How To Optimize VPS ?

[Muhammad_1]

Hello Cpanel Team and Respectable Members,

My website PINDITUBE****|****Pakistan Fastest Growing TV Program Videos Library with Wordpress CMS Run on VPS
My VPS Specification is
Total processors: 2
Processor #1
Vendor
GenuineIntel
Name
Intel(R) Xeon(R) CPU E5506 @ 2.13GHz
Speed
800.049 MHz
Cache
4096 KB
Processor #2
Vendor
GenuineIntel
Name
Intel(R) Xeon(R) CPU E5506 @ 2.13GHz
Speed
800.049 MHz
Cache
4096 KB

And ram is 2GB Dedicated and 2 GB SWAP
With 80 GB HDD ,

Problem is Website Load too much Slow ,i Have average Traffic is only 10000 ,
Can any one help me Step by Step how to Solve Website Loading issue
i have no experience
Regards

[cPanelTristan]

What is your PHP handler? You can grab that information from WHM > Apache Configuration > PHP and SuExec Configuration area.

Next, what are the current uptime and sar details for your machine to see the load and other details? You'd simply run "uptime" and "sar" commands in root SSH and paste that information.

What are your Apache settings for MaxClients, ServerLimit, etc? You can grab those from WHM > Apache Configuration > Global Configuration area.

[Muhammad_1]

Hello
This Information is Listed Below
Default PHP Version (.php files) 5
PHP 5 Handler suphp
PHP 4 Handler none

Apache suEXEC on

[Muhammad_1]

Hello
our settings have been saved.

The values are:

keepalivetimeout 6
maxrequestsperchild 4000
traceenable On
fileetag All
serversignature Off
keepalive On
minspareservers 5
maxclients 512
startservers 8
servertokens Full
maxkeepaliverequests 100
sslciphersuite ALL:!ADH:+HIGH:+MEDIUM:-LOW:-SSLv2:-EXP
root_options ExecCGI FollowSymLinks Includes IncludesNOEXEC Indexes -MultiViews SymLinksIfOwnerMatch
serverlimit 512
timeout 30
maxspareservers 10



and Server uptime is
erver Version: Apache/2.2.16 (Unix) mod_ssl/2.2.16 OpenSSL/0.9.8e-fips-rhel5 DAV/2 mod_bwlimited/1.4
Server Built: Sep 3 2010 08:28:58
Current Time: Wednesday, 27-Apr-2011 13:54:11 MDT
Restart Time: Wednesday, 27-Apr-2011 10:41:41 MDT
Parent Server Generation: 0
Server uptime: 3 hours 12 minutes 30 seconds
Total accesses: 21640 - Total Traffic: 155.7 MB
CPU Usage: u.42 s.17 cu125.6 cs0 - 1.09% CPU load
1.87 requests/sec - 13.8 kB/second - 7.4 kB/request
2 requests currently being processed, 8 idle workers
W.__.......W_...__...___........................................
................................................................
................................................................
................................................................
................................................................
................................................................
................................................................
................................................................

[cPanelTristan]

Do you have the sar information?

[Muhammad_1]

