You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Apr 22, 2023. It is now read-only.
Repository navigation
This repository was archived by the owner on Apr 22, 2023. It is now read-only.
AssertionError: false == true at SharedHandle.add (cluster.js:97:3) #9261
assert.js:86
throw new assert.AssertionError({
^
AssertionError: false == true
at SharedHandle.add (cluster.js:97:3)
at queryServer (cluster.js:480:12)
at Worker.onmessage (cluster.js:438:7)
at ChildProcess.<anonymous> (cluster.js:692:8)
at ChildProcess.emit (events.js:129:20)
at handleMessage (child_process.js:324:10)
at Pipe.channel.onread (child_process.js:352:11)
I'm getting the exact same problem and using node from master doesn't solve it. I'm only using UDP server sockets and I get the following when closing and restarting the UDP server:
AssertionError: false == true
at SharedHandle.add (cluster.js:99:3)
at queryServer (cluster.js:482:12)
at Worker.onmessage (cluster.js:440:7)
at ChildProcess.<anonymous> (cluster.js:694:8)
at ChildProcess.emit (events.js:131:20)
at handleMessage (child_process.js:326:10)
at Pipe.channel.onread (child_process.js:354:11)
It works fine without using the cluster API, but breaks with it.
$ node --version v0.13.0-pre
See the above referenced commit if you want to see some code.
The following code (the original test case posted in this issue) works in master, but not 0.12, as #8643 did not land in 0.12. Do you have a reduced test case, written in JavaScript, with no external dependencies, that reproduces the problem against the master branch? If you are using 0.12, you may want to look into using the exclusive option with your UDP sockets.
When is it going to be released? It's currently stopping me upgrading. I looked into the workaround but it didn't look like it was possible as the socket that causes the problems is created within the datagram code.
I'll take a look. If the backport isn't too significant then it shouldn't be a problem getting it in. That said, we don't have a clear timeline for v0.12.8 yet so we'll have to figure that out as well.
Activity
cjihrig commentedon Feb 22, 2015
Thanks for reporting this. Can you please provide a reduced test case, with no external dependencies, that reproduces the issue?
elliotstokes commentedon Feb 25, 2015
I'm also getting this error. Happens when I try to make a udp request from a cluster.
elliotstokes commentedon Feb 25, 2015
test case. It will work first request and fail for the second.
elliotstokes commentedon Feb 25, 2015
this doesn't look like it happens on 0.10.x only 0.12
FLYBYME commentedon Feb 25, 2015
I tracked it down to been UDP as elliotstokes did.
elliotstokes commentedon Feb 26, 2015
@FLYBYME did you find a workaround at all? i tried a few things but didn't get anywhere.
cjihrig commentedon Mar 1, 2015
The solution to this is in #8643. It landed in master, but not 0.12. A workaround would be to create an exclusive UDP socket in your worker.
First pass at adding cluster support. Broken for UDP connection due to …
rcrichton commentedon Mar 20, 2015
I'm getting the exact same problem and using node from master doesn't solve it. I'm only using UDP server sockets and I get the following when closing and restarting the UDP server:
It works fine without using the cluster API, but breaks with it.
$ node --version v0.13.0-preSee the above referenced commit if you want to see some code.
Worked around node bug in cluster.js, now our UDP audit server will n…
nexflo commentedon Mar 29, 2015
This is a major break for clustering (which is a core feature for node).
Is there an ETA in which release this will appear?
Thanks in advance.
pungoyal commentedon Apr 7, 2015
This has become a major problem for us. Our processes keep getting killed. Workaround is to put a monitoring tool/script, but we'd rather not do that.
cjihrig commentedon Apr 7, 2015
The following code (the original test case posted in this issue) works in master, but not 0.12, as #8643 did not land in 0.12. Do you have a reduced test case, written in JavaScript, with no external dependencies, that reproduces the problem against the master branch? If you are using 0.12, you may want to look into using the
exclusiveoption with your UDP sockets.elliotstokes commentedon Apr 7, 2015
When is it going to be released? It's currently stopping me upgrading. I looked into the workaround but it didn't look like it was possible as the socket that causes the problems is created within the datagram code.
cjihrig commentedon Apr 7, 2015
The master branch being released is probably still quite a while away. Here is a workaround in 0.12:
15 remaining items
cluster: allow shared reused dgram sockets
indutny commentedon Aug 25, 2015
Should be fixed by nodejs/node#2548
nodesocket commentedon Aug 25, 2015
Will this go out in
v0.12.8?indutny commentedon Aug 25, 2015
cc @jasnell
jasnell commentedon Aug 25, 2015
I'll take a look. If the backport isn't too significant then it shouldn't be a problem getting it in. That said, we don't have a clear timeline for v0.12.8 yet so we'll have to figure that out as well.
cluster: allow shared reused dgram sockets
cluster: allow shared reused dgram sockets
cluster: allow shared reused dgram sockets
cluster: allow shared reused dgram sockets
violet-day commentedon Sep 14, 2015
+1
austinkelleher commentedon Sep 14, 2015
Just a heads up, the issues I was experiencing with cluster have been resolved in 4.0.0.
Disable UDP socket sharing as it crash nodejs