Umbraco growing on disk and filling application pool
Hello!
About two months ago I released my first bigger umbraco site. It's got 6-700 subpages and about as many images. (Mostly cooking recipes). At first everything chugged along just fine, but soon I started to notice the umbraco folder growing on the server. At several occations filling the disk and making all the sites on that server unreachable. I "solved" it by cleaning up old files in other sites. I do have lots of high-res images that is scaled down to appropriate size using ImageGen, and my first guess would be that this is the culprit. How can I check/clear the cache?
Also, I have no proof its the umbraco site that is causing this, but I figure it is since I have never had the problem before: At times my server gives a "Service Unavailable" on all sites i have. The tech-support people chalks this error up to "Filling the application pool". This is easily remedied by restarting the iis service on the server.
Does anyone have any idea how to troubleshoot and (hopefully) fix these issues?
I'm using Umbraco 4.0.0 (started developing the site on umbraco 4 beta) on Windows Server 2003, IIS 6.0, .net 1.0-3.5, Sql Server Express 9.0.3042 on a virtual server (1 gig of ram ~2.8ghz).
Best regards /Hannes
By the way, this is not really an installation question, but I don't know what else fits better. Maybe add a Troubleshooting section?
- Add caching to your macro's - Check the umbraco log table, it can be safely emptied (although I think old versions of your documents will be gone) - The large images stay somewhere on the disk in the original format (if I'm not mistaken) - Disable debug mode in your web.config:
I hope others have more ideas on the application pool running out of memory, maybe you should set it to recycle more often (default is once every 24 hours), but it's not a solution. You should try to figure out why so much memory gets used.
Here a link how to configure the application pool, make sure to have a look on page 2 as well for how to set your IIS instance to another applicationpool
If you have a huge media folder, it is likely an issue with ImageGen. I have had several sites exploded in using ImageGen to resize images. It is a caching bug which I believe has been corrected. You may need to check you version and download the latest version from Doug's site.
To clear the cache folders, you can script it using a user control or various other ways. I have server access, so I did a search and deleted all cache folders when replacing the dll.
Also, regarding the application pool failing, there is likely an app or piece of code that is erroring quickly. By default, app pools are set to fail after a few errors in a short time frame. You can increase this number to help, but I would not increase it too much. The best thing to do is to find which sites are erroring by looking at the event logs and also segmenting your sites to find the culprit.
Also, depending upon which database you are using, sometimes some db tweaks can improve performance and alleviate some common issues.
I see you are using SQL Express. For each database, set the Auto Close setting to false. This setting causes some issues when running several databases. Also, consider limiting SQL's memory footprint, as SQL will use every byte it can get. Since you have 1GB, set the limit to something like 400 or 500 MB.
1 GB is only about enough to run about 10-15 sites stabily in my experience.
I got the exact same problem using umbraco 4.0.2 and imagegen 2.0.1 (newest download from the website) Is there a newer version of imagegen and If so, where to find it?
I'm trying things out, and so far it has worked. I updated ImageGen to a later version and the cache folder has not yet exploded. It's still kinda big, but i have a lot of resized images and dynamic image titles.
The application pool stuff seemed to happen about every 3-4 weeks. (on sundays which is a high load day for this site.)
I'm experiencing the same problem. I'm not sure about the disk size (haven't checked it) but I have problem with application pool growing (which seems to result in website going down). When I had a look in the Eventlog I got these errors:
"A process serving application pool '(domain)(2.0)(pool)'
suffered a fatal communication error with the World Wide Web Publishing
Service. The process id was '21992'. The data field contains the error
number."
"Event code: 3005
Event message: An unhandled exception has occurred.
"A process serving application pool '(domain)(2.0)(pool)'
suffered a fatal communication error with the World Wide Web Publishing
Service. The process id was '21992'. The data field contains the error
number."
"Event code: 3005
Event message: An unhandled exception has occurred.
I've installed the patch and it probably works. But my application pool still grows rapidly and approximately once a day the website goes down and the browser shows the message:
"Not enough storage is available to process this command"
This issue is getting frustrating...
I've now updated charset in http header to utf-8 as suggested in other forum thread and updated ImageGen to 2.2.1 as suggested in this article.
Umbraco growing on disk and filling application pool
Hello!
About two months ago I released my first bigger umbraco site. It's got 6-700 subpages and about as many images. (Mostly cooking recipes).
