Cannot connect to my Linode VPS (but can still ping to it)
I'm running my site on a Linode VPS, powered by nginx, easyengine and wordpress. Just started to try Linode few days ago and totally love the SSD here, it's awesome. My SSD got 1GB/s speed and it's about 3 times as fast what I got with DigitalOcean. However, there are few things that I'm struggling with, and really hope u guys help me figure it out.
1:
After reboot my VPS with "shutdown - r now", I can't connect to my site anymore, although I still can ping to it and have absolute no issue with access the root account in console. My site returns "Cannot connect" status with every browser I tried. It was not my domain provider also, cuz when I try to connect via my VPS ip address, still can't.
2:
In my Linode account dashboard, my storage indicator shows that my VPS is full now, though I'm so sure that it shouldn't, my site is small. However, when I check it with "df -l" command, it returned absolutely normal (around 14% used). Dunno is it an false indication or something like that.
3:
My site is an ecommerce site, powered by wordpress and woocommerce. I turned off all of my cache plugins (W3 total and Wordfence) but my shopping cart page and check out page still got cached and show previous session instead of current session. I also created a file named "woo-nginx.conf" just like the tutorial here:
Would u guys plz help me to figure it out. Thanks in advance.
4 Replies
2. This is an indication that you've allocated all of your disk space to disks, not that the disks themselves are full.
@glg:
1. It's better to reboot from linode manager, not within your linode. You should open lish and see what happened. Perhaps ssh shut down, but your server didn't or ssh didn't come up.
2. This is an indication that you've allocated all of your disk space to disks, not that the disks themselves are full.
Thanks, glg for response.
I found out what caused my site down, and sadly, it was the solution for the #3 that led to #1 issue. I put those line below in my nginx.conf and dunno why it didn't work:
"if ( $cookiewoocommerceitemsincart != "0" ) {
set $skip_cache 1;
}
}
if ($requesturi ~* "(/shop.|/gio-hang.|/my-account.|/thanh-toan.|/addons.|/wp-admin/|/xmlrpc.php|wp-..php|/feed/|index.php|sitemap(index)?.xml|[a-z0-9_-]+-sitemap([0-9]+)?.xml)") {
set $skip_cache 1;
}
if ( $arg_add-to-cart != "" ) {
set $skip_cache 1;
}"
I removed it and my site is on again, however, I'm still looking for the answer for #3 issue
service nginx reload
if you're on debian/ubuntu.
It looks like you're using nginx's fastcgi_cache. E-commerce and full page caching don't play well together and it's hard to get a setup like that working trouble-free out of the box unless you grab the cart contents on reload with ajax or something similar. It really depends on your store layout, etc.
If it were me, I'd turn off caching completely until you know what you're doing. You can do that by commenting out the line that starts with:
fastcgi_cache
Just put # in front of the line like:````
fastcgi_cache
````
I solved my problems and just wanna head back to provide information just in case anyone has the same issue.
1. Actually, it was a syntax error in my config.
2. It's ok, checked with df -l and fine.
3. I switched to redis for object cache and it is totally great now.
Thanks for support, my buddies.