helllo Here is Sar information

Linux 2.6.18-194.26.1.el5.028stab079.2 (us02345.sitevalley.com) 04/27/2011

12:00:01 AM CPU %user %nice %system %iowait %steal %idle
12:10:01 AM all 13.59 0.01 1.54 0.00 0.00 84.86
12:20:02 AM all 59.63 0.02 6.15 0.00 0.00 34.20
12:30:01 AM all 34.20 0.01 3.77 0.00 0.00 62.02
12:40:01 AM all 77.75 0.04 8.55 0.00 0.00 13.66
12:50:01 AM all 74.62 0.02 8.51 0.00 0.00 16.84
01:00:02 AM all 68.03 0.65 7.96 0.00 0.00 23.36
01:10:01 AM all 71.93 0.02 7.74 0.00 0.00 20.31
01:20:01 AM all 26.65 0.01 3.01 0.00 0.00 70.33
01:30:01 AM all 17.92 0.02 2.01 0.00 0.00 80.04
01:40:01 AM all 17.79 0.01 2.09 0.00 0.00 80.10
01:50:01 AM all 8.05 0.01 0.99 0.00 0.00 90.95
02:00:01 AM all 8.14 0.01 1.02 0.00 0.00 90.84
02:10:01 AM all 13.39 0.01 1.53 0.00 0.00 85.07
02:20:01 AM all 18.79 0.02 2.15 0.00 0.00 79.04
02:30:01 AM all 47.17 0.02 5.02 0.00 0.00 47.80
02:40:02 AM all 19.88 0.01 2.21 0.00 0.00 77.90
02:50:01 AM all 23.31 0.01 2.58 0.00 0.00 74.09
03:00:01 AM all 19.91 0.01 2.15 0.00 0.00 77.93
03:10:01 AM all 18.23 0.02 2.04 0.00 0.00 79.71
03:20:01 AM all 27.32 0.01 3.03 0.00 0.00 69.64
03:30:01 AM all 26.61 0.01 3.01 0.00 0.00 70.37
03:40:01 AM all 20.68 0.01 2.45 0.00 0.00 76.85

03:40:01 AM CPU %user %nice %system %iowait %steal %idle
03:50:01 AM all 22.34 0.01 2.60 0.00 0.00 75.04
04:00:01 AM all 17.32 0.02 2.11 0.00 0.00 80.54
04:10:01 AM all 18.67 0.01 2.21 0.00 0.00 79.11
04:20:01 AM all 23.84 0.01 2.88 0.00 0.00 73.28
04:30:01 AM all 22.26 0.01 2.62 0.00 0.00 75.11
04:40:01 AM all 20.33 0.01 2.18 0.00 0.00 77.48
04:50:01 AM all 15.98 0.02 1.74 0.00 0.00 82.26
05:00:01 AM all 22.12 0.01 2.40 0.00 0.00 75.47
05:10:01 AM all 15.62 0.01 1.73 0.00 0.00 82.63
05:20:01 AM all 18.44 0.01 2.05 0.00 0.00 79.50
05:30:01 AM all 31.75 0.01 3.66 0.00 0.00 64.58
05:40:01 AM all 21.77 0.02 2.45 0.00 0.00 75.76
05:50:01 AM all 19.85 0.01 2.25 0.00 0.00 77.88
06:00:01 AM all 36.59 0.01 4.38 0.00 0.00 59.02
06:10:01 AM all 21.69 1.17 2.44 0.00 0.00 74.70
06:20:01 AM all 20.50 0.01 2.28 0.00 0.00 77.21
06:30:01 AM all 18.74 0.02 2.05 0.00 0.00 79.19
06:40:01 AM all 21.55 0.01 2.30 0.00 0.00 76.14
06:50:01 AM all 27.40 0.01 2.88 0.00 0.00 69.70
07:00:01 AM all 21.22 0.01 2.44 0.00 0.00 76.33
07:10:01 AM all 19.23 0.01 2.14 0.00 0.00 78.62
07:20:01 AM all 23.23 0.02 2.46 0.00 0.00 74.28

