Differences

This shows you the differences between two versions of the page.

Link to this comparison view

irc:1454713200 [2017/05/27 13:44] (current)
Line 1: Line 1:
 +[03:31:09] <​saltuk>​ is there anyone ? i have question
 +
 +[03:37:30] <​jtruelove_>​ what's your question, not sure I can help but worth asking
 +
 +[03:39:40] <​saltuk>​ vertx is one threaded right ? ..  so .. if i have one database connection for single application ( create connection before listening) ​  ​.. ​  will it cause a performance ​ issue ?
 +
 +[05:37:23] <​jtruelove_>​ not technically,​ if you you start it up with only 1 instance then you'd have 1 event loop
 +
 +[05:37:38] <​jtruelove_>​ for the verticle you started
 +
 +[05:38:07] <​jtruelove_>​ but say you have 16 cores you probably would want to start 16 instances of your verticle, one event loop per core
 +
 +[05:39:07] <​jtruelove_>​ now when it comes to your DB connection pooling that is different, the JDBC interface is wrapped and all callback based (I haven'​t used it myself) but in standard JDBC fashion you can set you connection pool size
 +
 +[05:40:53] <​jtruelove_>​ so you could have each verticle have it's own jdbc DB session or share one between them all (probably what you want), take a look here http://​vertx.io/​docs/​vertx-jdbc-client/​java/​
 +
 +[05:41:05] <​jtruelove_>​ you'll notice a max_pool size option in that manual
 +
 +[05:41:26] <​jtruelove_>​ that is what gates how active simultaneous connections you can have to the DB
 +
 +[05:47:39] <​jtruelove_>​ saltuk ^ hope this helps
 +
 +[05:50:09] <​saltuk>​ thank you very much  for info
 +
 +[20:29:42] <​Lecherito>​ Is it possible to get the patch which the router was registered to when handling a request?
 +
 +[20:30:09] <​Lecherito>​ Like getting /users/:id when the actual request is /users/1
 +
 +[22:18:40] *** ChanServ sets mode: +o temporal_