Tracking high loads at certain time of the day

P

pmcwebs

Member
Hey all,

For about 22 hours a day my CPanel VPS runs brilliantly, however once a day, just after 5 pm (Australian DST) the server load goes up. I've logged a support ticket but nothing was found (and apparently the loads are within acceptable range).

The loads only go up to about 2.5 - 3, however the server becomes extremely slugglish with load times of a simple html page up to 10 -15 seconds, and WHM pages take about 20 seconds to load. The high loads go for up to 2 hours, but usually settle down within an hour or so.

It happens every day without fail. I've checked all my cron jobs and none of them run at this time. My backups run overnight so it's not backups either.

Are there any processes that Cpanel runs at a certain time that won't be listed in cron? What about the KH backups, could they be running at this time and impacting on performance?

I've sat here and refreshed apache logs, domlogs, error logs, all to no avail.

I have a second VPS here which gets higher loads at the same time, but the loads are quite acceptable and server response isn't affected.

The server has 512 Mb base ram and like I said, runs brilliantly at other times of the day.

'Something' is happening to the server at the same time each day, and I'm buggered if I can track it down. Any hints or suggestions gratefully accepted.

Pete M
 
Server backups will bog down the entire box, and being on the other side of the globe isn't going to be improving the situation. Fiber optics can only do so much.

Your second account may simply be on a more sparsely populated machine, or one in which more users have moved their backup time settings away from the default.
 
Pete,

What kind of processes do you see in first couple lines in "top" during the high load time?
Scheduled backups can slow down disk access a bit but won't block it as backup creation process is running with lowest possible priority. Also, we run backups every other day, not daily. So, I guess, VPS backup creation task isn't the actual source for the problem you're having. It might contribute to load every other day but not daily.

Regards,
Paul
 
KnownHost-P said:
Pete,

What kind of processes do you see in first couple lines in "top" during the high load time?
Scheduled backups can slow down disk access a bit but won't block it as backup creation process is running with lowest possible priority. Also, we run backups every other day, not daily. So, I guess, VPS backup creation task isn't the actual source for the problem you're having. It might contribute to load every other day but not daily.

Regards,
Paul
Click to expand...


From memory the only one that is up the top on a regular basis during these times is named, but I'll check it out later tonight (which will be in 6 hours from now)

As an example, I've uploaded a screenshot of my loadavg graph, which shows what happens at the specified time of the day compared with the load at other times. (On this occasion it was a load of just over one, but it regularly gets up over 2.5 - 3.0)


Thanks for the responses.
 

Attachments

  • load.jpg
    load.jpg
    13.2 KB · Views: 16
Well, just like clockwork it happened again tonight. Here is a graphic from my loadavg program:

[Broken External Image]:http://www.pmcwebs.com.au/images/load2.jpg


Here is the output from the load.sh program that support loaded for me a week ago.


