We're seeing a very odd issue with media items being "corrupted" in the media cache in a 6.1.3 site. Intermittently we get an error stating: "The valueDictionary is not formatted correctly and is missing any of the 'id,nodeId,__NodeId' elements".
We've grabbed 6.1.3 from GitHub and the error appears to be related to the GetTypedMedia method. When this is pulling the values from the media cache, some of the required Examine properties are missing or empty. The affected media items only seem to have umbracoName and umbracoAlias set in their Examine dictionary when it tries to convert them into a strongly typed media object.
Calls to umbraco.library.GetMedia and and the media service similarly fail, as they use the same underlying code.
Republishing the whole site seems to fix the issue for a while, but then after a few site visitors, the issue rears it's head again.
Oddly, this only seems to currently affect media referred to by nodes that are outside of the main site content tree (they're in a "settings" section). For example, when trying to get the media by id from the "image" property of something in "/settings/logos/test-logo". This may however be a red herring, we're not 100% sure yet.
Interestingly, if we bin off TypedMedia entirely, and just use Examine, using the SAME syntax as the underlying cache method does, it works. It appears that somewhere between Examine and the TypedMedia method returning, items are getting corrupted.......
There was an issue with examine where it didnt index media items which lived outside of the root, Told shannon at CodeGarden, and i saw a bug fix for it a while ago.
Cheers Ste! It turned out that the issue was to do with the InternalIndexSet being corrupted. The content cache (so calls to TypedContent) will still work, as if the call to InternalIndexSet fails it will hit up the XML Cache instead. However, for the media library, it will fall back to the in memory media cache, which was also not updating properly (I think that ultimately gets populated from InternalIndexSet as well).
Weird, weird stuff. It looks like everything was working fine on the site, then the IndexSet got corrupted and couldn't rebuild properly as a result, which meant media items were going missing from the site. We used the Examine manager to rebuild the index (which took a while, there's 5000 nodes on the site) and now everything appears to be working fine again. So far we haven't been able to break it again since (fingers crossed).
I've only seen one other person report this and it would be because for some reason there's the required information missing from that particular media item in the index. The index might be there but all of the information isn't and thus you get that error.
The only reason the data wouldn't get into the index properly would be some other error way higher up in the chain of events coming from Umbraco not examine itself. Unfortunately this isn't an easy on to replicate but if you come across it again please ping me on twitter again, would be great to iron out any of these subtle mysteries.
I've started working on this issue: http://issues.umbraco.org/issue/U4-2914 - for some odd reason the folder data for a newly created folder isn't getting into the index but these are all nuances with the Umbraco API that need to be fixed and there might be some other odd things going on.
Tim,
Your post has saved me some serious headache. Current site we are working on we uploaded images etc locally then checked in and pushed up to the dev server. we later updated some of the images however the old urls were being picked up. In code i was doing
@Shannon, we think this may have something to do with the way we have our dev environment set up. We've only experienced the issue on sites in local dev, it only made it onto a live server because the site was put live with a corrupted index.
We've had it happen twice now, but we're still trying to reproduce it consistently. If we manage, I'll let you know!
I have this bug on a 6.1.6 site.
The exception occurs when I use GetPropertyValue<T> on a typed media instance. (XmlPublishedCache?)
It does not happen when I use GetPropertyValue.
The reason however is that the property value actually doesn't exist in the database.
It happens on a field that has been added to the media type, and the media in question hasn't been saved since the field was added. To handle the exception, I check for existence of the property before attempting to get the value.
I guess the GetProperty method further down in the core could do the same. :)
We have had this problem twice in the last week or so on a live site. Running Umbraco v6.1.2 (Assembly version: 1.0.4927.23554) . I'm not really clear what steps to take to resolve this. Both times the website has been down for about 25 minutes, starting around the time the first of these messages appears in the log and ending around the time the messages cease to appear. Exception message: The valueDictionary is not formatted correctly and is missing any of the 'id,nodeId,__NodeId' elements... the problom occurs at the same line in the same partial view:
var mediaItem = Umbraco.TypedMedia(linkItem.GetPropertyValue("linkImage"));
This partial view is used several times on the base page. On each date when the problem occured all the messages were for the same media item on the same part of the page, however this was a different media item on a differnet part of the page for the second date this occured than on the first. The media items are next to each other in the folder structure, one is in folder .../media/17446/ the other is in folder .../media/1447/. Both items exists and have the correct id in the umbraco.config file.
The first time this happened I used the Examine Management and rebuilt the inices, as per the solution above. A couple of days later I got a report that half a dozen pdf links were broken, and found that these items physically were not in the media folder anywhere. I'm somewhat reluctant to re-run index rebuild in case anything else dissappears.
