Products

Solutions

Resources

Partners

Community

Blog

About

QA

Ideas Test

New Community Website

Ordinarily, you'd be at the right spot, but we've recently launched a brand new community website... For the community, by the community.

Yay... Take Me to the Community!

Welcome to the DNN Community Forums, your preferred source of online community support for all things related to DNN.
In order to participate you must be a registered DNNizen

HomeHomeOur CommunityOur CommunityGeneral Discuss...General Discuss...Web Farm IssuesWeb Farm Issues
Previous
 
Next
New Post
9/2/2008 6:04 PM
 

 I have set up a DotNetNuke site in a webfarm environment. I am using the "alternate" configuration with two webservers accessing a single database.

One server is serving pages averaging 7 seconds, and the other server is averaging 1 minute +! I have followed the DNN webfarm documentation to the T. Both server's IIS settings are identical. When I try accessing my site locally from the slow server, it is still slow. Any other html page or image is served immediately. Other web pages are fast. I even have the DNN site in its own application pool. Only the DNN pages are slow. I turned off file replication and still having the issue. Does anyone have any thoughts on what I should try??? 

 
New Post
9/4/2008 5:52 PM
 

So looking at the database traffic I was being hit with a lot of traffic from one of the webservers. I checked into the the scheduler and noticed that the broadcastpollingprovider scheduler was running every 30 seconds. The slow response was due to having to wait 1 minute + to get a database connection.

I turned off that scheduled job and now my site loads the same for both servers. However, I am still noticing very large growth of the Broadcasts and Servers table. My database size has gone from 94MB to 70GB! Is there any documentation on what these tables do? At a glance, they look like session logging Is there a clean up process or configuration I have missed?

 

(I have since figured out the problem, you will need to have the broadcastpollingprovider turned on. See post below for the rest of the solution.)

 
New Post
9/10/2008 9:18 AM
Accepted Answer 

Finally, I have the BroadcastPollingProvider implementation of the webfarm set up and functioning correctly. For some reason, one of my webservers kept making entries (around 20 per second) into the Servers table. Most of the entries in the LastPollDate where Null and occasionally one had a date. I was up to 1,000,000 + and growing when there should only be one entry per server. (Database size was 100Gig and increasing) I checked the provider code and could see why the entries where doing this yet, I don't know what started the issue. It lost its head somewhere when we were setting it up initially I guess. So, what I did to remedy the situation was to purge all entries for the problematic server in the Servers and Broadcasts table. So far after 24 hours I only have an entry per server in the servers table with an updating lastpolldate. I have no entries so far in the cached objects table.  The broadcasts table has been growing but I don't think it has grown to the point where the cleanup script starts purging. I will keep an eye on that though.

Also for those that are trying to implement the alternative webfarm environment, notice that in the web farm support document dated June 20, 2006 on page 10 the second bullet point reads:

Schedule Enabled" so it is checked. Click "Update".

I initially thought that maybe I should enable the DotNetNuke.Services.Cache.PurgeCache, DOTNETNUKE scheduled task since the first line in the "Supported Configuration" section on page 9 had some similar language:

 

 A scheduled task needs to be enabled. Log in as a SuperUser, go to the Host menu

and select “Schedule”. Edit the scheduled task

“DotNetNuke.Services.Cache.PurgeCache, DOTNETNUKE” by clicking on the edit

icon next to the scheduled task. Select the checkbox labeled “Schedule Enabled” so

it is checked. Click “Update”.

The DotNetNuke.Services.Cache.PurgeCache, DOTNETNUKE scheduled object does not need to be enabled. If it is enabled, then when you view its history it gives a message saying it is not necessary with the BroadcastPollingProvider.

I hope this brain dump may help someone who finds themselves in similar shoes. I could have used a post like this about a week ago!

 

 

 

Report Post 

 
New Post
9/16/2008 9:38 AM
 

After this last post, I noticed that I still had a problem with a growing Broadcasts table. I was up to 1,000,000 + records in just a matter of days. What I noticed was that the following SQL statement was supposed to be purging the broadcasts table every time the broadcastpollingprovider was scheduled to run (30 seconds by default):

 DELETE FROM {databaseOwner}{objectQualifier}Broadcasts
 WHERE BroadcastDate < (SELECT Min(LastPollDate) FROM {databaseOwner}{objectQualifier}Servers WHERE ServerID<>{databaseOwner}{objectQualifier}Broadcasts.SourceServerID)

Found in  DNNINSTALLROOT\Library\Providers\CachingProviders\BroadcastPollingCachingProvider\Install\Install.SQLDataProvider

I checked my minimum poll date of my servers table and it was several days ago. Turns out, I had an entry in the servers table from my staging server. I had not used the staging environment for several days (13 to be exact). What I found is that since I was not using the staging server, the poll date was not being updated. I deleted the staging server's entry in the servers table (ensuring no orphaned records in the cachedobjects and broadcasts table). I immediately went from 1,000,000 + records to around 70. I now have only my two production servers listed in the servers table. My Broadcasts table averages 200-300 rows. It is now under control.

So, the moral of the story is....

  •  Make sure that you only have your active production servers listed in the Servers table and nothing else.
  • DotNetNuke.Services.Cache.PurgeCache is not necessary for the BroadcastPollingProvider implementation of the WebFarm.
 
New Post
9/16/2008 10:40 AM
 

This is great information thank you for taking the time to share it with us!


-Mitchel Sellers
Microsoft MVP, ASPInsider, DNN MVP
CEO/Director of Development - IowaComputerGurus Inc.
LinkedIn Profile

Visit mitchelsellers.com for my mostly DNN Blog and support forum.

Visit IowaComputerGurus.com for free DNN Modules, DNN Performance Tips, DNN Consulting Quotes, and DNN Technical Support Services
 
Previous
 
Next
HomeHomeOur CommunityOur CommunityGeneral Discuss...General Discuss...Web Farm IssuesWeb Farm Issues


These Forums are dedicated to discussion of DNN Platform and Evoq Solutions.

For the benefit of the community and to protect the integrity of the ecosystem, please observe the following posting guidelines:

  1. No Advertising. This includes promotion of commercial and non-commercial products or services which are not directly related to DNN.
  2. No vendor trolling / poaching. If someone posts about a vendor issue, allow the vendor or other customers to respond. Any post that looks like trolling / poaching will be removed.
  3. Discussion or promotion of DNN Platform product releases under a different brand name are strictly prohibited.
  4. No Flaming or Trolling.
  5. No Profanity, Racism, or Prejudice.
  6. Site Moderators have the final word on approving / removing a thread or post or comment.
  7. English language posting only, please.
What is Liquid Content?
Find Out
What is Liquid Content?
Find Out
What is Liquid Content?
Find Out