I have transparent web-proxy with 400MB cache. I have 512/128Kbps DSL line. Hits take 5% from real web traffic.
Is it good idea to set up normal proxy, let say with 40GB cache.
Will I spend much more bandwith and can somebody give me output from real-world working web-proxy (ip web-proxy monitor)
10x in advance
[yogi@Proxy] ip web-proxy> mon
status: running
uptime: 9d3h10m54s
clients: 358
requests: 9292527
hits: 3728876
cache-size: 47185916 kB
received-from-servers: 74438326 kB
sent-to-clients: 84670863 kB
hits-sent-to-clients: 12122049 kB
[yogi@Proxy] ip web-proxy> pri
enabled: yes
src-address: x.x.x.x
port: 3128
hostname: proxy
transparent-proxy: yes
parent-proxy: 0.0.0.0:0
cache-administrator: x@x
max-object-size: 32768 kB
cache-drive: secondary-master
max-cache-size: 51200
status: running
reserved-for-cache: 51200 MB
What happens when cache is full or disk space limit is reached? Does it act like a buffer or just stop caching? Or does the whole box crash??
Trisc
free-memory: 429096 kB
total-memory: 905532 kB
cpu: Pentium
cpu-frequency: 993 MHz
It’s actually a dual 1GHz server. The HDD doesn’t ever reach the maximum specified. Nor does it just get full and crash
. What is troubling and I hope 2.9 will solve the issue. When the proxy has been running for a few weeks without a reboot/flush it starts running at 100% cpu-load. It doesn’t appear to have any adverse affect on traffic or latency, just doesn’t look healthy. I have not tried another server, for this setup does work well.
yogi,
how do you check if the the computer is running at 100% cpu-load?
regards
[yogi@Proxy] > sys reso pri
uptime: 4d14h18m33s
free-memory: 358948 kB
total-memory: 905532 kB
cpu: Pentium
cpu-frequency: 993 MHz
cpu-load: 97
free-hdd-space: 28548 kB
total-hdd-space: 60508 kB
write-sect-since-reboot: 1100
write-sect-total: 1214388
Today is a good day only 97% usage
![]()
You can view the system resources via winbox as well