<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Bryan Ward</title>
    <description>BryanWard.net</description>
    <link>http://bryanward.net/</link>
    <atom:link href="http://bryanward.net/feed.xml" rel="self" type="application/rss+xml"/>
    <pubDate>Fri, 20 Nov 2026 11:03:37 +0000</pubDate>
    <lastBuildDate>Fri, 20 Nov 2026 11:03:37 +0000</lastBuildDate>
    <generator>Jekyll v3.10.0</generator>
    
      <item>
        <title>Remove the Ads in Wireshark</title>
        <description>&lt;p&gt;It’s a sad day. There are now ads in Wireshark. 🙁 Here’s how to remove them.&lt;/p&gt;

&lt;!--more--&gt;

&lt;p&gt;Needless to say, I was shocked when I saw this new “feature”. Sure, it might be nice to have handy links to documentation and events like Sharkfest, but I definitely don’t want them inside the app I use daily for troubleshooting purposes.&lt;/p&gt;

&lt;p&gt;&lt;img src=&quot;/assets/img/posts/2026-05-29-image1.png&quot; alt=&quot;image&quot; /&gt;&lt;/p&gt;

&lt;h2 id=&quot;update&quot;&gt;Update&lt;/h2&gt;

&lt;p&gt;(Thanks Anders for pointing this out!)&lt;/p&gt;

&lt;p&gt;In the Edit&amp;gt;Preferences section, there is now an Appearance&amp;gt;Welcome Page section. You can go there and simply uncheck everything.&lt;/p&gt;

&lt;p&gt;&lt;img src=&quot;/assets/img/posts/2026-05-29-image2.png&quot; alt=&quot;image&quot; /&gt;&lt;/p&gt;

&lt;h2 id=&quot;original-post&quot;&gt;Original post&lt;/h2&gt;

&lt;p&gt;You’ll need to edit the preferences file for your Wireshark installation. On Windows, you can find this at &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;%APPDATA%\Wireshark\preferences&lt;/code&gt;. On Mac/Linux, you can find this at &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;~/.config/wireshark/preferences&lt;/code&gt;.&lt;/p&gt;

&lt;p&gt;Some other posts about this topic say to edit the &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;recent_common&lt;/code&gt; file, however that file gets regenerated every time Wireshark exits. You need to edit the &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;preferences&lt;/code&gt; file instead. This file will get overwritten any time you change the preferences in the app, and the changes we’re about to make might get erased, but that’s much less often an occurrence than just running and exiting Wireshark.&lt;/p&gt;

&lt;p&gt;Make sure Wireshark is not presently running before opening the preferences file. At the end of the file, add these lines:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;gui.welcome_page.sidebar.learn_visible: false
gui.welcome_page.sidebar.tips_visible: false
gui.welcome_page.sidebar.tips_events: false
gui.welcome_page.sidebar.tips_sponsorship: false
gui.welcome_page.sidebar.tips_tips: false
gui.welcome_page.sidebar.tips_interval: 0
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;And save the file.&lt;/p&gt;

&lt;p&gt;Remember to do this on any computer you use Wireshark on.&lt;/p&gt;
</description>
        <pubDate>Fri, 29 May 2026 11:48:28 +0000</pubDate>
        <link>http://bryanward.net/2026/05/29/remove-the-ads-in-wireshark.html</link>
        <guid isPermaLink="true">http://bryanward.net/2026/05/29/remove-the-ads-in-wireshark.html</guid>
        
        
        <category>Uncategorized</category>
        
      </item>
    
      <item>
        <title>SenseCap T1000 Data Rates</title>
        <description>&lt;p&gt;The Seeed SenseCap T1000 is an awesome little LoRaWAN device. A longer blog post about them will happen in the future, but for now I wanted to just quickly document the Data Rate setting.&lt;/p&gt;

&lt;!--more--&gt;

&lt;p&gt;This post contains affiliate links, through which I may earn a commission and/or referral fee on purchases made.&lt;/p&gt;

&lt;p&gt;Because this is not very clearly documented anywhere else, here’s what the LoRaWAN Data Rates in the &lt;a href=&quot;https://sensecap-mate-download.seeed.cn/&quot;&gt;SenseCraft App&lt;/a&gt; correspond to for LoRa Spreading Factors:&lt;/p&gt;

