Still getting these errors, anyone any ideas?

Are you using loadbalancing? Have you tasks configured in the umbracosettings.config?

Nope no load balancing and there is one task in the config - however when this is disabled, we still get the errors.

You are not the only one: http://issues.umbraco.org/issue/U4-4159

Dan Diplo seems to have a solution in this post.

Similar but not the same errors however could be related i guess - waiting for HQ to come back to me as well.

Any word?  I've got the exact same error happening every minute since upgrading to 6.2.2...

Not yet but Paul Sterling asked me to disable all the scheduled tasks in the umbraco settings file and check - I have tried that previously but it still happened but going to try again properly today

Hi Simon,

Just been scratching my head about this very issue. I have a client in pre-golive and this was filling their logs every minute. Turns out it's because I've put IP / Address filtering on IIS to stop Google et al finding the site prematurely... this has the side effect of blocking the scheduler from call itself on the server.

I've tried to find the correct IPv6 local IP to add to the execptions but it doesn't work. Guess I'll live with it until it goes live. 

Wonder if that is your issue?

Steve

PS I should note I'm on v7.1.6 so it might be very different.

Ah, it's only now noted I submitted a post with a similar issue. Have you resolved this?

Nope still happening and cannot get to the bottom of it!

Hi Simon

Have you written to support about this? You are a gold partner right? Then it might be worth a shot to get in touch with the guys directly if you have not done so already :)

/Jan

Certainly have, tried what they suggested but to no avail. Next steps are for me to download the source, rebuild against the project and trace the bloody thing through myself :(

I noticed my issue was due to IIS redirecting traffic to HTTPS using URL rewriting. Without the rewrite rules, both the KeepAlive and Scheduling pages are called succesfully.

The funny thing though is that both URL's are opened with a System.Net.WebClient. When I use that to open connections to both the URL's all is fine. From within Umbraco they fail with a "The underlying connection was closed" message. This is different from your "The operation has timed out" however.

I too am having the same issue although removing my HTTPS rewrites still doesn't seem to stop the errors:

http://our.umbraco.org/forum/umbraco-7/using-umbraco-7/57083-Error-with-SheduledPublishing,-underlying-connection-was-closed?p=0#comment197061

So we've been battling with the same errors in the logs for the past week and we've nailed it down to (in our setup at least) a continuous cycle of the site creating new app domains resulting in very poor performance on the sites themselves.

I'm adding what we found in the hope that it might confirm that you're having the same issue, this shouldn't be considered the "official" answer to the support case opened by Simon as we have no clear indication that this is the underlying issue. :-)

So, eventually what we found is that our sites kept creating new app domains for some strange reason. We haven't yet figured out why at all and that might take a while but the workaround we found is the following: http://stackoverflow.com/a/17645687/5018
What we've been doing is adding <httpRuntime fcnMode="Disabled" /> to the affected sites. Notice that this has some interesting side-effects and we don't know all of them yet, but one of them is that saving a .cshtml file will not automatically refresh the logic behind it, you HAVE to recycle the app pool/touch the web.config to make the changes have effect in the environment that fcnMode is disabled. So no more quick changes on the live site through the backoffice, you have to recycle. Other side-effects we don't know much about yet.

Note that the sites that had this problem in our case were using some classes in the App_Code folder, not sure if that is a contributing factor to the problem.

Again, I'm just sharing this information to see if it can help in your currently deployed sites and I would love to hear if the setups are similar so we can track down the actual source of the problem.

We have added some changes to 7.2.0 to at least not try to call non-existing URLs but I'd really like to make sure we're not just hiding the problem here. See issue: http://issues.umbraco.org/issue/U4-5728

We're currently in the process of upgrading a site from 6.1.6 to 6.2.4 and we are now seeing this issue. We've seen poor performance when attempting to configure our site to be load balanced and do not have any code running from App_Code.

