Hi all,
I get an exception in a forever start routine in one of my containers. Funny
thing is: It works in the original container but fails in clones of the
original container.
I am pretty sure this is no lxc problem but a nodejs.forever or even a
nodejs.fs problem, but maybe somebody has an idea for a hint.
My forever call starts a script and watches for file changes so it can restart
the script. This works fine in the original but fails in the clone:
events.js:85
throw er; // Unhandled 'error' event
^
Error: watch ENOSPC
at exports._errnoException (util.js:746:11)
at FSWatcher.start (fs.js:1172:11)
at Object.fs.watch (fs.js:1198:11)
at createFsWatchInstance
(/var/share/nodejs/lib/node_modules/forever/node_modules/forever-monitor/node_modules/chokidar/lib/nodefs-handler.js:37:15)
at setFsWatchListener
(/var/share/nodejs/lib/node_modules/forever/node_modules/forever-monitor/node_modules/chokidar/lib/nodefs-handler.js:80:15)
at EventEmitter.NodeFsHandler._watchWithNodeFs
(/var/share/nodejs/lib/node_modules/forever/node_modules/forever-monitor/node_modules/chokidar/lib/nodefs-handler.js:228:14)
at EventEmitter.NodeFsHandler._handleFile
(/var/share/nodejs/lib/node_modules/forever/node_modules/forever-monitor/node_modules/chokidar/lib/nodefs-handler.js:255:21)
at EventEmitter.<anonymous>
(/var/share/nodejs/lib/node_modules/forever/node_modules/forever-monitor/node_modules/chokidar/lib/nodefs-handler.js:468:21)
at FSReqWrap.oncomplete (fs.js:95:15)
Other people having issues with forever talked about
fs.inotify.max_user_watches=524288 in /etc/sysctl.conf
User watches are not accumulated over all containers, or are they?
Best regards
Björn
_______________________________________________
lxc-users mailing list
[email protected]
http://lists.linuxcontainers.org/listinfo/lxc-users
