Why 2 floating IP's are used to Weblogic managed servers?

In the configuration example discussed in this guide, two floating addresses
are used to each of the WebLogic Managed Servers and to the Administration
Server. The first type of floating IP address uses the IPoIB (BOND0) interface,
and the second type of floating IP uses the EoIB (BOND1) interface.
The above are excerpts from this document: http://docs.oracle.com/cd/E18476_01/doc.220/e18479.pdf page 3-10
My question is, why do I have to use 2 floating IP's, one uses bond0 and the other uses bond1? can I just create one floating on bond1 and for all purposes, for instance? For what reasons, do I have to use 2 floating IP's?
Thanks!
Edited by: 943714 on Jan 29, 2013 12:01 PM

Hi,
Two floating IP's should be used because on an Exalogic machine there are multiple network interfaces. As it says on that document, bond0 is for IPoIB and bond1 is EoIB.
bond1 is the interface available outside the Exalogic machine. But, for example, if you need to connect to an Exadata machine (using Weblogic with JDBC/GridLink) that must be done through InfiniBand (IPoIB - bond0 interface).
Hope this helps.
-Cris

Similar Messages

Weblogic managed servers connecting to the servers in different cluster

 Hi All,
 We have a weired problem going on for a while. We have a cluster configuration
 with an admin server and two managed servers. We have the similar configuration
 in DEV, TEST and PROD. The problem is that the managed server members in DEV cluster
 are making connections to managed servers which are member of PROD cluster for
 session replication. The same way TEST servers are trying to connect to PROD and
 DEV.
 Has anyone seen this kind of problem before. BEA seems to be cluless so far.
 Thanks in adavnce for your input.
 Udit


 Venkat,
 Thats a good suggestion but these things are too obvious to ignore. We have different
 multicast address in DEV and PROD and also hosts are on different sub net. I do
 not know if cluster name will make any differene though.
 Thanks for your input anyway,
 Udit
 "venkat" <[email protected]> wrote:
 >
 >Udit,
 > You can check the sub net, multicast address and the cluster name.
 >If the dev
 >and prod servers are in the same sub net with same multicast address,
 >then change
 >the multicast and try.
 >
 >Venkat
 >"venkat" <[email protected]> wrote:
 >>
 >>Udit,
 >>
 >>
 >>"Udit Singh" <[email protected]> wrote:
 >>>
 >>>Kumar,
 >>>Thanks for the reply.
 >>>The situation is that managed server in DEV try to replicate the session
 >>>to a
 >>>managed server in PROD and TEST and vice versa.
 >>>Let us say our dev managed servers are running on abc01 and abc02 and
 >>>prod managed
 >>>servers are running on xyz01 and xyz02. All the managed servers are
 >>running
 >>>on
 >>>port 7005.
 >>>If I do the netstat on abc01 or abc02 I could the see established connections
 >>>between abc01/02 and xyz01/02.
 >>>Why is that happening? We are running 6.1SP2.
 >>>
 >>>Udit
 >>>
 >>>Kumar Allamraju <[email protected]> wrote:
 >>>>We do not restrict intercluster communication as of 61 SP3.
 >>>>Once we get the IP from the cookie, we can safely make a
 >>>>connection to the other clustered node. We were not checking
 >>>>if the server is part of the same cluster or not. This is
 >>>>already fixed in 7.x and 61 SP4(not yet released) If you are
 >>>>on 61 Sp2 or SP3 then you should contact support and
 >>>>reference CR # CR089798 to get a one off patch.
 >>>>
 >>>>Regardless, are you traversing from DEV to PROD cluster and
 >>>>vice-versa. If not then this problem shouldn't happen unless
 >>>>plugin is routing the request to wrong cluster.
 >>>>
 >>>>--
 >>>>Kumar
 >>>>
 >>>>Udit Singh wrote:
 >>>>> Hi All,
 >>>>> We have a weired problem going on for a while. We have a cluster
 >>configuration
 >>>>> with an admin server and two managed servers. We have the similar
 >>>configuration
 >>>>> in DEV, TEST and PROD. The problem is that the managed server members
 >>>>in DEV cluster
 >>>>> are making connections to managed servers which are member of PROD
 >>>>cluster for
 >>>>> session replication. The same way TEST servers are trying to connect
 >>>>to PROD and
 >>>>> DEV.
 >>>>> Has anyone seen this kind of problem before. BEA seems to be cluless
 >>>>so far.
 >>>>>
 >>>>> Thanks in adavnce for your input.
 >>>>> Udit
 >>>>
 >>>
 >>
 >

