I think there must be something else going on with your server. The macro loads nominally slow things down, but shouldn't cause the problems you report unless there's some looping issue with the macros or problems that you're using. We have a testing lab with about 30 machines, which in peak times are hitting the server regularly, and we don't see wait times that are even seconds long (we do have a much bigger server, of course, but I think the observation is still valid).
Can you do some investigation on the server to see if there is some additional evidence of what's causing the slow down? e.g., top, uptime, and vmstat might give some information about where things are bogging down.
Gavin
I'm assuming that this is from a time when your server wasn't returning results from a gateway test. I don't see anything odd there, though I'm a little surprised that there are swapping events (for comparison, my server shows the following, albeit in an unloaded state:
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu----- r b swpd free buff cache si so bi bo in cs us sy id wa st 0 0 313632 5243360 766216 16049680 0 0 0 2 0 0 0 0 100 0 0
but I don't know if that's different enough that I would worry about it.)
I don't know of any combinations of the standard macros that would cause looping issues. The only cases where I've seen a problem is when I've gotten a problem with an infinite loop, and that generates timeouts from the server.
If you load the problems that you have in the test into a homework set and have multiple sessions submit them, do you get the same behavior? Can you shorten the gateway/quiz assignment to only one problem and elicit the same response?
Gavin
max_connections=5000to the
/etc/my.cnf file. That said, 5000 is certainly overkill, and we had the trouble many years ago (but have been setting max_connections ever since).
Gavin