[SOLVED] setting up chroot -- wrong architecture errors using pacman

Hi,
I'm following the WIKI to reinstall my chroot after a reinstall. I've done this 10s of times and am quite familiar with the process. I'm running into a problem with pacman syncing x86_64 packages for some reason. I'll paste what I have and wait for someone to tell me what would help...
# sed -e 's/\$arch/i686/g' /etc/pacman.d/mirrorlist > /opt/arch32/mirrorlist
# sed -e 's@/etc/pacman.d/mirrorlist@/opt/arch32/mirrorlist@g' /etc/pacman.conf > /opt/arch32/pacman.conf
# nano /opt/arch32/pacman.conf (change 'Architecture = auto' -> 'Architecture = i686'). Just to be sure:
# cat /opt/arch32/pacman.conf | grep Arch
Architecture = i686
# cat /opt/arch32 (extracted the one uncommented mirror here:)
Server = http://mirror.rit.edu/archlinux/$repo/os/i686
# mkdir -p /opt/arch32/var/{cache/pacman/pkg,lib/pacman}
# pacman --root /opt/arch32 --cachedir /opt/arch32/var/cache/pacman/pkg --config /opt/arch32/pacman.conf -Sy
:: Synchronizing package databases...
core 37.3K 2.1M/s 00:00:00 [###########################] 100%
extra 468.5K 2.0M/s 00:00:00 [###########################] 100%
community 427.9K 2012.6K/s 00:00:00 [###########################] 100%
But already, look at this (just picked a random package in core toward the top of the list):
]# cat /opt/arch32/var/lib/pacman/sync/core/acl-2.2.49-2/desc
%FILENAME%
acl-2.2.49-2-x86_64.pkg.tar.xz
And sure enough, the next step in the wiki fails:
# pacman --root /opt/arch32 --cachedir /opt/arch32/var/cache/pacman/pkg --config /opt/arch32/pacman.conf -S base base-devel
:: package [pkg-name]-x86_64 does not have a valid architecture
I've tried using
# linux32 pacman --root /opt/arch32 --cachedir /opt/arch32/var/cache/pacman/pkg --config /opt/arch32/pacman.conf -Sy
and
# pacman --root /opt/arch32 --cachedir /opt/arch32/var/cache/pacman/pkg --config /opt/arch32/pacman.conf --arch i686 -Sy
with no success.
The other posts (HERE, HERE, and HERE) seem to be about using pacman from inside the chroot or forgetting the 'Architecture = i686' bit in pacman.conf. I clearly have that.
Why is pacman still getting x86_64 packages?
Last edited by jwhendy (2011-03-23 18:20:21)

Found the issue. I never recall this happening before, but apparently there was a line in my mirrorlist from my install last night like this:
# Mirror used during installation
Server = http://mirror.rit.edu/archlinux/$repo/os/x86_64
Is that a new feature? Since the 'sed' command in the chroot wiki looks for '$arch' and makes it 'i686', this line got missed. I added a note about this in the WIKI (see note 3b).
Perhaps I didn't notice it or have this issue the last time since I installed with btrfs and did everything manually?

Similar Messages

[Solve]Set ibus as default IM when use other english language in gnome

Hello everybody,
I am satisfied with Arch + Gnome3 + Ibus in English language. With English language, Ibus is already set to default system-input-methods as well as I expected.
However, today I change language from english to vietnamese, System-input-methods change to other method, not Ibus.
I looked for programs to change input method such as im-settings, im-chooser and im-switch, but it's not exist in arch package or aur.
I add "export GTK_IM_MODULE="ibus"" to file ~/.bashrc , I can use Ibus only in firefox, libreoffice. To use Ibus with other programs as gedit, gnome-terminal, I must manual change input method by right click and choose Ibus.
So, How can I set Ibus as default system IM?
By the way, I can use Ibus in all Qt applications without trouble after set in qtconfig.
Thanks for reading.
Last edited by s8dtvt (2011-06-25 01:56:18)

Well, I set: LOCALE="vi_VN.UTF-8" instead of LOCALE="en_US.UTF-8" by default in file "/etc/rc.conf" and it's solved.