Weblogic managed servers question

Hi all,
I'm trying to set up a 3 machines , one having weblogic admin
server and two machines having managed weblogic servers.what i'm not
undersatnding is :
1. Do i have to install the weblogic software on all machines with the same
domain name .?
2. Do i have to deploy my components/web applications on all machines?
3. Do i have to put the above machines in a cluster to achieve load
balancing etc.
4. How will i access my manage servers considering that they will have 2
different ips (since they are installed on 2 different
machines)?
5.Does anyone know any dosc online which explains these matters.I have gone
through the weblogic server administrator
guide , but could not find the information i'm looking for (or have i
missed any section)?
Thanks in advance
Prashant

Hi.
My answers are embedded:
Prashant S wrote:
Hi all,
I'm trying to set up a 3 machines , one having weblogic admin
server and two machines having managed weblogic servers.what i'm not
undersatnding is :
1. Do i have to install the weblogic software on all machines with the same
domain name .?I'm not sure what you mean. The managed servers need to be in the same
weblogic domain (ie directory under config) if you are using clustering. Also
if you are using clustering the cluster members (managed servers) need to be on
the same LAN/subnet.
>
2. Do i have to deploy my components/web applications on all machines?You do not have to deploy your applications on your admin server. Typically I
install a webapp (or whatever) on the admin server then select the targets for
deployment to the managed servers.
>
3. Do i have to put the above machines in a cluster to achieve loadbalancing
etc.Yes.
>
4. How will i access my manage servers considering that they will have 2
different ips (since they are installed on 2 different machines)?For web applications the standard practice is to use a proxy front-end to the
cluster. We offer plugin's for Netscape, Apache and IIS that handle
loadbalancing and failover for weblogic clusters. We also offer the
HttpClusterServlet if you prefer to use an instance of WLS to proxy to your
cluster. Use your browser to hit the proxy, which will pass your request on to
a cluster member. See the following link for details:
http://edocs.bea.com/wls/docs61/adminguide/http_proxy_cluster.html
>
5.Does anyone know any dosc online which explains these matters.I have gone
through the weblogic server administrator
guide , but could not find the information i'm looking for (or have i
missed any section)?
Thanks in advance
Prashantnot sure if you have seen this but
http://edocs.bea.com/wls/docs61/cluster/index.html contains links to docs
discussing most cluster administration issues.
Hope this helps.
Michael Young
Developer Relations Engineer
BEA Support

Why SCOM 2012 R2 Agent version for Service Manager servers is 6.1.7221.49, why not 7.1.10184.

Hi ,
In SCOM 2012 R2 console if i look at all of my agent managed servers have agent builds of 7.1.10184 but my two service manager servers have agent builds of 6.1.7221.49
Can anyone let me know is this known issue or how can we upgrade the agent to correct version.
Thanks in advance
Bharath

Hi,
SCOM and SCSM are built on the same structure, Management Servers and DAta Warehouse etc. This means that when you install the SCSM Management Server, the Microsoft Monitoring Agent is also installed. When
you want to monitor Service Manager and the server it´s installed on, you need to configure the SCOM parameters from the control panel at the SCSM management server.
If you were to try and install the SCOM agent at the SCSM server, it would fail and tell you that you can´t install the agent on a server where Service Manager is installed. The agent will be upgraded when
you upgrade Service Manager so this is all expected and It´s as it should be :)
Regards,
Daniel
IT Consultant at Viridis IT