In addition to the performance problems, we are seeing are errors being logged from the keepalive service, ping and scheduled publishing service. On top of that we can't seem to get published content to appear on both web servers - i.e. Publish on one web server and the changes are not appearing on the second web server, this extends to examine indices as well.

Just having a poke around now and have found there does seem to be issues with Umbraco 6.2.X and 7.1.X with regards to running in a load balanced environment, I wonder if these are all related in some way:

• http://our.umbraco.org/forum/umbraco-7/using-umbraco-7/57083-Error-with-SheduledPublishing,-underlying-connection-was-closed
• http://our.umbraco.org/forum/getting-started/installing-umbraco/55867-Setting-up-distributedcall-for-Umbraco-load-balancing-with-Umbraco-7

Does adding <httpRuntime fcnMode="Disabled" /> to your site help at all? (question goes for everybody in this thread, by the way)

We are not running any load balanced sites but if they're not appropriately configured then I'm sure this could be a side effect.
Make sure to follow the documentation precisely: http://our.umbraco.org/documentation/Installation/load-balancing

After adding

<httpRuntime fcnMode="Disabled" />

the error seems to be gone. Error was:

ERROR Umbraco.Web.Scheduling.KeepAlive - [Thread 55] Error in ping System.Net.WebException: The operation has timed out...

Also my hosting provider sent me an email today asking why my site took 60% CPU load and a lot of memory... Perhaps this explains it...

Matthias

Yeah, we saw the same thing happening. My guess is that your hosting provider is using a SAN or something like that to host your files. Since Umbraco has quite a lot of files, they get scanned constantly (which can also cause higher load). My understanding is that disabling fcnMode will stop most of that scanning and will limit it to bin and web.config. Again, this means that if you DO do live edits on your razor (templates/views/macroscript) files then they will not reflect changes until you touch web.config.

I've still not completely figured out what other impact this has so make sure to read up on fcnMode and be aware that you might get some strange results.

I'm getting the same error after setting IIS to redirect traffic to HTTPS using URL rewriting. On 7.1.6 as well.

error:

Umbraco.Web.Scheduling.ScheduledPublishing
An error occurred with the scheduled publishing
System.Net.WebException: The operation has timed out
   at System.Net.WebClient.UploadDataInternal(Uri address, String method, Byte[] data, WebRequest& request)
   at System.Net.WebClient.UploadString(Uri address, String method, String data)
   at Umbraco.Web.Scheduling.ScheduledPublishing.Start(ApplicationContext appContext)

Rich

Just as an update, we're fixing this in an upcoming 7.1.9 release (should be out soon-ish) - http://issues.umbraco.org/issue/U4-5728

Thanks Sebastiaan - what about a v6 patch, is that coming at all?

Hi Sebastiaan, will you also be releasing a patch for version 6.2.x?

http://our.umbraco.org/forum/ourumb-dev-forum/bugs/59199-Errors-from-UmbracoWebSchedulingScheduledPublishing-and-UmbracoWebSchedulingKeepAlive http://issues.umbraco.org/issue/U4-5842

I've added a patch for 6.2.4 in this issue: http://issues.umbraco.org/issue/U4-5728

Thanks, Sebastiaan!

Thanks Seb - where were you on Friday at the festival, we all missed you :-)

My Umbraco 6.2.2 site's log is filled with similar errors.

From what I have read here people are claiming URL Rewriting and File Change Notification settings are to blame.

I'm using a rewrite rule force https so this could be my problem.

Am I right in thinking this has been fixed in a later version of 6.2?

If I cannot upgrage easily will <httpRuntime fcnMode="Disabled"/> fix it or will I need to find another way to force the use of SSL?