Hi Louise,
You should perhaps look at upgrading to the latest 6.1.6 version, patch releases contain many fixes and is always recommended to use the latest version so that the previous issues are fixed. For example, this issues probably directly affect what you are seeing:
http://issues.umbraco.org/issue/U4-2344
Hi. How will the upgrade fix the indices? This happened again this morning, so we are under a lot of pressure to put a proper fix in place quickly. I have tried doing an upgrade before to fix other issues and this has not worked, so reluctant to go down this route unless 100% sure it will work.
@Louise, If you read through that issue you'll see that it's listed as a 'major' issue which is contained in your current version. Re-indexing of media would never have worked in your current version.
Of course you should backup/test/etc before upgrading your live site but upgrading from your minor version to at least 6.1.3 (which this fix is contained in) should be no issue (though recommended to upgrade to 6.1.6)
Shannon, I'm seeing this intermittently on a load-balanced install. It happens after the media has worked fine for a while, then re-saving the media seems to help, but it's hard on pages w/ tons of media referenced because there's no good way (that I can see) to determine which media item is the problem. I haven't been able to isolate any admin activities that may be prompting the change, but will update here when I have more research.
I'm on a nightly build very close to 6.1.6, so the fix referenced above is there.
The problem is probably load balancing with Examine - depending on how your LB is setup, media is not distributed to all indexes unless you are handling that with your own code.
We plan on ensuring that media/member indexes is distributed like content among the servers involved in LB but will probably be scheduled for post 6.2/7.0.2
I am also seeing this problem on a live production site. It is running Umbraco v6.1.6 (Assembly version: 1.0.5021.24867). It is not load balanced.
The error occurs in the view on this line:
@Umbraco.Media(CurrentPage.heroBanner).Url
Stack trace is as follows:
System.FormatException: The valueDictionary is not formatted correctly and is missing any of the 'id,nodeId,__NodeId' elements
at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.DictionaryPublishedContent.ValidateAndSetProperty(IDictionary`2 valueDictionary, Action`1 setProperty, String[] potentialKeys)
at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.DictionaryPublishedContent..ctor(IDictionary`2 valueDictionary, Func`2 getParent, Func`2 getChildren, Func`3 getProperty, Boolean fromExamine)
at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.ConvertFromXPathNavigator(XPathNavigator xpath)
at Umbraco.Web.PublishedCache.ContextualPublishedCache`1.GetById(Boolean preview, Int32 contentId)
at Umbraco.Web.UmbracoHelper.DocumentById(Int32 id, ContextualPublishedCache cache, Object ifNotFound)
at Umbraco.Web.UmbracoHelper.DocumentById(String id, ContextualPublishedCache cache, Object ifNotFound)
I stopped IIS, deleted the App_Data\TEMP\ExamineIndexes folder, then started the website back up. It seems to have fixed the issue in the short term. We are still unsure what caused it in the first place.
Same problem here on a production site, occuring couple of times per week, on Umbraco.Media actions in the razorcode
Umbraco version: 6.1.6
-No loadbalancing -Not after making changes in the site -We don't use GetPropertyValue on media
Problem is gone by itself or by republishing entire site. Republishing entire site after update of one of the nodes doesn't prevent the problem from occuring
Having same issue inconsistently in production w/ v7. Are there any solutions?
Saw this thread (http://our.umbraco.org/forum/umbraco-7/using-umbraco-7/49903-Parsing-error-in-Umbraco-7-partial-view-missing-closing-brace-despite-all-brace-pairs-being-present) but nothing seemed to be 'the answer' as of yet.
We cannot know if everyone is having the 'same' issue unless people post what versions you are using. This issue (and many different forms of it) are solved in the latest Umbraco versions: 6.2.1 and 7.1.4
I have the same issue.
I'm using Umbraco 6.2.2. I update the website from 6.2.0 to 6.2.1, then to 6.2.2 to solve the issue. But it is not solved.
After many time I found a method to fire the issue.
The issue go up only in production server. Then I ask the server configuration for websites. The provider administrator say to me that the worker process is limited to use up to 256MB of Memory and 768MB of Virtual Memory.
Then I changed the website settings in my local instance.
Now it is necessary to navigate the website for some page. When IIS restart the website maybe the issue fire up.
I can temorary restore the website simply with a "Save and publish" of a ContentNode.
Error:
System.NullReferenceException: Riferimento a un oggetto non impostato su un'istanza di oggetto.
in Umbraco.Web.PublishedCache.XmlPublishedCache.XmlPublishedContent.
What did you try?
I simply try to navigate the website.
Load balance
No
Rebuilt indexes
Many times. It don't resolve the issue.
I am the issue in last months. I don't known how to patch it.
I can send you my project, so you will try it. To do it write me in private: [email protected]
Yip Guys, same problem here. Spent allot of time upgrading the problem site to from 6.1.2 to 6.2.4 and it just happened again.