Why two remote keys are genrated in data manager when i'm syndicating the same records twice

Hi Experts,
Please guide me that will there be any problem if two remot keys are getting generated for the same system thougt we are expecting to update the same Business Partner Number in ECC.
Also We are able to see two different business partner number(in ECC this is KUNNR in KNA1)
Please guide me how to proceed so that when we syndicate the same set of records we want to get update the records in ECC for the same KUNNR.
We don't want to get another KUNNR for the same records.
Thanks!
Pankaj

"barbara" wrote in message
news:506500000008000000FAC20000-1077587809000@exchange.ni.com...
> I'm using the simulation interface toolkit V2.0 but it works only the
> first time after the installation. When I restart again my computer it
> doesn't work any more. Why??????????
>
---snip
I am not familiar with the different software that you mentioned but as a
first step I would check compatibility which is done as below:
Get original operating system CD that was used to install Windows XP on the
machine in your post. When Windows XP is inserted into CD ROM drive it
normally will ask you to conduct compatibility check for the current
installation. See what it says.
HTH
Mohan Pawar
----snip

# of users are on the weblogic server

This is probably an application based question, but is there anyway to tell
on the Weblogic console how many people are using a Weblogic server? Based
on the size of the database pool? Threads?

Wouldn't the 'Connected Users' folder suffice? You could also look
under HTTP.Servlet and Web Applications.General.session. In it you'll
find the number of active sessions.
Jason
On Mon, 12 Jun 2000 15:40:28 -0400, "Mike Kendzierski"
<[email protected]> wrote:
This is probably an application based question, but is there anyway to tell
on the Weblogic console how many people are using a Weblogic server? Based
on the size of the database pool? Threads?

How to change the response of human task if we are using the skip rule

Hi,
We have developed the request workflow in OIM11g for this we are using the Beneficiary Manager approval composite to approve the request by his manager.
We are using the skip rule in the human task component to skip the user if the requester is beneficiary manager.
After the skip rule is executing the the request is going to failed state because of the response it is sending to the OIM
The skip rule is working as expected but the response sending to OIM is COMPLETED instead of APPROVE.
Please let us know the steps to change the response to Approve after the skip rule is executed or how to add the response COMPLETED in OIM.
Thanks

You will have to write a conditional block outisde the task where you would check if the skip condition is met and if yes then you will have to hard code the approved conditions into the payload for the task outcome.
<assign name="SetAutoApproveConditions">
 <copy>
 <from expression="string('APPROVE')"/>
 <to variable="ApprovalTask_1_globalVariable"
 part="payload"
 query="/task:task/task:systemAttributes/task:outcome"/>
 </copy>
 <copy>
 <from expression="string('COMPLETED')"/>
 <to variable="ApprovalTask_1_globalVariable"
 part="payload"
 query="/task:task/task:systemAttributes/task:state"/>
 </copy>
 </assign>HTH,
BB

Using a tag manager with the Elq. tracking script.

Has anyone had any luck using a tag managment tool to package the eloqua tracking scripts. We are using Google Tag Manager and not having any luck registering click-throughs on our emails.
Let me know if you're having success and what you're doing to get there. Thanks.