Code: 
LOAD STATUS RUN ON Thu Dec 13 17:30:05 EST 2007
*******************
*******************
 W Results
 17:30:06 up 4 days,  5:27,  1 user,  load average: 1.99, 1.25, 0.98
USER     TTY      FROM              LOGIN@   IDLE   JCPU   PCPU WHAT
root     pts/0    ppp59-167-72-209 17:23    1:19   0.85s  0.84s pico loadtest.t
*******************
*******************
PSTREE Results
*******************
init,1    \040\040
  |-antirelayd,32453
  |-chkservd,28270
  |-courierlogger,10233 -pid=/var/run/imapd.pid -start -name=imapd /usr/lib/courier-imap/libexec/couriertcpd -address=0-maxprocs=50
  |   `-couriertcpd,10234 -address=0 -maxprocs=50 -maxperip=30 -nodnslookup -noidentlookup 143/usr/lib/courier-imap
  |-courierlogger,11266 -pid=/var/run/imapd-ssl.pid -start -name=imapd-ssl /usr/lib/courier-imap/libexec/couriertcpd -address=0-max
  |   `-couriertcpd,11267 -address=0 -maxprocs=50 -maxperip=30 -nodnslookup -noidentlookup 993/usr/lib/courier-imap
  |-courierlogger,11273 -pid=/var/run/pop3d.pid -start -name=pop3d /usr/lib/courier-imap/libexec/couriertcpd -address=0-maxprocs=50
  |   `-couriertcpd,11275 -address=0 -maxprocs=50 -maxperip=4 -nodnslookup -noidentlookup 110/usr/lib/courier-imap/
  |-courierlogger,11280 -pid=/var/run/pop3d-ssl.pid -start -name=pop3d-ssl /usr/lib/courier-imap/libexec/couriertcpd -address=0-max
  |   `-couriertcpd,11281 -address=0 -maxprocs=50 -maxperip=30 -nodnslookup -noidentlookup 995/usr/lib/courier-imap
  |-courierlogger,11294 -pid=/var/spool/authdaemon/pid -facility=mail -start /usr/libexec/courier-authlib/authdaemond
  |   `-authdaemond,11295
  |       |-authdaemond,11296
  |       |   `-authProg,3264
  |       |-authdaemond,11297
  |       |   `-authProg,25871
  |       |-authdaemond,11299
  |       |   `-authProg,9686
  |       |-authdaemond,11300
  |       |   `-authProg,22469
 |       `-authdaemond,11301
  |           `-authProg,7739
  |-cpanellogd,28073
  |-cpbandwd,27669
  |-cpdavd,28117
  |-cphulkd.pl,28113
  |-cpsrvd-ssl,21641
  |   |-cpsrvd-ssl,26505
  |   |-cpsrvd-ssl,26545
  |   `-cpsrvd-ssl,26547
  |-crond,11713
  |   |-crond,26361
  |   |   `-sh,26514 -c /usr/local/cpanel/bin/dcpumon\040>/dev/null\0402>&1
  |   |       `-dcpumon,26520
  |   |-crond,26362
  |   |   `-sh,26515 /load.sh
  |   |       `-pstree,27657 -apu
  |   `-crond,26366
  |       `-sh,26513 -c /usr/local/bin/php\040-q\040/home/pmcwebs/public_html/stats/logger.php\040>/dev/null\0402>&1
  |           `-php,26518 -q /home/pmcwebs/public_html/stats/logger.php
  |-exim,32399,mailnull -bd -q60m
  |   `-exim,13724,root -q
  |       `-exim,13748 -q
  |           `-exim,13749,mailnull -q
  |-exim,32403,mailnull -tls-on-connect -bd -oX 465
  |-exim,1350,mailnull -oX 26 -bd
  |-httpd,5588 -k start -DSSL
  |   |-httpd,12012 -k start -DSSL
  |   |-httpd,30399,nobody -k start -DSSL
  |   |-httpd,3188,nobody -k start -DSSL
  |   |-httpd,5647,nobody -k start -DSSL
  |   |-httpd,19974,nobody -k start -DSSL
  |   |-httpd,26143,nobody -k start -DSSL
  |   |-httpd,30031,nobody -k start -DSSL
  |   |-httpd,25734,nobody -k start -DSSL
  |   |-httpd,25735,nobody -k start -DSSL
  |   |-httpd,28273,nobody -k start -DSSL
  |   |-httpd,17511,nobody -k start -DSSL
  |   `-httpd,26040,nobody -k start -DSSL
  |-lfd,32014
  |-mysqld_safe,3695 /usr/bin/mysqld_safe --datadir=/var/lib/mysql --pid-file=/var/lib/mysql/server7.pmcwebs.com.pid
  |   `-mysqld,3728,mysql --basedir=/ --datadir=/var/lib/mysql --user=mysql --pid-file=/var/lib/mysql/server7.pmcwebs.com.pid...
  |-named,9892,named -u named
  |-portsentry,12158 -tcp
  |-pure-authd,1533 -s /var/run/ftpd.sock -r /usr/sbin/pureauth
  |-pure-ftpd,1527                                                                                                              ...
  |   `-pure-ftpd,24532,hooked                                                                                                  ...
  |-sshd,9970
  |   `-sshd,5463
  |       `-login,5705
  |           `-bash,5733
  |               `-pico,16251 loadtest.txt
  |-syslogd,7754 -m 0
  `-xinetd,9986 -stayalive -pidfile /var/run/xinetd.pid
*******************
*******************
MYSQLADMIN Results
*******************
+-------+----------------------+-----------+-----------------+---------+------+------------------+------------------+
| Id    | User                 | Host      | db              | Command | Time | State            | Info             |
+-------+----------------------+-----------+-----------------+---------+------+------------------+------------------+
| 96875 | beauy_bfc            | localhost | beauy_footyclub | Sleep   | 2794 |                  |                  |
| 97030 | beauy_bfc            | localhost | beauy_footyclub | Sleep   | 2340 |                  |                  |
| 97171 | beauy_bfc            | localhost | beauy_footyclub | Sleep   | 1407 |                  |                  |
| 97655 | root                 | localhost |                 | Query   | 0    |                  | show processlist |
| 97656 | unauthenticated user | localhost |                 | Connect |      | Reading from net |                  |
+-------+----------------------+-----------+-----------------+---------+------+------------------+------------------+
*******************
*******************
MEMORY USAGE
*******************
             total       used       free     shared    buffers     cached
Mem:           512        130        381          0          0          0
-/+ buffers/cache:        130        381
Swap:            0          0          0
###################
 
