CF10 Datasource settings

I'm looking at settings for MS SQL Server data source, and I have a few questions about.

Connection limit

I see no reason to use this exception to help minimize the damage of a DOS attack.

Max pooled statements

I understand that this must be set on the amount of unique cfquery tags that use the cfqueryparam and cfstoredproc tags tag. It is great an all but why?

Timeout

That's how long he keeps each connection to this open data source. Is there a reason to change it, and how determine you the right setting?

Remote Query Timeout

It is by default 0 (no timeout), and I don't want a query running forever. I can't think the last time I had a query to run more than one second. Any reason why I wouldn't put in 5 seconds or less?

Thank you
Phil

Limit connections: why?

The main reason is that connections cost lots of processing overhead (CPU resources). If you have too much of themselves, things will finally slow down to a syrup.

Max Pooled statements: Why value it equal the number of unique cfquery tags that use the tag cfqueryparam and cfstoredproc tags?

The aim is efficiency. These query types involve a reuse. The underlying Java engine thus creates and maintains a pool of prepared statements to be reused by this each request. The number of unique queries is a good starting point. It is not necessary to allocate more memory space (for a larger pool) to. In addition, since Java creates grouped statements based on active connections, the connection must be maintained for the re-use of statements of pooling is possible.

Timeout: Is there a reason to change and how determine you the right setting?

That's how long ColdFusion keeps an idle connection before you destroy. I would like to leave 20 minutes by default. Which corresponds to the range generally accepted for session timeout in ColdFusion applications, namely, 20 to 30 minutes.

Remote Query Timeout: A reason to put it in 5 seconds or less?

Relational database management system is separated from ColdFusion. For reasons beyond control of ColdFusion, a query may run beyond the time that awaits the ColdFusion application. Coldfusion threads that wait on the request would continue to do so indefinitely. If it is essential to your application where a query to run less than 5 seconds, or well abandonment, then you put 'Remote Query Timeout' to 5 seconds. Charlie Arehart has written an informative on that article.

Tags: ColdFusion

HTTP 500.19 - internal error in the server error

The requested page cannot be accessed because the configuration data of the page are not valid.

Module	IIS Web Core
Notification	BeginRequest
Handler	Not yet determined
Error code	0 x 80070003
Config error	Cannot read the configuration file
Configuration file	\\?\C:\ColdFusion10\cfusion\wwwroot\ipmcenters_org\web.config

When you were on 10 ColdFusion, were serve you content in ColdFusion web root (\ColdFusion10\cfusion\wwwroot\...)? This seems to be the case based on the above error. If you use IIS, you should not use the sites off ColdFusion web root. If you have to, use IIS web root (C:\inetpub\wwwroot), although it is best to put your files on the web on a completely different disc of your OS (and ColdFusion must be it's own separate disk). So here's what I do (based on the different ColdFusion lockdown guides):

C: - OS file

D: - Installing ColdFusion (D:\ColdFusion10\...)

E: - websites (E:\WebSites\site1, E:\WebSites\site2, etc.)

The web.config file is an IIS configuration file (and the HTTP 500.19 error comes to IIS). This indicates that your IIS site configuration is probably at fault. You can take things apart and rebuild the site by lockdown CF11 guide and rebuild the site in IIS as a result.

