Bob, thanks for the gkrellm plug -- I am an old fan of it.
However, I'd never used the daemon/remote; only ever remote-X via ssh.
It is such a small app, do you think there is a substantial difference
between running the daemon
with a local gui client vs. tunneled/remote-X?
Ben
PS - the krells are very configurable, and I always changed the defaults --
I'd invert the charts for read-vs-write, in-vs-out, etc,
to make maximum data immediately readable. Anecdote: I was so impressed
when I first brought it up remotely, I was using my first laptop
and had never seen more than 1 CPU krell at once; this was on a UO system
which was just beginning to have some gnome goodies, over 5 years ago...
and those transparent themes are nice candy :)
On 12/22/06, Bob Miller <[EMAIL PROTECTED]> wrote:
Matthew Jarvis wrote:
> Learning something new...
>
> I'm trying to determine if a web server has enough RAM for the load. I
> was advised to use the ''free -m' command and to pay special attention
> to the Swap value.
Quentin already answered. You have plenty of RAM.
I'd like to recommend gkrellm as the *best* way to understand a Linux
system's workload. It continuously plots the last 60 seconds'
utilization of CPU (per-CPU and/or aggregate), disk (per-disk,
per-RAID, per-filesystem and/or aggregate), network I/O, run queue
length, memory used (with and without buffer cache) swap space used,
and swap I/O rate.
It has a client-server mode where you run gkrellmd on your server
and gkrellm on your X11 display.
If you run gkrellm on your systems all the time, you'll learn to
know at a glance exactly what's going on.
--
Bob Miller K<bob>
[EMAIL PROTECTED]
_______________________________________________
EUGLUG mailing list
[email protected]
http://www.euglug.org/mailman/listinfo/euglug
_______________________________________________
EUGLUG mailing list
[email protected]
http://www.euglug.org/mailman/listinfo/euglug
