I have already uploaded the pdf of my presentation "Datenbankupgrades mit Autoupgrade über 2 Nodes" at my Slide Download Center (you can find that on the right side or using this Link). The presentation is in german, but if you have any question, write a comment / contact me.
Showing posts with label DOAG Konferenz. Show all posts
Showing posts with label DOAG Konferenz. Show all posts
DOAG Conference 2020
What crazy times we have. Most conferences are online only, some are also on-premises, but without the necessary hygiene concept to avoid corona (or without enough control), as I have seen at twitter.
The big DOAG conference this year is an online conference AND an on-premises one - but with a big hygiene concept. Roundabout half of the presentations are streamed online, all others can only be seen at Nuremberg.
I was selected as a speaker again, which makes me still proud. This year, I will talk about migrating and upgrading a database using the autoupgrade tool over to nodes. I will show this with a Windows environment - because it's a little bit more to do manually, but I will also cover the things you need to do at Linux (as this means I will just say: you don't need to do that step on Linux 😉). Unfortunately, the presentation will not be streamed online, this means, if you want to see that, you have to go to Nuremberg (and speak german, as my presentation will be in german).
So, stay safe and see you there - hopefully!
Everything goes Auto? First experiences with the Oracle Autoupgrade-Tool - Part 5 - DBA still matters...
As this post is a part of a series of posts regarding Autoupgrade-tool, you might want to start with the first (configuration) or the second post (pre-upgrade analysis).
As every other DBA, I am not error free. Any good administrator is doing things wrong sometimes, this is normal, especially if you do a lot stuff in a very short time and you are very busy.
Well, to be honest, things happened also because I made some mistakes. E.g. at one of my tests (with GRP) I didn't used an administrator command for running the autoupgrade tool in MS Windows. MS Windows was quitting that with an OpenSCManager failed, Access denied error. (Slide 48).
As I tried to use "restore" in the console, Windows again throwed ne the error. What should I do now? I am lost in the middle of nowhere (ok, I could have went back to my VM snapshot, but I wanted to test the tool).
I exited the console (it was warning me that 1 job will stop) (Slide 49) and started the autoupgrade tool again in a administrator command window.
It attached automatically to the old job and I started a restore again. I don't know if there was something still in an undefined state, but first nothing happens.
Starting the "restore" a second time was successful. I have watched the job status saying it's flashing back parallel with the alert.log and after a while it was back on the pre-upgrade state (Slide 50-54).
That's a very robust behaviour, even if the first restore seemed to do nothing (I think it was able to fix a broken state, so the second execution could succeed).
A second error I did was arising while using the "add_after_upgrade_pfile" parameter (Slide 55). I had a stupid spelling mistake in that file. So autoupgrade finished, but the database was not started.
A look at the detailled log file and I have seen that the pfile wasn't copied / changed to spfile due to this error. I then copied the corresponding init.ora to the dbs/database subdirectory, created a spfile out of it and everything was fine. No need to roll back using the GRP...
If you are on MS Windows, one last thing happened while creating the Windows service in the new Oracle Home with oradim. I was lazy and just copied the whole command from the preupgrade.html.
The DIM-00003 error I got doing this is misleading.
There is no missing parameter - you can't specify the spfiles name after the "-spfile" switch (Slide 57).
My personal summary (Slide 58):
Always, always, always download the newest version of the autoupgrade-tool. As it evolves and improves fast, check again often, if there is a new version of it.
Well, "everything automatic" is something relative. You can still run in some hassles (as I did on MS Windows), you maybe try things which are non-production-use and so on - not only at the autoupgrade-tool things can happen, but also at the database level or because you as administrator are doing something wrong (like the most of my examples show). By the way, there was a whole session at the DOAG conference what can happen (from a database perspective) during an upgrade (and how you can find/fix the errors).
The tool is hands free, yes, but personally I would prefer to keep an eye on it while using it in production the first times. As always: If you test your configuration well in your environment and everything works there, then it works also in production! This is, where I used the VM snapshots in my test environments a lot to test as much as I can in a short timeframe.
What I was really impressed about is, how stable and robust a tool is, that is not that long on the market, how good resume and restore are working and, again, how detailled the logs are.
As we do have a lot customers at Switzerland running Windows with (Swiss) German or (Swiss) French language, we need to test the newer versions of the autoupgrade tool.
Linux environments of our customers are all in english and also don't have the MS Windows service problems, so it's easier for us to use the tool there. If you have more than 1 or 2 databases and these are similiar (especially the environment), it does really make sense to step into using the tool.
Hope you enjoyed the series. I will make some more tests next year. By the way, Mike Dietrich told at his session, that autoupgrade.jar will be the tool for upgrading exadata databases in the future!
As every other DBA, I am not error free. Any good administrator is doing things wrong sometimes, this is normal, especially if you do a lot stuff in a very short time and you are very busy.
Well, to be honest, things happened also because I made some mistakes. E.g. at one of my tests (with GRP) I didn't used an administrator command for running the autoupgrade tool in MS Windows. MS Windows was quitting that with an OpenSCManager failed, Access denied error. (Slide 48).
As I tried to use "restore" in the console, Windows again throwed ne the error. What should I do now? I am lost in the middle of nowhere (ok, I could have went back to my VM snapshot, but I wanted to test the tool).
I exited the console (it was warning me that 1 job will stop) (Slide 49) and started the autoupgrade tool again in a administrator command window.
It attached automatically to the old job and I started a restore again. I don't know if there was something still in an undefined state, but first nothing happens.
Starting the "restore" a second time was successful. I have watched the job status saying it's flashing back parallel with the alert.log and after a while it was back on the pre-upgrade state (Slide 50-54).
That's a very robust behaviour, even if the first restore seemed to do nothing (I think it was able to fix a broken state, so the second execution could succeed).
A second error I did was arising while using the "add_after_upgrade_pfile" parameter (Slide 55). I had a stupid spelling mistake in that file. So autoupgrade finished, but the database was not started.
A look at the detailled log file and I have seen that the pfile wasn't copied / changed to spfile due to this error. I then copied the corresponding init.ora to the dbs/database subdirectory, created a spfile out of it and everything was fine. No need to roll back using the GRP...
If you are on MS Windows, one last thing happened while creating the Windows service in the new Oracle Home with oradim. I was lazy and just copied the whole command from the preupgrade.html.
The DIM-00003 error I got doing this is misleading.
There is no missing parameter - you can't specify the spfiles name after the "-spfile" switch (Slide 57).
My personal summary (Slide 58):
Always, always, always download the newest version of the autoupgrade-tool. As it evolves and improves fast, check again often, if there is a new version of it.
Well, "everything automatic" is something relative. You can still run in some hassles (as I did on MS Windows), you maybe try things which are non-production-use and so on - not only at the autoupgrade-tool things can happen, but also at the database level or because you as administrator are doing something wrong (like the most of my examples show). By the way, there was a whole session at the DOAG conference what can happen (from a database perspective) during an upgrade (and how you can find/fix the errors).
The tool is hands free, yes, but personally I would prefer to keep an eye on it while using it in production the first times. As always: If you test your configuration well in your environment and everything works there, then it works also in production! This is, where I used the VM snapshots in my test environments a lot to test as much as I can in a short timeframe.
What I was really impressed about is, how stable and robust a tool is, that is not that long on the market, how good resume and restore are working and, again, how detailled the logs are.
As we do have a lot customers at Switzerland running Windows with (Swiss) German or (Swiss) French language, we need to test the newer versions of the autoupgrade tool.
