Just trying to disagnose some SQL load on my database server, I attached a profiler and saw a lot of queries for all the different umbraco sites we are hosting.
I'm well versed in SQL so I'll throw more hardware at the server if I need to, but I first need to understand why i'm getting load.
The queries were as follows: SELECT createDate, trashed, parentId, nodeObjectType, nodeUser, level, path, sortOrder, uniqueID, text FROM umbracoNode WHERE id = 1516 select xml from cmsContentXml where nodeID = 1516
And a lot of them, about 20,000 in about 5 min.
It looks like the XML cache is being refreshed, but it was across a lot of different websites simultanously.
I guess I can disable the continuouslyUpdateDiskCache (could be named something slightly different) app setting , but does anyone else know what scenarios cause this table to get queried when the disk cache is enabled?
If it was for one or two sites I wouldn't be alarmed but it was happening for a lot of them simultanously
1. People browsing the backoffice (expand a tree with loads of nodes and you will see this) 2. Are your sites maybe not using the nodefactory but the document class directly at some points?
I'm curious too, we have the same situation; lot's of queries from our different umbraco websites (we don't use the document API). As far as I know, umbraco is just not set-up the right way when it comes to SQL queries; there are just too many of them (Document, Media and Member api/objects use SQL queries).
Yes I know. I'm also using nodefactory all the way, most of our websites don't even create sql queries when visiting the homepage (after things are cached, like media)...
Diagnosing SQL load
Hi Guys
Just trying to disagnose some SQL load on my database server, I attached a profiler and saw a lot of queries for all the different umbraco sites we are hosting.
I'm well versed in SQL so I'll throw more hardware at the server if I need to, but I first need to understand why i'm getting load.
The queries were as follows:
SELECT createDate, trashed, parentId, nodeObjectType, nodeUser, level, path, sortOrder, uniqueID, text FROM umbracoNode WHERE id = 1516
select xml from cmsContentXml where nodeID = 1516
And a lot of them, about 20,000 in about 5 min.
It looks like the XML cache is being refreshed, but it was across a lot of different websites simultanously.
I guess I can disable the continuouslyUpdateDiskCache (could be named something slightly different) app setting , but does anyone else know what scenarios cause this table to get queried when the disk cache is enabled?
If it was for one or two sites I wouldn't be alarmed but it was happening for a lot of them simultanously
Gareth
2 things to consider:
1. People browsing the backoffice (expand a tree with loads of nodes and you will see this)
2. Are your sites maybe not using the nodefactory but the document class directly at some points?
I'm curious too, we have the same situation; lot's of queries from our different umbraco websites (we don't use the document API).
As far as I know, umbraco is just not set-up the right way when it comes to SQL queries; there are just too many of them (Document, Media and Member api/objects use SQL queries).
Seb, The volume of the transactions didn't really line up with people browsing the backoffice - I think it was more likely the xml cache updating
Kipusoep - V5 will fix this, but in V4 I'm pretty sure there's not much that can be done
I don't use new Document() at all, definitely all NodeFactory if anything
Yes I know. I'm also using nodefactory all the way, most of our websites don't even create sql queries when visiting the homepage (after things are cached, like media)...
is working on a reply...