Load Balance: No
I basically have to foce the app pool to recycle to fix. I take the site offline and bring it online. For now this has resolved. What is disturbing is that the site could be down for hours and I would not know and it would not repair itself. Interestingly this never happened on any sites running 4.7.2 or any of my sites that are running V7 (various versions). I might just upgrade this one to V7. But its going to be a big upgrade again.
Another "Me Too" for this, we were seeing this under 6.1.6 and have now upgraded to 6.2.4 and are still seeing it.
Loadbalanced: Yes, 2 nodes
Umbraco Version: 6.2.4
Error:
System.FormatException: The valueDictionary is not formatted correctly and is missing any of the 'id,nodeId,__NodeId' elements at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.DictionaryPublishedContent.ValidateAndSetProperty(IDictionary`2 valueDictionary, Action`1 setProperty, String[] potentialKeys) at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.DictionaryPublishedContent..ctor(IDictionary`2 valueDictionary, Func`2 getParent, Func`2 getChildren, Func`3 getProperty, Boolean fromExamine) at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.ConvertFromXPathNavigator(XPathNavigator xpath) at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.GetUmbracoMedia(Int32 id) at Umbraco.Web.PublishedCache.ContextualPublishedCache`1.GetById(Boolean preview, Int32 contentId) at ASP._Page_Views_Partials__NewsSlider_cshtml.<>c__DisplayClass1.<RenderHighlight>b__0(TextWriter __razor_helper_writer) in <PATH_REMOVED>\views\Partials\_NewsSlider.cshtml:line 42 at ASP._Page_Views_Partials__NewsSlider_cshtml.Execute() in <PATH_REMOVED>\views\Partials\_NewsSlider.cshtml:line 30 at System.Web.WebPages.WebPageBase.ExecutePageHierarchy() at System.Web.Mvc.WebViewPage.ExecutePageHierarchy() at System.Web.WebPages.WebPageBase.ExecutePageHierarchy(WebPageContext pageContext, TextWriter writer, WebPageRenderingBase startPage) at Umbraco.Core.Profiling.ProfilingView.Render(ViewContext viewContext, TextWriter writer) at System.Web.Mvc.Html.PartialExtensions.Partial(HtmlHelper htmlHelper, String partialViewName, Object model, ViewDataDictionary viewData) at ASP._Page_Views_HomePage_cshtml.Execute() in <PATH_REMOVED>\views\HomePage.cshtml:line 13 at System.Web.WebPages.WebPageBase.ExecutePageHierarchy() at System.Web.Mvc.WebViewPage.ExecutePageHierarchy() at System.Web.WebPages.WebPageBase.ExecutePageHierarchy(WebPageContext pageContext, TextWriter writer, WebPageRenderingBase startPage) at Umbraco.Core.Profiling.ProfilingView.Render(ViewContext viewContext, TextWriter writer) at System.Web.Mvc.ViewResultBase.ExecuteResult(ControllerContext context) at System.Web.Mvc.ControllerActionInvoker.<>c__DisplayClass1a.<InvokeActionResultWithFilters>b__17() at System.Web.Mvc.ControllerActionInvoker.InvokeActionResultFilter(IResultFilter filter, ResultExecutingContext preContext, Func`1 continuation) at System.Web.Mvc.ControllerActionInvoker.InvokeActionResultFilter(IResultFilter filter, ResultExecutingContext preContext, Func`1 continuation) at System.Web.Mvc.ControllerActionInvoker.InvokeActionResultWithFilters(ControllerContext controllerContext, IList`1 filters, ActionResult actionResult) at System.Web.Mvc.Async.AsyncControllerActionInvoker.<>c__DisplayClass25.<BeginInvokeAction>b__22(IAsyncResult asyncResult) at System.Web.Mvc.Controller.<>c__DisplayClass1d.<BeginExecuteCore>b__18(IAsyncResult asyncResult) at System.Web.Mvc.Async.AsyncResultWrapper.<>c__DisplayClass4.<MakeVoidDelegate>b__3(IAsyncResult ar) at System.Web.Mvc.Controller.EndExecuteCore(IAsyncResult asyncResult) at System.Web.Mvc.Async.AsyncResultWrapper.<>c__DisplayClass4.<MakeVoidDelegate>b__3(IAsyncResult ar) at System.Web.Mvc.MvcHandler.<>c__DisplayClass8.<BeginProcessRequest>b__3(IAsyncResult asyncResult) at System.Web.Mvc.Async.AsyncResultWrapper.<>c__DisplayClass4.<MakeVoidDelegate>b__3(IAsyncResult ar) at System.Web.HttpApplication.CallHandlerExecutionStep.System.Web.HttpApplication.IExecutionStep.Execute() at System.Web.HttpApplication.ExecuteStep(IExecutionStep step, Boolean& completedSynchronously)
We don't think this is related to publishing the items in question as this is happening on a news widget on the site and we have seen it even when nothing has been posted for many days.
