We have a customer that are facing long times for printing (from time to time), but it's hard to document the frequency and times. Users claim it might take up to 10min from clicking print until it starts printing.
The Print is spooled by a dedicated print server (TCP:445) and future sent to printers (TCP:9100). There are too many printers to be manually added one by one and there is a limitation of 5000 combinations per software definition line so we cannot define print server with random port numbers either.
Our goal is to both look at the time from user to print server as well as print server to printer. With this we can document if and how much time is caused by network, server etc.
How can DCRUM help us here?
Answer by Ryan C. ·
If the application is built using .NET and/or java, dynaTrace (a sister product to DCRUM) would be able to show you the transaction execution path and timings from the "print click" in the client/browser, across the network, into the web server, into the .NET/java app server(s), and down to the Print() method in the code.
It's basically sees everything from the user perspective (click) and the application perspective (print method) that executes.
Answer by Adam P. ·
I imagine this may be possible to create two Software Serivces (client->server and sever->printer), but I'm not sure if I correctly understand your the problem. I'm not sure what port you describe as random. Print server's client port or printer's one?
Base on this information we would be able to decide if monitoring communication between server and printers is possible.
Doing this between users and print server should be possible just by creating Software Service with print server's IP and 445 port. Analyzer "generic with transactions" should be used.