We are currently experiencing an issue in our Sybase environment. When we use Spotlight on Sybase 2.0.0.581 to help diagnose the problem, we find that the ASE CPU numbers are very high while the OS CPU numbers are low.
In the past these two metrics have always been the inverse. This we can understand because we have 15 Sybase engines running and when there isn't enough work to keep them all busy, they burn up OS CPU time looking for work. This is described in the Sybase manuals.
Can you give us an explanation of a situation that would cause the high Sybase but low OS CPU utilization?
We'd like to know what circumstances would cause Spotlight to show a high number of "active" users (70-90) when Sybase Central at the same time only shows that a dozen or so connections are running SQL commands.
The other thing we can't explain is how the ASE CPU can be so much higher than the total CPU on the server. The only thing running on our server is this Sybase instance.
We select our cpu numbers from the master..monEngine table and compute cpu use percentages based on those.
It is possible that we show a high percentage say of user time from the perspective of the ASE's internals while the OS level cpu numbers are very low.
For example, 99% of cpu crunching might be attributable to the ASE's user time while Sybase and everything else on the machine is totaling up to 22% of system cpu.
We can expand on the example above with this statement: Sybase is getting 22% cpu and of that 22%, 99% is being used by Sybase to process user requests.
The host cpu control can be drilled-down upon to examine what else the host cpu is doing and why Sybase is only getting 22%.
© ALL RIGHTS RESERVED. Terms of Use Privacy Cookie Preference Center