Very strange issue that's happened couple of times over the course of two days. On our production server the umbraco site app pool CPU goes Upto 100% and brings the whole server down.
Recycling the app pool brings everything back to normal and then stays normal.
Have looked at the logs but nothing helpful there it's really hard to replicate the issue as ther server is behaving normal now.
I have set the app pool to recycle every 30min but this is not an ideal solution.
I am looking to add some code that checks if CPU exceeds 60% then restart the app pool but then again it's a work around.
The site has behaved normal without any issues for the past 4-5 months.
Any help from the umbraco community will be we'll appreciated
I have checked and those two updates are not installed on the server.
However whenever the app pool uses 100% CPU I do see GoogleUpdate server being stopped and given as an error in the Server Manager dashboard any link between the two??
I am literally pulling my hair out with this. I am trying to recommned UMbraco to be used throughout the company and this issue is really really effecting that!!
Okay I have when it went above 60 I took a memory dump of that app pool before suspending it.
I have deleted two newly created pages yesterday republished the whole site and it's behaving and haven't gone down. But I am not 100% convinced that it was those two pages causing the high CPU.
I looked into the memory dump using windebug and found at least 4-5 threads that were taking upto 10min each all these threads were similar and looking at the stack trace I got the following. But just to let you know I deleted two newly created pages by the content admin on saturday mornign and after that I have stopped seeing the High CPU issue and the site hasn't gone done since. Makes me think if those two duplicated pages were the culprit. This memory dump was taken on Friday evening and below is the clrtstack trace from windebug. Any help or explanation is much appreicated..
OS Thread Id: 0x1008 (38)
Child SP IP Call Site
000000db9a9dcac0 00007ffb14611fbc System.Collections.Generic.Dictionary2[[System.__Canon, mscorlib],[System.Boolean, mscorlib]].Insert(System.__Canon, Boolean, Boolean)
000000db9a9dcb50 00007ffab7d1d45e Umbraco.Core.Models.EntityBase.TracksChangesEntityBase.OnPropertyChanged(System.Reflection.PropertyInfo)
000000db9a9dcba0 00007ffab7d16f2c Umbraco.Core.Models.EntityBase.TracksChangesEntityBase.SetPropertyValueAndDetectChanges[[System.__Canon, mscorlib]](System.Func2
Umbraco app pool suddenly jumps Upto 100%
Hi,
Very strange issue that's happened couple of times over the course of two days. On our production server the umbraco site app pool CPU goes Upto 100% and brings the whole server down.
Recycling the app pool brings everything back to normal and then stays normal.
Have looked at the logs but nothing helpful there it's really hard to replicate the issue as ther server is behaving normal now.
I have set the app pool to recycle every 30min but this is not an ideal solution.
I am looking to add some code that checks if CPU exceeds 60% then restart the app pool but then again it's a work around.
The site has behaved normal without any issues for the past 4-5 months.
Any help from the umbraco community will be we'll appreciated
Many thanks B
Hi Bilal,
To med it sounds like: http://issues.umbraco.org/issueMobile/U4-6338
What Server version is it?
Hi Rasmus,
I have to check when I am in the office but I believe it's 2012 r2 have to check if the update is applied or not.
But I don't get any log entry about configuration issue
Thanks
Hi Rasmus,
I have checked and those two updates are not installed on the server.
However whenever the app pool uses 100% CPU I do see GoogleUpdate server being stopped and given as an error in the Server Manager dashboard any link between the two??
I am literally pulling my hair out with this. I am trying to recommned UMbraco to be used throughout the company and this issue is really really effecting that!!
Have you tried looking at the "working process" under applications?
When it is above 60, you should be able to pin point the process that's taking longer then expected.
Okay I have when it went above 60 I took a memory dump of that app pool before suspending it.
I have deleted two newly created pages yesterday republished the whole site and it's behaving and haven't gone down. But I am not 100% convinced that it was those two pages causing the high CPU.
Could it be a dos or ddos attack??
Thanks Ad
Hi Rasmus,
I looked into the memory dump using windebug and found at least 4-5 threads that were taking upto 10min each all these threads were similar and looking at the stack trace I got the following. But just to let you know I deleted two newly created pages by the content admin on saturday mornign and after that I have stopped seeing the High CPU issue and the site hasn't gone done since. Makes me think if those two duplicated pages were the culprit. This memory dump was taken on Friday evening and below is the clrtstack trace from windebug. Any help or explanation is much appreicated..
OS Thread Id: 0x1008 (38) Child SP IP Call Site 000000db9a9dcac0 00007ffb14611fbc System.Collections.Generic.Dictionary
2[[System.__Canon, mscorlib],[System.Boolean, mscorlib]].Insert(System.__Canon, Boolean, Boolean) 000000db9a9dcb50 00007ffab7d1d45e Umbraco.Core.Models.EntityBase.TracksChangesEntityBase.OnPropertyChanged(System.Reflection.PropertyInfo) 000000db9a9dcba0 00007ffab7d16f2c Umbraco.Core.Models.EntityBase.TracksChangesEntityBase.SetPropertyValueAndDetectChanges[[System.__Canon, mscorlib]](System.Func2What packages so you have installed?
It sound and looks like the problem i mentioned earlier, I suggest you try and install the hotfix, just to see if that helps.
By looking at the dump. I see one that sticks out: Umbraco.Core.Cache.ICacheProvider, System.String, System.Func1
Each time I would publish, the Umbraco solution would hang. With the CPU 100%
http://stackoverflow.com/questions/28933078/application-domain-restarts-on-node-publish-umbraco/28943294#comment46410135_28943294
Packages I have installed? I don't think I have installed any new packages I will give you a list of all the existing packages.
Sorry what was the hotfix again? And apologise if I am asking stupid question just been working all night and my mind is a bit slow now
I have checked the production server and I couldn't find the the Microsoft update will have a look again to be absolutely sure!
Thanks for your help really appreciate it!
The hotfix should be part of windows update by now. So just check if your are up to date on all updates.
Try reporting it on issues.umbraco.com and see if shannon replies. Hes the best at such issues.
is working on a reply...