At first everything chugged along just fine, but soon I started to notice the umbraco folder growing on the server. At several occations filling the disk and making all the sites on that server unreachable. I "solved" it by cleaning up old files in other sites.
I do have lots of high-res images that is scaled down to appropriate size using ImageGen, and my first guess would be that this is the culprit. How can I check/clear the cache?
Also, I have no proof its the umbraco site that is causing this, but I figure it is since I have never had the problem before:
At times my server gives a "Service Unavailable" on all sites i have. The tech-support people chalks this error up to "Filling the application pool". This is easily remedied by restarting the iis service on the server.
Does anyone have any idea how to troubleshoot and (hopefully) fix these issues?
I'm using Umbraco 4.0.0 (started developing the site on umbraco 4 beta) on Windows Server 2003, IIS 6.0, .net 1.0-3.5, Sql Server Express 9.0.3042 on a virtual server (1 gig of ram ~2.8ghz).
Best regards
/Hannes
By the way, this is not really an installation question, but I don't know what else fits better. Maybe add a Troubleshooting section?
Some tips:
- Add caching to your macro's
- Check the umbraco log table, it can be safely emptied (although I think old versions of your documents will be gone)
- The large images stay somewhere on the disk in the original format (if I'm not mistaken)
- Disable debug mode in your web.config:
and:
- Disable tracing in your web.config:
I hope others have more ideas on the application pool running out of memory, maybe you should set it to recycle more often (default is once every 24 hours), but it's not a solution. You should try to figure out why so much memory gets used.
Clearing the umbracoLog table does not impact on previous revisions of your content.
Maybe not much of a help, but in IIS you should at least isolate each application in it's own application pool.
That way you can see which application is causing the error, and it will not have impact on the other applications.
Another advantage is that you can restart a single applicationpool instead of restarting IIS.
Nico
Here a link how to configure the application pool, make sure to have a look on page 2 as well for how to set your IIS instance to another applicationpool
http://www.developer.com/net/asp/article.php/2245511
Nico
hi kebabskal
If you have a huge media folder, it is likely an issue with ImageGen. I have had several sites exploded in using ImageGen to resize images. It is a caching bug which I believe has been corrected. You may need to check you version and download the latest version from Doug's site.
To clear the cache folders, you can script it using a user control or various other ways. I have server access, so I did a search and deleted all cache folders when replacing the dll.
Case
Also, regarding the application pool failing, there is likely an app or piece of code that is erroring quickly. By default, app pools are set to fail after a few errors in a short time frame. You can increase this number to help, but I would not increase it too much. The best thing to do is to find which sites are erroring by looking at the event logs and also segmenting your sites to find the culprit.
Also, depending upon which database you are using, sometimes some db tweaks can improve performance and alleviate some common issues.
I see you are using SQL Express. For each database, set the Auto Close setting to false. This setting causes some issues when running several databases. Also, consider limiting SQL's memory footprint, as SQL will use every byte it can get. Since you have 1GB, set the limit to something like 400 or 500 MB.
1 GB is only about enough to run about 10-15 sites stabily in my experience.
I got the exact same problem using umbraco 4.0.2 and imagegen 2.0.1 (newest download from the website) Is there a newer version of imagegen and If so, where to find it?
/Finn
You can best contact Douglas directly for this. I'm afraid he didn't have time yet to update the package on his site.
email me if you can't get hold of Doug, as I have a copy.
casey at neehouse dot com.
Case
Ahh.. Doug released the latest version... http://our.umbraco.org/projects/imagegen
Thanks for all the replies!
I'm trying things out, and so far it has worked.
I updated ImageGen to a later version and the cache folder has not yet exploded. It's still kinda big, but i have a lot of resized images and dynamic image titles.
The application pool stuff seemed to happen about every 3-4 weeks. (on sundays which is a high load day for this site.)
Looks like it's going to work now.
Thanks again!