Part 2:

And here is output from ps -auxh --sort=size | tac


Code: 
USER       PID %CPU %MEM   VSZ  RSS TTY      STAT START   TIME COMMAND
named     9892  0.0  0.8 69776 4320 ?        Ssl  14:26   0:02 /usr/sbin/named -u named
mysql     3728  0.0  2.2 20720 11748 ?       Sl   Dec10   2:38 /usr/sbin/mysqld --basedir=/ --datadir=/var/lib/mysql --user=mysql --pid-file=/var/lib/mysql/server7.pmcwebs.com.pid --skip-external-locking
root     28117  0.0  1.7 14784 8936 ?        S    Dec10   0:00 cpdavd - accepting connections on 2077 and 2078
root     28073  0.0  1.5 10952 8312 ?        SN   Dec10   0:44 cpanellogd - sleeping for logs
root     28113  0.0  1.2  9880 6604 ?        Ss   Dec10   0:00 cPhulkd - processor
nobody   19974  0.0  2.2 21300 11816 ?       S    16:49   0:00 /usr/local/apache/bin/httpd -k start -DSSL
root     32014  0.0  1.0  7480 5616 ?        S    Dec12   0:04 lfd - sleeping
nobody   30399  0.0  2.2 20956 11852 ?       S    16:12   0:01 /usr/local/apache/bin/httpd -k start -DSSL
nobody    3188  0.0  2.2 20892 11604 ?       S    16:43   0:00 /usr/local/apache/bin/httpd -k start -DSSL
nobody   25734  0.0  2.1 20792 11196 ?       S    17:19   0:00 /usr/local/apache/bin/httpd -k start -DSSL
nobody    5647  0.0  2.2 20816 11808 ?       S    16:44   0:00 /usr/local/apache/bin/httpd -k start -DSSL
nobody   26143  0.0  2.1 20728 11272 ?       S    16:51   0:00 /usr/local/apache/bin/httpd -k start -DSSL
nobody   30031  0.0  2.2 20748 11620 ?       S    16:52   0:00 /usr/local/apache/bin/httpd -k start -DSSL
nobody   25735  0.0  2.1 20656 11284 ?       S    17:19   0:00 /usr/local/apache/bin/httpd -k start -DSSL
nobody   28273  0.0  2.0 20624 10984 ?       S    17:20   0:00 /usr/local/apache/bin/httpd -k start -DSSL
nobody   26040  0.0  2.0 20600 10888 ?       S    17:29   0:00 /usr/local/apache/bin/httpd -k start -DSSL
root     28270  0.0  1.0  6848 5256 ?        S    Dec11   0:03 chkservd
root      5588  0.0  2.2 20328 11976 ?       Ss   Dec12   0:02 /usr/local/apache/bin/httpd -k start -DSSL
root     21641  0.0  2.0 21352 10508 ?       S    00:59   0:02 cpsrvd - waiting for connections
root     12012  0.0  1.5 19104 8040 ?        S    Dec12   0:00 /usr/local/apache/bin/httpd -k start -DSSL
root     27669  0.0  0.9  6412 4744 ?        S    Dec10   0:01 cpbandwd
root     22469  0.0  0.9  6664 5136 ?        S    07:46   0:03 /etc/authlib/authProg
root      9686  0.0  0.9  6664 4964 ?        S    07:42   0:03 /etc/authlib/authProg
root      7739  0.0  0.9  6664 5136 ?        S    07:41   0:03 /etc/authlib/authProg
root      3264  0.0  0.9  6664 5136 ?        S    07:40   0:03 /etc/authlib/authProg
root     25871  0.0  0.9  6664 5204 ?        S    Dec10   0:16 /etc/authlib/authProg
root     32453  0.0  0.2  3524 1504 ?        S    Dec11   0:05 antirelayd
mailnull  1350  0.0  0.2  7404 1140 ?        Ss   Dec11   0:00 /usr/sbin/exim -oX 26 -bd
mailnull 32403  0.0  0.2  7404 1188 ?        Ss   Dec11   0:00 /usr/sbin/exim -tls-on-connect -bd -oX 465
mailnull 32399  0.0  0.2  7404 1208 ?        Ss   Dec11   0:05 /usr/sbin/exim -bd -q60m
root     11713  0.0  0.1  2480  940 ?        Ss   Dec09   0:00 crond
nobody   32327  0.0  0.1  2308  804 ?        R    17:31   0:00 ps -auxh --sort=size
root      5463  0.0  0.4  7056 2280 ?        Ss   17:23   0:00 sshd: root@pts/0 
hooked   32062  0.0  0.2  3948 1080 ?        D    17:31   0:00 pure-ftpd (IDLE)                                                                                                                                                                                                                            
root      5705  0.0  0.2  2952 1320 ?        Ss   17:23   0:00 login -- root                                                     
root      1527  0.0  0.2  3840 1316 ?        Ss   Dec11   0:00 pure-ftpd (SERVER)                                                                                                                                                                                                                          
root      9970  0.0  0.1  4012 1028 ?        Ss   Dec09   0:00 /usr/sbin/sshd