Linux environments of our customers are all in english and also don't have the MS Windows service problems, so it's easier for us to use the tool there. If you have more than 1 or 2 databases and these are similiar (especially the environment), it does really make sense to step into using the tool.
Hope you enjoyed the series. I will make some more tests next year. By the way, Mike Dietrich told at his session, that autoupgrade.jar will be the tool for upgrading exadata databases in the future!
Everything goes Auto? First experiences with the Oracle Autoupgrade-Tool - Part 4 - Non-CDB to PDB conversion after upgrade
As this post is a part of a series of posts regarding Autoupgrade-tool, you might want to start with the first (configuration) or the second post (pre-upgrade analysis).
To convert a non-CDB to a PDB automatically after the upgrade, you can set "target_cdb" (first create an empty CDB or use an already existing one) in your configuration file.
While testing that, I was getting an error that the guaranteed restore point couldn't be dropped. But it wasn't existing, as I have set "restoration=no" in my configuration (Slide 42, 43). Well, I opened another SR that this step fails on Windows (on Linux I only had already CDBs, therefore I couldn't test that there). Mike Dietrich than told me that this feature is shown on his blog, but it does not exists in the documentation. As it is not official released yet, it should show the future of the tool.
As Mike is right, this is not a scenario for a SR, I closed mine immediately (Slide 44), for sure, there is no issue with a non-production-use of the tool. Only 2 or 3 days (!) later (now speaking with the development directly) I got a newer pre-version from Mike personally! So I was able to go forward with my tests.
The autoupgrade.jar tool then stepped over the GRP problem but was running into a plug-in violation (Slides 45/46). The, again, very good logfiles recorded an error, that the description.xml could not be read. The description.xml file did exist, so was it another non-production-use thing?
Development helped me analysing it and so we found out together that the problem seems to be related to the database version 19c (on Windows). If we put the description.xml into ORACLE_HOME/database the file can be found, but not at it's original place (Slide 47).
While the one thing is database bound, the fixes for the GRP will for sure be seen in the December or January release of the tool.
Edit (December, 2nd): What happened after the DOAG conference: With development feedback and some long testing runs over the weekend I have figured out, where the problem starts (it's a Windows/Oracle Database rights issue). I am waiting for the developers assessment on that, maybe I write another, different post, which is only a little bit related to autoupgrade.jar as the problem may hit you at every MS Windows environment.
Edit (May 14th, 2020): The newest version of autoupgrade.jar does now support Non-CDB to PDB conversion officially. I have not tested that yet.
The next post for my DOAG session can be found here.
To convert a non-CDB to a PDB automatically after the upgrade, you can set "target_cdb" (first create an empty CDB or use an already existing one) in your configuration file.
While testing that, I was getting an error that the guaranteed restore point couldn't be dropped. But it wasn't existing, as I have set "restoration=no" in my configuration (Slide 42, 43). Well, I opened another SR that this step fails on Windows (on Linux I only had already CDBs, therefore I couldn't test that there). Mike Dietrich than told me that this feature is shown on his blog, but it does not exists in the documentation. As it is not official released yet, it should show the future of the tool.
As Mike is right, this is not a scenario for a SR, I closed mine immediately (Slide 44), for sure, there is no issue with a non-production-use of the tool. Only 2 or 3 days (!) later (now speaking with the development directly) I got a newer pre-version from Mike personally! So I was able to go forward with my tests.
The autoupgrade.jar tool then stepped over the GRP problem but was running into a plug-in violation (Slides 45/46). The, again, very good logfiles recorded an error, that the description.xml could not be read. The description.xml file did exist, so was it another non-production-use thing?
Development helped me analysing it and so we found out together that the problem seems to be related to the database version 19c (on Windows). If we put the description.xml into ORACLE_HOME/database the file can be found, but not at it's original place (Slide 47).
While the one thing is database bound, the fixes for the GRP will for sure be seen in the December or January release of the tool.
Edit (December, 2nd): What happened after the DOAG conference: With development feedback and some long testing runs over the weekend I have figured out, where the problem starts (it's a Windows/Oracle Database rights issue). I am waiting for the developers assessment on that, maybe I write another, different post, which is only a little bit related to autoupgrade.jar as the problem may hit you at every MS Windows environment.
Edit (May 14th, 2020): The newest version of autoupgrade.jar does now support Non-CDB to PDB conversion officially. I have not tested that yet.
The next post for my DOAG session can be found here.
Everything goes Auto? First experiences with the Oracle Autoupgrade-Tool - Part 3 - Deploy mode
As this post is a part of a series of posts regarding Autoupgrade-tool, you might want to start with the first (configuration) or the second post (pre-upgrade analysis).
Most of the tests I did in the deploy mode, because I used snapshots (instead of backups) a lot while testing, so it was (also with SE2) easy to revert to the last state. Snapshots have been my best friends as I didn't want to wait for the restore/recovery...
As I said, I was sometimes a little bit confused that with "lsj" e.g. status showed "finished" while message said "starting" (Slide 26), so I preferred the detailed
status -job <jobno> to look at the status of the job.
There are a lot details you can see, e.g. the operation, what is already done, how many stages are pending, etc., the status command at a database with CDB architecture also shows all PDBs (including seed) which are in touch by the tool (Slide 27 and Slide 28).
As the first tests without archivelog mode succeeded, I tried to use "restoration=YES" in my configuration. This means, you use a guaranteed restore point (GRP, Enterprise Edition only) where you can roll back the database to, if something goes wrong.
For using GRP you need a Fast Recovery Area (FRA). Most of our customers don't use FRA by default, so this is something we need to introduce in our case, exclusively and only for the time of the upgrade.
Testing with GRP, after some minutes, I have seen with another "status" command, that there is a warning. The recovery area is nearly full and the upgrade could "appear hung" if the recovery area is filled up (Slide 29).
To find out, what happens, if the FRA is full, I started later on another autoupgrade run with a far too small FRA (Slide 30).
Thankfully, the tool did not hung, but throwed an unexpected exception error. In the very detailed logs (it's more a debug trace than a normal logfile) you can easily find the error: "MIN_RECOVERY_AREA_SIZE" is to small.
Fixing that can be done online with "alter system", but only with scope=MEMORY. The database is restarted a couple of times with a pfile while it runs through the deploy mode. (Slide 31)
You can either wait for this error happen more than once or, what I did, change the different pfiles, e.g. "during_upgrade_pfile_catctl.ora" (which is located at the dbupgrade subdirectory of your job) in addition. After resuming the job (Slide 33), the autoupgrade finished fine (Slides 34, 35, 38).
As the size of the FRA can be critical, it's good to set it big enough before you begin with the deploy mode. You could, e.g. use the "before_action" parameter in your configuration to change the FRA size or you can use add_during_upgrade_pfile, for example. (Slide 32)
While the tool is running, you can see with the status command that it's e.g. upgrading CDB$ROOT, upgrading PDB$SEED or your user PDBs (in parallel) (Slides 34/35).
There is also a "tasks" command (Slide 36) where you can see all the tasks that are running or waiting, but I didn't see anything there I think I would need at my daily work.
If you run the autoupgrade.jar on MS Windows, a temporary service in the target Oracle Home is created (Slide 37), dropped afterwards and you have to drop the old original service and create the new one manually, that's fine.
When the upgrade has finished, don't forget to drop the GRP by yourself, otherwise the FRA will be filled up completely (you could set a parameter in your configuration to do this automatically, but I prefer to test the upgraded environment first) (Slide 38).
