<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">On 5/20/14 7:03 AM, Hedonil wrote:<br>
</div>
<blockquote cite="mid:537B4474.4090502@action.ms" type="cite">
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
<div class="moz-cite-prefix">Multiple 500 errors & "no data
received" things from tools-webproxy again.. ^<br>
<br>
For the sake of reliability of tools services:<br>
Isn't there a way to provide a /stable/ & failesafe webproxy
- and do the testings on a dedicated test instance?<br>
<br>
</div>
</blockquote>
Right now the proxies lives in labs instances. We've discussed
moving them onto a proper production box for better performance, but
this is really the first time that we've hit any kind of load-based
failures. I'm not clear on if this is because the newer version of
NGINX has different/worse performance or if something is spiking in
tools.<br>
<br>
Yuvi, once you're done fighting the fire can you summarize what
you've learned about this?<br>
<br>
-Andrew<br>
<br>
</body>
</html>