Jim -
I've seen the scripts deployed and working properly across sites using various platforms from tag management to CMS that are truly tracking but where the tracking validator built into Eloqua doesn't validate the presences of the scripts. The best way to check it is two fold: 1. visit the pages on with an uncookied browser, submit a form, etc. to get a cookie and then surf around and check the contact record in eloqua about 20 min later. 2. look at the site map in eloqua to see if your pages are all visible.
Lauren

How to Calculate Overtime Using Negative Time Management

Hi Experts,
We are having PA,OM & PY in our project scope. We have just configured TM uptill workschedule.
But how to calculate the overtime when we are using negative time management.
We are having International Payroll Driver.
Regards,
Irfan

Hi,
Thanks for your prompt response.
But if the overtime is based on Basic , HRA and other wage components.
How the configuration of the wage type takes palce.
Regards,
Irfan

Why multiple log files are created while using transaction in berkeley db

we are using berkeleydb java edition db base api, we have already read/write CDRFile of 9 lack rows with transaction and
without transaction implementing secondary database concept the issues we are getting are as follows:-
with transaction----------size of database environment 1.63gb which is due to no. of log files created each of 10 mb.
without transaction-------size of database environment 588mb and here only one log file is created which is of 10mb. so we want to know how REASON CONCRETE CONCLUSION ..
how log files are created and what is meant of using transaction and not using transaction in db environment and what are this db files db.001,db.002,_db.003,_db.004,__db.005 and log files like log.0000000001.....plz reply soon

we are using berkeleydb java edition db base api, If you are seeing __db.NNN files in your environment root directory, these are environment's shared region files. And since you see these you are using Berkeley DB Core (with the Java/JNI Base API), not Berkeley DB Java Edition.
with transaction ...
without transaction ...First of all, do you need transactions or not? Review the documentation section called "Why transactions?" in the Berkeley DB Programmer's Reference Guide.
without transaction-------size of database environment 588mb and here only one log file is created which is of 10mb.There should be no logs created when transactions are not used. That single log file has likely remained there from the previous transactional run.
how log files are created and what is meant of using transaction and not using transaction in db environment and what are this db files db.001,db.002,_db.003,_db.004,__db.005 and log files like log.0000000001Have you reviewed the basic documentations references for Berkeley DB Core?
- Berkeley DB Programmer's Reference Guide
in particular sections: The Berkeley DB products, Shared memory regions, Chapter 11. Berkeley DB Transactional Data Store Applications, Chapter 17. The Logging Subsystem.
- Getting Started with Berkeley DB (Java API Guide) and Getting Started with Berkeley DB Transaction Processing (Java API Guide).
If so, you would have had the answers to these questions; the __db.NNN files are the environment shared region files needed by the environment's subsystems (transaction, locking, logging, memory pool buffer, mutexes), and the log.MMMMMMMMMM are the log files needed for recoverability and created when running with transactions.
--Andrei

In BAPI's why the structures are used in table parameters?

Hello sir,
what is BAPI sir? In BAPI's why the structures are used in table parameters? table parameters they are using structures but not using any customized tables ?
regards
rachu.

Hello Rachu
BAPIs provide RFC-enabled interfaces to SAP business objects (e.g. like customer, sales order, purchase order, etc.).
A BAPI does basically the same like you would need to do calling the corresponding transaction (e.g. BAPI_SALESORDER_CREATE -> VA01).
Since they are RFC-enabled they can be called from external systems.
BAPIs represent an external interface for the outside world. Very often you will find within the BAPI that there is a mapping done to the (SAP-)internal structures at the beginning of the coding and vice versa at the end of the coding. Thus, you will (almost) never find any DB table name used as type of a BAPI TABLES parameter.
Regards
Uwe

Why we are using dynamic actions in HR

why we are using dynamic actions in HR?? main purpose of it>>

Check the link -
Re: Dynamic Actions
Re: dynamic actions
Re: Dynamic Actions
www.hrexpertonline.com/downloads/Rehan%20Zaidi%20Online%20Portion.doc
Regards,
Amit
Reward all helpful replies.

While in zh|ja locale, zh|ja fonts are used even for Latin... why?