/Hannes
I'm experiencing the same problem. I'm not sure about the disk size (haven't checked it) but I have problem with application pool growing (which seems to result in website going down). When I had a look in the Eventlog I got these errors:
<!-- /* Font Definitions */ @font-face {font-family:"Cambria Math"; panose-1:2 4 5 3 5 4 6 3 2 4; mso-font-charset:1; mso-generic-font-family:roman; mso-font-format:other; mso-font-pitch:variable; mso-font-signature:0 0 0 0 0 0;} @font-face {font-family:Calibri; panose-1:2 15 5 2 2 2 4 3 2 4; mso-font-charset:0; mso-generic-font-family:swiss; mso-font-pitch:variable; mso-font-signature:-1610611985 1073750139 0 0 159 0;} @font-face {font-family:Consolas; panose-1:2 11 6 9 2 2 4 3 2 4; mso-font-charset:0; mso-generic-font-family:modern; mso-font-pitch:fixed; mso-font-signature:-1610611985 1073750091 0 0 159 0;} /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {mso-style-unhide:no; mso-style-qformat:yes; mso-style-parent:""; margin:0cm; margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:Calibri; mso-fareast-theme-font:minor-latin; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi; mso-fareast-language:EN-US;} p.MsoPlainText, li.MsoPlainText, div.MsoPlainText {mso-style-noshow:yes; mso-style-priority:99; mso-style-link:"Oformaterad text Char"; margin:0cm; margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.5pt; font-family:Consolas; mso-fareast-font-family:Calibri; mso-fareast-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi; mso-fareast-language:EN-US;} span.OformateradtextChar {mso-style-name:"Oformaterad text Char"; mso-style-noshow:yes; mso-style-priority:99; mso-style-unhide:no; mso-style-locked:yes; mso-style-link:"Oformaterad text"; mso-ansi-font-size:10.5pt; mso-bidi-font-size:10.5pt; font-family:Consolas; mso-ascii-font-family:Consolas; mso-hansi-font-family:Consolas;} .MsoChpDefault {mso-style-type:export-only; mso-default-props:yes; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:Calibri; mso-fareast-theme-font:minor-latin; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi; mso-fareast-language:EN-US;} @page Section1 {size:612.0pt 792.0pt; margin:70.85pt 70.85pt 70.85pt 70.85pt; mso-header-margin:36.0pt; mso-footer-margin:36.0pt; mso-paper-source:0;} div.Section1 {page:Section1;} -->
"A process serving application pool '(domain)(2.0)(pool)' suffered a fatal communication error with the World Wide Web Publishing Service. The process id was '21992'. The data field contains the error number."
"Event code: 3005
Event message: An unhandled exception has occurred.
Event time: 8/10/2009 10:26:38 AM
Event time (UTC): 8/10/2009 8:26:38 AM
Event ID: 7b24254fd8c340c18baacc012d459298 Event sequence: 864 Event occurrence: 1 Event detail code: 0 ...
Exception information:
Exception type: HttpException
Exception message: Invalid viewstate."
This made the website go down with the following errors shown when trying to access the website:
"not enough storage is available to process this command" ...or... "Service Unavailable" ...or just... "0x00000008"
I've looked up the "invalid viewstate" error which seems to be a IE with .Net AJAX bugg - I think I solved this by setting charset, read more about it in this article: http://stackoverflow.com/questions/728513/erratic-invalid-viewstate-issue-in-net-application
But the application pool still seems to be growing (rapidly from approx 100Mb to approx 800Mb) untill IIS restarts application pool.
I have the ImageGen 2.0.1 installed on a Umbraco 4.0.0.
The ImageGen 2.0.1 I have installed is created in 2008, there can't exist a newer version with same versionname, can it?!?
Any ideas would be appreciated!!!
Thanks in advance!
Regards, Tony H
"A process serving application pool '(domain)(2.0)(pool)' suffered a fatal communication error with the World Wide Web Publishing Service. The process id was '21992'. The data field contains the error number."
"Event code: 3005
Event message: An unhandled exception has occurred.
Event time: 8/10/2009 10:26:38 AM
Event time (UTC): 8/10/2009 8:26:38 AM
Event ID: 7b24254fd8c340c18baacc012d459298 Event sequence: 864 Event occurrence: 1 Event detail code: 0 ...
Exception information:
Exception type: HttpException
Exception message: Invalid viewstate."
Sorry about the crazy font-definitions in posts above - don't know how they got there and couldn't edit my post after submit...
There is a patch to ImageGen, which will be version 2.2. You can find it at http://our.umbraco.org/projects/imagegen
That may or may not be related to your problem, but it is a good patch to have nonetheless. Give it a try and see what happens.
cheers,
doug.
Thanks Douglas,
I've installed the patch and it probably works. But my application pool still grows rapidly and approximately once a day the website goes down and the browser shows the message:
"Not enough storage is available to process this command"
This issue is getting frustrating...
I've now updated charset in http header to utf-8 as suggested in other forum thread and updated ImageGen to 2.2.1 as suggested in this article.
Any other ideas?
Thanks in advance!
Regards, Tony H
is working on a reply...