Slow Hardware?
Server acting slow? There could be a lot of things going on. Once of the first things to check is hardware. Ask yourself, what is slow about the server? The network functions? Loading webpages? Disk access?
Example 1:
$ mii-tool eth0: negotiated 10baseT-HD, link ok
This example looks like the server's nic is not negotiating. Put in a support ticket so we can check the cable and switch.
Example 2:
$ hdparm -tT /dev/hda /dev/hda: Timing cached reads: 382 MB in 2.01 seconds = 190.21 MB/sec Timing buffered disk reads: 120 MB in 3.00 seconds = 39.95 MB/sec
These are about normal, if you are less than 40MB/sec on that second read, it could be and indication of a bad disk
Example 3:
$ free -m
total used free shared buffers cached
Mem: 948 816 131 0 6 483
-/+ buffers/cache: 326 621
Swap: 2933 33 2900
This example you can see that there are only 33 megabytes of swap used, which is fine. If this were high I would be worried about runaway processes or maybe think about upgrading my ram.
High Load?
see more at High Server Loads and You
If you can eliminate the possibility that the server's hardware is slow, then take a look at the load the server is currently experiencing. Are there a lot of simultaneous connections?
Example 1:
$ w 19:38:53 up 75 days, 22:13, 4 users, load average: 0.01, 0.01, 0.00 USER TTY FROM LOGIN@ IDLE JCPU PCPU WHAT
This server is under very low load (.01). If it is going slow, there may be something else wrong.
Example 2: (Taken from an apache status page)
.KWRRRRRRRRRRRRRRRRRRRRRRR.RRRRRRRRRRRR.RRRRRRRRRRRRRRRRRR_RRRRR RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR.RRR RRRRRRRRRRRRKRRRRRRRRRRRRRRRRRRRRRRRRRR.RRWRRRRRRRRRRRRRRRRRRRRR RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR.RRRRRRRRRRRRRRRRRRRR.R
This example is shows an apache server "Reading" a lot. Probably a sign of a bad disk, or a misconfigured apache. Symptoms of this would be extremely long load times on webpages.
Example 3: (Again from an apache status page)
WW.W.W_WKKW.KWWK_KKKK..K.KW.K..KWWKKKKKKKKW.WWKKKKKKK_KK..WWRWK WW.W.WK..K.KW.K..KWWKKKKKKK_KK..KWW_WKKKWK_WKKW.KWWK_KKKW.WKWKW. _KK..KWW_WKKKWKW.WKWW.W.WK..K.KW.K..KWWKKKKKKK_KKW.KWKW.KWKW.KW WKKKKKKKKW.WWKKKWKKKKKKKKW.WWKKKWKKKKKKKKW.WWKKKWKKKWKKKWKKKWKK
This apache stats page would make me think that keep-alives should be turn off, or that the number of max clients should be bumped. Almost all the connection slots are full, but turning off keep-alives would make them re-open quicker.