On MS Windows I had an unexpected error again on my german environment as stopping the service for the database to be upgraded was successful, but the tool was raising an error (Slide 40). I have opened another SR for that and I think it will be fixed soon (there is a bug open for that). After installing the US language pack and changing the Oracle user account to "US" language, everything worked fine.
The next post will be about the Non-CDB to PDB conversion with the autoupgrade tool.
Most of the tests I did in the deploy mode, because I used snapshots (instead of backups) a lot while testing, so it was (also with SE2) easy to revert to the last state. Snapshots have been my best friends as I didn't want to wait for the restore/recovery...
As I said, I was sometimes a little bit confused that with "lsj" e.g. status showed "finished" while message said "starting" (Slide 26), so I preferred the detailed
status -job <jobno> to look at the status of the job.
There are a lot details you can see, e.g. the operation, what is already done, how many stages are pending, etc., the status command at a database with CDB architecture also shows all PDBs (including seed) which are in touch by the tool (Slide 27 and Slide 28).
As the first tests without archivelog mode succeeded, I tried to use "restoration=YES" in my configuration. This means, you use a guaranteed restore point (GRP, Enterprise Edition only) where you can roll back the database to, if something goes wrong.
For using GRP you need a Fast Recovery Area (FRA). Most of our customers don't use FRA by default, so this is something we need to introduce in our case, exclusively and only for the time of the upgrade.
Testing with GRP, after some minutes, I have seen with another "status" command, that there is a warning. The recovery area is nearly full and the upgrade could "appear hung" if the recovery area is filled up (Slide 29).
To find out, what happens, if the FRA is full, I started later on another autoupgrade run with a far too small FRA (Slide 30).
Thankfully, the tool did not hung, but throwed an unexpected exception error. In the very detailed logs (it's more a debug trace than a normal logfile) you can easily find the error: "MIN_RECOVERY_AREA_SIZE" is to small.
Fixing that can be done online with "alter system", but only with scope=MEMORY. The database is restarted a couple of times with a pfile while it runs through the deploy mode. (Slide 31)
You can either wait for this error happen more than once or, what I did, change the different pfiles, e.g. "during_upgrade_pfile_catctl.ora" (which is located at the dbupgrade subdirectory of your job) in addition. After resuming the job (Slide 33), the autoupgrade finished fine (Slides 34, 35, 38).
As the size of the FRA can be critical, it's good to set it big enough before you begin with the deploy mode. You could, e.g. use the "before_action" parameter in your configuration to change the FRA size or you can use add_during_upgrade_pfile, for example. (Slide 32)
While the tool is running, you can see with the status command that it's e.g. upgrading CDB$ROOT, upgrading PDB$SEED or your user PDBs (in parallel) (Slides 34/35).
There is also a "tasks" command (Slide 36) where you can see all the tasks that are running or waiting, but I didn't see anything there I think I would need at my daily work.
If you run the autoupgrade.jar on MS Windows, a temporary service in the target Oracle Home is created (Slide 37), dropped afterwards and you have to drop the old original service and create the new one manually, that's fine.
When the upgrade has finished, don't forget to drop the GRP by yourself, otherwise the FRA will be filled up completely (you could set a parameter in your configuration to do this automatically, but I prefer to test the upgraded environment first) (Slide 38).
On MS Windows I had an unexpected error again on my german environment as stopping the service for the database to be upgraded was successful, but the tool was raising an error (Slide 40). I have opened another SR for that and I think it will be fixed soon (there is a bug open for that). After installing the US language pack and changing the Oracle user account to "US" language, everything worked fine.
The next post will be about the Non-CDB to PDB conversion with the autoupgrade tool.
Everything goes Auto? First experiences with the Oracle Autoupgrade-Tool - Part 2 - Pre-Upgrade-Analysis
This is the second post to the first experiences with the Autoupgrade tool. Please start with the first one.
You start the pre-grade analysis with the parameters -config <yourconfigfile> -mode analyze in console mode (Slide 20).
With "lsj" you can see the job number and some information, like stage, operation, status and message. As I was not able to understand all the different status combinations at the output, I started to prefer to select only the job number and issue a
"status -job <jobno>" to get the detailed information regarding a job.
The analysis job finishes after some minutes (1 to 3 depending on my virtualbox on the laptop or the VMWare server) (Slide 21).
At your autoupgrade log directory, you then get a number of files, the two I want to mention here are the preupgrade.html file and the checklist.cfg file (Slide 22).
As not all servers do have a web browser installed (mostly on linux there isn't a GUI), you really should copy the html file to your desktop!
There is soooo many information in this html and you should treat all messages serious, independend if they are "errors" or only "warnings", "recommended" or "infos" (Slides 23-25).
What you have to look at is what is the topic that raises you a message and if there is a fixup available. Fixup available yes means, the autoupgrade tool can correct this topic before (or after) the database is/was upgraded automatically by itself.
For our "customer-like" environment we had e.g. things in the dba recyclebin. This can be fixed automatically, but as DBA I like to prefer to have a look into the recycle bin first.Maybe there is something in there we already need?
Another topic found was 10G password version users are existing - they can't connect to the 19c database later on! Will be really bad if you don't fix that by yourself.
The "streams configuration" found in the database wasn't used, so I dropped that manually (the html shows you the DBMS-Packages you need as output).
Again, check all the things you see at the html carefully! Then decide to let it be fixed automatically (if there is a fixup available), to change the configuration or to ignore the issue (like ignoring some invalid procedures in our application schema depending on the type of software in use).
If you want to ignore a fix, you can tweak the checklist.cfg file by changing runfix from YES to NO. Keep in mind to NOT run the autoupgrade tool afterwards with the "deploy" parameter. You need to run the fixup and the upgrade modes manually.
Deploy creates a new pre-analysis, fixes things and upgrades the database in another job and does not know of things you may have changed in the older checklist.cfg which belongs to an old, finished job.
Another topic, we often see: Some of us do have underscore parameters at the spfile and the tool detects this and says, there is no fixup available. This is true, but there is a workaround.
If you want to get rid of all underscore parameters you can set a global parameter (remove_underscore_parameters=yes) and you can add the once you really need later with the add_to_pfile-files you can specify. Otherwise you can ignore this and proceed with the fixups or with the deploy mode (the underscore parameter are still in your spfile after the upgrade).
You start the pre-grade analysis with the parameters -config <yourconfigfile> -mode analyze in console mode (Slide 20).
With "lsj" you can see the job number and some information, like stage, operation, status and message. As I was not able to understand all the different status combinations at the output, I started to prefer to select only the job number and issue a
"status -job <jobno>" to get the detailed information regarding a job.
The analysis job finishes after some minutes (1 to 3 depending on my virtualbox on the laptop or the VMWare server) (Slide 21).
At your autoupgrade log directory, you then get a number of files, the two I want to mention here are the preupgrade.html file and the checklist.cfg file (Slide 22).
As not all servers do have a web browser installed (mostly on linux there isn't a GUI), you really should copy the html file to your desktop!
There is soooo many information in this html and you should treat all messages serious, independend if they are "errors" or only "warnings", "recommended" or "infos" (Slides 23-25).
What you have to look at is what is the topic that raises you a message and if there is a fixup available. Fixup available yes means, the autoupgrade tool can correct this topic before (or after) the database is/was upgraded automatically by itself.
For our "customer-like" environment we had e.g. things in the dba recyclebin. This can be fixed automatically, but as DBA I like to prefer to have a look into the recycle bin first.Maybe there is something in there we already need?
Another topic found was 10G password version users are existing - they can't connect to the 19c database later on! Will be really bad if you don't fix that by yourself.
The "streams configuration" found in the database wasn't used, so I dropped that manually (the html shows you the DBMS-Packages you need as output).
