Database Reference
In-Depth Information
+--------------------------------------------------------------+-----------+--------------+
| /u01/app/oracle/product/11.2.0.3/dbhome_1/install | INSTALL | INSTALL |
| Permission: public | | |
| Added by: root | | |
+--------------------------------------------------------------+-----------+--------------+
| /var/log | OS | OS |
| Permission: public | | |
| Added by: root | | |
'--------------------------------------------------------------+-----------+--------------'
The team behind TFA keeps adding and removing directories in the inventory as needed. They also provide a way
to manually add directories if needed. However, adding a directory is not recommended on the ODA unless doing so
has been discussed with Oracle Support.
The
add
directive is used to add directories or hosts. The TFA process is also intelligent, to avoid duplication.
Listing 7-23 shows an example of a directory that already exists attempting to be added. This tfactl will respond by not
adding the directory.
Listing 7-23.
tfactl Add Directory
# tfactl add directory /opt/oracle/oak/log
No new directories were added to TFA
TFA also provides a manual way to collect diagnostics information and allows for information to be trimmed as
needed. Listing 7-24 shows examples of collecting all diagnostics, as well as diagnostics for specific timeframes.
Listing 7-24.
tfactl Diagnostic Collection Examples
# tfactl diagcollect -all
Allows for diagcollect to be run an zip be collected in the initiating node
#tfactl diagcollect -crs node1,node2 -since 9h
Allows for diagcollect to be run for crs only and zip copied to initiating node
#tfactl diagcollect -for "Jun/29/2013 21:00:10"
Collect diagnostics for all files for all nodes and collect only for timestamp listed.
The TFA team determines the events that cause diagnostic collection. These events are listed in detail in the
TFA User Guide, accessible via MOS Note 1513912.1. Listing 7-25 provides a subset of the events that can trigger a
tfactl
collection.
Listing 7-25.
tfactl Diagnostic Collection Event Triggers
Search for:
.*ORA-297(01|02|03|08|09|10|40).*
In Files :
Alert Logs - ASM and RDBMS
Error Description:
29701, unable to connect to Cluster Synchronization Service
29702,error occurred in Cluster Group Service operation
29703, error occurred in global enqueue service operation
29708, error occurred in Cluster Synchronization Services
29709, Communication failure with Cluster Synchronization Services
29710, Current operation aborted by Cluster Synchronization Services
29740 ,evicted by instance number %s, group incarnation %s
29770,global enqueue process %s (OSID %s) is hung for more than %s seconds