What we have seen is that it seems to happen when the server is under memory pressure, however this may also not be the case as the AppPool was only 2.4GB at the time this happened and had been up for about 30 minutes.
I think I've seen it both fix without intervention and after a restart.
If it is in fact the same issue then your indexes are out of sync so ensure your indexes are in sync and this issue wont happen. And yes there is a fix for version six, see the issue tracker:
Thanks. Sorry, assumed tht fix was only for v7. How do I go about integrating this fix into my code? I'm guessing I have to download the source code, do the amendments and build the project?
After a lot of step by step debugging the issue related to a corrupt database entry specifically a media item in the cmsContent table that had a column contentType with a value of 0.
Removing this row and related rows, and recreating the item has banished the issue from our environments.
We ran these queries on the database to delete the orphaned nodes:
DELETE
FROM cmsContentVersion
WHERE ContentId IN (SELECT c.nodeId
FROM cmsContent c
WHERE c.contentType = 0)
DELETE
FROM cmsPropertyData
WHERE contentNodeId IN (SELECT c.nodeId
FROM cmsContent c
WHERE c.contentType = 0)
DELETE
FROM cmsContent
WHERE contentType = 0
DELETE
FROM umbracoNode
WHERE id IN (SELECT c.nodeId
FROM cmsContent c
WHERE c.contentType = 0)
This is a fix we knew about to resolve errors that we would see when we would try to reindex the site. We are hopeful that they will solve our intermittent homepage errors now as well.
Yes! Deleting the content from the database for nodes that didn't have contentTypes has solved our problem. We were able to reindex the Internal Indexer without errors and the site has stopped going down.
I'm still seeing the media cache falling apart on versions 7x, in this latest case, 7.1.6 on a single server w/ no load balancing and it's the only site on the server. For some reason the cache gets blitzed for no reason, ie., no users have been within Umbraco for hours, and the only way to fix it is to restart IIS, rebuild the indexes, and republish the site.
This issue brings down ALL media items, not just specific items here and there.
In our case, there are properties on the media items, via TypedMedia, that come back blank, particularly the Url/umbracoFile props.
So far, upgrading has not been the answer to the problem and there's nothing in the logs that points to any kind of issue in particular. So, to "get around" this, for our other sites that were having this issue, we started pulling the media from the DB rather than from cache. On those sites, we haven't had any more crashes like these, but we would like to get away from that method of pulling media items.
For those of us experiencing this issue do you have any example code that shows how to pull the media items directly from the DB, e.g. to get all children of a folder.
Issue With Media Cache in v6
We're seeing a very odd issue with media items being "corrupted" in the media cache in a 6.1.3 site. Intermittently we get an error stating: "The valueDictionary is not formatted correctly and is missing any of the 'id,nodeId,__NodeId' elements".
We've grabbed 6.1.3 from GitHub and the error appears to be related to the GetTypedMedia method. When this is pulling the values from the media cache, some of the required Examine properties are missing or empty. The affected media items only seem to have umbracoName and umbracoAlias set in their Examine dictionary when it tries to convert them into a strongly typed media object.
Calls to umbraco.library.GetMedia and and the media service similarly fail, as they use the same underlying code.
Republishing the whole site seems to fix the issue for a while, but then after a few site visitors, the issue rears it's head again.
Oddly, this only seems to currently affect media referred to by nodes that are outside of the main site content tree (they're in a "settings" section). For example, when trying to get the media by id from the "image" property of something in "/settings/logos/test-logo". This may however be a red herring, we're not 100% sure yet.
Interestingly, if we bin off TypedMedia entirely, and just use Examine, using the SAME syntax as the underlying cache method does, it works. It appears that somewhere between Examine and the TypedMedia method returning, items are getting corrupted.......
There was an issue with examine where it didnt index media items which lived outside of the root, Told shannon at CodeGarden, and i saw a bug fix for it a while ago.
http://issues.umbraco.org/issue/U4-2344
http://issues.umbraco.org/issue/U4-2297
Not sure if its relevant, but it may help!
Cheers Ste! It turned out that the issue was to do with the InternalIndexSet being corrupted. The content cache (so calls to TypedContent) will still work, as if the call to InternalIndexSet fails it will hit up the XML Cache instead. However, for the media library, it will fall back to the in memory media cache, which was also not updating properly (I think that ultimately gets populated from InternalIndexSet as well).