&lt;p&gt;At least for the US915 region, RP 1.0.3…&lt;/p&gt;

&lt;table&gt;
  &lt;thead&gt;
    &lt;tr&gt;
      &lt;th&gt;DR&lt;/th&gt;
      &lt;th&gt;SF&lt;/th&gt;
    &lt;/tr&gt;
  &lt;/thead&gt;
  &lt;tbody&gt;
    &lt;tr&gt;
      &lt;td&gt;1&lt;/td&gt;
      &lt;td&gt;SF9 BW125&lt;/td&gt;
    &lt;/tr&gt;
    &lt;tr&gt;
      &lt;td&gt;2&lt;/td&gt;
      &lt;td&gt;SF8 BW125&lt;/td&gt;
    &lt;/tr&gt;
    &lt;tr&gt;
      &lt;td&gt;3&lt;/td&gt;
      &lt;td&gt;SF7 BW125&lt;/td&gt;
    &lt;/tr&gt;
    &lt;tr&gt;
      &lt;td&gt;4&lt;/td&gt;
      &lt;td&gt;SF8 BW500&lt;/td&gt;
    &lt;/tr&gt;
  &lt;/tbody&gt;
&lt;/table&gt;

&lt;p&gt;If you want to buy a T1000, here are links to do so! Use Promo Code &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;TP0LTI8P&lt;/code&gt; for a discount!&lt;/p&gt;

&lt;p&gt;&lt;a href=&quot;https://www.seeedstudio.com/SenseCAP-Card-Tracker-T1000-A-p-5697.html?sensecap_affiliate=Kdtuj1U&amp;amp;referring_service=link&quot;&gt;T1000-A for LoRaWAN or Helium&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;a href=&quot;https://www.seeedstudio.com/SenseCAP-Card-Tracker-T1000-E-for-Meshtastic-p-5913.html?sensecap_affiliate=Kdtuj1U&amp;amp;referring_service=link&quot;&gt;T1000-E for Meshtastic&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;a href=&quot;https://www.seeedstudio.com/SenseCAP-Card-Tracker-T1000-E-for-LoRaWAN-p-6408.html?sensecap_affiliate=Kdtuj1U&amp;amp;referring_service=link&quot;&gt;T1000-E for LoRaWAN Open Source&lt;/a&gt;&lt;/p&gt;
</description>
        <pubDate>Tue, 22 Jul 2025 13:42:11 +0000</pubDate>
        <link>http://bryanward.net/2025/07/22/sensecap-t1000-data-rates.html</link>
        <guid isPermaLink="true">http://bryanward.net/2025/07/22/sensecap-t1000-data-rates.html</guid>
        
        <category>lorawan</category>
        
        
        <category>Uncategorized</category>
        
      </item>
    
      <item>
        <title>Setting and Changing Static WAN IP Addresses on the Juniper SSR</title>
        <description>&lt;p&gt;Once an SSR appliance is connected to the Mist cloud, it’s easy to make changes to the config. However, if your ISP has assigned a Static IP Address, you’ll need to know how to set that on the SSR using the console port before it will be able to phone-home to the Mist cloud.&lt;/p&gt;

&lt;!--more--&gt;