Again, check all the things you see at the html carefully! Then decide to let it be fixed automatically (if there is a fixup available), to change the configuration or to ignore the issue (like ignoring some invalid procedures in our application schema depending on the type of software in use).
If you want to ignore a fix, you can tweak the checklist.cfg file by changing runfix from YES to NO. Keep in mind to NOT run the autoupgrade tool afterwards with the "deploy" parameter. You need to run the fixup and the upgrade modes manually.
Deploy creates a new pre-analysis, fixes things and upgrades the database in another job and does not know of things you may have changed in the older checklist.cfg which belongs to an old, finished job.
Another topic, we often see: Some of us do have underscore parameters at the spfile and the tool detects this and says, there is no fixup available. This is true, but there is a workaround.
If you want to get rid of all underscore parameters you can set a global parameter (remove_underscore_parameters=yes) and you can add the once you really need later with the add_to_pfile-files you can specify. Otherwise you can ignore this and proceed with the fixups or with the deploy mode (the underscore parameter are still in your spfile after the upgrade).
Everything goes Auto? First experiences with the Oracle Autoupgrade-Tool - Part 1 - introduction and configuration
At the DOAG conference 2019 I had a presentation (due to the filled up room it was repeated the day after) regarding the Oracle Autoupgrade tool. The slides are in german and can be found at my Slide Download Center. As most slides are screenshots only, I will write in this series of posts a little bit about this new tool.
The first part will now cover an introduction and how to setup the configuration file.
Introduction:
After OOW 2018 I have heard the first time of the Autoupgrade Tool and I was very curious. I have downloaded the slides and what I have found was a thing I thought it should be there since years.
A tool which helps you upgrading your databases from one release to another without running all these scripts, pre-installation.jars and DBUA manually. With less hands on and with a lot of things automated. The more I have read the more curious I got.
Later on, I have seen autoupgrade as part of the documentation of Oracle 19c, but I wasn't really able to see how it could work and I also had no time to step into it in detail.
Then, Mid of June 2019, I have seen Mike Dietrichs (Mister Upgrade Guru) blog post about how to create a sample configuration file and with how less parameters an upgrade could be started.
So I decided (after a short conversation with Mike on twitter, June 13th) to test the tool and to send a proposal for a presentation at the DOAG conference. I already had installed a 19c Database software on my laptop with Windows 10, so I decided to make the first small test there.
Unfortunately, I've got a java exception at running autoupgrade.jar with a -version flag. Also the downloaded version from support.oracle.com had this problem. (Slide 11) It's so typical, that it hits me...😳
So I was really sorry to inform Mike that I couldn't test that due to (in my opinion) a problem with the german environment I had. I opened a SR to pass all the logs to the Oracle Support and Mikes Team.
July 1st (! I started testing mid of June! - Slide 13) I've got an email from Mike that the bug should be fixed in the version they have uploaded some days before. The fix included a "fallback" to default language (english) if the OS language is not supported. I downloaded that version and yes, it was working now. (Slide 15)
I asked my colleague to set up a "Swiss customer similar database environment" on a Windows 2012 server and I prepared a virtualbox with Oracle Linux and some CDBs.
As I was busy over the summer I only were able to test things in my spare time mostly. So I've started with reading the documentation and Mikes blog.
The more I read, the more I have seen some targets, that my employer could reach by using this tool (Slide 7):
As an ISV and System Integrator, we have a lot of databases to upgrade (most of them from 12.1 to 19c) next year. Some on Linux, some on Windows, some on ODA (but ODA is out of scope here).
The database editions are Standard Edition 2 and Enterprise Edition and we have all kind of licenses (Embedded, Application Specific and Full Use).
Main target for us is to get rid of any manual work as this is most error-prone, e.g. changing SPFile parameters, switching from Non-CDB to the new CDB architecture, maybe combining application upgrades with database upgrades (scripted), etc..
After the virtualbox and the Windows environments (on VMWare) have been installed, I was ready to start with the tests.
Preparation:
As a lot servers don't have (the right) java (version) installed, it's best to use the java from the ORACLE_HOME/jdk/bin directory of 18c or 19c (Slide 15). This works fine. Also, as release cycles of the autoupgrade tool are very short, you should download always the newest autoupgrade.jar from support.oracle.com. Even if you have checked the note a week ago, check again, before you start your work.
With that in mind, you can run the tool with the -create_sample_file config parameter (by the way, on my german environment it says unsupported language, defaulting to English) (Slide 16).
It's very easy to use the sample config file (Slide 17) to create your own configuration. A small thing that can be found on Windows is the global log directory misses "backslashes". This does not work out of the box but is easy to fix.On Linux, the default log directory is fine and points to /home/oracle. (Slide 18)
Next to do is to specify (Slide 19) e.g. the DB_UNIQUE_ID as "DBNAME", the "SID", "SOURCE HOME" and "TARGET HOME", "TARGET_VERSION" of the database and, if you wish so, even more (see the documentation for all parameters). The "START_TIME" parameter can be set to 'now' or a date/time - the format can be found in the documentation (no NLS_LANG setting). With "add_after_upgrade_pfile" (or delete or before or during instead of after) you can change automatically the SPFILE content.
One of the things I really like is to change is the tag "UPG1." to a tag that gives more information (e.g. EDM for our Energy Data Management DB, DAPHNE for the museums DB, etc.) (Slide 19).
So you can create one config file for all kind of databases and they are pretty named. Running "UTLRP" for recompilation is a nice thing, also have a look at the parameter for "TIMEZONE_UPG"rades as it is often forgotten in manual upgrades. You should also know the "RESTORATION" parameter, more information on this parameter will follow later.
If you have setup the config file, you are ready to start to work with your database(s).
The next step is the Pre-Upgrade analysis (next post).
The first part will now cover an introduction and how to setup the configuration file.
Introduction:
After OOW 2018 I have heard the first time of the Autoupgrade Tool and I was very curious. I have downloaded the slides and what I have found was a thing I thought it should be there since years.
A tool which helps you upgrading your databases from one release to another without running all these scripts, pre-installation.jars and DBUA manually. With less hands on and with a lot of things automated. The more I have read the more curious I got.
Later on, I have seen autoupgrade as part of the documentation of Oracle 19c, but I wasn't really able to see how it could work and I also had no time to step into it in detail.
Then, Mid of June 2019, I have seen Mike Dietrichs (Mister Upgrade Guru) blog post about how to create a sample configuration file and with how less parameters an upgrade could be started.
So I decided (after a short conversation with Mike on twitter, June 13th) to test the tool and to send a proposal for a presentation at the DOAG conference. I already had installed a 19c Database software on my laptop with Windows 10, so I decided to make the first small test there.
Unfortunately, I've got a java exception at running autoupgrade.jar with a -version flag. Also the downloaded version from support.oracle.com had this problem. (Slide 11) It's so typical, that it hits me...😳
So I was really sorry to inform Mike that I couldn't test that due to (in my opinion) a problem with the german environment I had. I opened a SR to pass all the logs to the Oracle Support and Mikes Team.
July 1st (! I started testing mid of June! - Slide 13) I've got an email from Mike that the bug should be fixed in the version they have uploaded some days before. The fix included a "fallback" to default language (english) if the OS language is not supported. I downloaded that version and yes, it was working now. (Slide 15)
I asked my colleague to set up a "Swiss customer similar database environment" on a Windows 2012 server and I prepared a virtualbox with Oracle Linux and some CDBs.
As I was busy over the summer I only were able to test things in my spare time mostly. So I've started with reading the documentation and Mikes blog.