One thing that has always bothered me is that my system uses funny fonts when it is running in a non-English locale, and I think I finally found the culprit.
When I check for font matching in English everything seems normal:
$ fc-match -s "sans"
DejaVuSans.ttf: "DejaVu Sans" "Book"
DejaVuSans-Bold.ttf: "DejaVu Sans" "Bold"
DejaVuSans-Oblique.ttf: "DejaVu Sans" "Oblique"
DejaVuSans-BoldOblique.ttf: "DejaVu Sans" "Bold Oblique"
iYaHei.ttf: "iYaHei" "Regular"
malgun.ttf: "Malgun Gothic" "Regular"
kartika.ttf: "Kartika" "Regular"
arialuni.ttf: "Arial Unicode MS" "Normal"
But, if I were to do it in Chinese or Japanese:
$ fc-match -s "sans:lang=zh|ja"
iYaHei.ttf: "iYaHei" "Regular"
DejaVuSans.ttf: "DejaVu Sans" "Book"
DejaVuSans-Bold.ttf: "DejaVu Sans" "Bold"
DejaVuSans-Oblique.ttf: "DejaVu Sans" "Oblique"
DejaVuSans-BoldOblique.ttf: "DejaVu Sans" "Bold Oblique"
malgun.ttf: "Malgun Gothic" "Regular"
kartika.ttf: "Kartika" "Regular"
arialuni.ttf: "Arial Unicode MS" "Normal"
DejaVuSans goes /after/ iYaHei. This is a problem, as iYaHei's Latin glyphs look horrible; its CJK glyphs, while not the best, are at least extensive, which is why I use it. For monospace it is even more of a problem, because then Arial MS Unicode, which isn't even a monospace font gets priority over DejaVu Sans Mono in zh|ja environments.
This is my current fonts.conf:
<fontconfig>
<alias>
<family>Monospace</family>
<prefer>
<family>DejaVu Sans Mono</family>
<family>Arial Unicode MS</family>
</prefer>
</alias>
<alias>
<family>sans-serif</family>
<prefer>
<family>DejaVu Sans</family>
<family>iYaHei</family>
<family>Malgun Gothic</family>
<family>Kartika</family>
<family>Arial Unicode MS</family>
</prefer>
</alias>
<alias>
<family>serif</family>
<prefer>
<family>Times New Roman</family>
<family>AR PL New Sung</family>
<family>MS Mincho</family>
<family>MingLiU</family>
<family>MingLiU-ExtB</family>
<family>MingLiU_HKSCS-ExtB</family>
<family>MS Song</family>
<family>UnBatang</family>
<family>RaghuMalayalam</family>
<family>Arial Unicode MS</family>
</prefer>
</alias>
<dir>/home/sokuban/.fontmatrix</dir>
<dir>/home/sokuban/.Fontmatrix/Activated</dir>
</fontconfig>
This is probably happening because while I'm in a zh|ja locale, the system is looking for the highest font on my priority list that says it is a zh|ja font and then uses it—even to cover Latin etc glyphs that fonts higher in my priority list can cover; I'm guessing this is because it is assuming that those fonts aren't suitable for use in a zh|ja environment.
So the simplest way to solve the problem that everyone would immediately think of is to say that DejaVu can do zh|ja. I (presume) it wouldn't cause any problems even though they can't, because then they will simply borrow from real zh|ja fonts and will now at least be used in a zh|ja locale. Actually, I would want this behaviour no matter what way I decide to solve the problem.
The other—and better—solution to the problem would be to specifically set a different priority list to be used for different languages. That way when the system looks for "sans:lang=zh_TW", it'll get DejaVu at the top of the list to handle Latin, and then a nice traditional Chinese font. Likewise if the system looked for "sans:lang=ja", it'll get DejaVu with a nice Japanese font. Before I didn't think this was possible, but now that I see how this font matching is working, it would be weird if it wasn't possible.
Right now I have these nice Chinese/Japanese fonts installed on my system, but I don't always get to use them (mostly while browsing, if the html is nice enough). I made my system default iYaHei (and the firefly font for serif) because iYaHei covers almost all of CJK and is readable. If I were to put one of the nice Chinese or Japanese fonts as my system default, I'd end up having some glyphs being borrowed from another font, and it looks really ugly when this happens. Monolinguals don't have this problem, because they can just set their system font to their language, and then be done with it, as they'll rarely need to display any glyphs unsupported.
If anyone can help me, especially with getting a different priority list for different languages, it would be appreciated.