Wrong Architecture

I have 4 mac's. A iMac 20, a iMac 24, a MacBook Pro 17 and a Mac Pro.
I updated my Mac's to Leopard. In the iMac 20, a iMac 24 and in the MacBook Pro 17 it works fine (I make a update install).
In the Mac Pro I installed a new hard drive and made a fresh install of Leopard. It works fine (in the surface...) and I migrate my account from another partition.
I begin to install my applications in Leopard. And the errors start to appear.
When I try to run Rise of Nations, I get a library error (wrong architecture). I found in the foruns how to workaround it and it works and let me start the application. But the game has all the menus with wrong fonts and colors and when I try to run the game it cancels (with the wrong architecture again).
The first errors occurs with lastest games from EA (C&C 3 and Battlefield 2142). When I try to start it, I get the wrong architecture error. I cannot start anything. Using the terminal and acessing cider directly it starts but all the video go down with wrong architecture error).
Reading the logs I find that the VideoOutput from Quicktime is the application that generates the "wrong architecture error".
I think that the error is because the migration brings some incorrect libraries from my older user. And I make a test, with another hard drive and another fresh install. But in this case I create a new account (with no relations with the old one) and started Leopard. With the fresh user, I try to install Battlefield (I try C&C and Rise of Nations) and the error was the same. With the results, my migration from my older account is not responsible for the wrong architecture message.
In other side, the machines that I migrate runs fine the same games that do not work in the fresh install.
Maybe the fresh install copy wrong libraries (the PPC libraries instead of Intel) ?

Hi all.
After hours installing many ways the Leopard, I found that the error in running my games (and in the libcryto.0.9.dylib in some posts here) exists only in case sensitive filesystems.
When you install 10.5 in a filesystem without case sensitive all works fine.
In my other systems I just upgrade a factory installation (and in this case the filesystem is not case sensitive) and this is why they are working fine.
Now I'm cleaning my installs (now I have 6 installation in 4 disks) and make a final install without case sensitive filesystem.
I hope to help someone with this.
Thank you to all.

Updated the wrong Architecture....[solved]

Hello, I am apparently retarded, but I was having a lot of problems today with pacman.
I ended up updated for the wrong architecture, I was supposed to update for i686 and I updated for x86_64.
Now I can't even use commands in the terminal because I just get a bash error saying I cannot execute the binary files. I cannot even open a terminal anymore actually after I closed it.
Am I screwed? Any help would be appreciated, but it sure seems like I am.
Last edited by efourseven (2011-03-19 01:21:40)

Thanks man, That's what I plan to try next, I just wanted to see if anyone had any tips before I reboot and go for it.
I have my /home directory on a separate drive, so I am thinking I might just go for a fresh install. I will try to get in somehow from the live cd and fix my pacman.conf file and try to upgrade again to the correct architecture (thats the only thing I can think of)... but I have no idea how that will work out.

Error message: The preset used by one or more sequences in this project requires third-party components that could not be located. These sequences will be modified to use a custom sequence setting instead. To continue editing using the original preset, qu

Hello all,
I can't open a PP project without this error message appearing:
"The preset used by one or more sequences in this project requires third-party components that could not be located. These sequences will be modified to use a custom sequence setting instead. To continue editing using the original preset, quit the application without saving the project, reinstall any third-party components that are required and reopen the project".
What I had been doing before this occurred was editing a PP project using ProRes sequences and multi-camera editing. Multi-camera was not working very well and after a few attempts to fix that I gave up and finished the job cutting 3 layers of video instead. AME refused to render the three finished sequences of over 1hr duration so rendered final videos from the PP sequences.
I tried deleting plists for AME, PP and QT, repairing disc permissions and rebooting but there was no improvement.
Client is satisfied for now but will want to come back to this job at a later date so decided to uninstall and re-install AME and PP in the hope both would be ready to work properly when needed again. AME works fine in conjunction with a different PP project but now I can't open the PP project in question without the above error message appearing.
Having searched the net it would seem that the problem may be caused by ProRes but I haven't been able to find a definitive solution for the problem. Does anyone know either:
1. How to fix this?
2. If I chose the "modified using a custom sequence setting", can I be sure that my sequences will look the same even if they don't use ProRes? The final deliverable format will be mp4 so as long as the overall look doesn't change then I can afford a change in the edit codec. It's just that with three sequences over an hour long, I don't want days of work to be ruined.
Other older PP projects of mine open and work fine.
Premiere Pro CC 8.2.0
Media Encoder CC 8.2.0.54
OSX 10.10.2
Any help would be greatly appreciated.
Duncan.