The more I read, the more I have seen some targets, that my employer could reach by using this tool (Slide 7):
As an ISV and System Integrator, we have a lot of databases to upgrade (most of them from 12.1 to 19c) next year. Some on Linux, some on Windows, some on ODA (but ODA is out of scope here).
The database editions are Standard Edition 2 and Enterprise Edition and we have all kind of licenses (Embedded, Application Specific and Full Use).
Main target for us is to get rid of any manual work as this is most error-prone, e.g. changing SPFile parameters, switching from Non-CDB to the new CDB architecture, maybe combining application upgrades with database upgrades (scripted), etc..
After the virtualbox and the Windows environments (on VMWare) have been installed, I was ready to start with the tests.
Preparation:
As a lot servers don't have (the right) java (version) installed, it's best to use the java from the ORACLE_HOME/jdk/bin directory of 18c or 19c (Slide 15). This works fine. Also, as release cycles of the autoupgrade tool are very short, you should download always the newest autoupgrade.jar from support.oracle.com. Even if you have checked the note a week ago, check again, before you start your work.
With that in mind, you can run the tool with the -create_sample_file config parameter (by the way, on my german environment it says unsupported language, defaulting to English) (Slide 16).
It's very easy to use the sample config file (Slide 17) to create your own configuration. A small thing that can be found on Windows is the global log directory misses "backslashes". This does not work out of the box but is easy to fix.On Linux, the default log directory is fine and points to /home/oracle. (Slide 18)
Next to do is to specify (Slide 19) e.g. the DB_UNIQUE_ID as "DBNAME", the "SID", "SOURCE HOME" and "TARGET HOME", "TARGET_VERSION" of the database and, if you wish so, even more (see the documentation for all parameters). The "START_TIME" parameter can be set to 'now' or a date/time - the format can be found in the documentation (no NLS_LANG setting). With "add_after_upgrade_pfile" (or delete or before or during instead of after) you can change automatically the SPFILE content.
One of the things I really like is to change is the tag "UPG1." to a tag that gives more information (e.g. EDM for our Energy Data Management DB, DAPHNE for the museums DB, etc.) (Slide 19).
So you can create one config file for all kind of databases and they are pretty named. Running "UTLRP" for recompilation is a nice thing, also have a look at the parameter for "TIMEZONE_UPG"rades as it is often forgotten in manual upgrades. You should also know the "RESTORATION" parameter, more information on this parameter will follow later.
If you have setup the config file, you are ready to start to work with your database(s).
The next step is the Pre-Upgrade analysis (next post).
Single Tenant ist unnütze Komplexität? / Single Tenant is only unnecessary complexity? - Relocate a PDB on-the-fly
Two parts of my three part blog posts are already available, so this is the third and (at the moment) last post about Single Tenancy and what you can do easily with pluggable databaseses using database links.
What you do need to use the scripts as they are can be found at my first blog entry.
Case 3 - Relocate a PDB on the fly:
There are different use cases for relocating a pdb on the fly, the most common are to move a pdb to another hardware or to the/another cloud infrastructure.
As long as the "copy" is not opened read-write, the old PDB stays where it is. So you can, e.g., create a remote_listener entry for the pdb at the new environment and let it point to the listener at the environment you want to move from. If you do so, the remote listener (at the original location) will work as some kind of proxy to the listener at the new environment. The new PDB then can be used, e.g. by an application server, without any changes at the tns/jdbc connect. Be aware, that the service name must be the same on both sides for the relocated PDB and that you need to change the application connect to the new location somewhere later. If you forget this and you start decommissioning the old environment, the application and users can not work with the moved pdb.
How do you have moved a database with the "good old times" tools? Export and import? Well, this means you get a nice, reorganised database, but this isn't a fast approach if you look at downtime. Creating a data guard environment or using any replication tools is sometimes not possible due to license issues or costs (Standard Edition, Golden Gate, Quest Shareplex, ...). Maybe you can clone the database using rman and recover the new environment manually, but this is a lot work and you may make a small mistake somewhere and the environment isn't recoverable. So the easiest and last chance is to shutdown the production and move everything (using compressing tools or maybe not) to the new environment. And the copy takes time and all the time the database is down...
What would you say to the following solution in short:
- Create a database link between two databases
- Enter one statement to move the database in the backgroud and let it be mounted (so you can work with the original production).
- Open the database at the new location (which means it is a last time synchronized with the original PDB and after that the original PDB is deleted automatically)
Sounds easys? IT IS that easy!
What are the steps you need to do in detail?
Again one last comment: I didn't tested the scripts as I have thrown things out. If something does not work, please come back and leave a comment, so I can fix it.
Thanks a lot!
What you do need to use the scripts as they are can be found at my first blog entry.
Case 3 - Relocate a PDB on the fly:
There are different use cases for relocating a pdb on the fly, the most common are to move a pdb to another hardware or to the/another cloud infrastructure.
As long as the "copy" is not opened read-write, the old PDB stays where it is. So you can, e.g., create a remote_listener entry for the pdb at the new environment and let it point to the listener at the environment you want to move from. If you do so, the remote listener (at the original location) will work as some kind of proxy to the listener at the new environment. The new PDB then can be used, e.g. by an application server, without any changes at the tns/jdbc connect. Be aware, that the service name must be the same on both sides for the relocated PDB and that you need to change the application connect to the new location somewhere later. If you forget this and you start decommissioning the old environment, the application and users can not work with the moved pdb.
How do you have moved a database with the "good old times" tools? Export and import? Well, this means you get a nice, reorganised database, but this isn't a fast approach if you look at downtime. Creating a data guard environment or using any replication tools is sometimes not possible due to license issues or costs (Standard Edition, Golden Gate, Quest Shareplex, ...). Maybe you can clone the database using rman and recover the new environment manually, but this is a lot work and you may make a small mistake somewhere and the environment isn't recoverable. So the easiest and last chance is to shutdown the production and move everything (using compressing tools or maybe not) to the new environment. And the copy takes time and all the time the database is down...
What would you say to the following solution in short:
- Create a database link between two databases
- Enter one statement to move the database in the backgroud and let it be mounted (so you can work with the original production).
- Open the database at the new location (which means it is a last time synchronized with the original PDB and after that the original PDB is deleted automatically)
Sounds easys? IT IS that easy!
What are the steps you need to do in detail?
- Create a common user and grant him rights (SYSOPER!), (let) open the PDB you want to move to another host/to the cloud
- Create a database link to the CDB$ROOT (!) you want to move from the CDB$ROOT you want the PDB moved to.
- Create the pdb with RELOCATE at the second CDB.
- Open the new PDB read write (this step synchronizes the PDBs and drops the old one afterwards).
Again one last comment: I didn't tested the scripts as I have thrown things out. If something does not work, please come back and leave a comment, so I can fix it.
Thanks a lot!
Single Tenant ist unnütze Komplexität? / Single Tenant is only unnecessary complexity? - Remote snapshot clone
Some people may wait for the second part of my three blog posts I am writing about Single Tenancy and why (some / a lot / most) people do think it is unnecessary complexity or - even worser - it is unnecessary at all.
What you do need to use the scripts as they are can be found at my first blog entry.
Case 2 - Remote snapshot clone of a PDB:
Snapshoting databases using a database link does have a lot of use cases.
The following two of them I do like most:
"Fast Provisioning", e.g. of nightly builds to testers or developers and application upgrades.
For remote snapshot pdb clones you either use file systems with snapshot technology included (like ACFS or ZFS) or you use traditional file systems (then you need to set CLONE_DB=true at your spfile and restart the database).