503 errors after about 3 months w / CF10

Server is not responding at all two days now. The record that they are running. Restarting temporarily solves.
My first thought was that he was a customer variables problem, 90 days purge people causes a lot of problems. But I have confirmed that the source code does not reference "ClientComm". I moved variable storage of customer in a database anyway (nothing is getting filled,) and disabled the global updates. It always crashes after doing this.
There is not a lot in newspapers (there is no hs*.log file and no exceptions in the regular CF logs), and so far Fusion reactor stressed not smoking guns. The only thing I can find is in the isapi_redirect.log file (ColdFusion10/config/wsconfig/1 /):
[18:44:05.215 Wed Jul 23 2014] [13144:10964] [info] ajp_process_callback::jk_ajp_common.c (2066): current number of reuse is 118 connection re-use max 250 and total number of endpoint 500
[18:44:05.217 Wed Jul 23 2014] [13144:6856] [info] ajp_process_callback::jk_ajp_common.c (2066): current number of reuse is 119 connection re-use max 250 and total number of endpoint 500
[19:48:52.294 Wed Jul 23 2014] [13144:6344] [info] jk_open_socket::jk_connect.c (626): connect to 127.0.0.1:8012 failed (errno = 61)
[19:48:52.295 Wed Jul 23 2014] [13144:6344] [info] ajp_connect_to_endpoint::jk_ajp_common.c (1047): failed to open socket (127.0.0.1:8012) (errno = 61)
[19:48:52.299 Wed Jul 23 2014] [13144:6344] [error] ajp_send_request::jk_ajp_common.c (1669): failed to connect to the back-end (cfusion). Tomcat is probably not started or is listening on the port incorrect (errno = 61)
[19:48:52.301 Wed Jul 23 2014] [13144:6344] [info] ajp_service::jk_ajp_common.c (2692): request to send (cfusion) Tomcat failed (recoverable), due to the error when sending the request (try = 1)
[19:48:53.402 Wed Jul 23 2014] [13144:6344] [info] jk_open_socket::jk_connect.c (626): connect to 127.0.0.1:8012 failed (errno = 61)
[19:48:53.406 Wed Jul 23 2014] [13144:6344] [info] ajp_connect_to_endpoint::jk_ajp_common.c (1047): failed to open socket (127.0.0.1:8012) (errno = 61)
This leads me to think the problem may be with Tomcat itself, or with the IIS connector, but I don't know where to go from here, if that's the case. We ran all CF10 update via update 12 and did not initially again the IIS connector through wsconfig.exe and restart IIS, but did about a month or two ago.
We have also adjusted the JVM settings a bit:
-XX: MaxPermSize =192 m-> 256 m

and set the size of the JVM min and max heap on 2048 (from 1024). We have 6 GB of RAM on the dedicated server.

Here is the use of the memory of the FMV of the time of the accident (green dots down indicate the restart of the server)
The rise without any garbage collection operation is the same as we saw the last time that the server crashed as well. I have logs of Fusion analytics since the last crash, if everyone would like to see a specific chart.
Any suggestions would be greatly appreciated. Thank you.

Hello

ISAPI log errors seems tomcat at the end of CF does not meet tomcat iis connector. Probably, you will need to make some adjustments or set values server.xml in cfusion\runtime\conf.

What are the values? This blog CF11 entry can help since the details also apply to the CF10.

http://blogs.ColdFusion.com/post.cfm/ColdFusion-11-IIS-connector-tuning

Keep in mind in making adjustments to the corresponding changes server.xml should be applied to workers.properties in config\wsconfig\N.

You could also do a surveillance with jdk free tools such as jconsole to know what values of adjustment would be based on better load type. Java and tomcat monitoring tho is time and effort if response on thread if you want to know more about how to track.

HTH, Carl.

accelerate CFHTTP on CF10 (5000 x slower than under CF8)

Hi, we have freshly installed a large application of CF8 to CF10 and see that CFHTTP performance is * much * slower on the new system.
We've checked by sequentially 50 CFHTTP to the small static active calls running on an Apache Server, measure the response time with gettickcount() and averaging on several tracks.
On the installation of our old (Ubuntu + JRun + CF 8), we are averaging about 1 ms per request.
On our new (Centos + CF 10 standalone) installation, we are averaging 5000 ms per request. Is 5,000 times slower.
Initially, we have assumed that this phenomenon was linked to questions of generation of entropy as the first query in a series of tests would be relatively fast back (a few hundred Ms total) compared to all requests to this term. However, try to change the JVM settings as prescribed is not solved the problem. We tried more far using entropy-generating alternative tools (such as recommended in the blog original post cities in the above thread), but that did not help either one.
Note that we also tested performance outside CF http using wget from a shell script, where we have seen good performance (about 1 ms per request).
Any help on this will be greatly appreciated, thanks!

