Open tbremer opened 9 years ago
Get exactly the same message when watch restarts. Happens on several projects with different configurations and on two different Macs.
Tested with those versions:
$ node v0.12.0
$ npm 2.3.0, 2.5.0, 2.7.0, 2.7.1
This happens for me too
+1 Ditto
Same here. Can't seem to find a solution...
+1 Yeah, same here. Watch task seems to actually work, but nonetheless the message is a bit annoying. I am on Yosemite 10.10.2
Same here...
I think the problem is node.js in v0.12.0, on an older version no problems..
Same problem. Upgraded to node.js v0.12.1, and it is still there.
Same problem.
Does anyone write an bugreport to the nodejs guys?
Same here
$ node v0.12.2
$ npm 2.7.4
OSX 10.10.2 (14C1514)
grunt[22532] (FSEvents.framework) FSEventStreamFlushSync(): failed assertion '(SInt64)last_id > 0LL'
If anyone wants to try this in io.js, should be fixed there.
Still happening for me as well
node v0.12.2 npm 2.7.5 grunt-contrib-watch 0.6.1 OS X 10.10.3
This seems to happen more when it has to run an uglify or concat task. But both are at the latest version.
Same here on
node v0.12.0 and v0.12.2 npm 2.5.1 and 2.7.4 grunt-contrib-watch 0.6.1 OS X 10.9.5
Happens only if I set the option spawn: false
or nospawn: true
. The tasks don't run correctly, it's not just the message for me.
Seams to work with node v0.10.26.
Same problem - Seems to be functioning fine, but always gives this error.
Node: v0.12.2,
NPM: 2.7.5,
grunt: 0.4.5,
grunt-contrib-concat: 0.5.1,
grunt-contrib-imagemin: 0.9.4,
grunt-contrib-uglify: 0.9.1,
grunt-contrib-watch: 0.6.1,
grunt-sass: 0.18.1,
OS X 10.10.2
(FSEvents.framework) FSEventStreamFlushSync(): failed assertion '(SInt64)last_id > 0LL'
Can confirm.
same here
Node v0.12.2
grunt-cli v0.1.13
grunt v0.4.5
grunt-contrib-watch 0.6.1
OS X 10.10.3
I'm also experiencing this issue
OSX: v10.10.2
Node v0.12.2
Npm v2.7.5
Packages:
├── grunt@0.4.5
├── grunt-contrib-connect@0.10.1
├── grunt-contrib-copy@0.8.0
├── grunt-contrib-less@1.0.1
├── grunt-contrib-uglify@0.9.1
├── grunt-contrib-watch@0.6.1
├── grunt-exec@0.4.6
└── grunt-express-server@0.5.1
Same here. node: v0.12.0
Removing spawn from Watch > Scripts > Options:{ } stops the error. What is spawn anyway again?
Might this potentially be why? https://github.com/bdkjones/fseventsbug/wiki/realpath()-And-FSEvents
@digitalcraftco solution worked with v0.12.7, removing all spawns from watch: {} task (even if they're false)
@juanbrujo Thanks for the follow up. Indeed this has fixed the issue for newer versions.
Interesting. Is 0.11.6 the earliest node version this manifests in? We've had the above linked io.js issue for this for a while but haven't really been able to debug it.
@digitalcraftco It defines whether a new task should be spawned or not. See https://github.com/gruntjs/grunt-contrib-watch#optionsspawn
Removing spawn
from options does avoid but not fix the issue.
+1 getting this as well still.
Interesting to note, as soon as you remove the spawn = false property the compile time jumps up dramatically, in my case from 0.3sec to 4.3sec !!
I just came across this error (with spawn set to false) maybe this is of help ->
(node) warning: possible EventEmitter memory leak detected. 11 change listeners added. Use emitter.setMaxListeners() to increase limit. Trace at StatWatcher.addListener (events.js:179:15) at Object.fs.watchFile (fs.js:1289:8) at Gaze._pollFile (/Applications/MAMP/htdocs/bla/node_modules/grunt-contrib-watch/node_modules/gaze/lib/gaze.js:329:10) at /Applications/MAMP/htdocs/bla/node_modules/grunt-contrib-watch/node_modules/gaze/lib/gaze.js:411:12 at Array.forEach (native) at /Applications/MAMP/htdocs/bla/node_modules/grunt-contrib-watch/node_modules/gaze/lib/gaze.js:409:11 at iterate (/Applications/MAMP/htdocs/bla/node_modules/grunt-contrib-watch/node_modules/gaze/lib/helper.js:52:5) at Object.forEachSeries (/Applications/MAMP/htdocs/bla/node_modules/grunt-contrib-watch/node_modules/gaze/lib/helper.js:66:3) at Gaze._initWatched (/Applications/MAMP/htdocs/bla/node_modules/grunt-contrib-watch/node_modules/gaze/lib/gaze.js:354:10) at Gaze._internalAdd (/Applications/MAMP/htdocs/bla/node_modules/grunt-contrib-watch/node_modules/gaze/lib/gaze.js:193:10)
I see similar messages in a C app using libuv and uv_fs_event* I wrote as well. Linked to https://github.com/nodejs/node/issues/854 which mentions this issue.
Absolutely nothing changed in my dependencies for the last month at least and I randomly just started seeing this
@maruf89 Did you change OS X versions?
No, I'm still on Yosemite 10.10.3 - The only thing that changed is I got a hard drive cable replaced but no software/npm deps changed
+1 having same issue
@maruf89, same by me, i had only changed the hard drive cable.
+1
What are the side-effects to this error? It started occurring in my project after I added imagemin
What are the side-effects to this error?
Unknown, probably nothing impactful? I think the OS just retries or something.
Strange that multiple people would see it after a harddrive cable, maybe it's an odd apple hardware issue?
Am sure it has nothing to do with hardware. As Fishrock123 said, no apparent(!) sideeffects, but still annoying. So a fix definetly would be nice, since also a lot of systems seem to be affected.
I'm having this issue with node v7.3.0 on OSX 10.11.6 (El Capitan) and grunt v0.4.5, grunt-cli v1.2.0
I can't make sense of why this worked, but I was having the exact same issue, and adding more specific selectors for the files I was watching resolved the issue. Here's a sample of my old Gruntfile:
watch: {
css: {
files: '**/*.scss',
tasks: ['sass']
},
js: {
files: '**/*.js',
tasks: ['rollup']
}
}
and here's my new Gruntfile
watch: {
css: {
files: './assets/**/*.scss',
tasks: ['sass']
},
js: {
files: './assets/**/*.js',
tasks: ['rollup']
}
}
What's more--my old grunt process was using a huge chunk of my CPU, and with my more specific file selection, I use only 2-3%.
I am getting the following error after
Running "watch" task
restartsHappens with these versions:
Related Grunt Config: