New on DevNet: Getting Started With FusionReactor

9 views
Skip to first unread message

Gianna Benson [FusionReactor Team]

unread,
Nov 17, 2009, 7:27:11 AM11/17/09
to FusionReactor
Getting Started With FusionReactor by Ajas Mohammed
=============================================

New to FusionReactor? Use this guide to help you Get Started.

View DevNet article: http://www.fusion-reactor.com/community/developers/autodev_detailed.cfm?article=FRS-238

charlie arehart

unread,
Nov 21, 2009, 1:47:19 PM11/21/09
to fusion...@googlegroups.com
Just wanted to share a little reminder to folks (which I tripped over
myself), when implementing a new server to be monitored.

If you implement a new instance of CF (such as CF9, or some other CFML
engine) and you had previously enabled the special undocumented feature that
can help prevent some instances of requests hanging up on
java.net.SocketOutputStream.socketWrite0, don't forget to enable that in the
newly installed instance. You have to remember to modify the reactor.conf
(in the [FusionReactor]\instance\instancename\conf) to add in the line:

crashprotection.buffering.active=true

For more on this feature, see a reply from Darren in this thread:

http://groups.google.com/group/fusionreactor/browse_thread/thread/c04e97d5b4
8b3d34/2e57c5fbab5305b1?#2e57c5fbab5305b1

/charlie

AMusella

unread,
Nov 21, 2009, 3:00:20 PM11/21/09
to FusionReactor
Does that still apply to Fusion Reactor version 3.5.1?
I am having the problem of my CF 7,0,2,142559 server randomly
stopping. Happens about once ever other week. When it does, FR
reports a lot of running requests, and when I try to kill them, they
won't die. I reboot and it works fine for a while..
Perhaps this " crashprotection.buffering.active=true" would help?
There is no monitoring feature in FR that would restart the Coldfusion
service when CF stops responding, is there?




On Nov 21, 1:47 pm, "charlie arehart" <charlie_li...@carehart.org>
wrote:
> http://groups.google.com/group/fusionreactor/browse_thread/thread/c04...
> 8b3d34/2e57c5fbab5305b1?#2e57c5fbab5305b1
>
> /charlie

Ajas Mohammed

unread,
Nov 21, 2009, 4:38:33 PM11/21/09
to fusion...@googlegroups.com
Hi Charlie,

The link you sent about Darren's reply is broken. Can you resend it again?

Thanks,

Ajas Mohammed.



--

You received this message because you are subscribed to the Google Groups "FusionReactor" group.
To post to this group, send email to fusion...@googlegroups.com.
To unsubscribe from this group, send email to fusionreacto...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/fusionreactor?hl=.



charlie arehart

unread,
Nov 21, 2009, 11:21:10 PM11/21/09
to fusion...@googlegroups.com
Well, I am assuming it still applies. I actually meant to ask for the
Intergral folks to confirm that (since indeed if one moves to CF9 they do
necessarily need to update to 3.5/3.5.1.)

As for your issue, it may or may not be related. The key is to see what the
stack trace shows for those requests that are hanging--either by clicking
the stack trace button on the running requests page, or viewing the stack
trace in the crash protection (CP) email notification.

The requests could be hung up in any of several other things not related to
what this fix addresses, such as talking to a database, doing a CFHTTP,
invoking a web service, etc. You can't interrupt a request doing that (not
with FR, not with CF or SeeFusion, nor the CF8 Server monitor).

Finally, as for a feature to restart CF when it stops responding, that
depends. The Enterprise edition of FR does offer the ability to have one
instance monitor others, and if it detect no reply, it can kick off a script
for that non-responding server. But that non-responsiveness is not measured
by whether the CF server can process CF template requests, but simply
whether the FR filter implemented within that CF server can respond to the
Enterprise client.

That raises an interesting question for the FR folks: it might be something
they could consider adding, since the FR CP mechanism does watch how many
requests are running, and whether they are running long. It could be useful
for there to be a way to say that if more than x requests are running for
more than y seconds, trigger a restart.

Hope that helps.

/charlie

charlie arehart

unread,
Nov 21, 2009, 11:34:37 PM11/21/09
to fusion...@googlegroups.com

Sigh. :-( With all due respect, Ajas (and you know I mean that), I have to say it mystifies me how people get tripped up by this.  The problem isn’t with the URL: it’s that many email clients split long lines, and that happens to some long URLs.