Regards, Matt

@Matt: (Most of) these error should be fixed in version 6.2.5. Also, in my case a lot of these errors disappeared when I removed all the rewrite rules from /config/UrlRewriting.config and replaced them with rewrite rules /rewriteRules.config (using IIS URL Rewrite 2.0). To force HTTPS it would look like this:

<?xml version="1.0" encoding="utf-8"?>
<rules>
    <clear />
    <rule name="ForceHttps" stopProcessing="true">
        <match url="(.*)" />
        <conditions>
            <add input="{HTTPS}" pattern="off" ignoreCase="true" />
        </conditions>
        <action type="Redirect" url="https://{HTTP_HOST}{REQUEST_URI}" redirectType="Permanent" appendQueryString="true" />
    </rule>
</rules>

Thanks Arjan,

I'm currently doing my rewrite with the following in my web.config:

<rewrite>
    <rules>
      <rule name="Redirect to HTTPS" stopProcessing="true">
        <match url="(.*)" />
        <conditions><add input="{HTTPS}" pattern="^OFF$" />
          </conditions>
        <action type="Redirect" url="https://{HTTP_HOST}/{R:1}" redirectType="Permanent" />
      </rule>
      </rules>
    </rewrite>

Is that what you mean?

Regards, Matt

Same problems here, 6.2.4

fcnMode="Disabled"

didn't help. I also installed Sebastiaan's patch for 6.2.4 which didn't resolve the issue.  I also force the entire site to HTTPS.  Really sucks because anytime I publish content the app restarts and locks up for 30 secs.  Any idea when 6.2.5 will be released?

Ah, so that's why my stite is taking 30 to 40 seconds to start up! :-S

@Matt: Yes, of course you can also include those rewrite rules in the web.config. I prefer storing these rewrite rules and mappings in separate config files (rewriteRules.config and rewriteMaps.config) which I then include in the web.config:

<system.webServer>
  <rewrite>
    <rewriteMaps configSource="rewriteMaps.config" />
    <rules configSource="rewriteRules.config" />
  </rewrite>
</system.webServer>

In my case these changes in combination with the 6.2.4 patch Sebastiaan provided fixed all the errors. Of course the 6.2.4 patch won't do you any good, since you're still on version 6.2.2. By the way, I also tried the fcnMode="Disabled" fix, but that didn't work for me either.

@Jeremy/Matt: I've never had issues with application pools restarting after publishing content on a forced HTTPS site. So I'm guessing these two issues are unrelated. I assume you configured this in your web.config?

<add key="umbracoUseSSL" value="true" />

I disabled the umbracoUseSSL value as I was running into random infinte loop redirects every couple weeks.  Didn't fix the issue.  Changed the rewrite rule to use {Request_URI} instead of {R:1}, no loops in a couple weeks so fingers crossed. umbracoUseSSL still set to false.

Publishing throws a ton of errors in the trace log.   Been using Umbraco since 2009, first time I've seen this issue.

Don't think the issue will be resolved until 6.2.5 released with the baseURL attribute, http://issues.umbraco.org/issue/U4-5965

For now I've just taken out the https rewrite rule in the web.config to see what effect it has.

I wouldn't be happy setting umbracoUseSS = false as you want to protect those log in credentials which will be in plain text otherwise.

If the logs clear up I'll try re-instating the rewrite rule and changing using {Request_URI} instead of {R:1} like you have.

Cheers, Matt

The result of removing my rewrite rule from the web.config was was that the errors disappeared from the log.

I then re-enabled the re-write rule but changed:

<action type="Redirect" url="https://{HTTP_HOST}/{R:1}" redirectType="Permanent" />

to

<action type="Redirect" url="https://{HTTP_HOST}{REQUEST_URI}" redirectType="Permanent" appendQueryString="true" />

Immediately the errors re-appeared. One every minute exactly to the second.