root     11301  0.0  0.0  1804  372 ?        S    Dec10   0:00 /usr/libexec/courier-authlib/authdaemond
root     11300  0.0  0.0  1804  372 ?        S    Dec10   0:00 /usr/libexec/courier-authlib/authdaemond
root     11299  0.0  0.0  1804  372 ?        S    Dec10   0:00 /usr/libexec/courier-authlib/authdaemond
root     11297  0.0  0.0  1804  368 ?        S    Dec10   0:00 /usr/libexec/courier-authlib/authdaemond
root     11296  0.0  0.0  1804  372 ?        S    Dec10   0:00 /usr/libexec/courier-authlib/authdaemond
root     11295  0.0  0.1  1804  608 ?        S    Dec10   0:00 /usr/libexec/courier-authlib/authdaemond
root      1533  0.0  0.1  3576  912 ?        S    Dec11   0:00 /usr/sbin/pure-authd -s /var/run/ftpd.sock -r /usr/sbin/pureauth
root      9986  0.0  0.1  2080  836 ?        Ss   Dec09   0:00 xinetd -stayalive -pidfile /var/run/xinetd.pid
root      5733  0.0  0.2  2160 1332 pts/0    Ss+  17:23   0:00 -bash
root         1  0.0  0.1  1628  596 ?        Ss   Dec09   0:00 init [3]      
root      3695  0.0  0.2  2152 1120 ?        S    Dec10   0:00 /bin/sh /usr/bin/mysqld_safe --datadir=/var/lib/mysql --pid-file=/var/lib/mysql/server7.pmcwebs.com.pid
nobody   32328  0.0  0.0  2148  428 ?        D    17:31   0:00 sh -c ps -auxh --sort=size | tac
nobody   32326  0.0  0.1  2148 1020 ?        S    17:31   0:00 sh -c ps -auxh --sort=size | tac
root      7754  0.0  0.1  1532  568 ?        Ss   Dec11   0:02 syslogd -m 0
root     12158  0.0  0.0  1488  396 ?        Ss   Dec09   0:00 /usr/sbin/portsentry -tcp
root     11281  0.0  0.0  1568  512 ?        S    Dec10   0:00 /usr/lib/courier-imap/libexec/couriertcpd -address=0 -maxprocs=50 -maxperip=30 -nodnslookup -noidentlookup 995 /usr/lib/courier-imap/bin/couriertls -server -tcpd /usr/lib/courier-imap/sbin/pop3login /usr/lib/courier-imap/bin/pop3d Maildir
root     11275  0.0  0.0  1568  512 ?        S    Dec10   0:01 /usr/lib/courier-imap/libexec/couriertcpd -address=0 -maxprocs=50 -maxperip=4 -nodnslookup -noidentlookup 110 /usr/lib/courier-imap/sbin/pop3login /usr/lib/courier-imap/bin/pop3d Maildir
root     11267  0.0  0.0  1568  512 ?        S    Dec10   0:00 /usr/lib/courier-imap/libexec/couriertcpd -address=0 -maxprocs=50 -maxperip=30 -nodnslookup -noidentlookup 993 /usr/lib/courier-imap/bin/couriertls -server -tcpd /usr/lib/courier-imap/sbin/imaplogin /usr/lib/courier-imap/bin/imapd Maildir
root     10234  0.0  0.0  1568  512 ?        S    Dec10   0:00 /usr/lib/courier-imap/libexec/couriertcpd -address=0 -maxprocs=50 -maxperip=30 -nodnslookup -noidentlookup 143 /usr/lib/courier-imap/sbin/imaplogin /usr/lib/courier-imap/bin/imapd Maildir
root     11280  0.0  0.0  1460  388 ?        S    Dec10   0:00 /usr/sbin/courierlogger -pid=/var/run/pop3d-ssl.pid -start -name=pop3d-ssl /usr/lib/courier-imap/libexec/couriertcpd -address=0 -maxprocs=50 -maxperip=30 -nodnslookup -noidentlookup 995 /usr/lib/courier-imap/bin/couriertls -server -tcpd /usr/lib/courier-imap/sbin/pop3login /usr/lib/courier-imap/bin/pop3d Maildir
root     11273  0.0  0.0  1460  388 ?        S    Dec10   0:01 /usr/sbin/courierlogger -pid=/var/run/pop3d.pid -start -name=pop3d /usr/lib/courier-imap/libexec/couriertcpd -address=0 -maxprocs=50 -maxperip=4 -nodnslookup -noidentlookup 110 /usr/lib/courier-imap/sbin/pop3login /usr/lib/courier-imap/bin/pop3d Maildir
root     11266  0.0  0.0  1460  388 ?        S    Dec10   0:00 /usr/sbin/courierlogger -pid=/var/run/imapd-ssl.pid -start -name=imapd-ssl /usr/lib/courier-imap/libexec/couriertcpd -address=0 -maxprocs=50 -maxperip=30 -nodnslookup -noidentlookup 993 /usr/lib/courier-imap/bin/couriertls -server -tcpd /usr/lib/courier-imap/sbin/imaplogin /usr/lib/courier-imap/bin/imapd Maildir
root     10233  0.0  0.0  1460  388 ?        S    Dec10   0:00 /usr/sbin/courierlogger -pid=/var/run/imapd.pid -start -name=imapd /usr/lib/courier-imap/libexec/couriertcpd -address=0 -maxprocs=50 -maxperip=30 -nodnslookup -noidentlookup 143 /usr/lib/courier-imap/sbin/imaplogin /usr/lib/courier-imap/bin/imapd Maildir
root     11294  0.0  0.0  1456  384 ?        S    Dec10   0:00 /usr/sbin/courierlogger -pid=/var/spool/authdaemon/pid -facility=mail -start /usr/libexec/courier-authlib/authdaemond
 
