Alex Rousskov wrote:
> In my formula, I have client_side.max_agent_load as a
> multiplier. Your file has client_side.max_host_load. There is a three
> orders of magnitude difference. :)
ah. missed that subtle point. Thanks
> FD_SETSIZE is an OS constant that Polygraph obeys on Linux. Rlimit is
> obtained using getrlimit(2) system call, I think. Perhaps the
> following URLs will help you to configure your Linux box.
> http://www.squid-cache.org/Doc/FAQ/FAQ-11.html#ss11.4
> http://squid.sourceforge.net/hno/linux-lfd.html
have visited the squid site before. I;ll visit it again. thanks
> > I get "server_side.hosts or its components not declared"
> > when I try to run the client.
>
> It's TheBench.server_side.hosts. I omitted a few prefixes to save
> some typing.
got it. thanks.
OK have nolb working with a client box and 3 of my 4 servers.
3 of the servers appear to be executing the different phases
just fine.
I don't know what the problem with the 4th server is.
The 4 servers were setup at different times but are approximately
similar in software (kernel, compiler, glibc versions etc)
and identical in hardware.
I can polyprobe between the client and the 4th server and
get the expected throughput. The server when running as a nolb
server is listening on 8080 (my polygraph port) by looking
with netstat -an and I can connect to it with telnet from the
client. The telnet connection hangs (I can't send "quit" ^C,
^D etc). When I fire up the nolb client, it has hundreds
of connections in the SYN state (by netstat -an) to the
4th server, while the connections to the
other 3 servers are in more normal states (ESTABLISHED etc).
At the bottom is the output of nolb startup on the 4th server
if you can give me any ideas as to what's happening.
On a slightly different topic, I'm trying to see if my L4 box
fits your test.
In the setup I'm running with nolb-14-4.pg, I have one client
which knows about 4 servers, all on different IPs, and on the
same network.
My L4 box expects the clients to access the servers through the
IP on the outside of the L4 box (which I call the VIP). The
servers behind the L4 box have their eth0 devices on private
addresses and are either listening on those addresses (different
for each server) or on the VIP (which is the same for all servers,
and is bound to a non arping device eg lo:0). In the first
case the requests are reverse masqueraded into the servers and
in the 2nd case the requests are routed to the VIP on the server
through each server's eth0 IP and the replies come back directly
to the client.
No matter which way I do it, the client sees only 1 IP no matter
how many servers are behind the L4 box and all the servers are on
a different network to the client (client comes from a public
IP, servers are on private IPs, L4 box has the public IP=VIP, that
a single server would have if there was only a single box as
server).
I know little about other L4 boxes. To me if 4 servers each have different
IP's and a client has a table of 4 IP's to connect to, then the
only thing a box in the middle can do is regular routing.
Is the setup I have testable in the switch-off?
Thanks Joe
-----------------------------------------------
#startup for server that doesn't work in nolb test.
sneezy:/snow/src/polygraph/polygraph-2.6.0# ./start_nolb_server.sh
000.04| Content distribution on server SrvLB-l4-4-Srv:
content planned% likely% error%
image 65.00 65.46 0.70
HTML 15.00 14.80 -1.31
download 0.50 0.49 -1.31
other 19.50 19.25 -1.31
expected average cachability: 80.01%
expected average object size: 10771.11Bytes
000.04| fyi: no interface name for address mask for Server agents; will not attempt to create agent addresses
000.14| created 1 out of 1 agents so far
000.14| Command: /usr/local/bin/polysrv --config /snow/src/polygraph/polygraph-2.6.0/workloads/my-nolb-14-4.pg --verb_lvl 10
--log /snow/src/polygraph/polygraph-2.6.0/sneezy.log
000.14| Configuration:
version: 2.6.0
host_type: i586-pc-linux-gnu
verb_lvl: 10
dump: <none>
dump_size: 1.000KB
notify: <none>
label: <none>
fd_limit: 983
config: /snow/src/polygraph/polygraph-2.6.0/workloads/my-nolb-14-4.pg
cfg_dirs:
console: -
log: /snow/src/polygraph/polygraph-2.6.0/sneezy.log
log_buf_size: 64.000KB
sample_log: /snow/src/polygraph/polygraph-2.6.0/sneezy.log
sample_log_buf_size:64.000KB
stats_cycle: 5.00sec
sync_phases: on
file_scan: poll
priority_sched: 5
fake_hosts:
idle_tout: 5.00min
rng_seed: 1
unique_world: on
ign_urls: off
000.14| Phases:
phase load_beg load_end rec_beg rec_end smsg_beg smsg_end goal flags
warm 0.01 0.01 0.09 0.09 0.10 0.10 3.00min
ramp 0.01 1.00 0.09 1.00 0.10 1.00 30.00min
plat 1.00 1.00 1.00 1.00 1.00 1.00 1.50hour
exit 1.00 0.01 1.00 1.00 1.00 1.00 30.00min
000.14| StatsSamples:
static stats samples: 0
dynamic stats samples: 0
000.14| FDs: 1024 out of 1024 FDs can be used; safeguard limit: 983
000.14| resource usage:
CPU Usage: 470msec sys + 12.73sec user = 13.20sec
Page faults with physical i/o: 309
000.14| group-id: 04c643d0.602400bb:00000002 pid: 187
000.14| current time: 992448032.912910 or Wed, 13 Jun 2001 16:00:32 GMT
000.14| fyi: PGL configuration stored (17681bytes)
000.14| fyi: current state (1) stored
000.15| starting 1 HTTP agents...
000.15| starting SrvLB-l4-4-Srv[1 / 04c643d0.602400bb:00000004] on 192.168.1.11:8080
000.23| i-warm 0 0.00 -1 -1.00 0 1
000.31| i-warm 0 0.00 -1 -1.00 0 1
000.40| i-warm 0 0.00 -1 -1.00 0 1
000.48| i-warm 0 0.00 -1 -1.00 0 1
000.56| i-warm 0 0.00 -1 -1.00 0 1
000.65| i-warm 0 0.00 -1 -1.00 0 1
000.73| i-warm 0 0.00 -1 -1.00 0 1
000.81| i-warm 0 0.00 -1 -1.00 0 1
000.90| i-warm 0 0.00 -1 -1.00 0 1
000.98| i-warm 0 0.00 -1 -1.00 0 1
001.06| i-warm 0 0.00 -1 -1.00 0 1
001.15| i-warm 0 0.00 -1 -1.00 0 1
001.23| i-warm 0 0.00 -1 -1.00 0 1
001.31| i-warm 0 0.00 -1 -1.00 0 1
001.40| i-warm 0 0.00 -1 -1.00 0 1
001.48| i-warm 0 0.00 -1 -1.00 0 1
001.57| i-warm 0 0.00 -1 -1.00 0 1
001.65| i-warm 0 0.00 -1 -1.00 0 1
001.73| i-warm 0 0.00 -1 -1.00 0 1
001.82| i-warm 0 0.00 -1 -1.00 0 1
This archive was generated by hypermail 2b29 : Tue Jul 10 2001 - 12:00:19 MDT