I just found out something very interesting. I'm using Application Initialization Module for IIS 7.5 to automatically start up application pools and web sites after a server reboot or an application pool recycle to improve web site responsiveness. IIS basically fakes request and pushes it directly into the IIS pipeline without hitting the network, and this causes Umbraco to save localhost:80/umbraco as the OriginalRequestUrl. And of course the ping and scheduled publishing won't working anymore, because they will both be making requests to localhost instead of the actual web site URL. Since I set all my application pools to do a recycle in the middle of the night, I kept getting these ping and scheduled publishing 404 errors in the Umbraco logs.

As of IIS 8 you should be able to configure the initializationPage for each web site, but this feature isn't available in IIS 7.5:

http://www.iis.net/configreference/system.webserver/applicationinitialization

I'm going to disable the pre-load feature for my Umbraco web sites, and implement a separate web application which does use the pre-load feature and fires a request to all the Umbraco web sites, making sure the proper OriginalRequestUrl is set for each web site.

We have been running into some mysterious application restarts and multiple appdomains recently and eventually stumbled upon the following blogpost for DNN: http://www.dnnsoftware.com/community-blog/cid/154980/aspnet-file-change-notifications-and-dnn

In this blog they mention issues very similar to ours and claim that it is caused by the large amount of directories used in their application. fcnMode="Disabled" eventually solved their solution and also solved ours.

Although I have not tried to reproduce it yet I get the feeling that our problem was also caused by large amounts of directories, which has rissen dramatically since we have been using ImageProcessor an d Slimmage in our recent projects.

Currently we are just sticking with the new fcnMode setting and haven't noticed any side affects yet.

i have a similar situation with Jeremy. 

using 6.2.4 but the site is still logging those errors. 

Updating to 6.2.5 cleared those errors.  Just make sure you add a hostname in backoffice, set umbracoUseSSL = true in web.config if forcing your site to use SSL, and to add a baseUrl attribute to umbracoSettings.config:  i.e. <scheduledTasks baseUrl="www.domain.com/umbraco/">

https://our.umbraco.org/contribute/releases/625/

@Jeremy before fixing the error, was umbraco hogging up  resources? our site kept going down so we had to keep restarting the app pool.

Memory usage was low, which is normal for our site.  During a save/publish on 6.2.4 the CPU would spike to 100% for 10-20 secs, but was normal when not making backoffice updates.

I was trying to replicate the error on a dev environment with ssl with umbraco 6.2.4. the error is not showing in the error log. this is so weird.

@avatart0ph did you configure host headers for the website in IIS?

we have not configured the host headers of the website in the production environment but in dev we have configured it like this 

https://technet.microsoft.com/en-us/library/cc753195%28v=ws.10%29.aspx

i do not touch the prod IIS. the client's IT does it. But i have told them to fill in the variable and see how it goes. thank you Jeremy

the client does not want named bindings. i guess i am stuck with those errors.

Why not just upgrade to the 6.2.5 version?  That cleared up all the ScheduledPublishing errors for us.  Just make sure to set the new baseUrl attribute in the config file, configure a hostname in the backoffice, as well as set umbracoUseSSL to the correct value (true if you want all backoffice resrouces served via SSL).

I'm using umbraco 7.2.1, IIS 7.5, no load balancing, no url rewrites, no SSL, no scheduled tasks,

yet, I still get the exceptions every minute.

has anyone in a similar configuration found a solution?

Thanks,

Gus

Upgrading to 6.2.5 seems to have cleared the errors for me too.

Hi there,

I got umbracoUseSSL = "true" in my web.config

and 

scheduledTasks baseUrl="https://www.mydomain.com/umbraco" in umbracoSettings.config

and the following errors still appear:

2015-05-08 14:50:39,257 [79] ERROR Umbraco.Web.Scheduling.ScheduledPublishing - [Thread 25] An error occurred with the scheduled publishing

System.Net.WebException: The remote name could not be resolved: 'https'

   at System.Net.WebClient.UploadDataInternal(Uri address, String method, Byte[] data, WebRequest& request)

   at System.Net.WebClient.UploadString(Uri address, String method, String data)

   at Umbraco.Web.Scheduling.ScheduledPublishing.Run()