Update: just tested the CFHTTP CF + http://coldfusion9.blogspot.com/2012/03/custom-cfhttp-tag.html Java-based alternative
It took a few minor bug fixes, but is greatly outperforming CFHTTP from CF 10, with an average of 33 ms by request (always much slower than CF 8, but can be used for most of the cases). However, I always look for a solution restore speed appropriate for CFHTTP!
Post edited by: Jon Obuchowski to add the test with the custom CFHTTP tag

Thank you. I have modified the code and added 2 additional calls to the CFC HTTPRequest and CFX_HTTP5. (I had to also modify the CFC to activate applications without the URL/form vars).

I tried this on ColdFusion 8, 9 & 10 Windows. I shot 50 files of 5 different internal web servers. (WARNING: the specifications are not the same for each server CF.)

Here are the averages.

10 ColdFusion CFHTTP: 9.04 CFC: 18,14 CFX_HTTP5: 1.26

ColdFusion 9 CFHTTP: 8.08 CFC: CFX_HTTP5 158.92: 2.4

ColdFusion 8 CFHTTP: 4.86 CFC: 149.14 CFX_HTTP5: 2.14

I posted a blog in this regard entry and shared the CFML that I used to compare performance:

http://gamesover2600.Tumblr.com/post/88683513184/ColdFusion-CFHttp-versus-httprequestcfc-v ersus

CF10 worked for months, now crashes. How to solve the causes?

We had CF10 vying for months now without any problems. On Tuesday, I begin to hear talk of the CF instance crashing every morning. I restart the service and it works again for the day. Wednesday he broke down again, so I restart the service twice before it will start work. Thursday service died again. all my efforts to restart the service or reboot the server only to allow him to work for minutes at a time. He starts to work quickly as usual, then slows down, then really lag, then the query time-out errors and then blocks the CF instance again. on and off like that all day and still today.
I try to tweak some settings in CF Admin, but nothing seems to give lasting assistance. Server seems to work very well (a Windows Server running IIS). I get the General to work continuously, HTML pages, even during the worst behavior of CF. I checked my SQL Server as well and it works fine, no strange or overwhelming requests. I diverted the traffic to the staging server instead of the Live production server and after a few minutes the server goes down too. It is almost like a slow memory leak, but analysis on CF Admin of the server shows the use of the stable memory, no dramatic increase.
I'm not a guy to server admin super knowledgeable, especially since I am a CF coder, but I am responsible to keep this server running.
How to solve this?
What should I look for?
Any ideas what causes this behavior and all of a sudden starting to do that now?
Thanks for any input in advance.

Thanks to all who have contributed.

To all those who might read this in the future, to respond to some of the suggestions that precede:
- No, there are no infinite loops. The error began last week and there had no code changes recently.
- No, we had not upgraded from an earlier version of the CF.
- This Tomcat IIS connection may have been relevant, it's a good thing for the future.
- I've never had to the isapi_redirect.log
- No, there are no scheduled task on the server.
- No, there was no unusual network traffic to indicate an attack by denial of service or something similar.
We hired some people of domestic support of the server, and two other suggestions came.

1. ColdFusion use memory Java virtual has been questioned as a possible problem

2. a person noted a previous problem with multiple server instances, you try to run both the cause of conflict.

We also noticed that the CPU on the server usage was remaining constant at 50% usage and fortification to 100% several times a minute. This is particularly remarkable, because we were see it during a period of low traffic. We didn't used to use the processor on a CF server, but that seemed odd.

With these suggestions in mind, I voted delete CF10 installation and reinstall a new copy. I saved a copy of the settings Admin, uninstalled, deleted the previous folder and reinstalled. After reconfiguring connections to sources of data and parameters the Server Admin work even better than ever. It's for three days now and the server is still running smooth with full traffic and even faster than before. Still not entirely sure what was the exact cause, but this resettlement was a relatively painless step to take while it was very worth it. Also the CPU usage is now closer to 10% with occasional spikes higher but nothing more by about 80%.