&lt;p&gt;Getting into the console of the SSR is similar to Junos devices. Connect to the serial port using 115200 baud 8N1. Then, log in using the &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;root&lt;/code&gt; username and the default out-of-box password of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;128tRoutes&lt;/code&gt;. This is case-sensitive. The password will be changed by Mist once it phones home. You’ll then be at the root shell. To get to the CLI, you now need to run &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;su admin&lt;/code&gt; instead of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;cli&lt;/code&gt; like you would on Junos. This will bring you to the PCLI interface, which functions similarly to the Junos CLI, but do note that this is not Junos and the command/config language is quite different.&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;sn1234567890 login: root
Password: 
Last failed login: Mon Jun 10 12:34:56 UTC 2025 on ttyS0
There was 1 failed login attempt since the last successful login.
Last login: Mon Jun  1 13:23:45 on pts/0
+---------------------------------------+
|                                       |
|    Welcome to:                        |
|                                       |
|     | .   . ,---. . ,---. ,---. ,--.  |
|     | |   | |   | | |---&apos; |---&apos; |     |
|     | `---&apos; &apos;   &apos; &apos; &apos;     `---&apos; &apos;     |
|  ---&apos;                                 |
|        __ ___       __   __       __  |
|  |\ | |_   |  |  | /  \ |__) |_/ (_   |
|  | \| |__  |  |/\| \__/ | \  | \ __)  |
|                                       |
| Session Smart Networking Platform ... |
+---------------------------------------+
[root@sn1234567890 ~]# 
[root@sn1234567890 ~]# cli
-bash: cli: command not found
[root@sn1234567890 ~]# su admin
Starting the PCLI...
admin@node0.900102abcdef#
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;If you wait too long at this prompt, you might get kicked back to the root shell. If this happens, just run &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;su admin&lt;/code&gt; again.&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;% Error: Session timed out after 900 seconds
Exiting the PCLI
[root@sn1234567890 ~]# su admin
Starting the PCLI…
admin@node0.900102abcdef#
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;Your prompt may say &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;admin@node.router&lt;/code&gt; instead of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;node0&lt;/code&gt; and the hostname, MAC Address, or Serial Number of the device. Adjust your commands accordingly to match the environment of your SSR. If your prompt says node.router, you’ll want to use commands like &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;configure authority router router node node&lt;/code&gt;. If your prompt has node0 and a MAC Address, you’ll want to use commands like &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;configure authority router 900102abcdef&lt;/code&gt; &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;node node0&lt;/code&gt;. I provide example using both throughout this article.&lt;/p&gt;

&lt;p&gt;Once at this prompt, setting a static IP Address is pretty straightforward. However, if your SSR comes out of the box with older code (such as version 5.4.4), the &lt;a href=&quot;https://www.juniper.net/documentation/us/en/software/mist/mist-wan/topics/topic-map/ssr-onboarding-with-static-ip.html&quot;&gt;documentation provided by Juniper&lt;/a&gt; is not a simple copy and paste. You’ll know if you’re in this situation if you see the following messages when you try to apply the config the documentation tells you to use:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;admin@node.router# configure authority router router node node device-interface
ge-0-0 network-interface ge-0-0-intf dhcp disabled
*admin@node.router# configure authority router router node node device-interface
 ge-0-0 network-interface ge-0-0-intf address 192.0.2.101 prefix-length 24
*admin@node.router# configure authority router router node node device-interface
 ge-0-0 network-interface ge-0-0-intf address 192.0.2.101 gateway 192.0.2.1

*admin@node.router# commit force
Validating, then committing…
% Error: Failed to commit:
1. PCI address, VMBus UUID or Interface Name is required for device-interface
ge-0-0 when type is ethernet.
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;The issue here is that the interfaces names may be different in the version of firmware your device is runnig that the version for which the documentation was written. You can check what version firmware the SSR is running by using the following command:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;admin@node.router#&amp;gt; show system version
Tue 2025-06-10 14:34:56 UTC

======== ====== ========= ====================== ==================
Router   Node   Version   Build Date             Package
======== ====== ========= ====================== ==================
router   node   5.4.4     2022-02-17T17:52:07Z   128T-5.4.4-9.el7

Completed in 0.05 seconds
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;admin@node0.900102abcdef# show system version
Mon 2025-06-10 15:45:07 UTC
Retrieving system version...

============== ======= ========= ======== ====================== ===============
 Router         Node    Version   Status   Build Date             Package
============== ======= ========= ======== ====================== ===============
 900102abcdef   node0   6.3.4     r2       2025-02-05T03:12:54Z   128T-6.3.4-7.
                                                                  r2.el7

Completed in 0.05 seconds
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;&lt;strong&gt;The solution is simple. Just remove (or add) the &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;-intf&lt;/code&gt; suffix from the name of the network-interface.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;You can see if an interface is configured for DHCP or a Static IP Address with the command &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show config run flat | grep -P &quot;ge-0-0 (dhcp|address)&quot;&lt;/code&gt;. Replace ge-0-0 with the name of the interface you want to check. If you see a single line that mentions dhcp, then the interface is configured to request a DHCP lease. If you see lines mentioning ip-address, prefix-length, and gateway config, then it’s set for the defined Static IP address.&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;admin@node0.900102abcdef# show config run flat | grep -P &quot;ge-0-0 (dhcp|address)&quot;
 