Adobe web chat come up with a solution.
Open the PP project.
Export your sequence or project as a Final Cut Pro XML file.
Set up a new PP project.
Import the Final Cut Pro XML file.
This will get you back up an running.
Since getting back to editing this project I have found that some things will be lost or change in using XML:
You will lose position key framing, black video clips, dissolve fx, audio levels, audio dynamic fx, the ability to open a multi camera clip and change the camera view.
You will keep cut points, crop fx.
Not perfect but only took a couple of hours to save 3 days work.

FATAL ERROR in native method: Wrong method ID used to invoke a Java method

When calling the same method second time , I get message ::
<FATAL ERROR in native method: Wrong method ID used to invoke a Java method>
void myFunction(int myVal)
JNIEnv *env = NULL;
jclass odbcconnls;
jint res;
printf("\nInitilaizing class ");
res = (jvm)->AttachCurrentThread((void **)&env,NULL);
if (res < 0) {
fprintf(stderr, "Can't get Env \n");
(jvm)->DestroyJavaVM();
return SQL_ERROR;
if(res == JNI_OK)
printf("\nThe env is initialized ");
if(*(&env) == NULL)
printf(" the env is NULL ");
printf("\nenv :::::: %s ", env);
// the jobject (dbc->actualConn) is a global reference
odbcconnls = (env)->GetObjectClass(dbc->actualConn);
if (odbcconnls == NULL) {
goto destroy;
switch(myVal){
case 1:
jmethodID methodId ;
jboolean jbool;
SQLINTEGER Val = (SQLINTEGER )Value;
SQLINTEGER val1 = *Val;
methodId = (env)->GetMethodID( odbcconnls,"myFun1","(Z)V");
if(methodId == NULL){
goto destroy;
if(val1 == SQL_FALSE )
jbool = 0;
else
jbool =1;
env->CallVoidMethod(dbc->actualConn,methodId,jbool);
env->DeleteLocalRef((jobject)res);
env->DeleteLocalRef((jobject)odbcconnls);
env->DeleteLocalRef((jobject)methodId);
jvm->DetachCurrentThread();
return ;
case 2 :
jmethodID methodId1 ;
SQLUINTEGER* Level;
methodId1 = (env)->GetMethodID( odbcconnls,"myFun2","(I)V");
if(methodId1 == NULL){
goto destroy;
Level = (SQLUINTEGER *)Value;
env->CallVoidMethod(dbc->actualConn,methodId1,(int)*Level);
dbc->txn_isolation = (SQLUINTEGER)Value;
env->DeleteLocalRef((jobject)res);
env->DeleteLocalRef((jobject)odbcconnls);
env->DeleteLocalRef((jobject)methodId1);
jvm->DetachCurrentThread();
return ;
case 3 :
SQLCHAR* Cate;
jmethodID methodId2 ;
jstring jStrCat;
methodId2 = (env)->GetMethodID(odbcconnls,"myFun3","(Ljava/lang/String;)V");
if(methodId2 == NULL){
goto destroy;
Cate = new SQLCHAR[20];
strcpy((char *)CCatalog,(char *)Value);
jStrCat = env->NewStringUTF((char *) Cate);
printf("\n got jSTring ");
env->CallVoidMethod(dbc->actualConn,methodId2,jStrCat);
printf("\n after called method ");
int len = strlen((char *)Cate);
dbc->Cate = new SQLCHAR[len+1];
strcpy((char *)dbc->Cate,(char *)Cate);
printf("\n copied result ");
env->DeleteLocalRef((jobject)res);
env->DeleteLocalRef((jobject)odbcconnls);
env->DeleteLocalRef((jobject)methodId2);
jvm->DetachCurrentThread();
return ;
destroy:
if ((env)->ExceptionOccurred()) {
(env)->ExceptionDescribe();
jvm->DetachCurrentThread();
(jvm)->DestroyJavaVM();
return SQL_ERROR;
When case 1 is called second time this error is thrown..
plz help me..
Thanx
MittalSunita.