07:20:01 AM CPU %user %nice %system %iowait %steal %idle
07:30:01 AM all 29.07 0.01 3.10 0.00 0.00 67.82
07:40:01 AM all 33.48 0.01 3.53 0.00 0.00 62.98
07:50:01 AM all 16.83 0.01 1.88 0.00 0.00 81.27
08:00:02 AM all 20.00 0.01 2.17 0.00 0.00 77.82
08:10:01 AM all 28.62 0.03 3.21 0.00 0.00 68.15
08:20:01 AM all 35.51 0.01 3.92 0.00 0.00 60.55
08:30:01 AM all 25.41 0.01 2.72 0.00 0.00 71.86
08:40:01 AM all 34.27 0.01 3.72 0.00 0.00 62.00
08:50:02 AM all 37.54 0.02 4.01 0.00 0.00 58.42
09:00:01 AM all 47.69 0.03 5.28 0.00 0.00 47.00
09:10:01 AM all 31.41 0.01 3.30 0.00 0.00 65.27
09:20:01 AM all 20.38 0.01 2.22 0.00 0.00 77.38
09:30:01 AM all 21.87 0.01 2.46 0.00 0.00 75.66
09:40:01 AM all 24.54 0.01 2.74 0.00 0.00 72.71
09:50:01 AM all 25.80 0.01 2.93 0.00 0.00 71.26
10:00:01 AM all 29.86 0.03 3.40 0.00 0.00 66.72
10:10:07 AM all 27.67 0.46 3.18 0.00 0.00 68.69
10:20:01 AM all 31.20 0.01 3.46 0.00 0.00 65.33
10:30:02 AM all 55.01 0.03 6.05 0.00 0.00 38.91
10:40:02 AM all 74.94 0.02 7.89 0.00 0.00 17.15
10:50:04 AM all 51.99 0.03 5.39 0.00 0.00 42.59
11:00:01 AM all 46.24 0.01 4.78 0.00 0.00 48.97

11:00:01 AM CPU %user %nice %system %iowait %steal %idle
11:10:02 AM all 35.18 0.01 3.73 0.00 0.00 61.08
11:20:01 AM all 41.94 0.02 4.70 0.00 0.00 53.35
11:30:01 AM all 28.92 0.01 3.15 0.00 0.00 67.92
11:40:01 AM all 22.32 0.02 2.51 0.00 0.00 75.14
11:50:01 AM all 25.82 0.01 2.81 0.00 0.00 71.35
12:00:01 PM all 20.28 0.01 2.25 0.00 0.00 77.46
12:10:01 PM all 20.78 0.01 2.32 0.00 0.00 76.90
12:20:01 PM all 20.60 0.01 2.23 0.00 0.00 77.15
12:30:01 PM all 22.38 0.02 2.41 0.00 0.00 75.19
12:40:01 PM all 31.20 0.01 3.63 0.00 0.00 65.15
12:50:03 PM all 39.26 0.01 4.33 0.00 0.00 56.40
01:00:01 PM all 20.63 0.01 2.30 0.00 0.00 77.06
01:10:01 PM all 16.97 0.01 1.90 0.00 0.00 81.12
01:20:01 PM all 26.28 0.03 2.91 0.00 0.00 70.78
01:30:01 PM all 36.96 0.01 4.09 0.00 0.00 58.94
01:40:01 PM all 30.82 0.01 3.39 0.00 0.00 65.77
01:50:01 PM all 22.40 0.01 2.39 0.00 0.00 75.20
02:00:01 PM all 20.84 0.01 2.16 0.00 0.00 77.00
02:10:01 PM all 21.83 0.02 2.31 0.00 0.00 75.84
02:20:01 PM all 19.13 0.01 1.97 0.00 0.00 78.89
Average: all 26.52 0.04 2.93 0.00 0.00 70.51

[k-planethost]