Hmm, that seems to work fine for one language, but I put 4 languages in there and only the first one I put works properly:
This would be my fontconfig
<fontconfig>
<match>
<test name="lang" compare="contains">
<string>zh_TW</string>
</test>
<test name="family">
<string>serif</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Serif</string>
<string>Bitstream Vera Serif</string>
<string>MingLiU</string>
<string>MingLiU-ExtB</string>
<string>MingLiU_HKSCS-ExtB</string>
</edit>
</match>
<match>
<test name="lang" compare="contains">
<string>zh_TW</string>
</test>
<test name="family">
<string>sans-serif</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Sans</string>
<string>Bitstream Vera Sans</string>
<string>Microsoft JhengHei</string>
</edit>
</match>
<match>
<test name="lang" compare="contains">
<string>zh_TW</string>
</test>
<test name="family">
<string>monospace</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Sans Mono</string>
<string>Bitstream Vera Sans Mono</string>
<string>WenQuanYi Zen Hei Mono</string>
</edit>
</match>
<match>
<test name="lang" compare="contains">
<string>zh_CN</string>
</test>
<test name="family">
<string>serif</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Serif</string>
<string>Bitstream Vera Serif</string>
<string>MS Song</string>
<string>WenQuanYi Bitmap Song</string>
<string>AR PL New Sung</string>
</edit>
</match>
<match>
<test name="lang" compare="contains">
<string>zh_CN</string>
</test>
<test name="family">
<string>sans-serif</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Sans</string>
<string>Bitstream Vera Sans</string>
<string>Microsoft YaHei</string>
<string>WenQuanYi Zen Hei</string>
</edit>
</match>
<match>
<test name="lang" compare="contains">
<string>zh_CN</string>
</test>
<test name="family">
<string>monospace</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Sans Mono</string>
<string>Bitstream Vera Sans Mono</string>
<string>WenQuanYi Zen Hei Mono</string>
<string>AR PL New Sung</string>
</edit>
</match>
<match>
<test name="lang" compare="contains">
<string>ja</string>
</test>
<test name="family">
<string>serif</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Serif</string>
<string>Bitstream Vera Serif</string>
<string>MS PMincho</string>
</edit>
</match>
<match>
<test name="lang" compare="contains">
<string>ja</string>
</test>
<test name="family">
<string>sans-serif</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Sans</string>
<string>Bitstream Vera Sans</string>
<string>Meiryo</string>
</edit>
</match>
<match>
<test name="lang" compare="contains">
<string>ja</string>
</test>
<test name="family">
<string>monospace</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Sans Mono</string>
<string>Bitstream Vera Sans Mono</string>
<string>MS Gothic</string>
</edit>
</match>
<match>
<test name="lang" compare="contains">
<string>ko</string>
</test>
<test name="family">
<string>serif</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Serif</string>
<string>Bitstream Vera Serif</string>
<string>Batang</string>
</edit>
</match>
<match>
<test name="lang" compare="contains">
<string>ko</string>
</test>
<test name="family">
<string>sans-serif</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Sans</string>
<string>Bitstream Vera Sans</string>
<string>Malgun Gothic</string>
</edit>
</match>
<match>
<test name="lang" compare="contains">
<string>ko</string>
</test>
<test name="family">
<string>monospace</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Sans Mono</string>
<string>Bitstream Vera Sans Mono</string>
<string>UnTaza</string>
</edit>
</match>
<alias>
<family>Monospace</family>
<prefer>
<family>DejaVu Sans Mono</family>
<family>Arial Unicode MS</family>
</prefer>
</alias>
<alias>
<family>sans-serif</family>
<prefer>
<family>DejaVu Sans</family>
<family>iYaHei</family>
<family>Malgun Gothic</family>
<family>Kartika</family>
<family>Arial Unicode MS</family>
</prefer>
</alias>
<alias>
<family>serif</family>
<prefer>
<family>Times New Roman</family>
<family>AR PL New Sung</family>
<family>MS Mincho</family>
<family>MingLiU</family>
<family>MingLiU-ExtB</family>
<family>MingLiU_HKSCS-ExtB</family>
<family>MS Song</family>
<family>UnBatang</family>
<family>RaghuMalayalam</family>
<family>Arial Unicode MS</family>
</prefer>
</alias>
<dir>/home/sokuban/.fontmatrix</dir>
<dir>/home/sokuban/.Fontmatrix/Activated</dir>
</fontconfig>
I had to put zh_TW at the very top, otherwise matching for zh_TW would pull up meiryo first. There is still one little problem in that matching for ko will pull up MS Mincho before Batang, but whatever, I don't really care that much. Meiryo and Malgun Gothic also still go before DejaVu Sans for some reason, but this again doesn't really matter, and might actually be a good thing.