 

So sure, you can’t just click the link as is.  But all one needs to do is grab both parts and put them on a URL, or start a reply to the message (even with no intention of sending it), and then you can edit the email to delete the line breaks to end up with the correct single line.

I suppose one may argue that someone who sends a long URL should consider a URL shortening service. Sure, we could, but like I said, it just mystifies me that we should have to worry about this. The solution to fixing long broken lines seems simple enough. What I don’t understand is why people seem to totally miss the split line. It seems pretty apparent to me.

Again, I don’t mean any disrespect, Ajas. More than anything, I really share this a) for others who see it as a problem and b) to ask if there’s something I’m missing in not understanding the problem.

Thanks.

/charlie

PS I actually blogged about this problem and another possible solution, putting brackets around long urls, at http://tipicalcharlie.blog-city.com/avoid_url_break_in_email.htm. Let’s see if it works:

<http://groups.google.com/group/fusionreactor/browse_thread/thread/c04e97d5b48b3d34/2e57c5fbab5305b1?#2e57c5fbab5305b1>

 

Ajas Mohammed.

>
> /charlie

--

You received this message because you are subscribed to the Google Groups "FusionReactor" group.
To post to this group, send email to fusion...@googlegroups.com.
To unsubscribe from this group, send email to fusionreacto...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/fusionreactor?hl=.

snake

unread,
Nov 22, 2009, 6:43:09 AM11/22/09
to fusion...@googlegroups.com
Hi Chaps,

I seem to have a problem by my cfadmin is getting aborted by FR on long
running pages. I have the feeling this problem is random is does not occur
all the time as sandbox actions always take a long time to run as it's a
busy server and do not always timeout.

The path I have in the exclusion list is

Request: regular expression
Mydomain.com/cfide/(.*)
Hostname: check
Params: ignore
Exclude from: all

I have also tried
Mydomain.com:8088/cfide/(.*)

/Russ


Al Musella, DPM

unread,
Nov 22, 2009, 10:49:56 AM11/22/09
to fusion...@googlegroups.com
Thanks. I will give it a try.
By the way - it might be nice to add a function to the Enterprise
dashboard that monitors a page on the server, and if the server
stops responding for a period of time, trigger a restart of the cold
fusion service, and if it still doesn't work in 2 minutes, a reboot
of the server, through Fusion reactor.

I have my home computer constantly monitoring my web server, but when
the server stops responding, there is no way for me to restart
the cold fusion service remotely automatically (is there?).. I just
have the monitor send me a text telling me to check the server.

For bonus points - to get fancy and add a few cool features to the
list.. my monitoring script checks 3 things on 1 page:
1. That my home page will load.
2. That my home page hasn't changed. I keep a copy of it in a
database table, and every time it changes, I add the entire page to
the database so I can always get the most recent good copy back (It
doesn't change often), and it shows me the changes so I know how
serious it is. I ignore the banner ad section.. This way if anyone
hacks me, I know immediately. I get a text whenever it changes.. when
I change it myself, I get the text which lets me know the system works.
3. I set up a database called Customers with a bunch of fake data
that never changes. My home page always checks that the email
address, first and last name of one of the records hasn't changed,
and alerts me if it did.


Darren Pywell

unread,
Nov 24, 2009, 11:47:05 AM11/24/09
to FusionReactor
Hi,

I posted something earlier but it seems like the post didn't appear.

You would have to set the crashprotection.buffering.active=true option
on new instances, even in 3.5.1.

FusionReactor Enterprise Edition can run a script/program when a
server goes down (or comes up!). The script can be used to restart the
CF service or the webserver/database etc.

Check out the following links for more details:

http://www.fusion-reactor.com/fr/featurefocus/monitoredServerScripts.cfm
http://www.fusion-reactor.com/fr/helpdocs/fusionReactor_enterprise_scripting.pdf
http://groups.google.com/group/fusionreactor/browse_thread/thread/c47882963b4c012b

Cheers,
Darren

On Nov 22, 4:49 pm, "Al Musella, DPM" <muse...@virtualtrials.com>
wrote:
Reply all
Reply to author
Forward
0 new messages