Well, how do you provision a nightly build today in 5 minutes to your testers and developers? On 40 machines? RMAN clone? Datapump export and import? What about the disk space you need for that? Like to spend money on it? Then tell me which shares I should buy before you throw the money out of the window. 😉
Or how do you make an application upgrade? With a non CDB database you can use offline backup the database, online backup but note down the last SCN, configure flashback database, ... And how long does it need to return to the before-upgrade-image if something goes wrong with the application upgrade?
Guys, years ago I was an app consultant and I had to go back with an installation of an application three times (at the same customer), because the business people did find some issues they haven't seen (tested) on the test system before. It took me hours and a lot nerves to restore everything to the right point-in-time and to setup the standby environments later on again...And if you do this three times in 4 months... 😟
So, the trick you will use for all of these uses cases is in the future - remote snapshot clone. Fast, reliable, you don't need to touch the "production", because all changes are done in your "new" snapshoted PDB.
What are the steps you need to do?
Script 2 to run at the target CDB$ROOT container where the snapshot clone should be created
Some notes to point 6: If you have dropped the snapshot clone PDB the source PDB can't be open read write. The error message you get is not telling you the truth, because the PDB files are on OS level also set to READ. So what you do need to do is to change the permission on OS level for the database files. So the right command will look like
sudo chmod 640 /u01/app/oracle/oradata/XFEE1/77F0ADE89C820516E0534738A8C0802B/datafile/*
After you have made these changes, the source PDB can be opened read write again. It's part of the script 2, but as the folder names are different on different databases, I have made a comment out of it.
One last comment: I didn't tested the scripts as I have thrown things out. If something does not work, please come back and leave a comment, so I can fix it.
Thanks a lot!
By the way - happy New Year 2019 everyone!
What you do need to use the scripts as they are can be found at my first blog entry.
Case 2 - Remote snapshot clone of a PDB:
Snapshoting databases using a database link does have a lot of use cases.
The following two of them I do like most:
"Fast Provisioning", e.g. of nightly builds to testers or developers and application upgrades.
For remote snapshot pdb clones you either use file systems with snapshot technology included (like ACFS or ZFS) or you use traditional file systems (then you need to set CLONE_DB=true at your spfile and restart the database).
Well, how do you provision a nightly build today in 5 minutes to your testers and developers? On 40 machines? RMAN clone? Datapump export and import? What about the disk space you need for that? Like to spend money on it? Then tell me which shares I should buy before you throw the money out of the window. 😉
Or how do you make an application upgrade? With a non CDB database you can use offline backup the database, online backup but note down the last SCN, configure flashback database, ... And how long does it need to return to the before-upgrade-image if something goes wrong with the application upgrade?
Guys, years ago I was an app consultant and I had to go back with an installation of an application three times (at the same customer), because the business people did find some issues they haven't seen (tested) on the test system before. It took me hours and a lot nerves to restore everything to the right point-in-time and to setup the standby environments later on again...And if you do this three times in 4 months... 😟
So, the trick you will use for all of these uses cases is in the future - remote snapshot clone. Fast, reliable, you don't need to touch the "production", because all changes are done in your "new" snapshoted PDB.
What are the steps you need to do?
- Create a common user and grant him rights, open the PDB you want to make a snapshot from read only (if you don't use a snapshot capable file system);
- Create a database link to the PDB you want to clone from the CDB$ROOT you want the PDB cloned to.
- Create the snapshot clone remote at the second CDB.
- Open the new PDB read write
- Make your application upgrade or develop or test
- if it is sucessful / if you need to keep the PDB - "alter pluggable database <xyz> materialize;"
- if it is not successful or before you create another nightly build - drop the new PDB and open the original source PDB read write again
Script 2 to run at the target CDB$ROOT container where the snapshot clone should be created
Some notes to point 6: If you have dropped the snapshot clone PDB the source PDB can't be open read write. The error message you get is not telling you the truth, because the PDB files are on OS level also set to READ. So what you do need to do is to change the permission on OS level for the database files. So the right command will look like
sudo chmod 640 /u01/app/oracle/oradata/XFEE1/77F0ADE89C820516E0534738A8C0802B/datafile/*
After you have made these changes, the source PDB can be opened read write again. It's part of the script 2, but as the folder names are different on different databases, I have made a comment out of it.
One last comment: I didn't tested the scripts as I have thrown things out. If something does not work, please come back and leave a comment, so I can fix it.
Thanks a lot!
By the way - happy New Year 2019 everyone!
Single Tenant ist unnütze Komplexität? / Single Tenant is only unnecessary complexity? - Remote refreshable clone
My presentation (in german) at the DOAG conference was about Single Tenancy and why (some / a lot / most) people do think it is unnecessary complexity or - even worser - it is unnecessary at all.
Yes, my first thoughts, as container based databases have been introduced in 12.1 were the same. I found that it is unnecessary at all. With 12.2 I changed my mind (a little bit or even a little bit more) as there are really nice features included, which can be used for free, even on Standard Edition 2.
I don't they, you should or must use these features, but as a Consultant and DBA I always want to help my customers doing things better or to free up the work life from boring stuff.
And what is more boring than staying in the office or working from home at the night or at a weekend when you could have time to play with your kids, make a trip to a cool event or meet some friends?
So if you can't live without spending times outside of your office hours with work, don't read this article further. 😏
For the DOAG I have created a live demo scenario on my notebook (with a Linux VM), so I would like to share not only my knowledge, but also part of my scripts, so that you can test everything on your own.
What you do need to use the scripts as they are, so you can copy paste them is not that much.
Preparation:
Some people do think about using the refreshable remote copy of a pdb as a poor men's standby database, but the best use case I can think of is setting up test environments (for applications or even for database updates) from your production. Another use case is to create a reporting PDB which is automatically refreshed ones a day because you can open the refreshable remote copy read only (and as long as it is opened read only no refresh happen).
Think about the following case: You do have your production database up and running and you should create a test environment. The business guys tell you, that they need the test environment in 3 days and the data should be the one after business hours or more horibble at saturday night.
What is your way to do this? Creating a standby database with Data Guard (not for SE2) and stop recover at the evening you should do? So you can prepare things and don't need to work a lot that evening? Or do you prepare an empty database and wait for that evening, then run an export on the production and an import at the test environment? How long does this take and how many hours do you spend that evening doing all this work?
A quick, fast solution is to create a refreshable remote copy of a pdb. All you need to do is to set it up the days before "that evening" comes, you let it refresh automatically all minutes and all you need to do at "that evening" is to stop the refresh and open the database. Maybe, due to application reasons, you need to run a SQL script which is always needed (also with Standby or with export/import). If you have tested that once or twice, you could even create a cronjob that does a last refresh and opens the database / run your application script when it is time to. So you don't need to work at the evening or weekend at all. 😊
If you want to test that, you need the things I have mentioned at the preparation part of this post. And then you can just run the following scripts.
The steps in the script are the following (you don't need to create the common user at the receiving database, I have just re-used the picture):
Script 2 to run at the CDB$ROOT where you want the PDB cloned to.
Maybe you try to refresh or set the refresh mode to none (see script 2) and you get an
ORA-65118: operation affecting a pluggable database cannot be performed from another pluggable database
Hope that will help you at your daily work. Stay tuned, there are more use cases where it is really nice to use PDBs instead of doing it "the old way". I will write a new post as soon as I have time to.
As always, comments are welcome.
You can find the second post regarding remote snapshot clone here.