Weird, weird stuff. It looks like everything was working fine on the site, then the IndexSet got corrupted and couldn't rebuild properly as a result, which meant media items were going missing from the site. We used the Examine manager to rebuild the index (which took a while, there's 5000 nodes on the site) and now everything appears to be working fine again. So far we haven't been able to break it again since (fingers crossed).
:)
Sorry for getting on to this one late
I've only seen one other person report this and it would be because for some reason there's the required information missing from that particular media item in the index. The index might be there but all of the information isn't and thus you get that error.
The only reason the data wouldn't get into the index properly would be some other error way higher up in the chain of events coming from Umbraco not examine itself. Unfortunately this isn't an easy on to replicate but if you come across it again please ping me on twitter again, would be great to iron out any of these subtle mysteries.
I've started working on this issue: http://issues.umbraco.org/issue/U4-2914 - for some odd reason the folder data for a newly created folder isn't getting into the index but these are all nuances with the Umbraco API that need to be fixed and there might be some other odd things going on.
Tim, Your post has saved me some serious headache. Current site we are working on we uploaded images etc locally then checked in and pushed up to the dev server. we later updated some of the images however the old urls were being picked up. In code i was doing
but as the internal index on the dev server was out synch image urls were incorrect. Did a rebuild in internal index and sorted it right out!
Ismail
@Shannon, we think this may have something to do with the way we have our dev environment set up. We've only experienced the issue on sites in local dev, it only made it onto a live server because the site was put live with a corrupted index.
We've had it happen twice now, but we're still trying to reproduce it consistently. If we manage, I'll let you know!
:)
I have this bug on a 6.1.6 site.
The exception occurs when I use
GetPropertyValue<T>on a typed media instance. (XmlPublishedCache?)It does not happen when I use
GetPropertyValue.The reason however is that the property value actually doesn't exist in the database.
It happens on a field that has been added to the media type, and the media in question hasn't been saved since the field was added. To handle the exception, I check for existence of the property before attempting to get the value.
I guess the
GetPropertymethod further down in the core could do the same. :)We have had this problem twice in the last week or so on a live site. Running Umbraco v6.1.2 (Assembly version: 1.0.4927.23554) . I'm not really clear what steps to take to resolve this. Both times the website has been down for about 25 minutes, starting around the time the first of these messages appears in the log and ending around the time the messages cease to appear. Exception message: The valueDictionary is not formatted correctly and is missing any of the 'id,nodeId,__NodeId' elements... the problom occurs at the same line in the same partial view:
var mediaItem = Umbraco.TypedMedia(linkItem.GetPropertyValue("linkImage"));
Hi Louise, You should perhaps look at upgrading to the latest 6.1.6 version, patch releases contain many fixes and is always recommended to use the latest version so that the previous issues are fixed. For example, this issues probably directly affect what you are seeing: http://issues.umbraco.org/issue/U4-2344
Hi. How will the upgrade fix the indices? This happened again this morning, so we are under a lot of pressure to put a proper fix in place quickly. I have tried doing an upgrade before to fix other issues and this has not worked, so reluctant to go down this route unless 100% sure it will work.
@Louise, If you read through that issue you'll see that it's listed as a 'major' issue which is contained in your current version. Re-indexing of media would never have worked in your current version. Of course you should backup/test/etc before upgrading your live site but upgrading from your minor version to at least 6.1.3 (which this fix is contained in) should be no issue (though recommended to upgrade to 6.1.6)
Shannon, I'm seeing this intermittently on a load-balanced install. It happens after the media has worked fine for a while, then re-saving the media seems to help, but it's hard on pages w/ tons of media referenced because there's no good way (that I can see) to determine which media item is the problem. I haven't been able to isolate any admin activities that may be prompting the change, but will update here when I have more research.
I'm on a nightly build very close to 6.1.6, so the fix referenced above is there.
The problem is probably load balancing with Examine - depending on how your LB is setup, media is not distributed to all indexes unless you are handling that with your own code.
There's been discussion about that here: http://our.umbraco.org/forum/core/general/47068-Distributed-publishing,-Examine-and-Media
We plan on ensuring that media/member indexes is distributed like content among the servers involved in LB but will probably be scheduled for post 6.2/7.0.2
I am also seeing this problem on a live production site. It is running Umbraco v6.1.6 (Assembly version: 1.0.5021.24867). It is not load balanced.
The error occurs in the view on this line:
Stack trace is as follows:
It seems to go away, either
Please advise why this is happening.
Thanks, Erik
I confirm. I've the same problem!
I stopped IIS, deleted the
App_Data\TEMP\ExamineIndexesfolder, then started the website back up. It seems to have fixed the issue in the short term. We are still unsure what caused it in the first place.I solved in the same manner ( see http://our.umbraco.org/projects/starter-kits/ublogsy/ublogsy-bugs/48868-Very-disappointed-for-error-Object-reference-not-set-to-an-instance-of-an-object?p=0#comment174316 ).
