升级到了tengine 2.1.0,发现
设置websocket日志"error_log" 记录到 syslog；
error_log出现很多如下错误：
[error] 839006#0: *2154531 upstream prematurely closed connection while reading response header from upstream
系统日志出现如下错误：
kernel: nginx[932167]: segfault at 0 ip 000000000044853b sp 00007ffff6102490 error 4 in nginx[400000+2c1000]
kernel: nginx[932168]: segfault at 0 ip 000000000044853b sp 00007ffff6102490 error 4 in nginx[400000+2c1000]
kernel: nginx[932172]: segfault at 0 ip 000000000044853b sp 00007ffff6102490 error 4 in nginx[400000+2c1000]
kernel: __ratelimit: 6 callbacks suppressed
；
查看了一下nginx官方文档，有点像 1.7.4 更新说明里的：
Bugfix: a segmentation fault might occur if logging of errors to
syslog was used; the bug had appeared in 1.7.1.
升级到了tengine 2.1.0,发现
设置websocket日志"error_log" 记录到 syslog；
error_log出现很多如下错误：
[error] 839006#0: *2154531 upstream prematurely closed connection while reading response header from upstream
系统日志出现如下错误：
kernel: nginx[932167]: segfault at 0 ip 000000000044853b sp 00007ffff6102490 error 4 in nginx[400000+2c1000]
kernel: nginx[932168]: segfault at 0 ip 000000000044853b sp 00007ffff6102490 error 4 in nginx[400000+2c1000]
kernel: nginx[932172]: segfault at 0 ip 000000000044853b sp 00007ffff6102490 error 4 in nginx[400000+2c1000]
kernel: __ratelimit: 6 callbacks suppressed
；
查看了一下nginx官方文档，有点像 1.7.4 更新说明里的：
Bugfix: a segmentation fault might occur if logging of errors to
syslog was used; the bug had appeared in 1.7.1.