Piping output...
config authority router 900102abcdef node node0 device-interface ge-0-0 network-interface ge-0-0 address 192.0.2.101 ip-address     192.0.2.101
config authority router 900102abcdef node node0 device-interface ge-0-0 network-interface ge-0-0 address 192.0.2.101 prefix-length  24
config authority router 900102abcdef node node0 device-interface ge-0-0 network-interface ge-0-0 address 192.0.2.101 gateway        192.0.2.1

admin@node0.900102abcdef# show config run flat | grep -P &quot;ge-0-1 (dhcp|address)&quot;
 
Piping output...
config authority router 900102abcdef node node0 device-interface ge-0-1 network-interface ge-0-1 dhcp                   v4
admin@node0.900102abcdef#
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;To change the address assignment, you must first remove the existing Static IP config like so:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;admin@node0.900102abcdef# configure authority router 900102abcdef node node0 device-interface ge-0-0 network-interface ge-0-0 delete address 192.0.2.101
% Warning: You are about to make configuration changes to a router that is
managed by the Mist Cloud. These changes could be overwritten at any time by the
Mist Cloud.
Are you sure you want to delete item &apos;[ip-address=192.0.2.101]&apos; [y/N]: y
*admin@node0.900102abcdef#

*admin@node0.900102abcdef# show config candidate flat | grep &quot;ge-0-0 address&quot;
Piping output...
Pipe failed: Command &apos;grep&apos; exited with non-zero exit status 1
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;(The error from ‘grep’ indicates that no config was found.)&lt;/p&gt;

&lt;p&gt;And then assign a different Static IP Address:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;config authority router 900102abcdef node node0 device-interface ge-0-0 network-interface ge-0-0 dhcp disabled
config authority router 900102abcdef node node0 device-interface ge-0-0 network-interface ge-0-0 address 192.0.2.111 prefix-length  25
config authority router 900102abcdef node node0 device-interface ge-0-0 network-interface ge-0-0 address 192.0.2.111 gateway        192.0.2.1
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;Or enable DHCP:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;*admin@node0.900102abcdef# config authority router 900102abcdef node node0 device-interface ge-0-0 network-interface ge-0-0 dhcp

# Nothing to show

*admin@node0.900102abcdef#
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;To “rollback” the candidate config to the version that is currently running, use:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;*admin@node0.900102abcdef# restore config running
% Warning: You are about to make configuration changes to a router that is
managed by the Mist Cloud. These changes could be overwritten at any time by the
Mist Cloud.
Are you sure you want to discard uncommitted changes from the candidate config f
or user admin? [y/N]: y
Discarding uncommitted changes...
Candidate configuration changes successfully discarded for admin
*admin@node0.900102abcdef#
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;The SSR equivalent of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show | compare&lt;/code&gt; is &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;compare config&lt;/code&gt;. It will display the difference between the running and the candidate in full stanzas, not in flat one-liner syntax.&lt;/p&gt;

&lt;p&gt;Once you’re satisfied with your config changes, run &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;commit&lt;/code&gt;. This checks the config for any warnings/errors before prompting you to apply it.&lt;/p&gt;
</description>
        <pubDate>Mon, 16 Jun 2025 11:33:56 +0000</pubDate>
        <link>http://bryanward.net/2025/06/16/setting-and-changing-static-wan-ip-addresses-on-the-juniper-ssr.html</link>
        <guid isPermaLink="true">http://bryanward.net/2025/06/16/setting-and-changing-static-wan-ip-addresses-on-the-juniper-ssr.html</guid>
        
        <category>juniper</category>
        
        <category>mist</category>
        
        <category>ssr</category>
        
        
        <category>Uncategorized</category>
        
      </item>
    
      <item>
        <title>Excel: Copy as JSON</title>
        <description>&lt;p&gt;A new twist on my existing &lt;a href=&quot;https://bryanward.net/wp/2020/01/17/excel-copy-as-csv/&quot;&gt;Copy As CSV&lt;/a&gt; macro, this code will copy the selected cells in an Excel spreadsheet as a JSON document to your clipboard.&lt;/p&gt;

