coredump in proxy module
iCloudDNS opened this issue · comments
Bugfix: coredump in proxy module
if an upstream block is placed after a proxy_pass direcitve, a coredump will happen.
我下载了最新的源码（标注已经修复了coredump BUG），编译替换了原来的 senginx 1.6.1 版本。
但使用中发现错误日志中依旧会出现 coredump
2014/12/23 16:09:06 [alert] 25598#0: worker process 25600 exited on signal 11 (core dumped)
2014/12/23 19:22:52 [alert] 25598#0: worker process 25604 exited on signal 11 (core dumped)
2014/12/23 19:22:58 [alert] 25598#0: worker process 25602 exited on signal 11 (core dumped)
2014/12/23 21:59:25 [alert] 25598#0: worker process 25605 exited on signal 11 (core dumped)
2014/12/23 21:59:26 [alert] 25598#0: worker process 25820 exited on signal 11 (core dumped)
2014/12/24 01:32:20 [alert] 25598#0: worker process 25694 exited on signal 11 (core dumped)
2014/12/24 01:32:20 [alert] 25598#0: worker process 25919 exited on signal 11 (core dumped)
我nginx.conf 文件中使用了 include proxy/*.conf，proxy目录下多个配置文件，每个配置文件中都包含一个
upstream 和 server 字段，每个server 字段中包含 proxy_pass
希望能尽快修复这个bug ，还有就是能升级到 nginx 1.7.8 版本。
因为现在版本中存在 header already sent 的BUG,nginx 新版已经修复此BUG
2014/12/24 01:32:30 [alert] 25823#0: *286349 header already sent, client: 66.249.75.163, server: _, request: "GET /includes/templates/nike/jscript/jscript_jquery-1.4.2.min.js HTTP/1.1", host: "www.nikeoutlet.net"
OK，我周末看一下
On 12/24/2014 09:54 AM, iCloudDNS wrote:
Bugfix: coredump in proxy module
if an upstream block is placed after a proxy_pass direcitve, a
coredump will happen.
我下载了最新的源码（标注已经修复了coredump BUG），编译替换了原来的
senginx 1.6.1 版本。
但使用中发现错误日志中依旧会出现 coredump
2014/12/23 16:09:06 [alert] 25598#0: worker process 25600 exited on
signal 11 (core dumped)
2014/12/23 19:22:52 [alert] 25598#0: worker process 25604 exited on
signal 11 (core dumped)
2014/12/23 19:22:58 [alert] 25598#0: worker process 25602 exited on
signal 11 (core dumped)
2014/12/23 21:59:25 [alert] 25598#0: worker process 25605 exited on
signal 11 (core dumped)
2014/12/23 21:59:26 [alert] 25598#0: worker process 25820 exited on
signal 11 (core dumped)
2014/12/24 01:32:20 [alert] 25598#0: worker process 25694 exited on
signal 11 (core dumped)
2014/12/24 01:32:20 [alert] 25598#0: worker process 25919 exited on
signal 11 (core dumped)
我nginx.conf 文件中使用了 include proxy/*.conf，proxy目录下多个配置文
件，每个配置文件中都包含一个
upstream 和 server 字段，每个server 字段中包含 proxy_pass
希望能尽快修复这个bug ，还有就是能升级到 nginx 1.7.8 版本。
因为现在版本中存在 header already sent 的BUG,nginx 新版已经修复此BUG
2014/12/24 01:32:30 [alert] 25823#0: *286349 header already sent,
client: 66.249.75.163, server: _, request: "GET
/includes/templates/nike/jscript/jscript_jquery-1.4.2.min.js
HTTP/1.1", host: "www.nikeoutlet.net"—
Reply to this email directly or view it on GitHub
#36.
valid_referers none blocked test.com *.test.com;
if ($invalid_referer)
{
return 403;
}
这里能不能修改一下，让 valid_referers 可以支持变量，在做反向代理时，大量的域名设置防盗链很不方便的。
最简单的就是用变量 $host 了，
valid_referers none blocked $host;
但默认nginx 配置$host 没有起到作用
hello，coredump那个能发一下具体的配置么
详细配置文件已发你qq邮箱
好的，有时间我看一下