Apache mode “prefork mpm” causes ImageMaster client to stall
If you use Apache httpd in your network environment, where you route ImageMaster requests through, you must not use the Apache “prefork” multi-processing module!
This will repeatedly block the client for end users, giving the impression that the client got stuck. Instead, consider one of these solutions:
-
Preferably, configure “event mpm” [Apache MPM Event].
-
Try to tweak the H2MinWorkers [Apache H2MinWorkers] (on your own risk, however).
Rationale
ImageMaster uses the HTTP long polling strategy to handle background requests that take up to 60 seconds. According to [Apache MPM Config]in prefork mode the internal module “mod_http2” will only process one request at a time per connection and other requests will stall. Prefork is an outdated strategy that is only chosen if you run processing engines that are not prepared for multi-threading.