<div dir="ltr"><div>Thanks for your response. Responses inline.</div><div><br></div><font color="#e06666">> Can you elaborate on which URLs your clients are accessing? Is this<br>> on an upstream SHA1 of bmcweb? Do said clients implement HTTP<br>> keep-alive? Keep-alive is important, because otherwise the bmc is<br>> negotiating TLS for every request, which can be quite slow, especially<br>> when done in parallel.</font><br><div><br></div><div>We are at the latest bmcweb upstream code.</div><div>We have 4 type of clients collecting set of URIs:</div><div>All thermals: <span style="color:rgb(10,48,105);font-family:ui-monospace,SFMono-Regular,"SF Mono",Menlo,Consolas,"Liberation Mono",monospace;font-size:12px;white-space:pre">/redfish/v1/Chassis/<str>/Thermal/</span></div><div>All other sensors: <span style="color:rgb(10,48,105);font-family:ui-monospace,SFMono-Regular,"SF Mono",Menlo,Consolas,"Liberation Mono",monospace;font-size:12px;white-space:pre">/redfish/v1/Chassis/<str>/Sensors/<str>/</span><br></div><div>All inventory: /<span style="color:rgb(10,48,105);font-family:ui-monospace,SFMono-Regular,"SF Mono",Menlo,Consolas,"Liberation Mono",monospace;font-size:12px;white-space:pre">redfish/v1/Chassis/<str>/ and </span><span style="color:rgb(10,48,105);font-family:ui-monospace,SFMono-Regular,"SF Mono",Menlo,Consolas,"Liberation Mono",monospace;font-size:12px;white-space:pre">/redfish/v1/Systems/<str>/Processors/<str>/</span></div><div>Log services: <span style="color:rgb(10,48,105);font-family:ui-monospace,SFMono-Regular,"SF Mono",Menlo,Consolas,"Liberation Mono",monospace;font-size:12px;white-space:pre">/redfish/v1/Systems/<str>/LogServices/EventLog/ and </span><span style="color:rgb(10,48,105);font-family:ui-monospace,SFMono-Regular,"SF Mono",Menlo,Consolas,"Liberation Mono",monospace;font-size:12px;white-space:pre">/redfish/v1/Systems/<str>/LogServices/EventLog/Entries/</span></div><div><span style="color:rgb(10,48,105);font-family:ui-monospace,SFMono-Regular,"SF Mono",Menlo,Consolas,"Liberation Mono",monospace;font-size:12px;white-space:pre">/redfish/v1/Systems/<str>/LogServices/EventLog/Entries/<str>/</span></div><div><br></div><div>All clients implement keep-alive so clients avoid the TLS handshake penalty.</div><div><br></div><div><font color="#e06666">> I'm not sure how a priority system would work in this case. Can you<br>> elaborate on your idea? There is currently no queue outside of the<br>> TCP kernel, and async tasks in io_context. Which queue were you<br>> planning on assigning priorities to?</font><br></div><div><br></div><div>We want to prioritize our thermal sensor URI.</div><div>Is there a way to prioritize HTTP requests over D-Bus response messages in bmcweb?</div><div><br></div><div>Thanks,</div><div>Sharad</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Wed, 16 Nov 2022 at 21:59, Ed Tanous <<a href="mailto:edtanous@google.com">edtanous@google.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On Wed, Nov 16, 2022 at 5:28 AM sharad yadav <<a href="mailto:sharad.openbmc@gmail.com" target="_blank">sharad.openbmc@gmail.com</a>> wrote:<br>
><br>
> Hi All,<br>
><br>
> We are facing a delay(till 6seconds) in bmcweb URIs responses when there are multiple clients(4 in our case) performing GET requests in parallel on different URIs.<br>
<br>
Can you elaborate on which URLs your clients are accessing? Is this<br>
on an upstream SHA1 of bmcweb? Do said clients implement HTTP<br>
keep-alive? Keep-alive is important, because otherwise the bmc is<br>
negotiating TLS for every request, which can be quite slow, especially<br>
when done in parallel.<br>
<br>
Can you put together an example script that reproduces the issue?<br>
<br>
> We have a very important telemetry URI, used for thermal monitoring. On standalone, telemetry URI response is within 200ms.<br>
> But with multiple clients, telemetry URI response reaches till 6seconds.<br>
><br>
> Can we get help on how we can assign priority to our telemetry URI over other URIs in bmcweb?<br>
<br>
I'm not sure how a priority system would work in this case. Can you<br>
elaborate on your idea? There is currently no queue outside of the<br>
TCP kernel, and async tasks in io_context. Which queue were you<br>
planning on assigning priorities to?<br>
<br>
><br>
> Thanks,<br>
> Sharad<br>
><br>
><br>
><br>
><br>
><br>
><br>
</blockquote></div>