-
-
You can download a perl script I wrote to replace the work of iscsi-ls here.
From the linux server:
wget http://oracle.3dub.com/iscsi-ls
chmod 755 iscsi-ls
./iscsi-lsTags: iscsi-ls red hat EL5, iscsi-ls redhat, iscsi-ls RHEL 5
-
- tmpfs needs to be expanded: /etc/fstab put size=4G in place of ‘default’ for the tmpfs. This is necessary to get isci to start! I beat myself up for 2 days on this
- openfiler host/netmask is somewhat misleading. Use the EXACT rac nodes IP addresses and the netmask 255.255.255.255 (specifying the host explicitly).
- ocfs2 rpms need to match ‘uname -a’ exactly or it wil NOT work. It will install without warning, but will NOT work.
- iscsi-ls is not available for RHEL5. You can run ocfs2console and it will show you the scsi ID on the local machine that the disks are attached to. You can then mount them manually.
- java not installed to /usr/local/java/bin, but at /bin/java/jre/1.4.2_06/bin so I had to symlink like so
cd /usr/local
mkdir java
cd java
ln -s /bin/java/jre/1.4.2_06/bin - vipca and srvctl have a wierd bug where it uses LS_ASSUME_KERNEL. They need to be commented out as explained in http://cs.felk.cvut.cz/10gr2/relnotes.102/b15659/toc.htm#CJABAIIF
- tmpfs needs to be expanded: /etc/fstab put size=4G in place of ‘default’ for the tmpfs. This is necessary to get isci to start! I beat myself up for 2 days on this
-
Guide at : http://www.mscs.dal.ca/~selinger/ups/backups.html
you will need to change the following line if you use dual NIC’s (as in the case of our RAC systems) in /usr/etc/upsd.conf
LISTEN localhost
-
Download from here: http://www.depicus.com/download.aspx?product=gui
-
OCFS2:
OCFS2 tools/console:
-
1. emca -config dbcontrol db -repos create
2. by default the sys user is NOT in the remote password file and will NOT be able to login via OEM. Add the user:
$ORACLE_HOME/bin/orapwd file=$ORACLE_HOME/dbs/orapw<INST> password=hullahoop -
select privname,object_name from ku$_objgrant_view join all_objects on object_id=obj_num where grantee=’THE_USER’;
-
If the server crashes while there are active nxclient connections nxclient will continue to detect the stale connection and will not allow you to connect. The easy way to get around this is by stopping the statistics daemon for NX:
/usr/NX/bin/nxstat stop
It is a good idea to restart it after you successfully login:
/usr/NX/bin/nxstat start
-
select group#,l.sequence#,archived,l.status running, lf.status os_status,member from v$log l join v$logfile lf using(group#) order by group#;
NOTES:
- The STATUS column of v$logfile (os_status) is NOT refreshed in real-time. Information is updated on log switch.
Category:
- Administration
- APPS
- Articles
- Backup & Recovery
- Certification
- Configuration
- DATABASE STRUCTURE
- DATAGUARD
- Documentation
- EBS
- Enterprise Manager
- FLASHBACK
- HACKING
- HEADS UP
- Installation
- JDeveloper
- Linux
- NXclient/server
- Online Resources
- PERL
- PERSONAL
- python
- RMAN
- Routing & Switching
- Scripts
- Serial-Parallel-Port-Testers
- SNMP
- Software
- Solaris
- Spiritual
- SQL & PL/SQL
- SQLPLUS
- Startup/Shutdown
- Streams
- Tuning
- Version Control
- VMWare
- VNC
- VPN
- Windows XP