So far setting it like this seems to work to some degree. More of Niconico Douga uses Meiryo now (the search bar and stuff still uses iYaHei, but I don't think much can be done about that), if I run a program with LANG=ja_JP.UTF-8 and it has Japanese support it'll use Meiryo too. But this doesn't work with Chinese. (I guess because I had country codes in my Chinese matching rules) Oh well, this is still a step up, and Trad Chinese with a Simp Chinese font isn't as bad as Japanese with a Simp Chinese font; I've gotten used to seeing it everywhere. (Heck, the book I'm using to study Chinese now is a translation from a book made to study Simplified - and although it is a traditional book, it still uses the same simplified font.)
EDIT: Actually, it doesn't seem to have anything to do with the program having Japanese support or not. Some programs will use Meiryo, others will completely ignore the fontconfig settings (and use iYaHei). Sonata for example, even though it had a working Japanese localization, used iYaHei, while thunar and mousepad used Meiryo.
I used to have this bug that pretty much made working in CJK locales impossible, it took a really long time to open up terminals because of the messed up font settings. This seems to be fixed in a Japanese locale, but not a Taiwanese. Again, I guess this has to do with the country code above. (If you use a country code do you need to put a different syntax? Because other than the fc-match -s "sans:lang=zh_TW" results, the Chinese lines don't seem to have any effect—though I guess if that command shows it, they must be affected?)
Also, I'm not sure if I've been blind and not noticed it until now, but after applying these Changes /all/ kana are using Meiryo. (Unless specified otherwise) This is both a good thing and a bad thing—no, I think it is a bad thing. Because Japanese text uses kana and kanji (CJK unified block). So now when I have a Japanese filename in thunar, the kana shows up using a proper Japanese font, but the Chinese Characters in the name show up using a Chinese font. While this is kind of okay of the fonts look similar, it isn't a good solution because not all fonts are designed to work well with each other, and my main goal in this quest is to avoid having to "borrow" from another font when the default font can't display a glyph properly. (Meiryo and iYaHei kiiinda look similar, so it isn't that bad, but if one day I decided to set my default gtk font to serif for some reason then I'll have problems for sure.) If somehow the system was able to tell if the text was Japanese or Chinese and use the appropriate font, that would be good, but I am aware that that sort of magic is impossible.
I want the computer to be able to use the correct font when some program specifies the text is in that language (perhaps by the locale, or perhaps by the other magic method niconico uses), but I want the english locale to use iYaHei for everything just to be safe because it works.
I have tried doing something like:
<match>
<test name="lang" compare="contains">
<string>en</string>
</test>
<test name="family">
<string>sans-serif</string>
</test>
<edit name="family" mode="prepend" binding="same">
<string>DejaVu Sans</string>
<string>iYaHei</string>
<string>Malgun Gothic</string>
<string>Kartika</string>
<string>Arial Unicode MS</string>
</edit>
</match>
Hoping that that would allow iYaHei to be used as the main fallback while using an English locale, but that doesn't happen; it still uses Meiryo for Kana, but it uses WenQuanYi Zen Hei for Chinese, which looks even worse than Meiryo and iYaHei.
So now I'm debating whether to use the new configurations or the old configurations. The new configurations fix a few things in Japanese locales, but cause a possibly grave problem in non-Japanese locales. I'm sure it has a few other effects too, but I didn't notice them yet.
What exactly is this fontconfig format supposed to do? Because it clearly isn't "setting a priority list for fonts to be used in a certain language". It seems more like "setting a priority list for fonts that are labelled as a certain language". Or maybe it just does some random stuff to the fonts described. I can't tell.
The core of the problem seems to be in the fonts themselves. Somehow, these fonts are claiming to cover "Chinese, Japanese, Korean" etc etc and while they probably do that, most can only cover 1 language in an acceptable way. Where do the fonts store this information? Is there a way to edit this?
Last edited by sokuban (2009-12-04 19:24:41)

