Re: [ANN] overload-middleware 0.1.1

On 16/11/13 21:22, Craig wrote: Very interesting. I have a similar requirement, but not in serving web requests. I haven't looked under the covers of your module, but wonder if it could be decoupled from web/ring? Craig I've just pushed 0.2.0-SNAPSHOT to Clojars and Github. This adds overl

Re: [ANN] overload-middleware 0.1.1

On 17/11/13 14:24, James Reeves wrote: I've added this library to clojure-toolbox.com . I can see this library being very useful for certain systems. Thanks! I hadn't heard of the Clojure Toolbox before, but there's a lot of interesting things in there that I now

Re: [ANN] overload-middleware 0.1.1

Hi Rob, Nice! Thanks for releasing this. I’m not sure when I’ll use this but I can pretty much guarantee that I will give it a go sooner or later. I’ve written a few of these in the past, it’s good to see someone make a library out of it. BTW, things like AngularJS and EmberJS should/could know

Re: [ANN] overload-middleware 0.1.1

On 17 November 2013 13:04, Clinton Dreisbach wrote: > People, you are not going to win a fight with a Level 65 Troll Wizard. > Back away slowly. > > Rob, this is a cool library: thanks for writing it. > Point taken. I've added this library to clojure-toolbox.com. I can see this library being ve

Re: [ANN] overload-middleware 0.1.1

On Sun, Nov 17, 2013 at 8:22 AM, Christian Romney wrote: > Maybe you should re-read this whole thread. Clinton is just pointing out > how impolite it is, particularly as the first response to an ANN post, to > poopoo all over someone else's work, > Please provide a citation to back up your claim

Re: [ANN] overload-middleware 0.1.1

No, no, I can assure you, it's not an insult, you are well beyond. :))) --> toward the edge of the universe Luc P > Is that intended as some sort of an insult aimed at me? I'm just pointing > out that "the web" != "everything that might conceivably be done with the > HTTP protocol"

Re: [ANN] overload-middleware 0.1.1

Maybe you should re-read this whole thread. Clinton is just pointing out how impolite it is, particularly as the first response to an ANN post, to poopoo all over someone else's work, shared freely and graciously with the rest of the community, simply because *you* can't see a use for it. Clearl

Re: [ANN] overload-middleware 0.1.1

Is that intended as some sort of an insult aimed at me? I'm just pointing out that "the web" != "everything that might conceivably be done with the HTTP protocol". "The web" is that thing you browse with Firefox, more or less by definition (even if sometimes tools like curl and wget are used as sho

Re: [ANN] overload-middleware 0.1.1

People, you are not going to win a fight with a Level 65 Troll Wizard. Back away slowly. Rob, this is a cool library: thanks for writing it. -- Clinton Dreisbach On Sun, Nov 17, 2013 at 7:53 AM, Cedric Greevey wrote: > So, when people here are talking about the web, they might not be talking

Re: [ANN] overload-middleware 0.1.1

So, when people here are talking about the web, they might not be talking about the web. Erm, okay, I guess ... On Sun, Nov 17, 2013 at 7:11 AM, James Reeves wrote: > On 17 November 2013 05:25, Cedric Greevey wrote: > >> On Sat, Nov 16, 2013 at 9:35 PM, James Reeves wrote: >> >>> On 17 Novembe

Re: [ANN] overload-middleware 0.1.1

On 17 November 2013 05:25, Cedric Greevey wrote: > On Sat, Nov 16, 2013 at 9:35 PM, James Reeves wrote: > >> On 17 November 2013 01:52, Cedric Greevey wrote: >> >>> The distribution will be narrow and peak at around 1 second, though, >>> which may not be what you want. Of course, the OP has sinc

Re: [ANN] overload-middleware 0.1.1

Even for a web site used exclusively by people, "fail fast" when overloaded can be good. You would rather give an error (one hopes a "gee, so sorry, we having some trouble right now" page) to some users rather than be extremely slow for many users. What do I do when I get a slow web page? I hit

Re: [ANN] overload-middleware 0.1.1

On Sat, Nov 16, 2013 at 9:35 PM, James Reeves wrote: > On 17 November 2013 01:52, Cedric Greevey wrote: > >> The distribution will be narrow and peak at around 1 second, though, >> which may not be what you want. Of course, the OP has since indicated that >> he meant non-web uses of HTTP rather

