This is the beginning of administrator documentation.
- Upgrading your server from the pre-August 2005 version to the new version
- Upgrading Tomcat from version 5.x to 5.y
- Upgrading Tomcat from version 4.x to 5.y
- Upgrading MySQL
All users are automatically placed in an "All User Test Project" where they can test BioCoRE. It is expected, however, that they will soon start creating their own projects and doing their work in these new projects. New private projects are only visible to members of the projects.
BioCoRE and Java Web Start
- Control Panel: The Control Panel application is designed to always be running while your users are logged in. It is a Java application that can be launched via Java Web Start. We recommend that you have your users install Java Web Start for the Control Panel. It is a much nicer way to use the "CP". We have installation instructions at: http://www.ks.uiuc.edu/Research/biocore/installation.shtml You are welcome to make your own instructions as well.
We highly suggest that you implement a backup procedure for your BioCoRE data. Cron and some variation of the following script can be used as a starter, if you wish. This script is invoked with an argument that is the name of the database that you wish to backup. Look over carefully and replace and modify items where needed. We suggest that you make a database user that only has read access to the databases in question, and can only access the database from the IP where you are running the cron job. Note that database passwords are being given in the script below, so secure the script as much as you can.
#!/bin/sh COLLDBDIR= << Root biocore directory >> COPYDIR= << Destination where to copy things to >> DBUSER= << USER >> DBPASS= << PASSWORD >> DUMPDIR=$COLLDBDIR/dump MYSQLROOT=$COLLDBDIR/dbms/$1 DATADIR=$MYSQLROOT/data EXECDIR=$MYSQLROOT/mysql/bin # get the database info $EXECDIR/mysqldump --defaults-file=$MYSQLROOT/scripts/my.cnf \ -u$DBUSER -p$DBPASS --flush-logs --lock-tables --quick \ $1 | gzip -9 > $DUMPDIR/$1/$1Dump`date +%m%d`.gz /bin/cp -f $DUMPDIR/$1/$1Dump`date +%m%d`.gz $COPYDIR/$1 # get rid of the old backups after they have been here a WHILE. No need # to delete these things right away. find $DUMPDIR -name "$1Dump[0-9]*.gz" -type f -mtime +90 | xargs rm -f
Various database changes you might be requested to make:
SELECT * from userInfo where login = 'oldloginname';(replace 'oldloginname' with their old login name) Make sure that it looks correct. There should only be a single line returned. Then, do the following:
UPDATE userInfo set login = 'newloginname' WHERE login = 'oldloginname';Afterwards, you can do another SELECT if you wish to verify that the change was correct.
Running the BioCoRE server without the "/biocore/" in the pathYou might wish to run the BioCoRE server at simply:
http://biocore.yourmachine.edu/biocore/To do this, you need to make a few changes. In the
$TOMCAT_HOME/conf/Catalina/localhost/directory, name your file
biocore.xmland the Context line at the top of the file should look like:
< Context path="" docBase="ROOT" >Then, when uncompressing the WAR file in the
webapps/directory, don't put it in
biocore/. It should go in a directory called
ROOT/. Finally, in the admin configuration ( http://biocore.yourmachine.edu/nonsecure/admin/ ) be sure that the webapp base settings don't include the trailing
biocore/. Other than these minor changes, you should be fine. If you have any problems, let us know.
Logging and your BioCoRE serverYou can dynamically enable and disable detailed logging for your BioCoRE server. This will typically be because someone from the BioCoRE team has asked you to turn on detailed logging to help with debugging a problem, or you have started delving into the code and want more output to help discern what is happening with the server. You can enable this logging by going to the admin page and choosing the Logging option. You will be presented with a page that requires some knowledge of how the code is structured. You can choose particular java classes that you want detailed logging for, and the changes will take place as soon as you submit the changes to the BioCoRE server. This is particularly useful for repeatable problems. You can turn on the logging, recreate the problem, and turn the extraneous logging off.