2015-05-08 14:50:39,257 [79] ERROR Umbraco.Web.Scheduling.KeepAlive - [Thread 15] Error in ping

System.Net.WebException: The remote name could not be resolved: 'https'

   at System.Net.WebClient.DownloadDataInternal(Uri address, WebRequest& request)

   at System.Net.WebClient.DownloadString(Uri address)

   at Umbraco.Web.Scheduling.KeepAlive.Start(ApplicationContext appContext, IUmbracoSettingsSection settings)

I am using Umbraco 7.2.1. I don't have any scheduled task that I need to run. Just trying to follow the steps above. 

I am also not using load balancing.

Does anybody have a suggestion?

Many Thanks,

Gus

Hi there,

Thanks for your feedback.

I did an Umbraco update to 7.2.5 and the ScheduledPublishing error disappeared.

The KeepAlive error persists though. I have tried many different configurations for the baseUrl tag mentioned above (including using IP address). No success.

Do you guys know how to eliminate this KeepAlive error?

Thanks,

Gus

Try this solution ;)

our.umbraco.org/forum/umbraco-7/using-umbraco-7/57083-Error-with-SheduledPublishing-underlying-connection-was-closed#226307

HI all,

Having the same issue on Umbraco V7.2.4 (no load balance, no schedule task too)

But both schedule task error and keepAlive keeps showing in the log every minute

Any solution to this yet?

Thanks

Umbraco 6.2.5

Seeing these errors, when using HTTPS redirection handled in IIS.

Have added:

• baseUrl into umbracoSettings.config
• enabled umbracoUseSSL in web.config

Application startup, MUCH faster, and so far no errors. Fingers crossed!

Hi everyone,

Is there a specific solution for this? I just upgraded an Umbraco 7.2.1 to a 7.4.3 and I'm getting these errors.

This is a staging server, production with 7.2.1 also has this issues. Thanks!

Any answer for this now???

I'm facing the same issue in Umbraco 7.6.3, any answer? Greetings.

Anybody get a resolution?

setting the umbracoApplicationUrl in the umbracosettings fixed this for me.

I'm also noticing this on 7.12.3, even with the following settings in place on umbracoSettings.config

scheduledTasks baseUrl="domain.com/umbraco/"

web.routing umbracoApplicationUrl="https://domain.com/umbraco"

ERROR Umbraco.Web.Scheduling.KeepAlive - Failed (at "https://domain.com/umbraco").

System.Net.Http.HttpRequestException: An error occurred while sending the request. ---> System.Net.WebException: The underlying connection was closed: An unexpected error occurred on a send. ---> System.IO.IOException: Unable to read data from the transport connection: An existing connection was forcibly closed by the remote host. ---> System.Net.Sockets.SocketException: An existing connection was forcibly closed by the remote host

Other 7.12.3 sites with the same settings are happy, so I wonder what might be causing this...

In my case, the site was running HTTPS and the hosting configuration was only accepting TLS 1.2, so I needed to make sure to run only TLS 1.2 over code:

System.Net.ServicePointManager.SecurityProtocol = System.Net.SecurityProtocolType.Tls12;

Hi all,

Added to Pedro's answer, a different version of the .NET framework has a different default TLS - please check it at here

Below is the solution that works for me for Umbraco 7

If you are running your Umbraco 7 site at .NET 4.5, here is the quick fix, thanks for @Paul Seal, you can install his package to make the treat - check out CodeShare.TLS1.2.Umbracoso

If you interest in how he fixed it, please check Pual'post at how-to-force-a-net-website-to-use-tls-12

For anyone that not familiar with the ApplicationStarted method that Pual is using, check out Umbraco's official doc for more info at here for Umbraco 7

Hopefully, this can solve your problem. Fingers crossed. :)

Cheers Jun

is working on a reply...