The intermittent, "HTTP ERROR: 504 Gateway Timeout" error, suggests there is a latent network problem. But diagnosis is going to be difficult. In an ideal world, it would trigger a 'traceroute' on the server but often we're seeing it at the browser: When it happened this morning, I ran a traceroute and took the above screenshot. Now that I have access, this is the current traceroute: $ traceroute PriusChattraceroute to PriusChat (199.195.144.4), 64 hops max, 40 byte packets 1 192.168.2.1 (192.168.2.1) 41.925 ms 0.520 ms 0.458 ms 2 10.170.200.152 (10.170.200.152) 315.477 ms 308.308 ms 308.918 ms 3 10.170.200.133 (10.170.200.133) 318.208 ms 308.415 ms 318.702 ms 4 10.184.81.140 (10.184.81.140) 377.899 ms 298.328 ms 318.607 ms 5 10.184.81.129 (10.184.81.129) 309.499 ms 345.457 ms 782.296 ms 6 10.170.200.5 (10.170.200.5) 72.546 ms 59.026 ms 57.893 ms 7 10.170.200.11 (10.170.200.11) 58.993 ms 68.529 ms 59.162 ms 8 10.184.6.189 (10.184.6.189) 60.679 ms 58.245 ms 49.179 ms 9 10.184.6.181 (10.184.6.181) 58.852 ms 58.566 ms 59.182 ms10 10.184.0.25 (10.184.0.25) 58.666 ms 58.639 ms 59.080 ms11 ae4-165.edge4.Atlanta2.Level3.net (4.59.12.97) 58.712 ms 58.453 ms 61.122 ms12 ROADRUNNER.edge4.Atlanta2.Level3.net (4.59.12.30) 56.934 ms ROADRUNNER.edge4.Atlanta2.Level3.net (4.30.53.62) 58.289 ms ROADRUNNER.edge4.Atlanta2.Level3.net (4.59.12.30) 55.589 ms13 107.14.19.12 (107.14.19.12) 108.669 ms ae-2-0.cr1.atl20.tbone.rr.com (66.109.6.176) 108.347 ms 107.14.19.12 (107.14.19.12) 108.062 ms14 107.14.17.142 (107.14.17.142) 107.867 ms 108.682 ms 108.398 ms15 107.14.17.173 (107.14.17.173) 108.648 ms ae-0-0.cr0.nyc20.tbone.rr.com (66.109.6.27) 108.379 ms 108.060 ms16 107.14.19.34 (107.14.19.34) 108.708 ms ae-8-0.cr0.chi10.tbone.rr.com (66.109.6.25) 108.497 ms 108.304 ms17 ae15.tr00.chcgildt.mwrtn.rr.com (66.109.6.205) 108.540 ms 108.039 ms 109.118 ms18 tge1-1.milwwignfl-dar1.twcwi.com (65.25.137.34) 108.660 ms 108.036 ms 109.121 ms19 rrcs-67-52-5-114.west.biz.rr.com (67.52.5.114) 98.597 ms 98.080 ms 99.151 ms20 static.66.185.29.193.cyberlynk.net (66.185.29.193) 98.682 ms 98.285 ms 99.046 ms21 * * *22 * *^CI was hoping to see a break in the trace before "static.66.185.29.193.cyberlynk.net" but no joy. Changing the port to 80 made no difference. This suggests the fault is behind the last, recognized hop, "static.66.185.29.193.cyberlynk.net." If you see the same fault, try to capture a traceroute and unlike my post, the date and time. We may be able to discover the duration as well as isolate the fault to a specific area. Bob Wilson