Run long-term stability tests
We should verify that the system can run stably for long periods of time (e.g. several days).
This should include various non-standard, but possible, scenarios, like triggers go away for a long period of time and then resume.
#1 Updated by Kurt Biery over 7 years ago
As part of testing V1495 user firmware version 2b, we are currently running a small system at the WH14N teststand. This system has one 1495 and one 1720. With a gate width of 50 usec and four types of triggers (random, pulser , laser , and external [many kHz]), we've successfully run for 45 hours (approx. 4.9 million events).
#2 Updated by Kurt Biery over 7 years ago
I just finished the weekend run at the WH14 teststand. This run was still going at 09:00 this morning. I had 6.67 million events, and the disk file was 237 GB in size. With the code being developed for endRunRobustness (Issue 4120), the file was successfully closed, and it can be read with an offline job.