Hi Florin, this is the vec len of vcm->workers[0].sessions in core: (gdb) p vl(vcm->workers[0].sessions) You can't do that without a process to debug. ... (gdb) p ((vec_header_t *)(vcm->workers[0].sessions) - 1)->len $3 = 1 (gdb) p ((vec_header_t *)(vcm->workers[1].sessions) - 1)->len $4 = 14118
I adjust nginx conf as below, seems like it does not work: keepalive_timeout 0 keepalive_requests 1 We upload the core in the jira ticket, it may help you, for testing, heapsize for this core is 64M, also, the issue still exists in a lower speed traffic (vcl heapsize 32M, 1 nginx worker + 1 nginx master), I test with python requests, probably 85 cps root@ubuntu:~# ./http_request.py begin time: 2021-12-01 18:22:38.829277 HTTPConnectionPool(host='10.0.1.1', port=80): Read timed out. (read timeout=10) end time: 2021-12-01 20:33:25.372029 finished request 668056 this is the "show session" output with the lower speed: vpp# show session Thread 0: 1 sessions Thread 1: 861 sessions In fact, we want to use nginx as a reverse proxy, when the traffic accumulating nginx will crash, we found that the web server has the similar problem, so we want to solve this first, Regards, Mercury
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#20573): https://lists.fd.io/g/vpp-dev/message/20573 Mute This Topic: https://lists.fd.io/mt/87398061/21656 Mute #vpp-hoststack:https://lists.fd.io/g/vpp-dev/mutehashtag/vpp-hoststack Mute #hoststack:https://lists.fd.io/g/vpp-dev/mutehashtag/hoststack Mute #vppcom:https://lists.fd.io/g/vpp-dev/mutehashtag/vppcom Mute #nginx:https://lists.fd.io/g/vpp-dev/mutehashtag/nginx Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