Re: [ANN] overload-middleware 0.1.1

FWIW, Netflix uses a sorta similar approach but the overload detection lives on the client-side since different clients may have varying definitions of "slow", may want finer grained control of fallback behavior, etc: http://techblog.netflix.com/2012/02/fault-tolerance-in-high-volume.html There

Re: [ANN] overload-middleware 0.1.1

On 17 November 2013 01:52, Cedric Greevey wrote: > The distribution will be narrow and peak at around 1 second, though, which > may not be what you want. Of course, the OP has since indicated that he > meant non-web uses of HTTP rather than serving web sites... > Web services are generally consi

Re: [ANN] overload-middleware 0.1.1

The distribution will be narrow and peak at around 1 second, though, which may not be what you want. Of course, the OP has since indicated that he meant non-web uses of HTTP rather than serving web sites... On Sat, Nov 16, 2013 at 7:10 PM, Colin Fleming wrote: > Web browsers don't "pick a rando

Re: [ANN] overload-middleware 0.1.1

> > Web browsers don't "pick a random sleep time before trying again", though; > they display a 500 error page and the user promptly clicks "reload" while > making an exasperated sigh. > That sounds like "picking a random sleep time before trying again" to me :-) > If the client is something ot

Re: [ANN] overload-middleware 0.1.1

On 16/11/13 21:22, Craig wrote: Very interesting. I have a similar requirement, but not in serving web requests. I haven't looked under the covers of your module, but wonder if it could be decoupled from web/ring? Craig Yes - I don't think that would be at all difficult. The only HTTP-sp

Re: [ANN] overload-middleware 0.1.1

On 16 November 2013 22:49, Cedric Greevey wrote: > On Sat, Nov 16, 2013 at 5:06 PM, James Reeves wrote: > >> Web servers are often used to serve information to clients other than web >> browsers. >> > > [citation needed] > Just google "web services". This is a *really* common use-case for HTTP.

Re: [ANN] overload-middleware 0.1.1

On 16/11/13 22:06, James Reeves wrote: I can see overload-middleware being useful in an internal network, where clients are submitting jobs to a central server. The clients can be programmed to wait a random length of time before retrying if the server is overloaded. I suspect that a better

Re: [ANN] overload-middleware 0.1.1

I've use many services within apps that are json or xml results served over HTTP. This is a common way to integrate third party data or functionality into an app (especailly webapps or mobile apps). On Saturday, November 16, 2013 2:49:46 PM UTC-8, Cedric Greevey wrote: > > On Sat, Nov 16, 2013 a

Re: [ANN] overload-middleware 0.1.1

On Sat, Nov 16, 2013 at 5:06 PM, James Reeves wrote: > Web servers are often used to serve information to clients other than web > browsers. > [citation needed] -- -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email t

Re: [ANN] overload-middleware 0.1.1

On 16 November 2013 21:45, Cedric Greevey wrote: > Web browsers don't "pick a random sleep time before trying again", though; > they display a 500 error page and the user promptly clicks "reload" while > making an exasperated sigh. > > If the client is something other than a web browser, then we'

Re: [ANN] overload-middleware 0.1.1

Web browsers don't "pick a random sleep time before trying again", though; they display a 500 error page and the user promptly clicks "reload" while making an exasperated sigh. If the client is something other than a web browser, then we're no longer talking about the web, but some other thing, li

Re: [ANN] overload-middleware 0.1.1

It may be useful for certain web services. If the server gets overloaded by a temporary spike, the clients could pick a random sleep time before trying again. - James On 16 November 2013 21:15, Cedric Greevey wrote: > I can think of very few web apps where this would be a desirable approach. >

Re: [ANN] overload-middleware 0.1.1

Very interesting. I have a similar requirement, but not in serving web requests. I haven't looked under the covers of your module, but wonder if it could be decoupled from web/ring? Craig -- -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post

Re: [ANN] overload-middleware 0.1.1

I can think of very few web apps where this would be a desirable approach. A user getting a spurious error in response to a URL that they *know* is valid is just going to hammer on the "reload" button until they get a correct response from the server. So the server will end up even more congested t

[ANN] overload-middleware 0.1.1

Hi all, I've just published the first working version of a Ring middleware that some of you might find useful. It's designed for web apps where, if you're overloaded, it's better to serve some requests quickly and fail the others than to try and serve all the requests and do it slowly. (My ba