I think that Umbraco team need to investigate in this problem and its cause.
Same problem here on a production site, occuring couple of times per week, on Umbraco.Media actions in the razorcode
Umbraco version: 6.1.6
-No loadbalancing
-Not after making changes in the site
-We don't use GetPropertyValue on media
Problem is gone by itself or by republishing entire site. Republishing entire site after update of one of the nodes doesn't prevent the problem from occuring
Don't know how to fix it, please advice,
Garret
Vote for the issue here http://issues.umbraco.org/issue/U4-4338
Any solution to the issue? I'm having the same issue right now
Having this same issue now, any working solutions yet?
Having same issue inconsistently in production w/ v7. Are there any solutions?
Saw this thread (http://our.umbraco.org/forum/umbraco-7/using-umbraco-7/49903-Parsing-error-in-Umbraco-7-partial-view-missing-closing-brace-despite-all-brace-pairs-being-present) but nothing seemed to be 'the answer' as of yet.
We cannot know if everyone is having the 'same' issue unless people post what versions you are using. This issue (and many different forms of it) are solved in the latest Umbraco versions: 6.2.1 and 7.1.4
I have the same issue.
I'm using Umbraco 6.2.2. I update the website from 6.2.0 to 6.2.1, then to 6.2.2 to solve the issue. But it is not solved.
After many time I found a method to fire the issue.
The issue go up only in production server. Then I ask the server configuration for websites. The provider administrator say to me that the worker process is limited to use up to 256MB of Memory and 768MB of Virtual Memory.
Then I changed the website settings in my local instance.
Now it is necessary to navigate the website for some page. When IIS restart the website maybe the issue fire up.
I can temorary restore the website simply with a "Save and publish" of a ContentNode.
@Flavio As Shannon already indicated, you're not providing enough details:
Hi @Sebastiaan.
Error:
System.NullReferenceException: Riferimento a un oggetto non impostato su un'istanza di oggetto. in Umbraco.Web.PublishedCache.XmlPublishedCache.XmlPublishedContent.
What did you try?
I simply try to navigate the website.
Load balance
No
Rebuilt indexes
Many times. It don't resolve the issue.
I am the issue in last months. I don't known how to patch it.
I can send you my project, so you will try it. To do it write me in private: [email protected]
Yip Guys, same problem here. Spent allot of time upgrading the problem site to from 6.1.2 to 6.2.4 and it just happened again.
Load Balance: No
I basically have to foce the app pool to recycle to fix. I take the site offline and bring it online. For now this has resolved. What is disturbing is that the site could be down for hours and I would not know and it would not repair itself. Interestingly this never happened on any sites running 4.7.2 or any of my sites that are running V7 (various versions). I might just upgrade this one to V7. But its going to be a big upgrade again.
Here is the log.
2014-11-13 17:32:13,040 [45] ERROR Umbraco.Core.UmbracoApplicationBase - [Thread 45] An unhandled exception occurred
System.FormatException: The valueDictionary is not formatted correctly and is missing any of the 'id,nodeId,__NodeId' elements
at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.DictionaryPublishedContent.ValidateAndSetProperty(IDictionary`2 valueDictionary, Action`1 setProperty, String[] potentialKeys)
at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.DictionaryPublishedContent..ctor(IDictionary`2 valueDictionary, Func`2 getParent, Func`2 getChildren, Func`3 getProperty, Boolean fromExamine)
at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.ConvertFromXPathNavigator(XPathNavigator xpath)
at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.GetUmbracoMedia(Int32 id)
at Umbraco.Web.PublishedCache.ContextualPublishedCache`1.GetById(Boolean preview, Int32 contentId)
at Umbraco.Web.UmbracoHelper.TypedDocumentById(Int32 id, ContextualPublishedCache cache)
at ASP._Page_Views_Shared_PageWithBanner_cshtml.Execute()
at System.Web.WebPages.WebPageBase.ExecutePageHierarchy()
at System.Web.Mvc.WebViewPage.ExecutePageHierarchy()
at System.Web.WebPages.StartPage.ExecutePageHierarchy()
at System.Web.WebPages.StartPage.ExecutePageHierarchy()
at System.Web.WebPages.WebPageBase.ExecutePageHierarchy(WebPageContext pageContext, TextWriter writer, WebPageRenderingBase startPage)
at Umbraco.Core.Profiling.ProfilingView.Render(ViewContext viewContext, TextWriter writer)
