I have a few questions here. I'm still way down on the learning curve...
During make test I saw this. Anything to worry about? Server seems to run
fine.
modules/stage.......FAILED test 1
Failed 1/2 tests, 50.00% okay
modules/status......Internal Server Error
dubious
Test returned status 9 (wstat 2304, 0x900)
DIED. FAILED tests 8-10
Failed 3/10 tests, 70.00% okay
Now, I'm comparing memory usage for light and heavy-build httpds, and have
two httpds running.
Here's the two servers and children after serving a requests a cgi script
(running under Apache::Registry):
(I edited out the other children with similar memory numbers.)
BTW -- this is SunOS
PID PPID %CPU %MEM VSZ RSS STIME TIME COMMAND
27737 1 0.0 0.6 6392 5344 12:12:08 0:00 httpd-heavy
27739 27737 0.0 0.8 8744 7216 12:12:09 0:00 httpd-heavy
28950 1 0.0 0.2 2200 1384 12:17:11 0:00 httpd-light
28974 28950 0.0 0.1 2208 1032 12:17:12 0:00 httpd-light
I'm trying to look at how much memory might be freed up by using a heavy
and light weight server proxy combination.
If all the heavy children have a RSS around 7M is that _each_? Or is that
number misleading in that RSS includes shared memory?
Does it really matter how big the processes are if they aren't being paged
out?
One last question, which I hope doesn't get lost way down here. Is there a
way to look at a server and determine what percent of the request could be
handled by a light-weight server? The URLs in the access log won't tell me
exactly as users could be running, say, Apache::Registry scripts via
.htaccess.
Thanks,
Bill Moseley
mailto:[EMAIL PROTECTED]
