| VPS/Apache HELP [message #16223] |
Wed, 29 August 2007 08:41  |
mcmuney
Messages: 10
Registered: July 2007
|
Junior Member |
|
|
I was on a VPS and noticed that every night between 11PM-1:30AM PST range, the site would become EXTREMELY SLOW and/or time out, extremely slow meaning a page that would normally take 1-2 seconds to load normally on cable connection would take 30+ seconds or timeout. After many exchanges with support and with no resolution, I decided to switch servers. Now, I'm finding a similar issue here as well.
Anyhow, below is the result for "ps -aux" command, does anything appear to be odd here (I'm thinking it could be apache config issue). During this time, i'm able to use Plesk/Virtuozzo normally. I've attempted VPS restart, but results are same. Under QoS Alerts, nothing is in black zone, shmpages sometimes hit yellow zone:
Warning: bad syntax, perhaps a bogus '-'? See /usr/share/doc/procps-3.2.5/FAQ
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 1 0.0 0.1 1672 628 ? Ss Aug26 0:01 init [3]
root 9381 0.0 0.1 1544 616 ? Ss Aug26 0:01 syslogd -m 0
named 9451 0.0 0.6 88808 3264 ? Ssl Aug26 0:00 /usr/sbin/named -u named -c /etc/named.conf -u named -t /var/named/run-root
root 9534 0.0 0.3 4328 1708 ? Ss Aug26 0:00 /usr/sbin/sshd
root 9560 0.0 0.1 2092 896 ? Ss Aug26 0:00 xinetd -stayalive -pidfile /var/run/xinetd.pid
root 9642 0.0 0.2 2228 1140 ? S Aug26 0:00 /bin/sh /usr/bin/mysqld_safe --defaults-file=/etc/my.cnf --pid-file=/var/run/mysqld/mysqld.pi
mysql 9699 0.0 3.1 87444 16668 ? Sl Aug26 0:09 /usr/libexec/mysqld --defaults-file=/etc/my.cnf --basedir=/usr --datadir=/var/lib/mysql --use
root 9819 0.0 0.2 4760 1212 ? S Aug26 0:00 /usr/lib/courier-imap/couriertcpd -address=0 -stderrlogger=/usr/sbin/courierlogger -stderrlog
root 9843 0.0 0.1 3536 836 ? S Aug26 0:00 /usr/sbin/courierlogger imapd
root 9857 0.0 0.2 4764 1212 ? S Aug26 0:00 /usr/lib/courier-imap/couriertcpd -address=0 -stderrlogger=/usr/sbin/courierlogger -stderrlog
root 9860 0.0 0.1 3528 828 ? S Aug26 0:00 /usr/sbin/courierlogger imapd-ssl
root 9866 0.0 0.2 4756 1208 ? S Aug26 0:00 /usr/lib/courier-imap/couriertcpd -address=0 -stderrlogger=/usr/sbin/courierlogger -stderrlog
root 9868 0.0 0.1 3528 828 ? S Aug26 0:00 /usr/sbin/courierlogger pop3d
root 9886 0.0 0.2 4760 1212 ? S Aug26 0:00 /usr/lib/courier-imap/couriertcpd -address=0 -stderrlogger=/usr/sbin/courierlogger -stderrlog
root 9889 0.0 0.1 3532 832 ? S Aug26 0:00 /usr/sbin/courierlogger pop3d-ssl
qmails 9898 0.0 0.0 1524 496 ? S Aug26 0:03 qmail-send
qmaill 9905 0.0 0.0 1480 464 ? S Aug26 0:01 splogger qmail
root 9906 0.0 0.0 1508 376 ? S Aug26 0:00 qmail-lspawn ./Maildir/
qmailr 9907 0.0 0.1 1900 788 ? S Aug26 0:01 qmail-rspawn
qmailq 9908 0.0 0.0 1472 356 ? S Aug26 0:00 qmail-clean
root 10184 0.0 2.5 27424 13524 ? Ss Aug26 0:27 /usr/sbin/httpd
root 11367 0.0 1.1 36332 5784 ? Ss Aug26 0:00 /usr/local/psa/admin/bin/httpsd
psaadm 11376 0.0 5.8 44376 30720 ? S Aug26 0:39 /usr/local/psa/admin/bin/httpsd
root 11491 0.0 0.2 4368 1204 ? Ss Aug26 0:00 /usr/sbin/saslauthd -m /var/run/saslauthd -a shadow -n 2
root 11492 0.0 0.2 4368 1204 ? S Aug26 0:00 /usr/sbin/saslauthd -m /var/run/saslauthd -a shadow -n 2
psaadm 11998 0.0 5.8 45240 30800 ? S Aug26 0:36 /usr/local/psa/admin/bin/httpsd
psaadm 23623 0.0 4.8 41608 25236 ? S Aug26 0:18 /usr/local/psa/admin/bin/httpsd
apache 28656 0.2 3.0 28480 15736 ? S 23:07 0:04 /usr/sbin/httpd
apache 15509 0.2 2.9 28256 15516 ? S 23:16 0:03 /usr/sbin/httpd
apache 20072 0.2 2.9 28260 15496 ? S 23:19 0:03 /usr/sbin/httpd
apache 20113 0.1 2.9 28540 15628 ? S 23:19 0:02 /usr/sbin/httpd
apache 20246 0.2 2.9 28272 15360 ? S 23:19 0:03 /usr/sbin/httpd
apache 21507 0.2 3.0 28572 15840 ? S 23:19 0:03 /usr/sbin/httpd
