We have a site which is on a server farm, and uses a single database with dfrs replication to handle the file system changes.
We get an issue occasionally where changes are made to content in the umbraco back end, they show up on the front-end but then within a day or so the front end has reverted to the old version of the content. The back end still shows the new version, and republishing the page usually clears it up.
I've read that the umbraco.config file is a cached version of the XML content structure, but when I look at this file it's completely empty. I've also been told that there's an in-memory cache of the site which is refreshed when you either restart the app pool or republish the entire site.
Given that the front and back end content seems to be getting out of sync, I'm wondering if the cache (in-memory and/or file system) is not getting updated correctly.
Is this plausible or am I looking in the wrong place?
I'm relaying this from a while ago and I'm not a server guy, but here's what I was able to gather.
In our configuration, the ips for each server didn't resolve to the website. I believe there's some configuration section in the web farms umbraco guide that talks about distributed calls(?). We had to set up internal dns entries so the ips in the distributed calls section could resolve to the website.
I don't know if this is any help, I'm just relaying what I was able to find out.
Cache or umbraco.config issue? front-end content is reverting, back-end is fine
We have a site which is on a server farm, and uses a single database with dfrs replication to handle the file system changes.
We get an issue occasionally where changes are made to content in the umbraco back end, they show up on the front-end but then within a day or so the front end has reverted to the old version of the content. The back end still shows the new version, and republishing the page usually clears it up.
I've read that the umbraco.config file is a cached version of the XML content structure, but when I look at this file it's completely empty. I've also been told that there's an in-memory cache of the site which is refreshed when you either restart the app pool or republish the entire site.
Given that the front and back end content seems to be getting out of sync, I'm wondering if the cache (in-memory and/or file system) is not getting updated correctly.
Is this plausible or am I looking in the wrong place?
I can't help you other than to say I am experiencing the same issue.
I believe we did get this fixed in the end.
I'm relaying this from a while ago and I'm not a server guy, but here's what I was able to gather.
In our configuration, the ips for each server didn't resolve to the website. I believe there's some configuration section in the web farms umbraco guide that talks about distributed calls(?). We had to set up internal dns entries so the ips in the distributed calls section could resolve to the website.
I don't know if this is any help, I'm just relaying what I was able to find out.
is working on a reply...