When calling the same method second time , I get
message ::
<FATAL ERROR in native method: Wrong method ID used
d to invoke a Java method>
void myFunction(int myVal)
JNIEnv *env = NULL;
jclass odbcconnls;
jint res;
printf("\nInitilaizing class ");
res = (jvm)->AttachCurrentThread((void **)&env,NULL);
if (res < 0) {
fprintf(stderr, "Can't get Env \n");
(jvm)->DestroyJavaVM();
return SQL_ERROR;
if(res == JNI_OK)
printf("\nThe env is initialized ");
if(*(&env) == NULL)
printf(" the env is NULL ");
printf("\nenv :::::: %s ", env);
// the jobject (dbc->actualConn) is a global
reference
odbcconnls = (env)->GetObjectClass(dbc->actualConn);
if (odbcconnls == NULL) {
goto destroy;
switch(myVal){
case 1:
jmethodID methodId ;
jboolean jbool;
SQLINTEGER Val = (SQLINTEGER )Value;
SQLINTEGER val1 = *Val;
methodId = (env)->GetMethodID(
( odbcconnls,"myFun1","(Z)V");
if(methodId == NULL){
goto destroy;
if(val1 == SQL_FALSE )
jbool = 0;
else
jbool =1;
env->CallVoidMethod(dbc->actualConn,methodId,jbool);
env->DeleteLocalRef((jobject)res);
env->DeleteLocalRef((jobject)odbcconnls);
env->DeleteLocalRef((jobject)methodId);
jvm->DetachCurrentThread();
return ;Why do you delete a local reference???
Did you ever call the get local reference?
case 2 :
jmethodID methodId1 ;
SQLUINTEGER* Level;
methodId1 = (env)->GetMethodID(
( odbcconnls,"myFun2","(I)V");
if(methodId1 == NULL){
goto destroy;
Level = (SQLUINTEGER *)Value;
env->CallVoidMethod(dbc->actualConn,methodId1,(int)*Le
el);
dbc->txn_isolation = (SQLUINTEGER)Value;
env->DeleteLocalRef((jobject)res);
env->DeleteLocalRef((jobject)odbcconnls);
env->DeleteLocalRef((jobject)methodId1);
jvm->DetachCurrentThread();
return ;
case 3 :
SQLCHAR* Cate;
jmethodID methodId2 ;
jstring jStrCat;
methodId2 =
(env)->GetMethodID(odbcconnls,"myFun3","(Ljava/lang/St
ing;)V");
if(methodId2 == NULL){
goto destroy;
Cate = new SQLCHAR[20];
strcpy((char *)CCatalog,(char *)Value);
jStrCat = env->NewStringUTF((char *) Cate);
printf("\n got jSTring ");
env->CallVoidMethod(dbc->actualConn,methodId2,jStrCat)
printf("\n after called method ");
int len = strlen((char *)Cate);
dbc->Cate = new SQLCHAR[len+1];
strcpy((char *)dbc->Cate,(char *)Cate);
printf("\n copied result ");
env->DeleteLocalRef((jobject)res);
env->DeleteLocalRef((jobject)odbcconnls);
env->DeleteLocalRef((jobject)methodId2);
jvm->DetachCurrentThread();
return ;
destroy:
if ((env)->ExceptionOccurred()) {
(env)->ExceptionDescribe();
jvm->DetachCurrentThread();
(jvm)->DestroyJavaVM();
return SQL_ERROR;
When case 1 is called second time this error is
thrown..
plz help me..
Thanx
MittalSunita.