apache 21522 0.1 2.9 28572 15660 ? S 23:19 0:03 /usr/sbin/httpd
apache 21545 0.2 2.8 28232 15204 ? S 23:19 0:03 /usr/sbin/httpd
apache 21554 0.1 2.8 28040 15120 ? S 23:19 0:02 /usr/sbin/httpd
apache 21556 0.1 3.0 28820 15900 ? S 23:19 0:03 /usr/sbin/httpd
root 12075 0.0 0.4 9068 2580 ? Rs 23:31 0:00 sshd: root@pts/0
root 13312 0.0 0.2 2240 1336 pts/0 Ss 23:31 0:00 -bash
root 24010 0.0 0.1 2332 808 pts/0 R+ 23:46 0:00 ps -aux
Thanks in advance.
[Updated on: Wed, 29 August 2007 08:42] Report message to a moderator |
|
|
|
| Re: VPS/Apache HELP [message #16225 is a reply to message #16223] |
Wed, 29 August 2007 08:57  |
khorenko
Messages: 533
Registered: January 2006
Location: Moscow, Russia
|
Senior Member |
|
|
Can it be a cron issue?
i mean, might be some cron jobs are scheduled on HN or/and in VE(s) (for example updatedb) which might slow the node dramatically, especially if they are going to run at the same time.
Could you check the node's load when it becomes slow? Which processes eat the resources?
Thank you.
If your problem is solved - please, report it!
It's even more important than reporting the problem itself...
|
|
|
|
| Re: VPS/Apache HELP [message #16227 is a reply to message #16225] |
Wed, 29 August 2007 09:17 |
mcmuney
Messages: 10
Registered: July 2007
|
Junior Member |
|
|
I don't think it's cron. I do have some crons running, but it doesn't start until 1:30, the problem usually ranges between 11-1:30. However, there is a system backup cron running (Minute=1,16,31,46), but if this is the problem, I'd expect this problem to occur every 15 minutes, right?
I just ran the "top" command for your review: http://rateometer.com/projects/ssh_top.jpg
However, this command was run after the site started functioning properly, so it potentially may not have captured the abnormality.
|
|
|
|
| Re: VPS/Apache HELP [message #16228 is a reply to message #16227] |
Wed, 29 August 2007 09:26 |
khorenko
Messages: 533
Registered: January 2006
Location: Moscow, Russia
|
Senior Member |
|
|
Yes, exactly, it makes sense to check what eats the resources when the node is slow. Please, try to find the weird process when the node is slow. You are also right about the 15 minutes period, so the backup doesn't look like the culprit.
If your problem is solved - please, report it!
It's even more important than reporting the problem itself...
|
|
|
|
| Re: VPS/Apache HELP [message #16235 is a reply to message #16227] |
Wed, 29 August 2007 11:57 |
Valmont
Messages: 225
Registered: September 2005
|
Senior Member |
|
|
Maybe troubles are with network part, not with cpu. Could you check network connections at this time on hardware node? ( netstat,tcpdump and so on ).
High network load can slow down httpd as high load averages.
// And, please, tell me, where you had checking top, ps aux? On hw or in vps?
[Updated on: Wed, 29 August 2007 12:03] Report message to a moderator |
|
|
|
|
|
|
|
|
|
| Re: VPS/Apache HELP [message #16306 is a reply to message #16300] |
Fri, 31 August 2007 09:28 |
khorenko
Messages: 533
Registered: January 2006
Location: Moscow, Russia
|
Senior Member |
|
|
| mcmuney wrote on Fri, 31 August 2007 02:09 |
Is there a better way to capture the screenshot? Any help will be greatly appreciated.
|
Your screenshots are fine, but they show info inside a VE and unfortunately i don't see any abnormals. :\
Could you please post the same info collected on a HN?
Yes, i understand, you cann't connect to the node when it slows down, but may be this can be done by putting some task to cron which should drop the info to the local file once in a 10-15 minutes?
If your problem is solved - please, report it!
It's even more important than reporting the problem itself...
|
|
|
|
|
|
|
|
OpenVZ Forum
Members
Search
Help
Register
Login
Home