at System.Web.Mvc.ViewResultBase.ExecuteResult(ControllerContext context)
at System.Web.Mvc.ControllerActionInvoker.<>c__DisplayClass1a.<InvokeActionResultWithFilters>b__17()
at System.Web.Mvc.ControllerActionInvoker.InvokeActionResultFilter(IResultFilter filter, ResultExecutingContext preContext, Func`1 continuation)
at System.Web.Mvc.ControllerActionInvoker.InvokeActionResultFilter(IResultFilter filter, ResultExecutingContext preContext, Func`1 continuation)
at System.Web.Mvc.ControllerActionInvoker.InvokeActionResultWithFilters(ControllerContext controllerContext, IList`1 filters, ActionResult actionResult)
at System.Web.Mvc.Async.AsyncControllerActionInvoker.<>c__DisplayClass25.<BeginInvokeAction>b__22(IAsyncResult asyncResult)
at System.Web.Mvc.Controller.<>c__DisplayClass1d.<BeginExecuteCore>b__18(IAsyncResult asyncResult)
at System.Web.Mvc.Async.AsyncResultWrapper.<>c__DisplayClass4.<MakeVoidDelegate>b__3(IAsyncResult ar)
at System.Web.Mvc.Controller.EndExecuteCore(IAsyncResult asyncResult)
at System.Web.Mvc.Async.AsyncResultWrapper.<>c__DisplayClass4.<MakeVoidDelegate>b__3(IAsyncResult ar)
at System.Web.Mvc.MvcHandler.<>c__DisplayClass8.<BeginProcessRequest>b__3(IAsyncResult asyncResult)
at System.Web.Mvc.Async.AsyncResultWrapper.<>c__DisplayClass4.<MakeVoidDelegate>b__3(IAsyncResult ar)
at System.Web.HttpApplication.CallHandlerExecutionStep.System.Web.HttpApplication.IExecutionStep.Execute()
at System.Web.HttpApplication.ExecuteStep(IExecutionStep step, Boolean& completedSynchronously)
Another "Me Too" for this, we were seeing this under 6.1.6 and have now upgraded to 6.2.4 and are still seeing it.
Loadbalanced: Yes, 2 nodes
Umbraco Version: 6.2.4
Error:
System.FormatException: The valueDictionary is not formatted correctly and is missing any of the 'id,nodeId,__NodeId' elements
at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.DictionaryPublishedContent.ValidateAndSetProperty(IDictionary`2 valueDictionary, Action`1 setProperty, String[] potentialKeys)
at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.DictionaryPublishedContent..ctor(IDictionary`2 valueDictionary, Func`2 getParent, Func`2 getChildren, Func`3 getProperty, Boolean fromExamine)
at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.ConvertFromXPathNavigator(XPathNavigator xpath)
at Umbraco.Web.PublishedCache.XmlPublishedCache.PublishedMediaCache.GetUmbracoMedia(Int32 id)
at Umbraco.Web.PublishedCache.ContextualPublishedCache`1.GetById(Boolean preview, Int32 contentId)
at ASP._Page_Views_Partials__NewsSlider_cshtml.<>c__DisplayClass1.<RenderHighlight>b__0(TextWriter __razor_helper_writer) in <PATH_REMOVED>\views\Partials\_NewsSlider.cshtml:line 42
at ASP._Page_Views_Partials__NewsSlider_cshtml.Execute() in <PATH_REMOVED>\views\Partials\_NewsSlider.cshtml:line 30
at System.Web.WebPages.WebPageBase.ExecutePageHierarchy()
at System.Web.Mvc.WebViewPage.ExecutePageHierarchy()
at System.Web.WebPages.WebPageBase.ExecutePageHierarchy(WebPageContext pageContext, TextWriter writer, WebPageRenderingBase startPage)
at Umbraco.Core.Profiling.ProfilingView.Render(ViewContext viewContext, TextWriter writer)
at System.Web.Mvc.Html.PartialExtensions.Partial(HtmlHelper htmlHelper, String partialViewName, Object model, ViewDataDictionary viewData)
at ASP._Page_Views_HomePage_cshtml.Execute() in <PATH_REMOVED>\views\HomePage.cshtml:line 13
at System.Web.WebPages.WebPageBase.ExecutePageHierarchy()
at System.Web.Mvc.WebViewPage.ExecutePageHierarchy()
at System.Web.WebPages.WebPageBase.ExecutePageHierarchy(WebPageContext pageContext, TextWriter writer, WebPageRenderingBase startPage)
at Umbraco.Core.Profiling.ProfilingView.Render(ViewContext viewContext, TextWriter writer)
at System.Web.Mvc.ViewResultBase.ExecuteResult(ControllerContext context)
at System.Web.Mvc.ControllerActionInvoker.<>c__DisplayClass1a.<InvokeActionResultWithFilters>b__17()
at System.Web.Mvc.ControllerActionInvoker.InvokeActionResultFilter(IResultFilter filter, ResultExecutingContext preContext, Func`1 continuation)
at System.Web.Mvc.ControllerActionInvoker.InvokeActionResultFilter(IResultFilter filter, ResultExecutingContext preContext, Func`1 continuation)
at System.Web.Mvc.ControllerActionInvoker.InvokeActionResultWithFilters(ControllerContext controllerContext, IList`1 filters, ActionResult actionResult)
at System.Web.Mvc.Async.AsyncControllerActionInvoker.<>c__DisplayClass25.<BeginInvokeAction>b__22(IAsyncResult asyncResult)
