SSL or not SSL?

[yusufu]

top - 17:01:35 up 4:46, 5 users, load average: 10.79, 11.00, 9.77
Tasks: 324 total, 7 running, 316 sleeping, 1 stopped, 0 zombie
Cpu(s): 87.6%us, 10.0%sy, 0.0%ni, 1.8%id, 0.1%wa, 0.1%hi, 0.5%si, 0.0%st
Mem: 8310168k total, 8100680k used, 209488k free, 327372k buffers
Swap: 2031608k total, 80k used, 2031528k free, 2459160k cached



this is what I get when I run top command.

Meanwhile, management wants me to activate SSL on the IMAP and POP ports.
I know this is good for security but is this really a good idea going by my present RAM utilization?

Y does Zimbra use up so much RAM resources and so lil' SWAP space?

Help me!!!

[phoenix]

What else is running on that server? You show no details of the running tasks on there, why do you think it's Zimbra causing the problem? I run a Centos5.1 server with Zimbra installed and it shows only 135 tasks and 1-5 running, why do you have so many task shown in top?

[phoenix]

Originally Posted by yusufu

Y does Zimbra use up so much RAM resources and so lil' SWAP space?

Your top output above shows only 80k swap space used, that would be quite normal and doesn't indicate that anything is using swap..

[gnyce]

RAM looks ok - unused RAM is wasted, so Linux utilizes as much as it can - but that load average is nuts (10.79, 11.00, 9.77)! Run top to see what is taking the majority of your CPU. And I agree, that seems like quite a bit with regard to the number of tasks.

[yusufu]

top - 10:28:29 up 22:13,  2 users,  load average: 0.56, 0.40, 0.32
Tasks: 134 total,   1 running, 129 sleeping,   4 stopped,   0 zombie
Cpu(s): 27.6%us,  4.1%sy,  0.0%ni, 66.2%id,  1.7%wa,  0.1%hi,  0.4%si,  0.0%st
Mem:   8310168k total,  8122856k used,   187312k free,   436856k buffers
Swap:  2031608k total,       80k used,  2031528k free,  2832052k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
 6190 zimbra    18   0 2652m 2.5g 5424 S   28 31.0  83:46.94 java
29352 postfix   16   0  7512 1980 1584 S    2  0.0   0:00.29 cleanup
 5790 zimbra    18   0 1812m 1.7g 4140 S    1 21.1  20:02.94 mysqld
13477 postfix   15   0  7508 1884 1460 S    1  0.0   0:00.06 lmtp
26868 postfix   15   0  7436 1720 1328 S    1  0.0   0:00.06 pickup
  467 root      10  -5     0    0    0 S    1  0.0   4:44.07 kjournald
 2082 named     25   0 68952 3904 2024 S    0  0.0   2:27.65 named
 6507 root      15   0  7380 1616 1276 S    0  0.0   1:15.86 master
 6516 postfix   16   0 11580 5940 1372 S    0  0.1   1:12.59 qmgr
25136 postfix   15   0  7516 1984 1584 S    0  0.0   0:00.46 cleanup
28192 postfix   15   0  7516 2108 1640 S    0  0.0   0:00.04 cleanup
    1 root      15   0  2032  640  552 S    0  0.0   0:01.19 init
    2 root      RT   0     0    0    0 S    0  0.0   0:00.53 migration/0
    3 root      34  19     0    0    0 S    0  0.0   0:00.07 ksoftirqd/0
    4 root      RT   0     0    0    0 S    0  0.0   0:00.00 watchdog/0
    5 root      RT   0     0    0    0 S    0  0.0   0:00.55 migration/1
    6 root      34  19     0    0    0 S    0  0.0   0:00.09 ksoftirqd/1
    7 root      RT   0     0    0    0 S    0  0.0   0:00.00 watchdog/1
    8 root      RT   0     0    0    0 S    0  0.0   0:00.58 migration/2
    9 root      39  19     0    0    0 S    0  0.0   0:00.07 ksoftirqd/2
   10 root      RT   0     0    0    0 S    0  0.0   0:00.00 watchdog/2
   11 root      RT   0     0    0    0 S    0  0.0   0:00.57 migration/3
   12 root      34  19     0    0    0 S    0  0.0   0:00.09 ksoftirqd/3
   13 root      RT   0     0    0    0 S    0  0.0   0:00.00 watchdog/3
   14 root      10  -5     0    0    0 S    0  0.0   0:00.00 events/0
   15 root      10  -5     0    0    0 S    0  0.0   0:00.00 events/1
   16 root      10  -5     0    0    0 S    0  0.0   0:00.00 events/2
   17 root      10  -5     0    0    0 S    0  0.0   0:00.00 events/3
   18 root      10  -5     0    0    0 S    0  0.0   0:00.00 khelper
   19 root      10  -5     0    0    0 S    0  0.0   0:00.00 kthread
   25 root      10  -5     0    0    0 S    0  0.0   0:00.66 kblockd/0
   26 root      10  -5     0    0    0 S    0  0.0   0:00.13 kblockd/1
   27 root      11  -5     0    0    0 S    0  0.0   0:00.08 kblockd/2
   28 root      10  -5     0    0    0 S    0  0.0   0:00.10 kblockd/3
   29 root      14  -5     0    0    0 S    0  0.0   0:00.00 kacpid
  143 root      14  -5     0    0    0 S    0  0.0   0:00.00 cqueue/0
  144 root      14  -5     0    0    0 S    0  0.0   0:00.00 cqueue/1
  145 root      14  -5     0    0    0 S    0  0.0   0:00.00 cqueue/2

