Release Notes: Support for HTTP keep-alive was added on the client side. Health check status reporting has been refined again to help with troubleshooting server issues. The stats interface has been improved with embedded links to ease monitoring of large setups. The config parser has been strengthened to detect more corner cases. A lot of regressions were fixed, and many minor improvements were made.
Release Notes: The stats page was improved a lot with the report of precise health check status. It is now possible to reset the counters without restarting and to change a server's weight live. Consistent hashing was implemented as an optional hashing method. The static round-robin algorithm was reintroduced for people running with more than 4000 servers in a farm. Some ACLs were added to match on queue length.
Release Notes: The rework of its internal achitecture for more flexibility was continued. This release adds features such as CLF log format, RDP protocol, a new, improved interactive CLI and HTML stats page, support for HTTP+SSL on the same port, more than 64k concurrent connections, smart TCP optimizations to reduce the number of packets in a session, runtime-adjustable buffer size, and support for forcing the TCP MSS on frontends. Keep-alive development has started and SSL integration is planned. The code has stabilized.
Release Notes: This release brings new long-awaited features, among which are TCP splicing support, conditional redirection, TCP content filtering, session rate reporting and limiting, invalid request/response capture, binding to specific network interfaces, per-process affinity for frontends and backends, a monotonic internal clock, and many others. The internal architecture has been reworked in layers to ease development, enhance reliability, and improve performance. Performance gains of about 10% are to be expected compared to 1.3.15.
Release Notes: Since 220.127.116.11, problems were fixed with hot-restart where the old process would not always stop, cookie capture was fixed (as it was broken since the frontend/backend split), critical startup errors are now reported on the console, and a failing server will no longer consume pending requests from the global queue. This last problem was the cause of many 503 errors for some people.
Release Notes: The connections regulation mechanism could send more connections than the allowed limit to a server when this server had just rejected a connection attempt; for servers with very low connection limits such as mongrel, this could easily trigger timeouts. The slowstart mechanism could return a limit of zero connections for very low maxconn values, causing servers to re-enter the farm a bit late.
Release Notes: Several timeouts were not correctly updated on some rare occasions. A possible segfault when using "url_param check_post" has been fixed. regparm is disabled on gcc versions < 3. "use_backend" would not correctly consider "unless". This last change should probably be the real reason to upgrade, as check_post is rarely used.
Release Notes: The queue/dequeue logic was reworked; using a low server maxconn value to serialize requests could result in some requests never being served. Users are encouraged to upgrade, especially those running haproxy in front of mongrel. All versions are affected. Version 18.104.22.168 was also released to address this problem in the previous branch.
Release Notes: Building with gcc 4.3 has been fixed. A condition where statistics could be truncated if exactly zero bytes remained in the buffer has been fixed. An event starvation that caused connection acceptance to slow down when approaching 40000 concurrent connections at gigabit speed has been fixed.
Release Notes: Many new features, including stats updates (HTTP, UNIX, and SNMP), enhancements to server checks such as tracking and dynamic intervals, the addition of the leastconn load-balancing algorithm and POST parameter analysis, a fully transparent mode on Linux, better handling of connection failures (dead server avoidance and turn-around state), support for inter-site off-loading through redirects, updates to the build process, and large documentation updates.