Hop T1 T2 T3 IP
1. 5.60ms 5.86ms 5.68ms 66.249.16.2 Reverse IP | Ping | DNS Lookup
2. 5.59ms 8.10ms 5.60ms 64.246.161.201 Reverse IP | Ping | DNS Lookup
3. 6.59ms 7.12ms 7.60ms igr1-pe2.blh.fibercloud.net Reverse IP | Ping | DNS Lookup
4. 10.61ms 9.99ms 12.59ms 12.119.199.41 Reverse IP | Ping | DNS Lookup
5. 32.55ms 31.86ms 34.57ms cr1.st6wa.ip.att.net Reverse IP | Ping | DNS Lookup
6. 31.56ms 31.86ms 32.53ms cr2.sffca.ip.att.net Reverse IP | Ping | DNS Lookup
7. 31.62ms 31.32ms 30.59ms 12.122.136.205 Reverse IP | Ping | DNS Lookup
8. 31.57ms 32.41ms 31.69ms 192.205.34.34 Reverse IP | Ping | DNS Lookup
9. 31.54ms 31.87ms 31.47ms te2-4.ccr02.sjc03.atlas.cogentco.com Reverse IP | Ping | DNS Lookup
10. 32.59ms 32.49ms 31.56ms te4-2.ccr02.sjc01.atlas.cogentco.com Reverse IP | Ping | DNS Lookup
11. 33.53ms 33.10ms 33.63ms te0-1-0-5.ccr22.sfo01.atlas.cogentco.com Reverse IP | Ping | DNS Lookup
12. 78.58ms 79.62ms 79.57ms te0-4-0-1.ccr22.mci01.atlas.cogentco.com Reverse IP | Ping | DNS Lookup
13. 68.59ms 68.41ms 68.51ms te0-5-0-4.ccr22.ord01.atlas.cogentco.com Reverse IP | Ping | DNS Lookup
14. 159.56ms 159.05ms 158.54ms te0-5-0-6.ccr22.bos01.atlas.cogentco.com Reverse IP | Ping | DNS Lookup
15. 158.52ms 159.32ms 158.51ms te0-4-0-1.ccr22.lpl01.atlas.cogentco.com Reverse IP | Ping | DNS Lookup
16. 160.50ms 160.03ms 159.48ms te4-2.ccr01.man01.atlas.cogentco.com Reverse IP | Ping | DNS Lookup
17. 159.48ms 159.94ms 159.49ms 149.6.11.170 Reverse IP | Ping | DNS Lookup
18. 160.51ms 160.91ms 160.49ms lv0202.1b0701.mnuk01.hostnoc.eu Reverse IP | Ping | DNS Lookup
19. * * * Request Timed Out
20. * * * Request Timed Out

this is a traceroute for your domain
for a USA virtual its ok i think.
maybe is cause of the distance that the DC is located from your country that cause issues.
the provider that you deal with has a DC available in UK also in Miami , LA and Scrapton
for my location loads good for a USA vps

[Muhammad_1]

Thanks k-planethost
u can see how much VPS Used Memory
System information

Server load 4.81 (7 CPUs)
Memory Used 12.73% (533944 of 4194304)
Swap Used 0.00% (0 of 1)
i wonder why CPU Dont use more memorey
is any restraction on Vps to use memorey?

[JawadArshad]

VPS' parent OS has several CPU QoS parameters that can be used to cap your CPU. Your host might be capping CPU. Run 'sar -q' and paste the results. Secondly, check "Daily Process Log" in WHM to find the most resource intensive processes as well as monitor your server via top to find the cause of high CPU usage.

Thanks k-planethost
u can see how much VPS Used Memory
System information

Server load 4.81 (7 CPUs)
Memory Used 12.73% (533944 of 4194304)
Swap Used 0.00% (0 of 1)
i wonder why CPU Dont use more memorey
is any restraction on Vps to use memorey?

[Muhammad_1]

Here is 'sar -q result
Linux 2.6.18-194.26.1.el5.028stab079.2 (server1.pinditube.com) 04/30/2011

12:00:01 AM runq-sz plist-sz ldavg-1 ldavg-5 ldavg-15
12:10:01 AM 3 64 2.53 2.43 2.60
12:20:02 AM 4 65 2.71 2.40 2.47
12:30:02 AM 3 71 2.34 2.56 2.54
12:40:02 AM 6 69 1.94 2.38 2.51
12:50:01 AM 3 67 2.37 2.40 2.43
01:00:01 AM 3 64 2.50 2.45 2.47
01:10:01 AM 3 62 2.91 2.62 2.53
Average: 4 66 2.47 2.46 2.51
Here is Daily process log data

You have 9 CPUs; therefore, these CPU percentages are divided by 9 to indicate the true percentage of all CPU power used.

