I have been struggling with this issue for literally weeks now, but seem to have found a resolution after trying every option on the net. However i'm still at a loss to explain it.
We have a web service with a method that performs a long running SQL query and takes roughly 7 minutes to complete. However, on intermittant runs we get the error;
"The underlying connection was closed: An unexpected error occurred on
a receive."
Now it turns out that in IIS7 it was the Process Model section on the application pool causing this. Even if we set the 'Ping Maximum Response Time' to something like 600 seconds (10 minutes) the error would still appear at random times... sometimes even after only 100 seconds or so.
The only way to stop this error from happening is to set the 'Ping Enabled' feature to false.
I am concerned that this might not be the best way to fix the issue... but the client needs it fixed so this will do for now. Can anyone shed any light on this? We have other sites using older versions of IIS and SQL that do not have this issue. It is only happening on a server running SQL 2008 R2 Express and IIS 7.5.
It is built on the .NET framework v2.0.
Thanks for any insight anyone could give.
Related
this quite weird happen to our server and this happen last week ago and I don't know where this issue came from, The testing procedure that we do is to check in localhost if this module part is working well and check each task if works well. The conclusion of our test is working on the local side. However on our staging and production side some of the api is not working and based on our observation this is on the AWS problem issue.
First Testing Process :
What we did right now is to check the Idle timeout of load balancer.
From 60 we increase it by 4000 as testing process if this will work but on our testing process still same issue.
Second Testing Process:
To check again each api if working well and all our api is working well.
Error:
I have a fusebox application setup on coldfusion 9 and it is in production mode. When application got started, it is giving me following error.
A timeout occur while attempting to local fusebox.
I have increased the timeout but that not help. Also tried other solution but nothing help.
There is one more thing. When i increased the JVM heap size from 1-2 gb to 2-3 gb. Then after restart of coldfusion service, Error remain for 1-2 hour and then site start working.
It also stop working after running for more than half day correctly. Then i get 504 gateway error and i have to restart the service again.
Can any one guide me to solve this problem?
Which FB version are you running? Is it set to production or development mode? If set to development, try setting it to production mode.
On loading the file from Apex application, I'm getting the Network Error (tcp_error).
Error Desc: "A Communication error occurred. The Web Server may be down, too busy, or experiencing other problems preventing it from responding to requests. You may wish to try again at a later time".
Please help me in resolving this issue. If increasing the timeout period fixes the issue, please let me know where exactly to do.
I have the same experience before. This might solve your problem. I re-install my Oracle APEX with firewall turn off.
An exception occurred when setting up mail server parameters. This
exception was caused by:
coldfusion.mail.MailSpooler$SpoolLockTimeoutException: A timeout
occurred while waiting for the lock on the mail spool directory..
Recently i started to get this nasty exception in my mail.log file. Once this exception shows up, every mail that is sent from that coldfusion instance throws the same exception.
The only thing that seems to work is to restart the coldfusion server. After (usually) a day or two the same exception pops up again and we're back in the same situation.
I am aware of the hotfix to control the mailspool timeout but all it does is increase the timeout from 30 to 60 seconds. Since the mails are sent successfully until i get the exception, i don't think this is my solution.
Also i read the thread in the adobe forum where people have installed the hotfix, but still get the error.
I also tried a script to restart only the mailservice when this exception showed up, but this didn't work for me, as it didn't for others with this problem. This would also not be a concrete solution.
The mails that i send arre simple html mails.
The number of mails sent spreaded over a day is not more then 30.
I've sent mails from
the exact same coldfusion server many times before, but with
<cfmail>. This is the first time i'm sending them in cfscript. I
don't know if this has anything to do with it, but it's only since
i'm using the cfscript equivalent of <cfmail> that i started to get
this exception.
All related blog posts that i could find are all unanswered but also pretty old. I thought that someone might have a solution by now.
Thanks.
(using coldfusion 9.0.1 server on windows 2008 server)
We were also experiencing this mail spool lock issue. After the issue occurred a fourth time in 2 months, we started reviewing these forums and found no solution.
This made me think that perhaps the solution and problem are not really CF at all, so I went into the server's virus protection and excluded the CF mail spool directory so that the virus protection does not touch the spool directory at all. So far, we have not had the problem again.
So I am not sure that this is the permanent fix, but it has worked so far for us. No outside entities create emails within our systems, so the directory should be relatively safe but not having email-outs work is not an option.
this chain from talkingtree might give some light:
http://www.talkingtree.com/blog/index.cfm?mode=entry&entry=67FD4A34-50DA-0559-A042BCA588B4C15B
what they are saying is that it could be an issue with disk activity taking to long. you can increase the mail spool timeout with the jvm argument: -Dcoldfusion.spooltimeout=120
oh.... one more thing. if you're using cfmail to email dumps when an error occurrs, make sure to add 'format="text"' to the cfdump tags. some of the emails can get pretty big and might be causing the error.
This question already has answers here:
Closed 11 years ago.
Possible Duplicate:
How to restart Coldfusion Application Server when application is timeout?
Currently I have an ColdFusion application that causes server issues. After 1-2 days that serve doesn't respond until a manual restart is done.
I know that I had to find what is going wrong in my scripts and I spend time and time for several weeks.
But pending I would make a script that restart automatically ColdFusion service if it is bugged.
I have not many knowledge in batch script etc.. but I guess that the test would be a request to a .cfm and the response would be serve until a timeout. ?
Has anyone ever met a script like this ?
Config: Win 2k8 Server R2 - Coldfusion 9(.0.0)
Thank you
Two things here
The real way is to fix the issue and you can do that with Fusion Reactor - http://www.fusion-reactor.com/fr/ It will help you monitor and restart and self heal as it needs.
You could create a batch file, and create a Scheduled Task in Windows that ran it.
Using Net Start / Net Stop Commands
net stop "Macromedia JRun CFusion Server"
net start "Macromedia JRun CFusion Server"
Thought this may not always work so I have a batch file:
c:\JRun4\uninstall\KillJRun.exe
net start "Macromedia JRun CFusion Server"
Which works for me.
Your best bet is to use Pingdom or another server monitoring tool. When the server goes down (responds with a 503 error, service unavailable) you may be able to have Pingdom send a response to a PHP script on the server that calls a batch file. I am not sure if Pingdom supports pinging another server is one is down, but you could have Pingdom email to an inbox that your PHP can check every few minutes.
This may end up being more work than figuring out what is wrong with your script though.
Edit: You may want to look at this question. This will only work if the service has stopped, whereas usually when a script crashes ColdFusion it is hanging. If you run the script that crashes the server, then look at the service, if it says stopped, then this may work for you.
The other thing that I would check is the JVM memory. Often times crashes are due to processing large amounts of data from files or the database and the JVM doesn't have the memory to do that.
Nope. It cannot be restart automatically when your CF services/server is hanging. The only one way is to restart by windows schedule.
You could also use Nagios+Plugins to fire a restart script when the service hangs. But following the previous advice & finding out what the problem is is your best bet.