&lt;!--more--&gt;

&lt;p&gt;&lt;img src=&quot;/assets/img/posts/2025-04-09-image1.gif&quot; alt=&quot;image&quot; /&gt;&lt;/p&gt;

&lt;p&gt;Install this in Excel using the same instructions as provided in my &lt;a href=&quot;https://bryanward.net/wp/2020/01/17/excel-copy-as-csv/&quot;&gt;Copy as CSV&lt;/a&gt; post, but just use the below code instead. It will return a list of dicts, with keys based on the first row of the selected cells. Each row in the selection becomes an item in the list. All values are treated as strings for simplicity.&lt;/p&gt;

&lt;script src=&quot;https://gist.github.com/a8f9a31440c74cbf9112f2a1ccf7f04f.js&quot;&gt; &lt;/script&gt;

</description>
        <pubDate>Wed, 09 Apr 2025 12:01:46 +0000</pubDate>
        <link>http://bryanward.net/2025/04/09/excel-copy-as-json.html</link>
        <guid isPermaLink="true">http://bryanward.net/2025/04/09/excel-copy-as-json.html</guid>
        
        <category>code</category>
        
        <category>excel</category>
        
        
        <category>Uncategorized</category>
        
      </item>
    
      <item>
        <title>Protocol Translator</title>
        <description>&lt;p&gt;Here’s a quick one. Sometimes you need to get two apps to talk to each other, but one only supports TCP and the other only supports UDP. (In my case, I’m hooking up my &lt;a href=&quot;https://www.axis.com/products/axis-a8105-e/support&quot;&gt;Axis Video Doorbell&lt;/a&gt; to trigger sound effects (and other things) in &lt;a href=&quot;https://qlab.app/&quot;&gt;QLab&lt;/a&gt; for some Halloween fun.)&lt;/p&gt;

&lt;!--more--&gt;

&lt;p&gt;The Axis doorbell allows you to send raw TCP messages when the button is pressed. It does not support sending raw UDP. QLab can listen for raw UDP, but it only listens for a specific and more complicated protocol (&lt;a href=&quot;https://ccrma.stanford.edu/groups/osc/index.html&quot;&gt;OSC&lt;/a&gt;) via TCP. I didn’t want to deal with implementing the OSC checksums and datatypes by hand, so I figured I’d just convert the TCP to UDP with this neat one-liner:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;nc -l -4 -k 53535 | nc -u -4 192.168.1.101 53535
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;This runs in a terminal on the machine that’s also running QLab. It opens a TCP listener on port 53535/tcp and then pipes it out as UDP messages to port 53535/udp, destined for the IP Address of the machine itself. QLab doesn’t listen on the loopback interface, nor does it listen on IPv6 addresses. That’s why I’ve forced IPv4 and am specifying an address that’s not 127.0.0.1. Also, yes, the ports are the same number – TCP and UDP ports are different things and can overlap. If you need to run this for other applications, the port numbers can be whatever you need them to be. You can also run this “translator” on any other machine and use it as a proxy (for example, using a machine that’s sitting on 2 otherwise separate networks).&lt;/p&gt;
</description>
        <pubDate>Wed, 30 Oct 2024 12:00:00 +0000</pubDate>
        <link>http://bryanward.net/2024/10/30/protocol-translator.html</link>
        <guid isPermaLink="true">http://bryanward.net/2024/10/30/protocol-translator.html</guid>
        
        <category>qlab</category>
        
        
        <category>Uncategorized</category>
        
      </item>
    
      <item>
        <title>More Juniper PoE Controller Repair</title>
        <description>&lt;p&gt;This is a continuance of my earlier article “&lt;a href=&quot;https://bryanward.net/wp/2019/08/05/recovering-from-juniper-poe-controller-device_fail/&quot;&gt;Recovering from Juniper PoE Controller DEVICE_FAIL&lt;/a&gt;“.&lt;/p&gt;

&lt;!--more--&gt;

