Actually it's working perfectly. A request that comes in will be handled by
the first available redirector. If redirector 1 is busy handling a request
then redirector 2 will be checked, and so on. I suppose it would be possible
to do some type of round-robin system, but the current implementation seems
to make the most sense. It gives you some idea of how many concurrent users
are requesting objects. The low avg service time indicates that your
redirector does not do much ( it handles requests quickly ). Our redirector
does about 4msec avg and we do not see quite the ratio that you are getting.
Even though redir 1 serves significantly more ( about 10-15 times ) requests
than redir 2.
James Porter wrote:
>
> I get the following from the cache manager. There seems to be
> disproportionate load to Redirector 1 as opposed to the rest. Should it be
> more average compared to the rest?
>
> Redirector Statistics:
> requests: 1029065
> replies: 1029064
> queue length: 0
> avg service time: 0 msec
> number of redirectors: 16
> use histogram:
> redirector #1: 1019131 (1034090 rewrites)
> redirector #2: 7544 (8276 rewrites)
> redirector #3: 1299 (1723 rewrites)
> redirector #4: 561 (816 rewrites)
> redirector #5: 300 (435 rewrites)
> redirector #6: 155 (218 rewrites)
> redirector #7: 39 (53 rewrites)
> redirector #8: 11 (13 rewrites)
> redirector #9: 5 (5 rewrites)
> redirector #10: 4 (5 rewrites)
> redirector #11: 3 (3 rewrites)
> redirector #12: 3 (3 rewrites)
> redirector #13: 3 (3 rewrites)
> redirector #14: 3 (3 rewrites)
> redirector #15: 2 (2 rewrites)
> redirector #16: 2 (2 rewrites)
-- ------------------------------------------------------------------------------- Rodney D. Holm rodneyh@apexxtech.com Apexx Technology, Inc. http://www.apexxtech.com -------------------------------------------------------------------------------Received on Wed Nov 18 1998 - 13:14:01 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:43:09 MST