what should be the System Usage ?

[Shazam]

Originally Posted by mrugesh78

at present my memory_limit inside PHP configuration editor is : 256M

I think before two month when my VPS was going to the black zone, I informed to support department and they have increased it. Bcos they replied me that : "we have increased memory_limit" So I do not know what was the default. May be it was 128M. May be.

But this can affect mysql process ? bcos I think PHP is different then mysql and the proess consuming is of mysql. Can you guide me please ?

Way too high. Go to 64M and Save the settings. That explains a lot.

[mrugesh78]

thank you Shazam for your quick reply.
I have set it to 64M and saved it and restarted VPS. Now lets wait for some hours so that we can know the final result.

by the way, today I closely noted one thing that, my vps's system usage goes to high only after 2.5 hours. Like just I restarted VPS, so now it is near 40 to 43%. It will remain maximum 48% up to next two hours. But after that It increasing trend will start. It will go to 50, then 55, then 60.... and at the end of 4.30 hours, it will reach to 95%. This is fixed pattern I am observing.

Somthing after two hours is happening.

my.conf has following settings :

[mysqld]
skip-bdb
skip-innodb
concurrent_insert = 1
connect_timeout = 10

delayed_insert_timeout = 30

join_buffer_size = 1M

interactive_timeout = 30

# memory=key_buffer+(sort_buffer_size+read_buffer_size)*max_connections
max_allowed_packet = 16M
max_connections = 150
max_connect_errors = 10
max_delayed_threads = 20
max_user_connections = 50
max_heap_table_size = 64M
max_tmp_tables = 128
myisam_sort_buffer_size = 64M

key_buffer = 16M

low_priority_updates = 1

query_cache_limit = 1M
query_cache_size = 16M
query_cache_type = 1

read_buffer_size = 1M
read_rnd_buffer_size = 1M

safe-show-database
skip-locking
socket = /var/lib/mysql/mysql.sock
sort_buffer_size = 1M

table_cache = 1000
thread_cache_size = 128
thread_concurrency = 2
tmp_table_size = 64M

wait_timeout = 20

[mysqld_safe]
log-error = /var/log/mysqld.log 
nice = 10
open_files_limit = 8192

[mysqldump]
max_allowed_packet = 16M
quick

[mysql]
no-auto-rehash
safe-updates

[mysqlhotcopy]
interactive-timeout

please if you can help.... please

[Shazam]

I'm not a MySQL expert, but for now just see if reduced PHP memory helps your situation. It was way too high.

[mrugesh78]

I have reduced PHP memory to 64M. but still after 2 hours, now, my system usage inside VPS is showing me 66%. It was at 43% when restarted. And as I said, now two hours is over, so it will start increasing and will reach to 100% in next two hours.

do not know why this is happening. spend the whole week but not able to get the solution. please help.

[Shazam]

Originally Posted by mrugesh78

I have reduced PHP memory to 64M. but still after 2 hours, now, my system usage inside VPS is showing me 66%. It was at 43% when restarted. And as I said, now two hours is over, so it will start increasing and will reach to 100% in next two hours.

do not know why this is happening. spend the whole week but not able to get the solution. please help.

The other possibility is to pay for maintenance and have someone at HostICan work on it for you.

[mrugesh78]

What is the charge for the maintence to do this job ? is it on job basis or hour basis ?

[Shazam]

It starts at $19.95. Call support and have them assess what you need.

[itwasntme]

I have noticed the same problem with two VPS accounts I have. Resources on both creep up to 100%--but stays all green--over the course of a few hours, but there are no QoS alerts and no problems I have noticed with either site. Restarting MySQL drops the resource usage for a few hours.

I first noticed this with the update of cPanel to 11.23.x. Both these VPS accounts are running PHP 5.2.2 and MySQL 5.0.67-community, and were updated to cPanel 11.24.4-Release last week without any change in the resource usage.

I haven't had any crashes or alerts, so I'm wondering if the 100% resource may be a Virtuozzo/cPanel issue? If I was really using 100% of my resources, it seems like I would get some QoS alerts.

I, too, have my PHP memory limit set to 256 MB, and have had it that way for over a year without any problems.

[Shazam]

That's way way too high. Even HostiCan's own forums are at 128M.

[itwasntme]

Originally Posted by Shazam

That's way way too high. Even HostiCan's own forums are at 128M.

In looking into my own high resource meter in Virutozzo, I discovered that I had three plugins in WHM that were broken: Logview, ConfigServer Mail Queues, and ConfigServer Mail Manage. I don't know what database Logview uses, but if it's MySQL it might account for my problem. I uninstalled and reinstalled all three, restarted MySQL, and my resources are down to 34%. I'll have to wait a few hours to see if that's the fix, but so far, so good.

Shazam, what are the consequences of having your PHP memory limit too high?

I will wait a day or so to see if my repair of the WHM plugins has fixed my resource problem, then I'll reduce the memory limit to 128 if that's a good idea.