We are having a problem where for one of our clients our
application that uses ArcGIS is failing to connect to your services. We
have looked into this issue and realised that every time we request a
token from within our application the HOST IP address keeps on changing
e.g look at the calls list below. Same call but different IP address
each time. The clients environment is highly secure and they need
upfront to allow access for certain IP addresses but now if it’s going
to be changing for the same URL how do we get around this. Unfortunately
they can not just open up their network. I thought you would have
static IP addresses for your URLs but this is not the case. If you may
please let me know a workaround to this it would be great.
The url we are using is
HTTPS-SERVER-CIPHER: TLS1_CK_ECDHE_RSA_WITH_AES_128_CBC_SHA
HTTPS-SERVER-SESSIONID: 70 44 98 7F 6E 73 FF 09 43 5A D4 F0 6A 18 81 A5 CF A2 99 76 18 1E 03 18 63 4C C1 B8 D9 C3 0C 39
X-CLIENTIP: 127.0.0.1
X-CLIENTPORT: 63800
X-EGRESSPORT: 63801
X-HOSTIP: 54.243.139.226
HTTPS-CLIENT-SNIHOSTNAME: www.arcgis.com
X-CLIENTIP: 127.0.0.1
X-CLIENTPORT: 63547
X-EGRESSPORT: 63548
X-HOSTIP: 54.225.191.183
HTTPS-CLIENT-SNIHOSTNAME: www.arcgis.com
X-CLIENTIP: 127.0.0.1
X-CLIENTPORT: 64528
X-EGRESSPORT: 64529
X-HOSTIP: 54.225.173.54
For
the routing calls it seems you only use 2 IP addresses which is better
as they can allow access to these. The addresses we have picked up are
If you do a reverse name lookup on all of those IPs, they all point back to Amazon's Elastic Compute Cloud, which is where ArcGIS.com resides. You are getting different host IPs because of load balancing in the cloud. What are the ranges of IP addresses that participate in the load balancing? I don't know, and I doubt Esri will publish them. That said, you may be able to get some information if you open an Esri Support case.