SQL queries causing slowdown and outages

DRE

DRE

Well-known member
I had lots of downtime with xFHost.me so I left and when I did I was given a chance to have my site hosted at a VPS for a limited time. During my stay my site went down again and they gave me a list of what MySQL queries were slowing down my server and deleted my site. So I'm off to looking for another host again. I think 8thos will be down for longer period since I will have to find someone to look at this .txt file to see what's causing issues with my database. I don't know who does anything like that so I will have to take some time this weekend to research.
 
Sort by date Sort by votes
tenants said:
For the servers that you were on, do you know which of these were using litespeed webserver:


I hazard a guess that xFHost and Knownhost do (but the others don't)?
Click to expand...
ahaha @ me mis-numbering that list. Sorry I didn't see your reply or any of your replies. I think something's wrong with my xenforo account cause I don't always get alerts. Anyway, I know for sure URLJet used Litespeed. I can't remember if MDDHosting did. I know xFHost did at some point. I've removed over a dozen addons and two styles due to the problems I had a couple weeks ago. Here's an updated list of the addons I have currently installed. Please reply there instead of here if you have any questions: http://xenforo.com/community/threads/8thos-com.31074/page-3#post-498534
 
Upvote 0 Downvote
Well I am not going to point fingers and blame anyone as I don't know what happened and what didn't. I know I am hosted with xFHost and the only problems I have had is when litespeed started acting up and then one other time when they were getting the hourly backups setup and apparently the server didn't restart like it was supposed to and Mike wasn't alerted. Other than that I have had no other issues. Sorry you had issues DRE and hopefully your site will continue to run fine on Nimbus. Not all hosting companies are created equal. What may work for one person may not work for someone else. I have used alot of hosting companies throughout the years and so far the one I am using now has been the best.
 
Upvote 0 Downvote
Sadly this thread is the sole reason I would never go back to shared hosting

If a single site can bring a server to it's knees, it isn't worth the risk.
 
Upvote 0 Downvote
There's still no evidence to suggest it is the fault of DRE's site, to be fair.

Yeah, ok, the introduction of his site caused problems on that VPS, but when it was introduced onto another, it hasn't.

So, ultimately, could be a bad server config perhaps not helped by certain aspects of the site, but I think it's unfair to blame everything on the site.
 
Upvote 0 Downvote
AdamD said:
If a single site can bring a server to it's knees, it isn't worth the risk.
Click to expand...

Agree, that is why we moved from Lite Speed to Nginx, Lite Speed does not seem to work correctly with Cloud Linux under PHP 5.4. Ever since moving to Nginx, no site has been able to go over it's shared resource amount. That is why Dre continued to have issues as were Ernest had none.


Chris Deeming said:
There's still no evidence to suggest it is the fault of DRE's site, to be fair.
Click to expand...

On the flip of that coin, once Cloud Linux was working like it should, it was only Dre's site that continued to have out of memory and MySQL failure errors, due to him running 90/100% resources 24/7. Also since him switching, average resource usage as been down by a lot. We have sites with millions of posts and 400+ online not using the amount of resources Dre used. I think the evidence points to one of the add-ons he uninstalled.

Also, just checking on how fast his site was loading now. I noticed a member stating in shout box earlier today based on time stamp "Keep getting errors cannot go into any threads"
 
Upvote 0 Downvote
Well true to form, DRE has been playing with add-ons today including removing several so that was the potential cause of errors.

But certainly the VPS' general performance and stability hasn't been affected.

It's just going to be one of those unexplained things.
 
Upvote 0 Downvote
K this is my last update to this thread since the problem seems to have resolved itself.

I still have no idea what caused the issue but I now have 95 addons installed but only 85 addons are actively enabled.

I am back to averaging around 10 to 15 members online a day and the server I'm on seems to be stable.

I don't know if the site is stressing out the host or anything like that. I would have to ask Chris.
 
Upvote 0 Downvote
Today I moved from nimbushosting to XenFormula and the database issues occured again, but Slavik fixed it.

I hope it doesn't occur again if I ever go on another host geez wtf.

But yeah my site's going blazing fast now. Wish I knew what the heck keeps popping up. I've reduced all my addons by 1/2 man.
 
  • Like
Reactions: rdn
Upvote 0 Downvote
Every night, my site goes down around 1AM, so now I know the problem is with a cron that my site's running and not the actual server. This has persisted on every host I've been on since I left xFHost.
 
Upvote 0 Downvote
DRE said:
Every night, my site goes down around 1AM, so now I know the problem is with a cron that my site's running and not the actual server. This has persisted on every host I've been on since I left xFHost.
Click to expand...

I often see a server error at that same time - but the server still works. This is only on my large site.....the small one probably does not have enough entries or add-ons to overload. I never bothered to look up which cron is doing it (or maybe it's another backup script running and overtaxing the server)....since it does not throw off an httpd error - that is, the site still works...

Code: 
Zend_Db_Adapter_Mysqli_Exception: Too many connections - library/Zend/Db/Adapter/Mysqli.php:333
Generated By: Unknown Account, Wednesday at 1:03 AM

EDIT - I think this is because an entire db dump is done from a unix cron job at that time - so XF probably get's a momentary error.

Uptime is my #1 requirement for my big site and very important for ALL sites. As an indication of my expectations- in the last 12 years, my site has been down maybe a total of 10 hours. So one hour a year is OK.....if my site was a complete hobby site (no expenses or revenue of any means or no intention of), then 99.9% would satisfy me - that's about 10 hours per year.
 
  • Like
Reactions: DRE
Upvote 0 Downvote
These are all the crons that are running in AdminCP/Tools. As you can see, none of them run around 1AM which is when the site goes down every day.

8thoscrons1.webp
8thoscrons2.webp
 
Upvote 0 Downvote
You must log in or register to reply here.

Similar threads

Zig
  • Question Question
MG 2.2 Media Gallery causing excessive database load
Replies
0
Views
392
Zig
Zig
H
  • Question Question
These will sound like silly questions, probably...
Replies
5
Views
527
CedricV
CedricV
AndrewSimm
  • Question Question
XF Cloud Question
Replies
2
Views
440
AndrewSimm
AndrewSimm
I
  • Question Question
XF 2.2 Logging Max Connections and IP Address
Replies
1
Views
420
Paul B
Paul B
Black Tiger
  • Solved
XF 2.2 [solved] 403 but no block present?
Replies
5
Views
401
Black Tiger
Black Tiger
Top Bottom