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
I'm running a bot that does mostly https requests, sitting only on around 5 https requests per second MAX. My bot scrapes 5 different reliable websites.
I've found that every 100 or so https requests I get"
{ Error: connect ETIMEDOUT 216.58.200.106:443
at Object._errnoException (util.js:1024:11)
at _exceptionWithHostPort (util.js:1046:20)
at TCPConnectWrap.afterConnect [as oncomplete] (net.js:1182:14)
code: 'ETIMEDOUT',
errno: 'ETIMEDOUT',
syscall: 'connect',
address: '216.58.200.106',
port: 443 }
Some deeper searching into the stacks and things tell me that its the underlying TCPWrap . syscall connect() is returning a -110 status error (which means timeout).
To my understanding the tcp socket is not making it to the remote server, the timeout happens when Linux cannot allocate a tcp socket to nodejs.
I've also seen this v8 error as well, which I don't think I would see unless it was a node bug.
To be clear, I'm quite certain it's not a poor internet connection because it happens on different web endpoints randomly, also I have a fibre connection (as good as it gets in Australia).
I've been running this exact program for months now, but this only started happening when I installed Ubuntu 17.10, but I keep my node versions up to date too, perhaps this was included in a newer release.
I've also attached a chrome devtools stack trace, if that helps
The text was updated successfully, but these errors were encountered:
ETIMEDOUT errors are a fact of life, you just need to handle them like you would any other run-time error. They mean the TCP handshake didn't complete within a sufficient time frame.
Do you get that assertion error while debugging with the inspector? That should be fixed in the next release.
I have no issue with network timeout errors, I know how to handle them and have been handling them for 5 years.
But I don't think that's what this is, because it's not a network or remote server issue. It's a localised issue that only happens on this server. This is a timeout between nodejs and the linux syscall connect(), not between my local server and a remote server on the network layer.
I've been running this software for 4 months, as soon as I switched to Ubunutu 17.10 I started getting these issues.
And yes, I did get that v8 trace while running inspector, it's great to see you guys are on top of it :).
Hi All,
I'm running a bot that does mostly https requests, sitting only on around 5 https requests per second MAX. My bot scrapes 5 different reliable websites.
I've found that every 100 or so https requests I get"
Some deeper searching into the stacks and things tell me that its the underlying TCPWrap . syscall connect() is returning a -110 status error (which means timeout).
To my understanding the tcp socket is not making it to the remote server, the timeout happens when Linux cannot allocate a tcp socket to nodejs.
I've also seen this v8 error as well, which I don't think I would see unless it was a node bug.
To be clear, I'm quite certain it's not a poor internet connection because it happens on different web endpoints randomly, also I have a fibre connection (as good as it gets in Australia).
I've been running this exact program for months now, but this only started happening when I installed Ubuntu 17.10, but I keep my node versions up to date too, perhaps this was included in a newer release.
I've also attached a chrome devtools stack trace, if that helps
The text was updated successfully, but these errors were encountered: