Have you discovered that ASM is not writing to its alert log? This came up recently at work when our monitoring was upgraded to detect stale log files (log files that haven’t been written too for a certain period of time.) In this case it was one of the databases I was responsible for that the “problem” was detected on. The log file had not been written to on any of the eight nodes for almost three months.
I did some research and didn’t find much about it. Unlike the Oracle database, ASM is not a chatty Cathy. It was clear that ASM was working fine because it was processing I/O for the RAC instance. I could login to asmcmd and look around.
As it turned out, ASM just doesn’t talk much if it doesn’t need to. I tried creating an alias using the “mkalias” command to see if anything would happen. I accidentally made a mistake in the command and got a syntax error. I took a look at the alert log file and sure enough the error had been written to it. I then started a rebalance operation and ASM wrote to the file.
So, if you’re worried that ASM is not writing to the log files, don’t worry, do a quick test and make sure it is.