Failure to connect to Cloud9 instance - never get past spinning icon

0

I have tried several times, always accepting all default options.

As per the trouble shooting page, I enabled 3rd party cookies, etc.

As an AWS user for about 15 years, I have never seen such a poorly rolled out product on AWS. If there are configuration problems, some form of warning or logging would be appreciated.

Edited by: Mark L. Watson on Dec 2, 2017 6:09 AM

已提问 6 年前3520 查看次数
3 回答
0

Hey Mark,

I was having the same issue (just spinning forever on "Connecting..." and managed to resolve it. My issue was that I was trying to launch the EC2 instance in a private subnet. Once I added a route to the Internet Gateway it connected within a few seconds. So I would double-check that your instance is reachable.

Hope that helps!

Cheers,
Timo

已回答 6 年前
0

Thanks Timo, that did it!

I deleted the first VPC I created, made a new one, and everything works :-)

已回答 6 年前
0

Note: This behavior also happens when the instance is out of CPU credits and still processing.
How this happened to me:
I created an instance and then downloaded and started building boost (c++), since it doesn't come with, like the c9.io machines do.
Boost took very, VERY long to compile, but was going, but after a while the IDE stopped responding and timed out.
I went to EC2 console and noticed that the machine was down to 10% cpu usage and cpu credits were over (actually, stuck at 5 credits).
The next day the machine was still like this and still wouldn't connect.
No matter how many times I tried to shut it down, restart, etc... once it got into this state once, something bugged up the Cloud9 environment, I guess; because I couldn't connect back to it ever again.

After 3 or 4 attempts with similar end-results (I tried different config combinations, etc... including turning on T2 unlimited); although a couple of these I did manage to connect back again (t2 unlimited); but since there were boost build errors ("fatal error out of memory" or such).
So in the end I created an environment using a larger EC2 type (don't remember which, exactly, but not the default free-tier 1gb type; I think it was 1 or 2 above that, with at least 2gb ram) WITH t2 unlimited. This time it ran to success... not sure if T2 unlimited would even be required in this case, since it built MUCH faster...

So, anyway, to sum up:
If you wanna build boost c++; do NOT use the default free-tier instance, use a higher instance with at least 2gb and turn off the cpu credit thinggy by switching to t2 unlimited.
Once boost is built, you can switch back to free-tier machine, depending on your needs, of course.

lflfm
已回答 6 年前

您未登录。 登录 发布回答。

一个好的回答可以清楚地解答问题和提供建设性反馈,并能促进提问者的职业发展。

回答问题的准则