I run top again today, a public holiday, and this is what i get.
I have openfire also running on this server but that shouldn't be taking up so much, although it uses java too

[yusufu]

Y do u think all those extra processes were running,
I'm not toooooo comfy with linux yet, how do find out those extra processes and stop the unneeded ones.
Even the openfire is running right now.

Note: I have 5000+ users.

[gnyce]

I'm not the "high load on linux" expert, but I'll put some things down. Likely you will have to watch this and see if the same behavior happens again.

- Swap is slow, so when you see no (or very little) swap being used, that is a _good_ thing. You want to fully utilize your RAM. If you see swap being used, then you either have too little RAM, or your server is running too much stuff. So from your first post, knowing nothing more, it does not _seem_ like a low-RAM situation.

- watch your load average... from your first posting, a load avg. of 9-10 is too high. Stated simply, your CPU(s) are not keeping up with the work queued up for it. You state a large number of users, but haven't listed what kind/number of CPUs in this machine.

- I find it interesting that the first post listed 324 processes, and the most recent one listed 134. I haven't been running Zimbra enough to know if based on load/# of users, if more processes/threads are spawned or what. I know nothing about openfire.

- You also dont' say anything about disk... what kind, how many, is it RAID, etc. This could also very much be part of the equation.

Keep an eye on it.... but your issue, at face value, does not appear to be lack of RAM.

[p24t]

I've got a server with about 75 AJAX client users. Right now, top reports 287 processes, so those numbers don't seem out of whack.

What kind of processor does this system have? It looks like whatever you have can't handle all your users. Are your 5K users IMAP/POP based? Do many use the webclient?

SSL will increase your CPU load more than it will your RAM usage. That's your major concern here, it seems. You might need more horsepower if you're seeing loads reaching 10+. Are there any speed problems right now?

If you want to see RAM usage, you should run "free" instead of top. It's arranged a little better, and it's easier to see how much of your RAM is just sitting in buffers/cache (not really being used, but Linux still has ahold of it and will use it when it needs it)

             total       used       free     shared    buffers     cached
Mem:      16432296   16268328     163968          0     602404    7497476
-/+ buffers/cache:    8168448    8263848
Swap:      2031608        172    2031436

Mine shows about 8G of usage, there's another 8G just sitting around waiting to be used.

[duran.goodyear]

I guess knowing what kind of CPU's are on the box will help explain things...

Am I reading it right also that you are running all of zimbra's functions on one server?

I'm not the expert either, but everything I've seen done so far, for a larger user group (in the 5000+ range) has had at least 2 servers splitting the load between different functions.

Just a thought.