User Domain % CPU % MEM MySQL Processes
pinditube 3.30 2.01 1.9
mysql 1.32 0.82 0.0
root 0.02 5.98 1.9
named 0.00 0.10 0.0
nobody 0.00 0.03 0.0
cpanel 0.00 0.01 0.0
dlinksco 0.00 0.00 0.0
dbus 0.00 0.00 0.0
dovecot 0.00 0.00 0.0
mailnull 0.00 0.00 0.0
mailman 0.00 0.00 0.0
68 0.00 0.00 0.0
avahi 0.00 0.00 0.0
sshd 0.00 0.00 0.0
wwww 0.00 0.00 0.0
xfs 0.00 0.00 0.0
unauthenticated 0.00 0.00 0.0
eximstats 0.00 0.00 0.1
leechprotect 0.00 0.00 1.0


User Domain % CPU Process
Top Processes
pinditube 75.0 /usr/bin/php /home/pinditub/public_html/index.php
pinditube 66.0 /usr/bin/php /home/pinditub/public_html/index.php
pinditube 62.0 /usr/bin/php /home/pinditub/public_html/index.php
root 4.8 cpanellogd - updating bandwidth for pinditub
root 3.0 /usr/bin/perl /scripts/restartsrv_cpanellogd --check
root 14.0 /usr/bin/perl /scripts/restartsrv_tailwatchd --check
mysql 12.3 /usr/sbin/mysqld --basedir/ --datadir/var/lib/mysql --usermysql --log-error/var/lib/mysql/us02345.sitevalley.com.err --pid-file/var/lib/mysql/us02345.sitevalley.com.pid
mysql 12.2 /usr/sbin/mysqld --basedir/ --datadir/var/lib/mysql --usermysql --log-error/var/lib/mysql/us02345.sitevalley.com.err --pid-file/var/lib/mysql/us02345.sitevalley.com.pid

[k-planethost]

max clients i believe are too much for a value vps
256 i believe should be good
well in this forum there are many posts how to optimize myscl and apache configuration that will feed your needs such as
http://forums.cpanel.net/f189/http-h...on-202021.html

[dutaspace]

I really do this and it is really very helpful.

[jackg164]

Hello,

I will explain how to calculate maxclients:

------------------------------------
MaxClients = Total RAM / Max child process size
------------------------------------

Calculate memory used by Apache:

ps -ylC httpd --sort:rss

Example of Output:

S   UID   PID  PPID  C PRI  NI   RSS    SZ WCHAN  TTY          TIME CMD
S    99  6859  5230  0  75   0 19616  25081 semtim ?        00:00:01 httpd

19616 / 1024 = 19,15 MB per Apache proccess


Now you need see the total number of processes

lsof -i | grep httpd | wc -l

Example of Output:

52

Well, the conclusion of this example:

19,15 MB used per Apache proccess x 52 Apache processes on memory = 995,8 MB of memory used

Then if you have 2 GB of RAM, in this case you should set MaxClients to 93, why 93?

2048 MB (Total RAM) - 256 MB (This memory reserved for OS and cPanel) / 19,15 MB (Memory used per Apache proccess) = 93,57....


Also, if you have 10000 daily visitors, you should upgrade your VPS or move to dedicated.

I hope this can help you, if I'm wrong on this, someone correct me.

[srpurdy]

10,000 doesn't seem like much for a vps. I get about half on one of my vps's (30K raw) with only 512 memory, and have zero issues. (load is maybe 0.2 at peak hours) I do have a 16 cores but limited performance as well. Seems the majority of your load is coming from pinditube 3.30 2.01 1.9 user. Is the the primary domain with traffic? Looking at this website, you have very low scores in page speed. This is where I would start, as you can probably improve the load on the server by optimizing the website itself.

[Sannin]

All the optimizing guides i have read talk about calculating maxclients based on apache memory size. From my calculation, i set maxclients at 75 and i did a benchmark with ab and loadimpact.com on my wordpress site. After 20 concurrent connections memory was full because of the forked php processes that spawned with the apache processes.

So shouldn't we include php process size into the calculation?