I am having problems transefering contacts to my I pod when syncing the message comes back with contact in use by I pod please close and try again, please could you let me know how to close contacts on the pod or is there a setting that is wrong etc,

I am having problems transfering contacts to my I pod when syncing the message comes back with contacts in use by I pod please close and try again, please could you let me know how to close contacts on the pod or is there a setting that is wrong etc, i have restored the device. if i go to contact via settings the screen comes on for about 2 seconds and then goes off - Help

Single click on the Home button. Then double click on the Home button. In the recently used row, touch and hold the Contact app till they start to wiggle and then tap the minus sing by the contacts app.
Also make sure the Contact program on your computer is closed.

[SOLVED] Error message every time using pacman

Hi.
I was installing foomatic-db when my laptop died. Since then every time I use pacman i get this message
error: could not open file /var/lib/pacman/local/foomatic-db-1:4.0.15_20120406-1/desc: No such file or directory
I can still use pacman without problems but is there a way to get rid of the error-message coming every time?
Last edited by Jompa89 (2012-06-11 15:34:28)

When I try to reinstall I get a lot of "exists in filesystem"-messages. This is the bottom part of the list because it goes on for long.
foomatic-db: /usr/share/foomatic/db/source/printer/Xerox-WorkCentre_XK35c.xml exists in filesystem
foomatic-db: /usr/share/foomatic/xmlschema/driver.xsd exists in filesystem
foomatic-db: /usr/share/foomatic/xmlschema/option.xsd exists in filesystem
foomatic-db: /usr/share/foomatic/xmlschema/printer.xsd exists in filesystem
foomatic-db: /usr/share/foomatic/xmlschema/types.xsd exists in filesystem
foomatic-db: /usr/share/licenses/foomatic-db/COPYING exists in filesystem
Errors occurred, no packages were upgraded.
Last edited by Jompa89 (2012-06-11 15:19:29)

Warnings/Errors using getClass().newInstance()

I'm getting some warnings & errors using getClass().newInstance() on an object. I can't however see what's wrong.
First a little background, Bar is a class that manages a set of beans, including creation. One of the actions is to create a special copy of bean, the method that does this does some bookkeeping (such that a particular instance of Bar will only work with objects it created).
This class was originally written in 1.4 and I'm now trying to genericify it.
First attempt:
public class Bar<T extends Foo> {
    public T getFlaggedBean(T fromBean) throws InstantiationException, IllegalAccessException {
        T newBean = fromBean.getClass().newInstance();
        // do some stuff
        return newBean;
}This yielded: "Type mismatch: cannot convert from capture#1-of ? extends Foo to T"
I can cast:
        T newBean = (T) fromBean.getClass().newInstance();Which drops me to a warning a can suppress: "Type safety: Unchecked cast from capture#1-of ? extends Foo to T".
The code runs fine like that, but I'd rather not even have a warning to suppress, I don't like suppress them unless I know that it really is safe.
The code above looks perfectly reasonable to me, I can't figure out why it could be problematic, which leads me to wonder if I'm missing something. We've got a parameter that is of type T or a subclass of it, calling getClass().newInstance() creates an new object such that origobj.getClass() == newobj.getClass() *. Since that class is T or a subclass of it, it should be assignable to T.
The code above when, compiled and subjected to type erasure, effectively produces a method that looks like this:
    public Foo getFlaggedBean(Foo fromBean) throws InstantiationException, IllegalAccessException {
        Foo newBean = (Foo) fromBean.getClass().newInstance();
        // do some stuff
        return newBean;
    }Note that the cast to Foo is strictly unnecessary.
There's a couple ways I could solve this, such as:
    public <S extends T> T getFlaggedBean(Class<S> useClass, S fromBean) throws InstantiationException, IllegalAccessException {
        T newBean = useClass.newInstance();
        // do some stuff
        return newBean;
    }I'd rather not change the API, though. Can anybody give me an idea where the initial code, with warnings could be unsafe?
Neil
* Unless you're mucking about with class loaders, but that problem would be an issue with non-generic code too.

