IIS issues after K2 Five upgrade in place

  • 29 October 2019
  • 3 replies
  • 24 views
B
Badge +8

I have recently performed an update in place from K2 4.6.11 to 4.7 and then to K2 Five. The upgrade was performed in a stable environment. Once the upgrade was complete I started having a daily issue with the application pools in IIS shutting down on a daily basis. The WAS service also shuts down. The only thing that has changed in this environment was the K2 upgrade. I am also getting an error in Blackpearl Server when started. I am not sure if these two issues are related but the error is as follows.

 

Failed to start socket service: Only one usage of each socket address <protocol/network address/port> is normally permitted.

 

I cloned the original environment. That environment was also stable. Upon upgrading that environment to K2 Five I have the same issue. I shut down this environment to ensure there were no issues with network configuration.

 

Does anyone have any ideas as to what is causing IIS to fail? 

 

K2 Developers if you are listening. We truely hate this upgrade process. It is dangerous to our processes. If K2 is going to be central to our business processes there needs to be ZERO room for failure upon upgrade. From my point of view if the upgrade fails our business processes stop. This means the business that depends on them stop. This means production isn't moving and we are not making money. How are we supposed to depend on K2 to govern our processes when we can't even depend on K2 to provide an upgrade path that has zero percent chance of failure. We are afraid to to put any processes from the manufacturing floor into K2 because we do not want to have to redevelop the processes upon future upgrades and we don't know if a future upgrade will cause failure to our current processes.

 

Thanks,

Bryan Peters


16460i786AFE365676D378.png

3 replies

C

Hello,


 


Please take a look at the following KB article regarding your reported error:  https://help.k2.com/support-services/kbt144385.  This can often times be seen when there are multiple instances of the K2 Server service running. 


 


I also understand that issues with upgrades can be frustrating.  K2 is trying to optimize the install/upgrade process everyday, and K2 Five installers tend to be a lot more user-friendly.  Additionally, have you considered a migration to K2 Cloud?  I would recommend speaking to your K2 Technical Account Manager regarding obtaining more information on this.  


 


Thanks


 


 

S
Badge +6

Hi Bryan,

 

Lets seperate these issues:

  • IIS app pools and WAS service shutdown
  • K2 logs show "Only one usage of each socket address <protocol/network address/port> is normally permitted"

Regarding the first issue: Did you see the Application and System event log to determine why the App Pools are stopped/crashing. There is usually an event logged there with more details. There are several settings for app pool that can be configured which IIS uses to take automatic corrective actions. By default, IIS sets the app pools to Suspend as Idle Time-out Action. Check if it is set to Terminate. Also verify the "Start Mode", it should be AlwaysRunning. 

 

Regarding the second issue, it is a generic Windows message that is logged if multiple services are trying to use the same port on the same IP address. Did you happen to use the same IP address or port for multiple VMs?

M
Badge +11

Good place to start is Administrative Events veiws on machine which hosts IIS server. I'm sure there will be an error message or event indicating why IIS was stopped/crashed. Once you have specific even like unhandled exception you can work with that to get it fixed. From my side I saw couple of scenarios where IIS app pool was crashing in V5 and all of those already addressed in current CU/FPs, so you can just try to upgrade to the latest FP/CU and if your issue is still persist try to find for specific error/eception logged on IIS/WAS shut down - I assume they probably crashing with some unhandled exception, in any case we need to see what's logged at the time of shut down.


Also be sure to check if any Windows Updates were applied on the server, you should not assume that, you need to verify :) But most logical thing is to always start with check on exact events associated with service shut downs.


 


For example to check on WAS service stops/crashes - you may use this PowerShell command:


 


 Get-EventLog -LogName system -Source "service control manager" |?{$_.message -match "WAS"} | select timegenerated, message -First 10

This will give you the following output:



If you don't see "stopped unexpectedly" in the output then it means service has been stopped regular way after receiving stop command (user action or machine reboot).


 


I believe you cannot do the same trick for individual app pools as they run as individual w3wp.exe processe not corresponding to a service, but I'm 100% sure you can find some unhandled exception in Administrative Event view or Application log.

Reply


View our Community Site Map

Nintex Community Sitemap
Terms & ConditionsCookie settings