Hello Pmcwebs,

Another place to look is in WHM | Server Status | CPU/Memory/MySQL usage. There you will also find links to look at the previous days too.

Hope that helps
 
Try top rather than ps; nothing in that last one is actively being scheduled for any CPU time at all that I can see, and sorting by "size" only tells you which process is using the greatest (approximate) amount of swap space on the drive--not what you're interested in at all.

Also might want to try turning off automatic updates in cPanel and see if that has anything to do with it. If not, try disabling some of these statistics loggers. Right now you've got

|-crond,11713
| |-crond,26361
| | `-sh,26514 -c /usr/local/cpanel/bin/dcpumon\040>/dev/null\0402>&1
| | `-dcpumon,26520
| |-crond,26362
| | `-sh,26515 /load.sh
| | `-pstree,27657 -apu
| `-crond,26366
| `-sh,26513 -c /usr/local/bin/php\040-q\040/home/pmcwebs/public_html/stats/logger.php\040>/dev/null\0402>&1
| `-php,26518 -q /home/pmcwebs/public_html/stats/logger.php

All firing at once, and nothing chews up CPU time like parsing gigantic text files full of tab-delimited statistics.
 
Hey guys,

Thanks for your suggestions. I'll fiddle with those scripts in crond and wipe them out for a while to see if that helps.

Things turned from bad to worse last night, after my last post I got hit with loads in excess of 20 and the server sites were inaccessible for most of the night.