How to use getAncestor(n),getdescendant(child,child) in sql why they are use please give me example

how to use getAncestor(n),getdescendant(child,child) in sql why they are use please give me example
jitendra

Below are a few examples that demonstrate these hierarchyid data type methods. See the hierarchyid reference in the SQL Server Books Online for more examples of these and other methods (http://technet.microsoft.com/en-us/library/bb677193.aspx).
--return a child of this node ('/1/')
SELECT CAST('/' AS hierarchyid).GetDescendant(NULL, NULL).ToString();
--returns a child of this node greater than child node '/1/' ('/2/')
SELECT CAST('/' AS hierarchyid).GetDescendant(CAST('/1/' AS hierarchyid), NULL).ToString();
--returns a child of this node less than chold node '/1/' ('/0/')
SELECT CAST('/' AS hierarchyid).GetDescendant(NULL, CAST('/1/' AS hierarchyid)).ToString();
--returns a child of this node greater than child node '/1/' and less than child node '/2/' ('/1.1/')
SELECT CAST('/' AS hierarchyid).GetDescendant(CAST('/1/' AS hierarchyid), CAST('/2/' AS hierarchyid)).ToString();
--return ancestor of node 1 level up from this node ('/1/2/')
SELECT CAST('/1/2/3/' AS hierarchyid).GetAncestor(1).ToString();
--return ancestor of node 2 levels up from this node ('/1/')
SELECT CAST('/1/2/3/' AS hierarchyid).GetAncestor(2).ToString();
Dan Guzman, SQL Server MVP, http://www.dbdelta.com

Why we are using cluster tables mainly HR

can any plz tell me why we are using cluster tables mainly HR???

Nice question -
Am making my guess based on whatever little I know;
PCLn are file systems and not a traditional (RDBMS) table system. Each File may contain one or more data clusters. SAP Defines data clusters thus -
A data cluster is a grouping of several data objects. Elementary fields, field strings and internal tables can be grouped in a data cluster.
Example:Clusters B1 and B2 in files PCL1 and PCL2 are relevant to time evaluation, as is cluster PS, which stores the generated schema.
The reason why file system of data storage is used instead of DB Table system may be for the purpose of storing voluminous data (Payroll and time) and ease of retrieval during processing (RDBMS may hv tough time in this). Also probably because of SAP's origin from Mainframe.
Why data clusters are used -? Probably data clusters are an offshoot (or part) of File system
Pls feel free to contradict the above. Actually DB experts can throw more light on this..
Regards
Chandra

Why 2 floating IP's are used to Weblogic managed servers?

Similar Messages

Maybe you are looking for