Hi, I see that there have been replies to this post but I am not receiving them to this email account? Do I need to something so that I get these replies and not have to go look at the mail-archive.
As to the replies I would be able to help test 3.1.10+ for any items or improvements for CPU utilization. I have just in the past week started working with 3.2 and I can also test and maybe work on some items. I do have 3.1.10 running in lower volume envoriment at 630+ RPS and I will post the data tomorrow. Regards, Jack ----- Original Message ---- From: Jack Quinlin <[email protected]> To: [email protected] Sent: Fri, January 7, 2011 9:09:20 AM Subject: Re: Follow up on Benchmarking results for 3.1.10 vs 2.7.9 Hi, I have some more info on the 3.1.10 line I had to remove it from service and roll back to 2.7.9 because of performance issue. I noticed that during some peak usage CPU would spike to 100% then come down to the high 70s but this caused numerous issues and errors. I reset the 3.1.10 instance on Dec 06, 2011 and let it run until it hit this threshold again and it took about a day I have 3.1.10 running in a data center with have the load and it is doing ok. I do see a high number of page faults for the short period of time. I have the 3.1.10 instance still in place with all the logs and well review things to see if I can determine the issue. I think I might try using a small cache-dir and adjusting some mem settings. I am also including some config info and if someone has ideas or suggestions pleas let me know. Resource usage for squid: UP Time: 45842.506 seconds CPU Time: 35590.969 seconds CPU Usage: 77.64% CPU Usage, 5 minute avg: 99.01% CPU Usage, 60 minute avg: 100.16% Process Data Segment Size via sbrk(): 1929256 KB Maximum Resident Size: 0 KB Page faults with physical i/o: 6 Squid Config info: Squid Cache: Version 3.1.10-20101227 configure options: '--prefix=/usr/local/squid-3.1.10-20101227' '--disable-maintainer-mode' '--disable-dependency-tracking' '--enable-async-io=60' '--enable-storeio=ufs,aufs' '--enable-cache-digests' '--enable-epoll' '--with-pthreads' '--enable-snmp' '--with-large-files' '--enable-large-cache-files' '--enable-follow-x-forwarded-for' '--with-maxfd=16384' '--disable-ident-lookups' '--disable-wccp' '--enable-wccpv2' '--enable-removal-policies=heap,lru' '--enable-htcp' '--enable-auto-locale' --with-squid=/usr/local/src/squid-3.1.10-20101227 --enable-ltdl-convenience [root@cach1 local]# cat squid/etc/squid.conf # # Recommended minimum configuration: # acl manager proto cache_object acl localhost src 127.0.0.1/32 ::1 acl to_localhost dst 127.0.0.0/8 0.0.0.0/32 ::1 # Example rule allowing access from your local networks. # Adapt to list your (internal) IP networks from where browsing # should be allowed acl localnet src 10.0.0.0/8 # RFC1918 possible internal network acl localnet src 172.16.0.0/12 # RFC1918 possible internal network acl localnet src 192.168.0.0/16 # RFC1918 possible internal network acl localnet src fc00::/7 # RFC 4193 local private network range acl localnet src fe80::/10 # RFC 4291 link-local (directly plugged) machines acl SSL_ports port 443 acl Safe_ports port 80 81 # http acl Safe_ports port 21 # ftp acl Safe_ports port 443 # https acl Safe_ports port 70 # gopher acl Safe_ports port 210 # wais acl Safe_ports port 1025-65535 # unregistered ports acl Safe_ports port 280 # http-mgmt acl Safe_ports port 488 # gss-http acl Safe_ports port 591 # filemaker acl Safe_ports port 777 # multiling http acl CONNECT method CONNECT acl PURGE method PURGE follow_x_forwarded_for allow localhost # # Recommended minimum Access Permission configuration: # # Only allow cachemgr access from localhost http_access allow manager localhost http_access allow manager localnet http_access deny manager http_access allow PURGE localhost http_access deny PURGE # Deny requests to certain unsafe ports http_access deny !Safe_ports # Deny CONNECT to other than secure SSL ports http_access deny CONNECT !SSL_ports # We strongly recommend