Paul,
So does that bit of the JLS mean that my compiler is doing something wrong when it finds the getClass() call, in the code below, to return Class<? extends Foo> rather than Class<? extends T>? If the latter were returned it would compile fine.
public class Bar<T extends Foo> {
    public T getFlaggedBean(T fromBean) throws InstantiationException, IllegalAccessException {
        T newBean = fromBean.getClass().newInstance(); // doesn't compile
        // do some stuff
        return newBean;
}I was initially distracted by the "capture#1-of" and thought that was culprit, but as the code below demonstrates calling getInstance() on an instance of a class the compiler notes as "Class<capture#1-of ? extends Foo>" assigns to Foo just fine. Interestingly in this code my IDE says that getClass() is returning "Class<? extends Foo>", both javac and the IDE spat out "Class<capture#1-of ? extends Foo>" when the 1st line of the method had a warning or error. I assume that "Class<? extends Foo>" and "Class<capture#n-of ? extends Foo>" are equivalent and that "capture#n-of" is something fairly internal to the compiler. Can anyone shed light on that?
public class Bar<T extends Foo> {
    public Foo getFlaggedBean(T fromBean) throws InstantiationException, IllegalAccessException {
        Foo newBean = fromBean.getClass().newInstance(); // this works
        // do some stuff
        return newBean;
}

Blackberry Z10,post security wipe...BBM Not working..Showcasing error Unable to complete Set Up and Temporary Server Error"

I would request you to please help fixing the BBM error - Unable to complete Set Up and Temporary Server Error”.
I am using a BB Z10 with Software version 10.1.0.4181 and OS version 10.1.0.4633
Already tried resetting device to Factory Settings and Security Wipe...
Please help.

Hey KaranPrabhakar,
Welcome to the BlackBerry Support Community Forums.
Thanks for the question.
Follow the steps in this KB article to resolve the issue: www.blackberry.com/btsc/KB33780
Let me know if you have any more questions.
Cheers.
-ViciousFerret
Come follow your BlackBerry Technical Team on Twitter! @BlackBerryHelp
Be sure to click Like! for those who have helped you.
Click Accept as Solution for posts that have solved your issue(s)!

SQL Error using cachedrowSet.populate.

Hi,
I am trying to use cachedrowset to pass my result set into my jsp.
I use request.setAttribute and pass my cachedrowset and in my jsp I use <jsp:usebean> tag to retrieve it.
This usually works for me. However, I have one query which when I am trying to populate the cachedrowset with its result set (that is cr2 in code below )
rs2 = stmt.executeQuery(semSql);
rs3 = stmt.executeQuery(countSql);
crs2 = new CachedRowSet();
crs3 = new CachedRowSet();
crs2.populate(rs2);
crs3.populate(rs3);
is throwing an sql error.
I ran the query in my query analyzer tool and it runs fine. But when I try to populate it in cachedrowset it throws sql error.
I have more queries in the same java which I am passing to the jsp as cachedrowset (for example cr3 in the above code) And they all are working fine. Only this particular one is causing the issue.
I was also looking at some other posts with the same issue. One post said if the data is a negative number then it can cause issue while trying to populate a CRS. However, in my case there is no negative data.
I am clueless. Please help !!

Apparently noone replied to this.
Well, this one is odd so cant blame you guys.
I finally found the answer after losing lots of my hair.
It looked like somewhere my connection was getting closed. I dont know why it was getting close at the first place because i didnt have any rs.close() or so statement in there.
One of the reasons that I have in mind is that after getting the results of my first query in a cachedrowset (CRS) , I moved the cursur of CRS and that may have caused some problem. But then also my second query was totally independant of my first query. And my third query which apparently was working fine and didnt thru an error on crs.populate was dependant on the first one,.
Finally, I rearraged the code, so that the operations i was doing on the first CRS was after I populate my second CRS. And that solved the problem.
Hope this solution helps somebody like me out there.

