this crap down the line...
You might try copying some stuff off of /usr (if you can find it) and try again.

Websphere 4.0.4 crash on heavy load after solaris 8 upgrade

We have our application running on solaris 8 with websphere 4.0.4 and IBM HTTP Server. Things seem to be fine since one year. We recetly went with a solaris patch upgrade (may 2005) applying
108528-29 SunOS 5.8: kernel update and Apache patch
116973-01 SunOS 5.8: Apache Patch
Since than websphere is crashing with the following error
java.lang.OutOfMemoryError: unable to create new native thread
at java.lang.Thread.start(Native Method)
at com.ibm.ws.util.CachedThread.start(ThreadPool.java:126)
at com.ibm.ws.util.CachedThread.wakeup(ThreadPool.java:111)
at com.ibm.ws.util.ThreadPool.getThread(ThreadPool.java:453)
at com.ibm.ejs.oa.pool.ThreadPool.startWorkerThread(ThreadPool.java:83)
at com.ibm.CORBA.iiop.IIOPConnection.processInput(IIOPConnection.java:879)
at com.ibm.CORBA.iiop.StandardReaderThread.run(StandardReaderThread.java:187)
libthread panic: cannot create new lwp : dumping core (PID: 7313 LWP 2)
stacktrace:
ff361c90
0
Not sure if the solaris patch has caused it.
Any suggestions will be very helpful.
Thanks,
Praveen

No specific ideas but take a look at this trouble-shooting article.
You could try trashing the preferences again as described in it and if that does no good it might be worth uninstalling it exactly as described and reinstalling .
http://support.apple.com/kb/TA24195?viewlocale=enUS&locale=enUS

Recommended Solaris package upgrade procedure

Similar Messages

Maybe you are looking for