Show Posts
1
Web Proxy Filtering and Caching / Squid Proxy "503 Service Unavailable" During Load Testing
« on: September 06, 2024, 08:12:48 pm »
Hi ALL,
It's my first post here !
I use Google cloud and in my case egress traffic is allowed via squid proxy . Here my API details that's makes egress calls
API Usage Details: All requests are POST requests with JSON payloads.Responses are also Json each around 400 KB in size.Traffic Volume: The API handles around 300,000 requests per day.Concurrency: During load testing, the peak load is 15-20 concurrent users. I am using cloud run for my API.
When I perform load testing with 12 peak concurrent users, everything runs smoothly for about 5-6 minutes. However, after this period, Response time is slow and Squid begins to return " I/O error Unable to tunnel through proxy. Proxy returns "HTTP/1.1 503 Service Unavailable". This happens intermittently but frequently after 5-6 minutes and response time slows down as well
I have tried to disable caching initially since my data is dynamic so caching is no use.
I am new to GCP and squid . Any help would be appreciated.
Squid conf :
acl localnet src xxx.xx.xx.xx/xx
acl Safe_ports port 80 # http
acl Safe_ports port 443 # https
http_access deny !Safe_ports
http_access allow localhost manager
http_access deny manager
acl allowsites dstdomain "/etc/squid/approvedsites.lst"
acl exempt_ips src "/etc/squid/approved_ips.lst"
http_access allow allowsites exempt_ips
#http_access allow all
http_access deny all
http_port 3128
coredump_dir /var/spool/squid
# refresh_pattern ^ftp: 1440 20% 10080
#refresh_pattern ^gopher: #1440 0% 1440
#refresh_pattern -i (/cgi-bin/|\?) 0 0% 0
#refresh_pattern . 0 20% 4320
I tried only disable caching and I am reading and learning what more to do here. I increased the vm for squid to 4 cpu and 8 Gb. How many I for my usecase better the squid settings
It's my first post here !
I use Google cloud and in my case egress traffic is allowed via squid proxy . Here my API details that's makes egress calls
API Usage Details: All requests are POST requests with JSON payloads.Responses are also Json each around 400 KB in size.Traffic Volume: The API handles around 300,000 requests per day.Concurrency: During load testing, the peak load is 15-20 concurrent users. I am using cloud run for my API.
When I perform load testing with 12 peak concurrent users, everything runs smoothly for about 5-6 minutes. However, after this period, Response time is slow and Squid begins to return " I/O error Unable to tunnel through proxy. Proxy returns "HTTP/1.1 503 Service Unavailable". This happens intermittently but frequently after 5-6 minutes and response time slows down as well
I have tried to disable caching initially since my data is dynamic so caching is no use.
I am new to GCP and squid . Any help would be appreciated.
Squid conf :
acl localnet src xxx.xx.xx.xx/xx
acl Safe_ports port 80 # http
acl Safe_ports port 443 # https
http_access deny !Safe_ports
http_access allow localhost manager
http_access deny manager
acl allowsites dstdomain "/etc/squid/approvedsites.lst"
acl exempt_ips src "/etc/squid/approved_ips.lst"
http_access allow allowsites exempt_ips
#http_access allow all
http_access deny all
http_port 3128
coredump_dir /var/spool/squid
# refresh_pattern ^ftp: 1440 20% 10080
#refresh_pattern ^gopher: #1440 0% 1440
#refresh_pattern -i (/cgi-bin/|\?) 0 0% 0
#refresh_pattern . 0 20% 4320
I tried only disable caching and I am reading and learning what more to do here. I increased the vm for squid to 4 cpu and 8 Gb. How many I for my usecase better the squid settings