MichalKleiner

just saw your comment on my throughput question yesterday.. so it should persist in some ways… however, still worth checking that SS can create a cache folder with write access to it.. and that such folder persists..

harvs1789uk

i have a silverstripe-cache folder in my web root and it is being written to (so presumably also read from)

harvs1789uk

I am getting close to all out of ideas tbh, my current plan is to spin up the site on a VM with comparable CPU/RAM etc and (hopefully) prove to myself that it is not just that SS / my implementation is really slow generally.

Pretty confident there is something funny going on with the AppService set up, but can't for the life of me work out what

MichalKleiner

@harvs1789uk probably not the best channel, but I can’t think of any other better one for this question. Did you look into what @brett.tasker suggested re: the caching and checked whether the infrastructure is ephemeral or persist in some ways between requests? If a new computing unit is spun up every time, the boot time will be slow.

harvs1789uk

I am seriously struggling to identify what about the AppService would cause SS to take ages (7 seconds or so) to start processing requests (after whuch, the actual processing of requests seems pretty quick/normal)

harvs1789uk

Seeing some seriously slow HTTP requests for anything which goes via SilverStripe on an Azure AppService for Linux based deployment…other HTTP requests are fine.

I’ve installed the SS debug bar and it reports that 80-90% of these slow SS requests are ‘framework boot’ (which seems to be the time between the reported $_SERVER['REQUEST_TIME_FLOAT'] and the time in micro seconds that debug bar ties into the HTTPMiddleware process)

harvs1789uk

Would this be the best forum (well channel) to discuss SS hosting requirements/issues?

Chris

Hi there 👋. Is it possible to trigger dev/build on a CI deployment when in live mode?