Exercise 3: Why Swap Wasn't the Active Problem in the Worked Example — Possible Solution ==================================================================== WHY THE HIGH SWAP FIGURE LOOKED CONCERNING AT FIRST ------------------------------ Per this chapter, the machine showed "swap at 1890.3 of 2048.0 MiB used" - nearly the entire swap allocation in use, which "at first glance, looks like a serious active problem." WHAT THE vmstat CHECK ACTUALLY SHOWED ------------------------------ Per this chapter's own worked example, checking Chapter 2's vmstat sample from the same machine showed "si and so are both 0 across every sample - nothing is actively swapping in or out right now." Per Exercise 2's own reasoning, that means the high swap-used total is historical, not an indication of current, ongoing memory pressure. WHY IT WAS JUDGED NOT TO BE THE ACTIVE PROBLEM ------------------------------ Per this chapter, "this machine's high swap-used figure is a leftover from an earlier period of pressure, not an active crisis." Since nothing was actively moving in or out of swap, memory was not the thing currently causing the machine's symptoms - the swap total was a secondary, historical detail rather than live evidence of a current bottleneck. WHAT THE REAL, CURRENTLY-ACTIVE BOTTLENECK WAS ------------------------------ Per this chapter, "combined with Chapter 2's own finding of 0.0 wa, the picture is consistent and honest... the actual, currently-active bottleneck is exactly what Chapter 2 already found - genuine CPU contention." The real problem was the CPU overload Chapter 2 identified (load average of roughly 2.1 per core), not memory. WHY THIS WORKS AS AN ANSWER ------------------------------ It walks from the initial concerning appearance, through the specific vmstat evidence that ruled out active swapping, to the chapter's own final conclusion that CPU contention (from Chapter 2) was the real, currently-active cause - correctly tying the two chapters' findings together rather than treating them as unrelated.