[plug] Query: Very problematic memory behaviour with SSH/Debian

[James Devenish]

[This is a response to an off-list post but since it contains
additional information I am posting it back to the list.]

Thanks for your kind response, but the behaviour of the machine is so
abnormal that I doubt adding additional swap space would help.

The machine reported about 900 MB of real memory and a dedicated swap
partition (2 GB, apparently). The only mounted partition was a large
(100 GiB) partition for /, and no others (!).

While the nature of the hard drive usage and lack of responsiveness is
reminiscent of heavy swapping, and vmstat does indicate that io activity
is occurring, this was not reported as swapping. And none of the usual
reporting tools seemed to indicate any process as holding the memory.

Debian's vmstat appears to output in this format:

   procs                      memory    swap          io     system         cpu
 r  b  w   swpd   free   buff  cache  si  so    bi    bo   in    cs  us  sy  id
(There should be two non-blank lines above this one, not wrapped).

At the point of crisis (when real RAM reported as being in use), the io
columns showed continual activity with occasional large bursts, but the
swap columns were zero.

I wonder if there could be a fault with the disk that would lead to
these symptoms.

James.