Page 1 of 1

ETIMEDOUT 52.4.157.82:443

PostPosted: Sat Dec 02, 2017 12:30 pm
by jagsd
Over the last couple of days I have been getting numerous ETIMEDOUT errors from the host 52.4.157.82. Would you please look into this? Thanks.

EDIT: Now also getting timeouts from 52.4.222.218. pings work. I have FiOS gigabit which is working elsewhere. These timeouts appear to be somewhat sporadic. I have never seen this before with json.schedulesdirect.org.

Re: ETIMEDOUT 52.4.157.82:443

PostPosted: Sat Dec 02, 2017 3:44 pm
by rkulagow
You're not actually connecting to a host - you connect to an Amazon load balancer, and you should be using the DNS name because IP addresses are ephemeral.

That being said, there's no service-level issue right now. Load is normal.

Re: ETIMEDOUT 52.4.157.82:443

PostPosted: Sat Dec 02, 2017 4:24 pm
by jagsd
rkulagow wrote:You're not actually connecting to a host - you connect to an Amazon load balancer, and you should be using the DNS name because IP addresses are ephemeral.

That being said, there's no service-level issue right now. Load is normal.


Of course I am using the domain name json.schedulesdirect.org. I was just reporting which ip address was generating the ETIMEDOUT error. I also know that these are AWS servers. And in fact I was rather surprised. I have gigabit FiOS and I have not seen any such errors on any other site. Maybe there has been some problem in the internet path between me and them. It is hard to tell. I have tried pings and they seem to work even when getting the error. And the error seems to be intermittent though I get 3 to 6 of them in a single update.

I have never had these ETIMEDOUT errors before but I have seen A LOT of them the last 24 to 36 hours.

CLOSED: ETIMEDOUT 52.4.157.82:443

PostPosted: Tue Dec 12, 2017 4:51 pm
by jagsd
The timeouts happened for another 2 days and then just stopped. No problem for the last week and none ever before this started. Something was flaky for about 4 days. I had no noticeable net problem with my FiOS gig connection other than this. I cannot explain it but problem seems to be solved nonetheless.