Jaspal Khanna
asked on
Connection Timeout - Max pool size reached on System Center 2016 - Blank page on login ( Randomly
System Center 2016 giving error SyncInvoke--------- Failed with system.invalidoperationexc
This error is happening in many of the method which we didnt make any changes. approx 20-25 of the, may be more.
Normally we can login without issue after app pool reset, however once it gives the above error in the system center even the default page of the webserver shows blank page, then after sometime when we just refresh the default page would be displayed.
This code has been running for more than 2 years and we didnt have any problem , recently we did the deployment and the problem started. The website works for sometime but after say 20 minutes, it starts giving the above timeout issue and sometime even the default page does not show up and then later when we just refresh the default page will display.
Number of users that can cause this is even 15 users , earliernormal traffic is about 300 users at a time, the issue can happen even when 5 users are present.
DB SQL Server : There are around 150-250 sleeping connection most of the time with the text "Text" only, it reduces automatically.
The layout is as follows:
Client are redirected through load balancer on any of the following, web server are connected to app directly and APP is connected to SQLDB
External Server
__________________________
Web server 1 Web server 2 Database : SQLDB
WebApp 1 Web App 2
we have internal server as well and when we test on that with 15 users it does not have any problem whatsoever. we have placed 30 min app pool reset on both App 1 and 2.
Internal server
__________________________
Web Internal server 1 Database : SQLDB ( Database is same as production on top )
WebApp Internal 1
ListofError.png
This error is happening in many of the method which we didnt make any changes. approx 20-25 of the, may be more.
Normally we can login without issue after app pool reset, however once it gives the above error in the system center even the default page of the webserver shows blank page, then after sometime when we just refresh the default page would be displayed.
This code has been running for more than 2 years and we didnt have any problem , recently we did the deployment and the problem started. The website works for sometime but after say 20 minutes, it starts giving the above timeout issue and sometime even the default page does not show up and then later when we just refresh the default page will display.
Number of users that can cause this is even 15 users , earliernormal traffic is about 300 users at a time, the issue can happen even when 5 users are present.
DB SQL Server : There are around 150-250 sleeping connection most of the time with the text "Text" only, it reduces automatically.
The layout is as follows:
Client are redirected through load balancer on any of the following, web server are connected to app directly and APP is connected to SQLDB
External Server
__________________________
Web server 1 Web server 2 Database : SQLDB
WebApp 1 Web App 2
we have internal server as well and when we test on that with 15 users it does not have any problem whatsoever. we have placed 30 min app pool reset on both App 1 and 2.
Internal server
__________________________
Web Internal server 1 Database : SQLDB ( Database is same as production on top )
WebApp Internal 1
ListofError.png
ASKER
We are sure that none of the new connection were made in the code, All code are from previous deployment, the changes for this deployment was adding new fields in the table and related control on the aspx page with JS changes.
Its with the deployment, I suppose but unable to understand whats the root cause and how to resolve it?
Its with the deployment, I suppose but unable to understand whats the root cause and how to resolve it?
the changes for this deployment was adding new fields in the table and related control on the aspx pageIs it possible that some new DataReaders or functions returning DataReaders were introduced? Connection will never close outside the function that created them unless they are properly discarded/closed.
SCCM 2016 is a red herring here. The problem has nothing to do with SCCM or SCOM they are simply informing you of errors in other programs/applications.
Somewhere in your code you are not closing connections and just keep opening more connections. This article explains a bit.
https://blogs.msdn.microsoft.com/tolong/2006/11/20/max-pool-size-was-reached/ you can also in your web.config increase the # of connections.
Somewhere in your code you are not closing connections and just keep opening more connections. This article explains a bit.
https://blogs.msdn.microsoft.com/tolong/2006/11/20/max-pool-size-was-reached/ you can also in your web.config increase the # of connections.
ASKER
Thanks Icohan, Code had been changed in such a way that it has utilized the method which we use as always and those method are the same method used in previous deployment. so connection are getting closed and seems this msg might be not the real cause and actual root cause might be different.
ASKER
Thanks David, These are the same code for opening and closing connection what we had the last deployment, we have only added many new controls, some are linked to the database to fetch the data to populate the dropdown list / multiple checkbox but those methods are the same ones which we have used before in previous deployment.
I was thinking that it might be a misleading error could be network / connection related issues on the server.
I heard the TCP issues can also output similar connection error, Not sure about it.
I was thinking that it might be a misleading error could be network / connection related issues on the server.
I heard the TCP issues can also output similar connection error, Not sure about it.
This question needs an answer!
Become an EE member today
7 DAY FREE TRIALMembers can start a 7-Day Free trial then enjoy unlimited access to the platform.
View membership options
or
Learn why we charge membership fees
We get it - no one likes a content blocker. Take one extra minute and find out why we block content.
Can someone with authority perform a compare between old and new code as most likely somewhere in the new version some connections are maybe not closed/discarded as they should? Or maybe check for some DataReaders that are not closed? But all this is in line whit the deployment as it is obviously from the description that this issue was introduced with that deployement - whatever that was.