Yes, my first thoughts, as container based databases have been introduced in 12.1 were the same. I found that it is unnecessary at all. With 12.2 I changed my mind (a little bit or even a little bit more) as there are really nice features included, which can be used for free, even on Standard Edition 2.
I don't they, you should or must use these features, but as a Consultant and DBA I always want to help my customers doing things better or to free up the work life from boring stuff.
And what is more boring than staying in the office or working from home at the night or at a weekend when you could have time to play with your kids, make a trip to a cool event or meet some friends?
So if you can't live without spending times outside of your office hours with work, don't read this article further. 😏
For the DOAG I have created a live demo scenario on my notebook (with a Linux VM), so I would like to share not only my knowledge, but also part of my scripts, so that you can test everything on your own.
What you do need to use the scripts as they are, so you can copy paste them is not that much.
Preparation:
- 2 Container databases XFEE1 and XFEE2 (XF is my naming convention on test systems for XFS file systems, AC for ACFS and AS for ASM, EE means Enterprise Edition, SE means Standard Edition - the tests do work also with Standard Edition Databases and on ACFS/ZFS without any changes).
- 1 Pluggable database EE1PDB1 inside of XFEE1
- TNSNames.ora entries to reach XFEE1, XFEE2, EE1PDB1 and EE2PDB1 (EE2PDB1 is later on used for cloned/snapshot cloned PDB1 inside of XFEE2)
- You need to set CLONE_DB true (alter system set clone_db=true scope=spfile;) for non-ACFS/ZFS file systems. Restart the database after setting it.
- To not violate any license rules you can also set MAX_PDBS to one (alter system set max_pdbs=1 scope=both;)
- Set all passwords to "oracle" (or change the scripts)
Some people do think about using the refreshable remote copy of a pdb as a poor men's standby database, but the best use case I can think of is setting up test environments (for applications or even for database updates) from your production. Another use case is to create a reporting PDB which is automatically refreshed ones a day because you can open the refreshable remote copy read only (and as long as it is opened read only no refresh happen).
Think about the following case: You do have your production database up and running and you should create a test environment. The business guys tell you, that they need the test environment in 3 days and the data should be the one after business hours or more horibble at saturday night.
What is your way to do this? Creating a standby database with Data Guard (not for SE2) and stop recover at the evening you should do? So you can prepare things and don't need to work a lot that evening? Or do you prepare an empty database and wait for that evening, then run an export on the production and an import at the test environment? How long does this take and how many hours do you spend that evening doing all this work?
A quick, fast solution is to create a refreshable remote copy of a pdb. All you need to do is to set it up the days before "that evening" comes, you let it refresh automatically all minutes and all you need to do at "that evening" is to stop the refresh and open the database. Maybe, due to application reasons, you need to run a SQL script which is always needed (also with Standby or with export/import). If you have tested that once or twice, you could even create a cronjob that does a last refresh and opens the database / run your application script when it is time to. So you don't need to work at the evening or weekend at all. 😊
If you want to test that, you need the things I have mentioned at the preparation part of this post. And then you can just run the following scripts.
The steps in the script are the following (you don't need to create the common user at the receiving database, I have just re-used the picture):
- Create a common user and grant him rights
- Create a database link to the PDB you want to clone from the CDB$ROOT you want the PDB cloned to.
- Create the refreshable PDB clone and let it run until you need it
- Open the automatically refreshed PDB read-write
Script 2 to run at the CDB$ROOT where you want the PDB cloned to.
Maybe you try to refresh or set the refresh mode to none (see script 2) and you get an
ORA-65118: operation affecting a pluggable database cannot be performed from another pluggable database
If so, just set the container to the pdb ( alter session set container=EE2PDB1; ) and try again.
Hope that will help you at your daily work. Stay tuned, there are more use cases where it is really nice to use PDBs instead of doing it "the old way". I will write a new post as soon as I have time to.
As always, comments are welcome.
You can find the second post regarding remote snapshot clone here.
DOAG Conference 2018
Hi Folks,
last year I have written a small review for the DOAG conference and I want to do the same for this years conference.
I was - again - a happy speaker at the DOAG conference 2018. My presentation was about "Single Tenant is only unnecessary complexity?" and I held it at the last conference day at noon. Despite the key note before my presentation and despite of the time where all people wanted to go to lunch, I think I had more than 100 people attending it. In 45 minutes (I had a slight overtime of some seconds only) I presented some features of Single Tenancy on Oracle databases - including a live demo to show how easy it is to do in reality. As a lot people after the presentation asked me how to do this or that (they started to think about the requirements they have at home), I think it was a valuable presentation at all.
It was good to do this, even a week before I have held the presentation, my VM crashed and was unreparable and I was thinking (for some minutes only) to skip the live demo part. But - luckily - I had a one week old backup of the system so I needed only two night shifts to get everything up running again. You could have smelled my fear minutes before I started the demo - I really hoped that nothing crashes at the presentation...
Beside my own presentation the 3 days were fully packed with know how provided by well prepared speakers. A trend I have seen last year also continued this year: In my opinion the number of live demos is dropping more and more. Unfortunately, in my opinion, because I like presentations where you can really see "things are running". So, if you plan to speak somewhere at a conference: Yes, it is a shitload of work preparing everything and testing and preparing more and testing again - but it's worth it guys (by the way, my live demo was a one week workload - including the crash)!
What else, hm... the community: Even if some guys don't think it, but Oracle is "still alive" and the community is also. More than 2'000 people found their way to Nuremberg and more than 400 speakers including really famous international speakers (and Oracle product managers) were there. And the best is they are approachable - all of them can be found at the coffee lounge or somewhere else. So you can discuss, you can ask, you can get solutions for free (what is especially worthful when the Oracle support does not seem to make the job very well like it was told us at a session regarding support satisfaction). Stefan Köhler, e.g., was just sitting two afternoons at the coffee lounge - ready to discuss about performance with everyone.
Another thing I have seen at the conference - the interest to know more about alternative databases is high, so open source database talks were fully booked out. Especially when PostgreSql was mentioned at the conference planner, the rooms were filled until the last place. In my opinion, open source databases will get a bigger piece of the cake (so robotron and therefore I also will do up to 24/7 support on PostgreSql), nevertheless, they are not soooo free at all if you make a TCO calculation. While Oracle database is a big, feature-rich database monster out of one hand, you do have in open source one main project, but if you want to use additional features you need some additional software here, a plugin there, ...
At the end, you either throw some money into a consulting company, to get help for the community software, or you throw some money into a couple of internal guys who will find out what you need and how you get it from the communities - or you buy a "open source software" by one of the vendors around the communities. The problem then is the same as with Oracle. You will have some kind of vendor lock and knowbody knows if these companies do exist in 5 or 10 years or are bought by someone and the "product" maybe is not supported anymore. It is not so easy at the moment to make the right decisions (at least I think Oracle will still exist in 10 years also). And the big companies like Amazon or Salesforce only give a small part of the development, they do, back into the community versions. As I said, there will be a rising number of open source database installations, but I think also Microsofts SQLServer and Oracles database will be the two biggest fishes in the next couple of years.
In addition to technical details there are a lot things you can hear about makeing better projects, about AI or Machine Learning, about development or DSGVO, Cloud, ...
Some last words to the organisation: It is astonishing for me how this all works year for year - especially if you know that a big part of work is done in the spare time of the DOAG members. Organizing stuff for more than 2'000 people, including food, the key notes, the party, all community actions (like apex, PL/SQL, ...), incredible. And the location - with it's different rooms and restaurants at each floor it is really perfect.
My conclusion: if you didn't had time to join the conference (it would be good to speak german as most presentations are hold in german) - try to come to Nuremberg next year (19.-22. November 2019). You hopefully find me also there again...
