Recital 10 enhances the SQL optimizer. Now, production indexes with a FOR <conditions> will be used to optimize SQL SELECT statements. If a WHERE <condition> on a SELECT statement matches a FOR <condition> on an index tag, this index will be used to optimize the query. The WHERE <condition> must be an exact match with the FOR <condition>. For example;
USE accounts INDEX on account_no TAG outstanding FOR balance > 0 EXPLAIN SELECT * FROM accounts WHERE balance > 0 Optimized using for condition on tag 'OUTSTANDING'
A good article describing the configuration of samba for file/record locking can be found here.
You can further tune samba by following this guide.
Basically you must add these directives to the smb.conf file:
[data] oplocks = False level2 oplocks = False
The default oplock type is Level1. Level2 oplocks are enabled on a per-share basis in the smb.conf file. Alternately, you could disable oplocks on a per-file basis within the share:
veto oplock files = /*.dbf/*.DBF/*.ndx/*.NDX/*.dbx/*.DBX/*.dbt/*.DBT/
You can further tune samba by following this guide.
If you specify the Common Internet File System (CIFS) when you mount the samba share then you must specify the following options
mount -t cifs {mount-point} -o username=name,pass=pass,directio
The directio option is used to not do inode data caching on files opened on this mount. This precludes mmaping files on this mount. In some cases with fast networks and little or no caching benefits on the client (e.g. when the application is doing large sequential reads bigger than page size without rereading the same data) this can provide better performance than the default behavior which caches reads (readahead) and writes (writebehind) through the local Linux client pagecache if oplock (caching token) is granted and held. Note that direct allows write operations larger than page size to be sent to the server.
If you get the following error when trying to mount the {mount-point}
Apr 22 16:57:39 bailey kernel: Status code returned 0xc000006d NT_STATUS_LOGON_FAILURE Apr 22 16:57:39 bailey kernel: CIFS VFS: Send error in SessSetup = -13 Apr 22 16:57:39 bailey kernel: CIFS VFS: cifs_mount failed w/return code = -13The you need to create the Samba user specified on the mount command
smbpasswd -a usernameFYI - Make sure you umount all the Samba {mount-point(s)} before shutting down Samba.
- For building shared libraries on the MAC the following need to be set
-
- The shared library file extension should be .dylib
- The compile flag is -dynamic
- For accessing the shared libraries at runtime
-
- DYLD_LIBRARY_PATH needs to be set to the location of the shared libraries
- Useful utilities for shared library support
-
- The following command will display the table of contents of the dynamically linked library
otool -TV sharedlibraryfile.dylib
In Recital 10, you can declare anonymous classes and call anonymous methods in these classes.
// declare some simple procedures
proc display(cArg)
echo "display=" + cArg
endproc
proc show(cArg)
echo "show=" + cArg
endproc
// create an object based on an anonymous class
myobj = new object()
// add some properties
myobj["name"] = "barry"
myobj["company"] = "recital"
// now declare an anonymous method
myobj["mymethod"] = display
// call the method
myobj.mymethod("hello world") // displays "display=hello world"
// redeclare the method
myobj["mymethod"] = show
// call the method
myobj.mymethod("hello world") // displays "show=hello world"
Where this becomes particularly useful is when you have a procedure that calls anonymous methods in order to process data. This technique can be used to call anonymous procedures in your code.
proc processdata(oArg)
oArg.mymethod(oArg.name)
endproc
proc show(cArg)
echo "show=" + cArg
endproc
myobj = new object()
myobj["name"] = "barry"
myobj["mymethod"] = show
processdata(myobj) // displays "show=barry"
If you use Eclipse Ganymede with large projects on linux you may run out of memory. To prevent this happening, you can specify the amount of memory to be allocated to Eclipse in the eclipse.ini file which is located in the eclipse directory.
Specifying this seems to reslove the problem:
Specifying this seems to reslove the problem:
-Xmx512m -XX:MaxPermSize=512m
A number of people have complained about data loss after a system crash when using Ext4. A bug report was logged. According to multiple posts by a developer of Ext4, Theodore Tso, this is due to differences in approach to security and performance issues between the two file systems.
