1 2 Previous Next 18 Replies Latest reply on Feb 3, 2018 2:04 PM by AndrewSayer Go to original post
      • 15. Re: Database Crashing Due To Memory Issues
        JohnWatson2

        How do you know this, " it is it is like the system is not releasing memory, and it builds up until it runs out."?

        Your output from free is not showing the lines that matter.

         

        And "MEMORY_TARGET not supported on this system" is because you have not configured enough shared memory segments or tmpfs, as I said before.

        • 16. Re: Database Crashing Due To Memory Issues
          3625952

          I apologise...I said 'it is it is like the system is not releasing memory, and it builds up until it runs out' simply as a way to try and describe what happens.  There are no errors in the alert log at all.  Everything runs along fine, and then suddenly we start seeing these j000 and m000 .  The database crashes at that point.  This is a very lightly used system, and the database usually goes down overnight when there is no load on the database at all. 

           

          Currently the total memory on the VM is 4G, and the tmpfs is set to 4G as well.  The memory_target is set at 3G.  I thought this would be more than enough, but you are probably right right.  Thanks again.

          • 17. Re: Database Crashing Due To Memory Issues
            Pavan Kumar

            Hi

             

            1. Strange to see that your alert is not having any stuff dumb when your instance got crashed
            2. Have you verified from OS level error messages /logs ? since they should point out why some process was killed or cause of issue. Think in related approach what next how to verify the stuff (identifying problem) then further resolution.
            3. Secondly, as other suggested, I believe you should govern the memory, that is opt for ASMM instead of AMM

             

             

            Hope it helps,

             

            - Pavan Kumar N

            • 18. Re: Database Crashing Due To Memory Issues
              AndrewSayer

              3625952 wrote:

               

              I apologise...I said 'it is it is like the system is not releasing memory, and it builds up until it runs out' simply as a way to try and describe what happens. There are no errors in the alert log at all. Everything runs along fine, and then suddenly we start seeing these j000 and m000 . The database crashes at that point. This is a very lightly used system, and the database usually goes down overnight when there is no load on the database at all.

               

              Currently the total memory on the VM is 4G, and the tmpfs is set to 4G as well. The memory_target is set at 3G. I thought this would be more than enough, but you are probably right right. Thanks again.

              What do you see in the alert log around the time of it going down?

               

              Have you googled the process is dead lines? Are they a normal occurrence. Usually if a process does because of a memory related issue the alert log line is much more explicit. With zero knowledge because I haven’t googled, it could be that your job processes have run out of stuff to do so have decided to end rather that staying around.

              1 2 Previous Next