We’ve just upgraded Oracle 220.127.116.11 GI to 18.104.22.168, and found clusterware logs have been relocated from $GI_HOME/log/.
According to Oracle support Doc (ID 1915729.1), from Oracle 22.214.171.124 on Oracle clusterware (part of Grid Infrastructure) uses Oracle database fault diagnosability infrastructure to manage diagnostic data and its alert logs. As a result, most diagnostic data resides in the Automatic Diagnostic Repository (ADR).
Please note :
1)More space is required for GI $ORACLE_BASE in case of file system full , because Clusterware logs are moved to ADR which is part of ORACLE_BASE.
2)When opening SR, instead of diagcollection.pl, TFA should be used to collect diagnostics:
$/u01/app/126.96.36.199/grid/bin/tfactl diagcollect -from “Jan/20/2015 08:00:00” -to “Jan/23/2015 13:00:00”
Collecting data for all nodes
Scanning files from Jan/20/2015 08:00:00 to Jan/23/2015 13:00:00
Repository Location in hx415 : /u01/app/grid/tfa/repository
2015/01/23 14:42:55 EST : Running an inventory clusterwide …
2015/01/23 14:42:55 EST : Collection Name : tfa_Fri_Jan_23_14_42_49_EST_2015.zip