Dear All, we have an issue after a graceful restart in a chrooted uwsgi application.
After the touch, the app is trying to restart but fails because can't find the INI file of the vassal. The INI file is outside the chroot, while the touchfile not. After the error (no such file or directory), the vassal dies and the master process waits long time before restarting it. Should the INI file be in the chroot as it's the uwsgi binary or we need to add an extra conf directive? Here the extract from log: Tue Sep 1 09:21:48 2015 - *** /home/user/domain/uwsgi.ini has been touched... grace them all !!! *** ...gracefully killing workers... Gracefully killing worker 2 (pid: 18661)... worker 2 buried after 1 seconds binary reloading uWSGI... chdir() to /opt/uwsgi closing all non-uwsgi socket fds > 2 (max_fd = 1024)... found fd 3 mapped to socket 0 (/home/user/.socks/uwsgi-domain-app.sock) running /opt/uwsgi/uwsgi *** has_emperor mode detected (fd: 12) *** *realpath() of user-domain-app.ini failed: No such file or directory [core/utils.c line 3607]* *** Starting uWSGI 2.0.9 (64bit) on [Tue Sep 1 11:24:01 2015] *** compiled with version: 4.8.2 on 14 March 2015 23:23:42 os: Linux-3.14.32-xxxx-grs-ipv6-64 #1 SMP Sat Feb 7 11:35:27 CET 2015 nodename: <hostname> machine: x86_64 clock source: unix pcre jit disabled detected number of CPU cores: 8 current working directory: /opt/uwsgi writing pidfile to /home/user/tmp/uwsgi-domain-app.pid detected binary path: /opt/uwsgi/uwsgi *** dumping internal routing table ***
_______________________________________________ uWSGI mailing list [email protected] http://lists.unbit.it/cgi-bin/mailman/listinfo/uwsgi
