Can't understand why my Linode is so slow
My Linode 512 is hosting this Wordpress site: http://www.corstorphinehill.co.uk
It's not getting much traffic.
Sometimes, maybe every 15th page, it will take the page up to 40 seconds to load. It seems to spend a lot of time "waiting for website" in Firefox.
My YSlow score is Grade A (93/100)
My PageSpeed score is 95/100
I've ruled out my firewall (ufw) by temporarily disabling it. The load never seems to spike. I'm at a bit of a dead end. Can anyone please help? Is there any way to tell exactly why my browser is waiting?
Here's my apache config:
ServerRoot "/etc/apache2"
LockFile /var/lock/apache2/accept.lock
PidFile ${APACHE_PID_FILE}
Timeout 300
KeepAlive Off
MaxKeepAliveRequests 100
KeepAliveTimeout 2
<ifmodule mpm_prefork_module="">KeepAliveTimeout 2
StartServers 5
MinSpareServers 5
MaxSpareServers 5
MaxClients 20
MaxRequestsPerChild 1000</ifmodule>
<ifmodule mpm_worker_module="">StartServers 1
MinSpareThreads 25
MaxSpareThreads 75
ThreadLimit 64
ThreadsPerChild 25
MaxClients 8
MaxRequestsPerChild 0</ifmodule>
<ifmodule mpm_event_module="">StartServers 1
MaxClients 8
MinSpareThreads 25
MaxSpareThreads 75
ThreadLimit 64
ThreadsPerChild 25
MaxRequestsPerChild 0</ifmodule>
User ${APACHE_RUN_USER}
Group ${APACHE_RUN_GROUP}
AccessFileName .htaccess
<files ~="" "^\.ht"="">Order allow,deny
Deny from all
Satisfy all</files>
DefaultType text/plain
HostnameLookups Off
ErrorLog /var/log/apache2/error.log
LogLevel warn
Include /etc/apache2/mods-enabled/*.load
Include /etc/apache2/mods-enabled/*.conf
Include /etc/apache2/httpd.conf
Include /etc/apache2/ports.conf
LogFormat "%v:%p %h %l %u %t \"%r\" %>s %O \"%{Referer}i\" \"%{User-Agent}i\"" vhost_combined
LogFormat "%h %l %u %t \"%r\" %>s %O \"%{Referer}i\" \"%{User-Agent}i\"" combined
LogFormat "%h %l %u %t \"%r\" %>s %O" common
LogFormat "%{Referer}i -> %U" referer
LogFormat "%{User-agent}i" agent
logfile
CustomLog /var/log/apache2/other_vhosts_access.log vhost_combined
Include /etc/apache2/conf.d/
Include /etc/apache2/sites-enabled/
20 Replies
James
I think the problem might only be evident if I'm logged in to Wordpress, although it seems OK today. I'll do some more experimenting.
Thanks again!
@DrookitMunter:
I think the problem might only be evident if I'm logged in to Wordpress, although it seems OK today. I'll do some more experimenting.
I don't know WordPress specifically, but it wouldn't be surprising if caching was largely or completely disabled for logged-in users.
Thanks for everyone's help and input, it's appreciated
front page has three 150kb images in the slideshow
google analytics(13kb) and google pagespeed(54kb) and facebook js(43kb) loading in the head and 170kb of other facebook js loading later
3 web fonts
It looks nice, but it's not exactly built for speed.
Just to let you know, I loaded the front page a few times and I ended up getting a logged in page for DrookitMunter. I didn't actually have admin access, but you should check that you have your caching system properly configured, because it's caching admin users.
These things are tough to track down. Unfortunately, the tools are pretty technical. xhprof for profiling, mysql slow query log and EXPLAIN if it's in the database query. I've heard good things about New Relic free trial.
You might want to go through your wordpress modules and research each one and see if others are having problems with any.
Every refresh takes about 20 seconds.
@bradleyg:
Yeah
http://corstorphinehill.co.uk/category/blog/ is slow for me.Every refresh takes about 20 seconds.
It took a LONG time to load for me initially but then refreshing was just as fast as could be. The first load though was at least 30-45 seconds if not more.
Try disabling all wordpress plugins and themes and see if it fixes it, if it does enable each theme and plugin one by one and re-test until it goes slow again then you'll have found the culprit.
I also hired linode 512, and is taking around 20 seconds to load !! fucking insane, then when the client is connected is lighting fast, i think this problem is not related to our apache configuration instead is related to linode networking itself, and we can't do nothing about that !
i will make a support ticket about this ASAP, i am losing clients…
As obs said, it looks like a configuration problem in the web app software, not a Linode issue, but that's just my opinion.
Edit: PS My sites are done in GWT and they run fine.
(Are folks forgetting to set their KeepaliveTimeouts to 1 second or so?)
So? The problem is still reproducible.