/usr/local/lib/libgfortran.3.dylib: mach-o, but wrong architecture

Hello every one,
I am an absolutely new mac user. I need to run a couple of softwares for my research but I faced errors that I really can not understand. There is an Unix executable and a .f Fortran file in each folder that I use. As I run the executable the following error appears:
dyld: Library not loaded: /usr/local/lib/libgfortran.3.dylib
Referenced from: /Users/somaiehbeladi/Desktop/100601stereo/combine_segments/combinesegs
Reason: no suitable image found. Did find:
/usr/local/lib/libgfortran.3.dylib: mach-o, but wrong architecture
/usr/local/lib/libgfortran.3.dylib: mach-o, but wrong architecture
Trace/BPT trap
logout
However, the libgfortran.3.dylib is already in the mentioned directoty and I used the File command on it which gives
Somaieh-Beladis-MacBook:lib somaiehbeladi$ File libgfortran.3.dylib
libgfortran.3.dylib: Mach-O dynamically linked shared library ppc
Then I tried the otool -L command
Somaieh-Beladis-MacBook:combine_segments somaiehbeladi$ otool -L combine_segs
-bash: otool: command not found
So, could you please let me know how I should find out the proper architecture for my executable and then how to change the architecture.
Many thanks in advance.

It looks as though you are using an executable compiled for an Intel processor but have the PowerPC versions of the gfortran libraries installed.
How did you install this software?
Which version of OS X are you using?
Ar you using a PowerPC-based Mac, or an Intel-based Mac?

Architecture for using MDB with a server cluster

          Hi,
          I would like to get some advice on the architecture that is the most desirable
          for a scenario that I have here.
          I have 3 machines, each running a managed server belonging to the same cluster.
          My admin server runs on a 4th machine. I have deployed my MDB on the cluster and
          I'm trying to find a way to configure the 3 + 1 machines to give the best end-to-end
          time for processing the JMS messages.
          To investigate this, I'm running some simple benchmark tests with a client application
          that sends JMS messages to the system. The client repeatedly sends messages to
          the MDB's queue and the MDB puts processed messages to another queue that the
          client is listening to.
          Now, my question is: where should I set the target for my JMS server (and hence
          my queues) and the connection factories? I can think of a few possiblities:
          1. Connection factories target: cluster; JMS server target: one of the servers
          in the cluster
          -> Potential drawback - The server with the JMS server will be handling and redistributing
          the JMS messages to other servers in the cluster. This means that a portion of
          its processing power is used to do this instead of actually having the MDB process
          the JMS message. (Please correct me if i'm wrong)
          2. Connection factories target: cluster; JMS server target: have one JMS server
          for each server in the cluster and make use of distributed destinations
          -> Potential drawback - My client establishes connection with the MDB's queue
          only once before it sends its messages to it. Probably as a result of this and
          the way WebLogic clusters load-balance themselves, all the messages end up being
          routed to the same server. This option appears to be out since 2 of the 3 servers
          are not utilized at all.
          3. Connection factories target: admin server; JMS server target: admin server
          -> Potential drawback - The MDB has to maintain a queue connection with a server
          that is not part of the cluster. (Again, please correct me if I'm wrong.) I'm
          not sure if this introduces extra time taken for the MDB to receive its messages
          and for it to send the processed messages to the queue.
          I'd appreciate it if someone could advice me on the most desirable architecture
          to use here. From my understanding of the problem, option 3 seems to be the answer,
          but I may be wrong. Perhaps there is no significant difference in terms of performance
          that 3 can give, compared to 1 and 2.
          One last question. I'd like to understand, in option 1, if the admin server plays
          any part in load-balancing the cluster. Are the JMS messages received on the cluster's
          JMS queue forwarded to the admin server before they are rerouted to the server
          that is supposed to process it?
          Cheers,
          C.Y.


