Hi, I've recenly switched a system over from 1.6.9, which has been running fine for years, to 1.9.4. I've updated the configuration to use nbthread instead of nbproc, and cleaned up the config a lot.
A few times now, however, i've seen haproxy using all available CPU on the system, even when traffic is mostly idle (or when the loadbalancer isn't even active anymore after a failover to the 2nd node). There is some output on the udp syslog, and the proxy still seems to work fine, but i'm only seeing a small subset of the requests. Is there any thing that can point me in the right direction? echo "show info" | nc 127.0.0.1 14567 Name: HAProxy Version: 1.9.4 Release_date: 2019/02/06 Nbthread: 8 Nbproc: 1 Process_num: 1 Pid: 20931 Uptime: 0d 2h31m20s Uptime_sec: 9080 Memmax_MB: 0 PoolAlloc_MB: 27 PoolUsed_MB: 27 PoolFailed: 0 Ulimit-n: 223153 Maxsock: 223153 Maxconn: 100000 Hard_maxconn: 100000 CurrConns: 36 CumConns: 390500 CumReq: 5700138 MaxSslConns: 0 CurrSslConns: 36 CumSslConns: 182620 Maxpipes: 11500 PipesUsed: 0 PipesFree: 2 ConnRate: 0 ConnRateLimit: 0 MaxConnRate: 689 SessRate: 0 SessRateLimit: 0 MaxSessRate: 689 SslRate: 0 SslRateLimit: 0 MaxSslRate: 168 SslFrontendKeyRate: 0 SslFrontendMaxKeyRate: 145 SslFrontendSessionReuse_pct: 0 SslBackendKeyRate: 0 SslBackendMaxKeyRate: 0 SslCacheLookups: 48562 SslCacheMisses: 4962 CompressBpsIn: 0 CompressBpsOut: 0 CompressBpsRateLim: 0 ZlibMemUsage: 0 MaxZlibMemUsage: 0 Tasks: 657 Run_queue: 4294967285 Idle_pct: 51 node: grtzlb2 Stopping: 0 Jobs: 96 Unstoppable Jobs: 0 Listeners: 59 ActivePeers: 0 ConnectedPeers: 0 DroppedLogs: 0 BusyPolling: 0 -- Mark Janssen -- [email protected] Unix / Linux Open-Source and Internet Consultant