Thanks a lot for helping me solve this problem.

Web services CF10, components not found - the seller Java problem?

Did someone having problems web services in Oracle Java CF10, IIS 7 calls,? We have recently improved, and web services are now back the error message depending on whether the component is not found:
The web service operation has caused an exception of invocation.
The root cause was that: coldfusion.runtime.CfJspPage$ NoSuchTemplateException: could not find the ColdFusion component or interface

The web service must be created in cfadmin? It won't let me. I was understanding the cfinvoke webservice call would create the webservice, but this isn't.

Any information outside of Adobe documentation I've read would be useful.

We have this same file on four different servers. It works on two of the four. The only difference I see in the server settings in addition to the CF and the operating system version, is the vendor/version of Java.

1 CF10, Java Sun, IIS 7.5.7600.16385 - work
2 CF8, Sun Java - work
3 CF10, Java, Oracle - does not
4 CF10, Java, Oracle, IIS 7.5.7600.16385 - does not

Thank you.

It turns out that it was a matter of nomenclature of directory. Directory has a period in it (for example, domaine.org). Once the period has been removed from the directory name, it worked. For some reason, it worked fine in CF8 but not CF10. I hope this helps someone else.

CF10 WSDL question

I am runing the CF9 tracking trouble code; I have question runing the same code in CF10

< cftry >

< cfinvoke

"WebService ="http://www.webservicex.net/CurrencyConvertor.asmx?WSDL"

method = "ConversionRate.

returnVariable = "ToCurrencyValue."

< name cfinvokeargument = "FromCurrency" value = "USD" / >

< name cfinvokeargument = "ToCurrency" value = "EUR" / >

< / cfinvoke >

< cfcatch type = "any" >

< cfset UpdateExchangeRate = "NOINTERNET" >

< / cfcatch >

< / cftry >

I added the following code - and he works in one of my CF10 environment but I get error access WSDL when you try in a new CF10.

< cfset args = {refreshWSDL = true, wsversion = 1} >

< cfset local.service = createObject ("webservice"

, "http://www.webservicex.net/CurrencyConvertor.asmx?WSDL".

args) >

The web site, you access has met an unexpected error.

Please contact the site administrator.

The following information is for the creation of Web sites for debugging purposes.

Error occurred while processing request

Cannot read the WSDL to the URL: http://www.webservicex.NET/CurrencyConvertor.asmx?WSDL.
Error: java.net.SocketException: permission denied: connect. It is recommended to use a web browser to retrieve and review the required WSDL document to ensure that it is correct.

Any idea?

Thank you

Kam

Sounds like a question JVM. Add this additional option:

-Djava .net .preferIPv4Stack = true

your JVM arguments in the ColdFusion Administrator Java/JVM settings and give that a try please. What there is a space before the d so that the new argument is separated. Then restart ColdFusion and try again.

in place upgrade from 32 bit to 64 bit CF10?

We initially installed CF10 32-bit because we had among the existing COM objects needed to run. We finally managed to get rid of these items and I would like to convert to 64-bit. Is there an easy way to convert/upgrade that would preserve all the most / some of my configuration? Or what I need to do a complete uninstall/reinstall?
Windows Server 2008 R2
IIS 7.5
CF10 Standard
Thank you!

I recommend doing a clean uninstall/install. You can make a copy of files * .xml-neo found in \ColdFusion10\cfusion\lib somewhere first. Most of the configuration settings that you set in the CF administrator in these .xml files. After you install the 64-bit version, make a clean backup of .xml files somewhere safe, then copy those you entered your installation of 32-bit. In this way, if ColdFusion restart no problem, you're good to go. If this isn't the case, you have to restore .xml files backups from.

-Carl V.

CF10 Datasource settings

Similar Questions

HTTP 500.19 - internal error in the server error

The requested page cannot be accessed because the configuration data of the page are not valid.

The web service operation has caused an exception of invocation.

Cannot read the WSDL to the URL: http://www.webservicex.NET/CurrencyConvertor.asmx?WSDL.

Maybe you are looking for