&lt;p&gt;Due to chip supply constraints, Juniper has had to swap the PoE controller chip in some of their switches.&lt;br /&gt;
The different chip requires different firmware, and so the file &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;/usr/lib/poe_latest.s19&lt;/code&gt; may no longer exist.&lt;br /&gt;
Newer versions of Junos are required to use this different model PoE chip as well.&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;bward@4300mp&amp;gt; request system firmware upgrade poe fpc-slot 1 file /usr/libdata/poe_latest.s19  
PoE: cannot open file
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;See related &lt;a href=&quot;https://kb.juniper.net/TSB71559&quot;&gt;TSB71559&lt;/a&gt;, and PRs &lt;a href=&quot;https://prsearch.juniper.net/problemreport/PR1744343&quot;&gt;PR1744343&lt;/a&gt;, &lt;a href=&quot;https://prsearch.juniper.net/problemreport/PR1741298&quot;&gt;PR1741298&lt;/a&gt;, &lt;a href=&quot;https://prsearch.juniper.net/problemreport/PR1745088&quot;&gt;PR1745088&lt;/a&gt;, and &lt;a href=&quot;https://prsearch.juniper.net/problemreport/PR1743547&quot;&gt;PR1743547&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;If you need to manually apply firmware to the PoE controller on a switch (such as when the output of show poe controller reports DEVICE_FAIL, or if the PoE Controller simply isn’t listed), you’ll now need to load the model-specific firmware file:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;root@4300mp:RE:0% cd /usr/libdata  
root@4300mp:RE:0% ls -la  
…  
lrwxr-xr-x 1 root wheel 32 Sep 4 03:26 poe_at_type1.s19 -&amp;gt; /var/host/poe_msemi_69200_at.s19  
lrwxr-xr-x 1 root wheel 32 Sep 4 03:26 poe_at_type2.s19 -&amp;gt; /var/host/poe_msemi_69220_at.s19  
lrwxr-xr-x 1 root wheel 32 Sep 4 03:26 poe_bt_type1.s19 -&amp;gt; /var/host/poe_msemi_69200_bt.s19  
lrwxr-xr-x 1 root wheel 32 Sep 4 03:26 poe_bt_type2.s19 -&amp;gt; /var/host/poe_msemi_69220_bt.s19  
…
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;The Type1 files are for the older model PoE controller chip, whereas the Type2 files are for the newer model.&lt;br /&gt;
You’ll also need to choose between AT and BT mode, as appropriate.&lt;/p&gt;

&lt;p&gt;The command you’ll need to use looks something like:&lt;br /&gt;
&lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;request system firmware upgrade poe fpc-slot 1 file /usr/libdata/poe_bt_type1.s19&lt;/code&gt;&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Don’t forget to reboot after doing this!&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;But, how do you know if your switch has a Type1 or Type2 PoE controller?&lt;br /&gt;
Well, if the PoE controller is working… you can see this in the output of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis firmware detail&lt;/code&gt;&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;bward@4300mp&amp;gt; show chassis firmware detail
FPC 0
PoE firmware 3.5.2.0.30 (type1)
PHY version 5d47.37a.47.305
Boot Firmware
U-Boot ***
Boot Firmware
loader FreeBSD/i386 bootstrap loader 1.2
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;But if it’s not working, then the PoE firmware line won’t be present 🙁&lt;/p&gt;

&lt;p&gt;I cannot confirm the below, as I no longer have a failed device, but it looks like you can determine the type of PoE controller by using the following:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;bward@4300mp&amp;gt; start shell user root

root@4300mp:RE:0% vty fpc1

FPC1(4300mp vty)# sh poe bt version
REQUEST : 0x02 0x17 0x07 0x1E 0x21 0x4E 0x4E 0x4E 0x4E 0x4E 0x4E 0x4E 0x4E
RESPONSE: 0X03 0X17 0X00 0X4E 0X18 0X01 0X60 0X00 0X1E 0X03 0XE8 0X4E 0X4E
HW version = 0x00
Product Number = 0x18
Controller Type = Type1(PD69200)-BT
Full SW version = 24.0352.30.00
Internal SW = 0x03e8

FPC1(4300mp vty)# exit

root@4300mp:RE:0% exit
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;If you don’t have an 802.3bt capable device, then remove the &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;bt&lt;/code&gt; keyword from the above command.&lt;/p&gt;