Not sure if it was totally related to the 'same time of the day' load issue, but it was certainly pretty bad for the server.



[Broken External Image]:http://www.pmcwebs.com.au/images/load4a.jpg
 
We had the same problem on ours for a while as well. Basically every night at the same time, loads would go through the roof, and everything would take forever to load, start timing out, etc. I was never able to run top or auwwwwfx because SSH would just time out.

Support tried telling me that the timing out was a network connectivity issue (even though pings and tracert and such showed nothing unusual), and couldn't happen due to the high loads. Ummmm sure. They even tried questioning what port I was using for SSH, even though SSH, HTML, PHP, cPanel, everything wouldn't load.

At first, KH thought that the cpaddons CRON was causing the issue, and changed its time, but that ultimately didn't help. I'm really not sure what the issue was, as we never really got anything resolved over in the support tickets, but the problem just seems to have gone away. I haven't had any issues recently (the last ticket was from Oct. 31), so I'm keeping my fingers crossed that KH found something on the server and fixed it.
 
Bryan said:
We had the same problem on ours for a while as well. Basically every night at the same time, loads would go through the roof, and everything would take forever to load, start timing out, etc. I was never able to run top or auwwwwfx because SSH would just time out.

Support tried telling me that the timing out was a network connectivity issue (even though pings and tracert and such showed nothing unusual), and couldn't happen due to the high loads. Ummmm sure. They even tried questioning what port I was using for SSH, even though SSH, HTML, PHP, cPanel, everything wouldn't load.

At first, KH thought that the cpaddons CRON was causing the issue, and changed its time, but that ultimately didn't help. I'm really not sure what the issue was, as we never really got anything resolved over in the support tickets, but the problem just seems to have gone away. I haven't had any issues recently (the last ticket was from Oct. 31), so I'm keeping my fingers crossed that KH found something on the server and fixed it.
Click to expand...

Thanks for the info Bryan. A couple of nights recently the loads have cracked 20, and as you point out, troubleshooting is no good because ssh times out etc. I've had support look at it but no joy. I did notice last night that syslogd was always coming to 'top' when the loads went up, so I killed the process and loads came down. It's too early to tell if syslog is the problem, but I'll leave it off for a day or so to monitor. If I work out that syslog IS the problem, it will then become an issue to find out WHY syslog is the problem, as the answer obviously isn't just to leave it turned off.
 
Pmcwebs,

What I would suggest is that you actually dump your log files every month. The possiblity is that your log files are full, could be beyond 500mb, and when they run, it basicly uses up alot of CPU to parse through that 500mb of files. I would also suggest turning off AWSTATS, and WEBALIZER, as these stat programs can actually cause high CPU usage when you are dealing with sites that have higher then 50 people on the site at one time. Some of the other things I would suggest is actually limiting apache and tweaking it, such as start servers be lowered to 3 and max severs is lowered down to 5.

for example:
KeepAlive Off
MaxKeepAliveRequests 256
KeepAliveTimeout 2
MinSpareServers 2
MaxSpareServers 5
StartServers 3
MaxClients 100
MaxRequestsPerChild 150

has allowed me to have over 200 people online at once in which the VPS was still stable to hold more, with a CPU AVG of less then 1.75.

I would also suggest looking at your MySQL, and checking the # of connections or number of query runs. You should also monitor how much information in Megs does the MySQL server output. You should look at your my.cnf in your /etc/ folder and actually go over the configurations to limit the number of information that can fly out. These query runs can actually put load on your CPU!

-Andrew
 
Andrew,

Thanks for your post. I've been monitoring my log files for the last couple of weeks, none of them are above 10 mb and they get rotated daily. I've turned off awstats but I keep running Webalizer.

I ended up requesting that my VPS be moved to another node and once that was done (Thanks Paul for being so prompt about it) - the loads have dropped immensely and the server is running great. There is no difference in loads at any time of the day (except when I run my own backups) and the server is always lightning quick.

Having said that, I have no idea what the issue was, but it is the second time in 12 months that moving host servers has fixed an erratic load problem. The other time was 6 months ago with another provider. There was obviously some conflict between my VPS and the host. Whether that was because of other VPS's that were on that host or not, I'm not sure. My 5 minute load average is now 0.02 for the entire day and the server is lightning fast.

Thanks everyone for your help.

Pete
 
You must log in or register to reply here.
Top