at System.Web.Mvc.Controller.<>c__DisplayClass1d.<BeginExecuteCore>b__18(IAsyncResult asyncResult)
at System.Web.Mvc.Async.AsyncResultWrapper.<>c__DisplayClass4.<MakeVoidDelegate>b__3(IAsyncResult ar)
at System.Web.Mvc.Controller.EndExecuteCore(IAsyncResult asyncResult)
at System.Web.Mvc.Async.AsyncResultWrapper.<>c__DisplayClass4.<MakeVoidDelegate>b__3(IAsyncResult ar)
at System.Web.Mvc.MvcHandler.<>c__DisplayClass8.<BeginProcessRequest>b__3(IAsyncResult asyncResult)
at System.Web.Mvc.Async.AsyncResultWrapper.<>c__DisplayClass4.<MakeVoidDelegate>b__3(IAsyncResult ar)
at System.Web.HttpApplication.CallHandlerExecutionStep.System.Web.HttpApplication.IExecutionStep.Execute()
at System.Web.HttpApplication.ExecuteStep(IExecutionStep step, Boolean& completedSynchronously)
We don't think this is related to publishing the items in question as this is happening on a news widget on the site and we have seen it even when nothing has been posted for many days.
What we have seen is that it seems to happen when the server is under memory pressure, however this may also not be the case as the AppPool was only 2.4GB at the time this happened and had been up for about 30 minutes.
I think I've seen it both fix without intervention and after a restart.
I hope that's at least slightly helpful.
I've updated the issue here with a description of what is probably happening:
http://issues.umbraco.org/issue/U4-5711
Hi,
Is there also a related ticket for Umbraco 6x?
Thanks,
Paul
As Paul said, any fix for Umbraco 6x.
I'm using version 6.1.6, not loadbalanced. Need a fix that doesn't require an upgrade...
If it is in fact the same issue then your indexes are out of sync so ensure your indexes are in sync and this issue wont happen. And yes there is a fix for version six, see the issue tracker:
http://issues.umbraco.org/issue/U4-5711 http://issues.umbraco.org/issue/U4-5992
Thanks. Sorry, assumed tht fix was only for v7. How do I go about integrating this fix into my code? I'm guessing I have to download the source code, do the amendments and build the project?
If anyone is still following this...
After a lot of step by step debugging the issue related to a corrupt database entry specifically a media item in the cmsContent table that had a column contentType with a value of 0.
Removing this row and related rows, and recreating the item has banished the issue from our environments.
So I don't believe it to be version specific.
Thanks, Darren. This has really helped us link two issues we have been experiencing. https://our.umbraco.org/forum/getting-started/installing-umbraco/57118-622-InternalIndexer-error-ArgumentNullException-Parameter-name-contentType
We ran these queries on the database to delete the orphaned nodes:
This is a fix we knew about to resolve errors that we would see when we would try to reindex the site. We are hopeful that they will solve our intermittent homepage errors now as well.
Yes! Deleting the content from the database for nodes that didn't have contentTypes has solved our problem. We were able to reindex the Internal Indexer without errors and the site has stopped going down.
I'm still seeing the media cache falling apart on versions 7x, in this latest case, 7.1.6 on a single server w/ no load balancing and it's the only site on the server. For some reason the cache gets blitzed for no reason, ie., no users have been within Umbraco for hours, and the only way to fix it is to restart IIS, rebuild the indexes, and republish the site.
This issue brings down ALL media items, not just specific items here and there.
In our case, there are properties on the media items, via TypedMedia, that come back blank, particularly the Url/umbracoFile props.
I've got another thread, https://our.umbraco.org/forum/developers/razor/57936-TypedMedia-URL-is-empty, that may be related to this as well.
So far, upgrading has not been the answer to the problem and there's nothing in the logs that points to any kind of issue in particular. So, to "get around" this, for our other sites that were having this issue, we started pulling the media from the DB rather than from cache. On those sites, we haven't had any more crashes like these, but we would like to get away from that method of pulling media items.
Any help would be appreciated.
Thanks.
Sean
Hi Sean,
For those of us experiencing this issue do you have any example code that shows how to pull the media items directly from the DB, e.g. to get all children of a folder.
This would help until a fix is found.
cheers,
Nick
is working on a reply...