For the guys looking for the scripts of my presentation: You can find them here.
last year I have written a small review for the DOAG conference and I want to do the same for this years conference.
I was - again - a happy speaker at the DOAG conference 2018. My presentation was about "Single Tenant is only unnecessary complexity?" and I held it at the last conference day at noon. Despite the key note before my presentation and despite of the time where all people wanted to go to lunch, I think I had more than 100 people attending it. In 45 minutes (I had a slight overtime of some seconds only) I presented some features of Single Tenancy on Oracle databases - including a live demo to show how easy it is to do in reality. As a lot people after the presentation asked me how to do this or that (they started to think about the requirements they have at home), I think it was a valuable presentation at all.
It was good to do this, even a week before I have held the presentation, my VM crashed and was unreparable and I was thinking (for some minutes only) to skip the live demo part. But - luckily - I had a one week old backup of the system so I needed only two night shifts to get everything up running again. You could have smelled my fear minutes before I started the demo - I really hoped that nothing crashes at the presentation...
Beside my own presentation the 3 days were fully packed with know how provided by well prepared speakers. A trend I have seen last year also continued this year: In my opinion the number of live demos is dropping more and more. Unfortunately, in my opinion, because I like presentations where you can really see "things are running". So, if you plan to speak somewhere at a conference: Yes, it is a shitload of work preparing everything and testing and preparing more and testing again - but it's worth it guys (by the way, my live demo was a one week workload - including the crash)!
What else, hm... the community: Even if some guys don't think it, but Oracle is "still alive" and the community is also. More than 2'000 people found their way to Nuremberg and more than 400 speakers including really famous international speakers (and Oracle product managers) were there. And the best is they are approachable - all of them can be found at the coffee lounge or somewhere else. So you can discuss, you can ask, you can get solutions for free (what is especially worthful when the Oracle support does not seem to make the job very well like it was told us at a session regarding support satisfaction). Stefan Köhler, e.g., was just sitting two afternoons at the coffee lounge - ready to discuss about performance with everyone.
Another thing I have seen at the conference - the interest to know more about alternative databases is high, so open source database talks were fully booked out. Especially when PostgreSql was mentioned at the conference planner, the rooms were filled until the last place. In my opinion, open source databases will get a bigger piece of the cake (so robotron and therefore I also will do up to 24/7 support on PostgreSql), nevertheless, they are not soooo free at all if you make a TCO calculation. While Oracle database is a big, feature-rich database monster out of one hand, you do have in open source one main project, but if you want to use additional features you need some additional software here, a plugin there, ...
At the end, you either throw some money into a consulting company, to get help for the community software, or you throw some money into a couple of internal guys who will find out what you need and how you get it from the communities - or you buy a "open source software" by one of the vendors around the communities. The problem then is the same as with Oracle. You will have some kind of vendor lock and knowbody knows if these companies do exist in 5 or 10 years or are bought by someone and the "product" maybe is not supported anymore. It is not so easy at the moment to make the right decisions (at least I think Oracle will still exist in 10 years also). And the big companies like Amazon or Salesforce only give a small part of the development, they do, back into the community versions. As I said, there will be a rising number of open source database installations, but I think also Microsofts SQLServer and Oracles database will be the two biggest fishes in the next couple of years.
In addition to technical details there are a lot things you can hear about makeing better projects, about AI or Machine Learning, about development or DSGVO, Cloud, ...
Some last words to the organisation: It is astonishing for me how this all works year for year - especially if you know that a big part of work is done in the spare time of the DOAG members. Organizing stuff for more than 2'000 people, including food, the key notes, the party, all community actions (like apex, PL/SQL, ...), incredible. And the location - with it's different rooms and restaurants at each floor it is really perfect.
My conclusion: if you didn't had time to join the conference (it would be good to speak german as most presentations are hold in german) - try to come to Nuremberg next year (19.-22. November 2019). You hopefully find me also there again...
For the guys looking for the scripts of my presentation: You can find them here.
DBT-00007 User does not have the appropiate write privilege when starting dbca
At the moment I am preparing some virtualboxes to create the live demos for my presentation about "Single Tenancy is only more complexity!?" at the DOAG Conference 2018.
For this, I have setup a fresh new Linux and installed 18c Grid Infrastructure into it (using role separation, so GI is installed with the grid user, the databases should be installed with the oracle user).
After I have created my 2 ORACLE_HOMEs with the Standard Edition 2 and the Enterprise Edition database (Software only) I wanted to set up the first databases, but unfortunately the dbca (Database configuration assistant) struggled with "a [DBT-00007] User does not have the appropriate write privilege when starting dbca" error. Since I NEVER have seen that error before, I really was wondering, what happens (and I don't think I have missed some steps). And no, I don't want to check this with the Oracle support, like mentioned at the popup window details.
I have setup my ORACLE_BASE and ORACLE_HOME before I have started dbca, so this isn't the cause for this error. As the popup window does not shown anything in addition, I tried to check the log of the dbca at $ORACLE_BASE/cfgtoollogs/dbca - but there wasn't any.
So the problem seemed to by correlated to this directory.
The owner of $ORACLE_BASE/cfgtoollogs was grid (group oinstall) and the rights set were 755 - so only the grid user was allowed to write, group members weren't.
To change this I have first changed the owner of the cfgtoollogs directory:
chown -hR oracle:oinstall /u01/app/oracle/cfgtoollogs
and afterwards I have given write rights to all subdirectories, so also the grid user would still be able to add his stuff to the log directory:
chmod 775 /u01/app/oracle/cfgtoollogs/
chmod 770 /u01/app/oracle/cfgtoollogs/dbca
chmod 770 /u01/app/oracle/cfgtoollogs/asmca
chmod 770 /u01/app/oracle/cfgtoollogs/netca
That's it folks, the dbca started after I have made these changes.
For this, I have setup a fresh new Linux and installed 18c Grid Infrastructure into it (using role separation, so GI is installed with the grid user, the databases should be installed with the oracle user).
After I have created my 2 ORACLE_HOMEs with the Standard Edition 2 and the Enterprise Edition database (Software only) I wanted to set up the first databases, but unfortunately the dbca (Database configuration assistant) struggled with "a [DBT-00007] User does not have the appropriate write privilege when starting dbca" error. Since I NEVER have seen that error before, I really was wondering, what happens (and I don't think I have missed some steps). And no, I don't want to check this with the Oracle support, like mentioned at the popup window details.
I have setup my ORACLE_BASE and ORACLE_HOME before I have started dbca, so this isn't the cause for this error. As the popup window does not shown anything in addition, I tried to check the log of the dbca at $ORACLE_BASE/cfgtoollogs/dbca - but there wasn't any.
So the problem seemed to by correlated to this directory.
The owner of $ORACLE_BASE/cfgtoollogs was grid (group oinstall) and the rights set were 755 - so only the grid user was allowed to write, group members weren't.
To change this I have first changed the owner of the cfgtoollogs directory:
chown -hR oracle:oinstall /u01/app/oracle/cfgtoollogs
and afterwards I have given write rights to all subdirectories, so also the grid user would still be able to add his stuff to the log directory:
chmod 775 /u01/app/oracle/cfgtoollogs/
chmod 770 /u01/app/oracle/cfgtoollogs/dbca
chmod 770 /u01/app/oracle/cfgtoollogs/asmca
chmod 770 /u01/app/oracle/cfgtoollogs/netca
That's it folks, the dbca started after I have made these changes.
Subscribe to:
Posts (Atom)