3. Connection factories target: admin server; JMS server target: admin          server
          > -> Potential drawback - The MDB has to maintain a queue connection with a
          server
          > that is not part of the cluster. (Again, please correct me if I'm wrong.)
          I'm
          > not sure if this introduces extra time taken for the MDB to receive its
          messages
          > and for it to send the processed messages to the queue.
          Admin server is not supposed to participate in the cluster. I wouldn't
          deploy anything on the admin server.
          I think my personal preferene would be connection factories to the cluster
          and use distributed destinations.
          Regards...

How can I set up a custom 404 error page on OSX Server?

I moved my web site to a local server and changed the structure drastically. Unfortunately I am getting hits for information that was on the old server which I haven't put back yet. I'd like to set up a custom 404 Error Page to let people know what's up. In server.app I can set up a 500 error page but not a 404 and when I tried hand coding it into the sites .conf file I really messed things up. It took me a while to get that all corrected!
Thanks for any advice,
Bill W

Found it!
Under the web server Advanced Setting set "Allow overrides using .htaccess" then create an .htaccess file in the root directory with the line:
ErrorDocument 404 notfounderror.html
Use whatever HTML/PHP/etc. document you have created.

Set multi user column with PowerShell used to work in 2010 but no longer works in 2013

In SharePoint 2010 I used PowerShell to set the value of multi user people columns and it worked really well. I attempt to use the same PowerShell to set the same column type in SharePoint 2013 and it fails.
here is the PowerShell that I use in 2010:
$web = Get-SPWeb "http://intranet"
$list = $web.lists["TestList"]
$item = $list.items.add()
$item["Title"] = "Test multi user column"
$users = @("Domain\user1", "Domain\user2")
$userList = new-object Microsoft.SharePoint.SPFieldUserValueCollection
foreach($user in $users)
$spUser = $web.EnsureUser($user)
$userValue = new-object Microsoft.SharePoint.SPFieldUserValue($web, $spUser.ID, $spUser.Name)
$userList.Add($userValue)
$item["MultiUserColumn"] = $userList
$item.update()
I have used this on three SharePoint 2013 farms with differing results. On two of them I receive an error when running $item.update(): Exception calling "Update" with "0" argument(s): "Invalid look-up value. A look-up
field contains invalid data. Please check the value and try again."
If I take one of the users out of the $users list then it works fine, but it will not allow multiple users to be set with PowerShell. I can use the GUI to add more than one user but not PowerShell.
Does anyone know if these methods have changed in 2013? I haven't been able to find anyone else with this issue.
mmm... coffee...

Not sure but maybe something to do with casting. Below is the code snippet from one of the blogs. Try modifying your script like below and see if still you get the error
[Microsoft.SharePoint.SPFieldUserValueCollection]$lotsofpeople = New-Object Microsoft.SharePoint.SPFieldUserValueCollection
$user1 = $w.EnsureUser("domain\user1");
$user1Value = New-Object Microsoft.SharePoint.SPFieldUserValue($w, $user1.Id, $user1.LoginName)
$user2 = $w.EnsureUser("domain\user2");
$user2Value = New-Object Microsoft.SharePoint.SPFieldUserValue($w, $user2.Id, $user2.LoginName);
$lotsofpeople.Add($user1Value);
$lotsofpeople.Add($user2Value);
$i["lotsofpeoplefield"] = $lotsofpeople;
$i.Update();
#-or-
$l.Fields["lotsofpeoplefield"].ParseAndSetValue($i,$lotsofpeople);
$i.Update();
Reference to the link
http://social.technet.microsoft.com/wiki/contents/articles/20831.sharepoint-a-complete-guide-to-getting-and-setting-fields-using-powershell.aspx
Geetanjali Arora | My blogs |

[SOLVED] setting up chroot -- wrong architecture errors using pacman

Similar Messages

Maybe you are looking for