the following be uncommented to protect innocent # web applications running on the proxy server who think the only # one who can access services on "localhost" is a local user #http_access deny to_localhost # # INSERT YOUR OWN RULE(S) HERE TO ALLOW ACCESS FROM YOUR CLIENTS # # Example rule allowing access from your local networks. # Adapt localnet in the ACL section to list your (internal) IP networks # from where browsing should be allowed http_access allow localnet http_access allow localhost # And finally deny all other access to this proxy http_access deny all # Squid normally listens to port 3128 http_port 3128 icp_access allow localnet icp_access deny all icp_port 3130 # We recommend you to use at least the following line. hierarchy_stoplist cgi-bin ? # Uncomment and adjust the following to add a disk cache directory. cache_dir aufs /usr/local/squid/var/cache 75000 64 512 # Leave coredumps in the first cache dir coredump_dir /usr/local/squid/var/cache # Add any of your own refresh_pattern entries above these. 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 access_log none cache_store_log none cache_swap_low 98 cache_swap_high 99 cache_mem 512 MB cache_peer xxxx sibling 3128 3130 proxy-only cache_peer xxxx sibling 3128 3130 proxy-only prefer_direct off digest_generation on digest_rebuild_period 10 minutes digest_rewrite_period 10 minutes memory_replacement_policy heap GDSF cache_replacement_policy heap LFUDA maximum_object_size 32 MB maximum_object_size_in_memory 2 MB request_header_max_size 64 KB reply_header_max_size 64 KB shutdown_lifetime 10 seconds request_timeout 1 minutes forward_timeout 2 minutes positive_dns_ttl 300 seconds negative_dns_ttl 29 seconds negative_ttl 0 seconds ipcache_size 40480 ipcache_low 97 ipcache_high 99 fqdncache_size 512 cache_mgr visible_hostname walter unique_hostname walter logfile_rotate 10 forwarded_for off pipeline_prefetch on logfile_rotate 20 httpd_suppress_version_string on Jack ----- Original Message ---- From: Jack Quinlin <[email protected]> To: [email protected] Sent: Thu, January 6, 2011 10:44:21 AM Subject: Follow up on Benchmarking results for 3.1.10 vs 2.7.9 Hi, I have some benchmarking numbers that I would like to share with the community. These numbers are from 2 servers that have the exact same hardware configuration in the same data center receiving the same type of requests (not the best setup for a SQUID server but that is a different story). The server spec is below follwed by the benchmark numbers. I am looking to improving the 3.1.10 config or adjust RHEL4 in efforts to reduce the CPU consumption and improve hits. I will follow up with any improvements or findings that find interesting. Server Spec: CPU: Dual Intel(R) Xeon(R) CPU L5310 @ 1.60GHz Quad Core Mem: 4GB HD: 450 SATA OS: Red Hat Enterprise Linux AS release 4 (Nahant Update 4) 32-bit, OS has been tuned for 32-bit & RHEL 4.4 Squid Object Cache: Version 3.1.10-20101227 Users: 57 RPS: 1409 Hit Ratio Request: 36.3%, 36.4% , Byte: 47.6%, 48.6% CPU Usage: 51.63% Squid Object Cache: Version 2.7.STABLE9-20100923 Users: 55 RPS: 1419 Hit Ratio Request: 44.6%, 44.9% , Byte: 47.7%, 48.5% CPU Usage: 33.42% All comments and questions welcomed. Thanks, Jack ----- Original Message ---- From: Jack Quinlin <[email protected]> To: [email protected] Sent: Tue, December 28, 2010 1:12:08 PM Subject: Benchmarking results for Squid-3.1.9-20101210 Hi, I wanted to share some benchmarking results from one of the nodes in live system running the 3.1.9 build line. I also plan in installing the 3.1.10 line within the next week after verification of the newer build. I will share the results if the show improvements. If you have any questions or comments please feel free to ask. CPU Intel(R) Xeon(R) CPU L5310 @ 1.60GHz (dual-core) RAM 8 GB HDD 1 x 400G SAS OS Red Hat Enterprise Linux AS release 4 (Nahant Update 4) 32-bit, OS has been tunned for 32-bit & RHEL Users 54 RPS 615 Hit Ratio Request 41.7%-42.8% , Byte 49.8.7%-46.9% CPU Usage 31.69% Thanks, Jack
