Hi All,
I am working on Ubuntu 8.04.3 OS, with this I am getting a problem, Daily my server is down on same time at 4:00 PM
I seems server is down by "kswapd0" process, I am not sure,
As I run top command, I got below out put

top - 16:26:10 up 23:48, 3 users, load average: 60.29, 47.39, 24.40
Tasks: 116 total, 43 running, 73 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.0%us, 28.1%sy, 0.0%ni, 22.3%id, 0.0%wa, 0.0%hi, 0.0%si, 49.6%st
Mem: 1747660k total, 1739396k used, 8264k free, 284k buffers
Swap: 917496k total, 917496k used, 0k free, 1868k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
78 root 10 -5 0 0 0 D 11.1 0.0 1:01.49 kswapd0
1754 root 23 0 929m 21m 368 S 4.0 1.3 0:16.09 java
1224 mysql 16 0 156m 31m 80 S 2.5 1.9 44:08.96 mysqld
13728 kpmrs 18 0 29252 1124 76 R 0.9 0.1 0:01.64 php
14831 www-data 18 0 45700 12m 76 R 0.9 0.7 0:03.14 apache2
15196 www-data 18 0 44640 11m 76 R 0.6 0.7 0:03.06 apache2
15641 www-data 18 0 35148 2032 64 R 0.3 0.1 0:02.71 proftpd
14696 www-data 18 0 36304 3304 76 R 0.9 0.2 0:02.52 apache2
1570 www-data 18 0 20668 740 72 R 0.7 0.0 0:01.86 apache2
1226 root 18 0 1716 56 44 R 0.4 0.0 0:00.06 logger
14784 www-data 18 0 36584 3296 72 R 0.4 0.2 0:01.84 apache2
14831 www-data 18 0 45700 12m 76 R 0.4 0.7 0:03.11 apache2
14837 www-data 18 0 37104 4424 80 R 0.4 0.3 0:01.59 apache2
14923 www-data 18 0 36632 4056 76 R 0.4 0.2 0:02.09 apache2
15654 root 18 0 2964 128 56 R 0.4 0.0 0:01.25 killall
24159 kpmrs 18 0 30544 1808 76 R 0.4 0.1 0:02.93 php
77 root 15 0 0 0 0 S 0.2 0.0 0:00.23 pdflush
2484 kpmrs 18 0 26836 2476 80 R 0.2 0.1 0:03.93 php
13518 www-data 18 0 36632 3180 76 R 0.2 0.2 0:03.57 apache2
15196 www-data 18 0 44640 11m 76 R 0.2 0.7 0:03.04 apache2

Please help and guide me
sorry for my poor english
Thanks
Regards,

I don't know much about servers, but the output above shows clearly that your memory and swap are completely full. And kswapd is the kernel swap daemon, which is probably just trying to deal with the lack of memory.

in/on my 804 i can clean out the cache files
applications, other, cache

hope this helps

I know this is a little late but if you can use Gentoo I highly recommend it for a server.

I find ubuntu (Even without a GUI) to be WAY to slow for a server.

I can run a gentoo server on less than 32 MB of RAM and a P2 CPU.

I don't know why on earth you would WANT to but... you can.

Probably just not getting enough CPU - what are those other guests doing at that time (and/or the host).

Well you're load avgs are extremely high, and, as noted, both your RAM and your swap partition are completely full.
You seem to have about 1.7Gb oRAM and 1GB swap (approx). I'd want swap at least as large as RAM on a server.
However, that won't solve your problem, you need look at what those processes are doing; probably start with Apache/PHP.

I remember of my old box running slackware. At 4AM Slackware have crontab stuff running.. Maybe you also have something...

Hi
I have checked cron, but I didn't get any solution.
what I need to check more on the server?

Please help...

Thanks

We need to figure out why your swap space is full..
When you swap is full your kswapd goes crazay

One of your running apps is eating your memory and then your swap. What is happening is that sometime your machine tries to get rid of all that trash. You need to monitor which one is doing this

Hi,
Okay today I will monitor and try to get something

Thanks

Hope you're not using any java app.. Java sux. lol

Well, as others have pointed out - you're very short on RAM; but
you ALSO have the problem of 49.6%st ... the hypervisor chooses
to run something else instead of your guests half the time.

Those facts together (not enough actual grunt from the hypervisor,
plus massive memory constraints) to me explain the load satisfactorily ;}



Cheers,
Tink