Ext3 commits writes to disk within approximately 5 seconds - Ext4 can take from 40-150 seconds. In addition, if a system is using Ext3 and crashes before the commit takes place you will still have the previous contents of a file where under Ext4 the file will be empty. Theodore Tso feels that this is a failure at the application level and that the file system is behaving as designed and as specified by the POSIX spec (which apparently does not specify what is supposed to happen when a system is not shut down cleanly). His solution to the issue is to suggest proper use of fsync() and lists various scenarios/examples in post 54 of the bug report (linked above). In addition he wrote a patch that recognize the rename() situation mentioned in his post 54 yet retains the normal Ext4 behaviors and performance in the majority of cases. Also a more "proper" solution has been provided which allows the behavior of Ext3 to be retained under Ext4 by mounting it with alloc_on_commit.
A somewhat related topic is the use of on-board caching by hard drives. This behavior can be modified on most drives by using hdparm.
Ext3 commits writes to disk within approximately 5 seconds - Ext4 can take from 40-150 seconds. In addition, if a system is using Ext3 and crashes before the commit takes place you will still have the previous contents of a file where under Ext4 the file will be empty. Theodore Tso feels that this is a failure at the application level and that the file system is behaving as designed and as specified by the POSIX spec (which apparently does not specify what is supposed to happen when a system is not shut down cleanly). His solution to the issue is to suggest proper use of fsync() and lists various scenarios/examples in post 54 of the bug report (linked above). In addition he wrote a patch that recognize the rename() situation mentioned in his post 54 yet retains the normal Ext4 behaviors and performance in the majority of cases. Also a more "proper" solution has been provided which allows the behavior of Ext3 to be retained under Ext4 by mounting it with alloc_on_commit.
A somewhat related topic is the use of on-board caching by hard drives. This behavior can be modified on most drives by using hdparm.
Occasionally as a Linux administrator you will be in the situation where working on a remote server and you are left with no option but to force a reboot the system. This may be for a number of reasons, but where I have found it most frequent is when working on Linux clusters in a remote location.
When the "reboot" or "shutdown" commands are executed daemons are gracefully stopped and storage volumes unmounted.
This is usually accomplished via scripts in the /etc/init.d directory which will wait for each daemon to shut down gracefully before proceeding on to the next one. This is where a situation can develop where your Linux server fails to shutdown cleanly leaving you unable to administer the system until it is inspected locally. This is obviously not ideal so the answer is to force a reboot on the system where you can guarantee that the system will power cycle and come back up. The method will not unmount file systems nor sync delayed disk writes, so use this at your own discretion.
To force the kernel to reboot the system we will be making use of the magic SysRq key.
The magic_SysRq_key provides a means to send low level instructions directly to the kernel via the /proc virtual file system.
To enable the use of the magic SysRq option type the following at the command prompt:
echo 1 > /proc/sys/kernel/sysrq
Then to reboot the machine simply enter the following:
echo b > /proc/sysrq-trigger
Voilà! Your system will instantly reboot.
{linkr:related;keywords:linux;limit:5;title:Related Articles}
{linkr:bookmarks;size:small;text:nn;separator:%20;badges:2,1,18,13,19,15,17,12}
Each Recital table can have one or more data dictionaries to provide a central repository for constraints and other metadata.
Here's how to set up field validation for a field with a small static number of acceptable values.
Using the example.dbf table from the southwind sample database, validation can be added to the title field to ensure it matches one of a list values.
If you have access to the Recital Workbench, you can use the modify structure worksurface to add and alter your dictionary entries, including a customized error message if required.
Here's how to set up field validation for a field with a small static number of acceptable values.
Using the example.dbf table from the southwind sample database, validation can be added to the title field to ensure it matches one of a list values.
open database southwindThe inlist() function checks whether the specified expression exists in the comma-separated list which follows. An attempt to update title with a value not in the list will give an error: Validation on field 'TITLE' failed.
alter table example add constraint;
(title set check inlist(alltrim(title),"Miss","Mr","Mrs","Ms"))
If you have access to the Recital Workbench, you can use the modify structure worksurface to add and alter your dictionary entries, including a customized error message if required.
Recital 10 enhances the APPEND FROM command. The enhancement added the following syntax ;
APPEND FROM TYPE CSV <file-name.csv>The TYPE keyword has now been enhanced to support a comma separated values (CSV) format