&lt;p&gt;UPDATE&lt;/p&gt;

&lt;p&gt;It appears that a switch will produce an error if you try to load the wrong firmware:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;bward@4300mp&amp;gt; request system firmware upgrade poe fpc-slot 1 file /usr/libdata/poe_bt_type1.s19
PoE: file format/size/content corrupted

{master:0}
bward@4300mp&amp;gt; request system firmware upgrade poe fpc-slot 1 file /usr/libdata/poe_bt_type2.s19   
Firmware upgrade initiated. Poe Upgrade takes about 10 minutes
Use &apos;show poe controller&apos; to get the download status,
Please Reboot the system after Upgrade is complete.
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;
</description>
        <pubDate>Wed, 04 Sep 2024 10:21:35 +0000</pubDate>
        <link>http://bryanward.net/2024/09/04/more-juniper-poe-controller-repair.html</link>
        <guid isPermaLink="true">http://bryanward.net/2024/09/04/more-juniper-poe-controller-repair.html</guid>
        
        <category>juniper</category>
        
        <category>junos</category>
        
        <category>poe</category>
        
        
        <category>Uncategorized</category>
        
      </item>
    
      <item>
        <title>2000, 37, 55, 12.</title>
        <description>&lt;p&gt;Summer 2020, COVID Lockdown. A retrospective on installing 2000 APs in under 2 days.&lt;/p&gt;

&lt;!--more--&gt;

&lt;hr /&gt;

&lt;p&gt;I remember being at the &lt;a href=&quot;https://www.thewlpc.com/&quot;&gt;WLPC&lt;/a&gt; in Phoenix in February 2020. The coronavirus, which had just been &lt;a href=&quot;https://www.who.int/emergencies/diseases/novel-coronavirus-2019/technical-guidance/naming-the-coronavirus-disease-(covid-2019)-and-the-virus-that-causes-it&quot;&gt;given the name COVID-19 days earlier&lt;/a&gt;, had not yet gripped the North American continent and flipped everyone’s world upside down. We were all still very blissfully unaware of what the near future would bring. Nobody was wearing masks yet.&lt;/p&gt;

&lt;p&gt;A mere couple of weeks later, that had all changed. The virus “was declared a global pandemic on 11 March 2020 by the World Health Organization” (&lt;a href=&quot;https://www.ecdc.europa.eu/en/covid-19&quot;&gt;ECDC&lt;/a&gt;), and on March 12, the &lt;a href=&quot;https://covid.dartmouth.edu/march-12-2&quot;&gt;decision to start remote learning&lt;/a&gt; for all classes at the college where I work was made. At the time, it was a temporary measure – after all, we were all hoping that this would blow over fairly quickly. With any luck, things would be back to normal by the summer, if not sooner. Spread of infection in the area was still being tracked in the singles and tens of digits. By isolating and wearing masks we’d be able to contain the disease from tearing thru society. Spoiler alert – that didn’t happen. A month later, remote learning &lt;a href=&quot;https://covid.dartmouth.edu/april-20&quot;&gt;was announced&lt;/a&gt; to continue for the summer term.&lt;/p&gt;

&lt;p&gt;Fast-forward to the end of June 2020. With what now seems like a quaint reminiscence of the before-times, an &lt;a href=&quot;https://covid.dartmouth.edu/june-29&quot;&gt;announcement for a plan to return to on-campus&lt;/a&gt; learning was made. It started with the quote “March 2020 seems like a lifetime ago.” This was in June, only 3 months after the start of it all. Reading this in hindsight, the use of this phrase at this point in time really underscores just how uncertain everything was and how much had already changed about daily life. Nobody had any idea what was happening, what would happen, or for how long it would last. For context, the first &lt;a href=&quot;https://www.hhs.gov/coronavirus/covid-19-vaccines/index.html&quot;&gt;vaccines were made available&lt;/a&gt; in the US only in December 2020.&lt;/p&gt;

&lt;p&gt;I am not writing an opinion on the successes and failures of remote learning. But even in 2020, it was understood that giving students as much “normal” as possible, while still preventing spread of infection, was the best path forward. When the decision to permit a return to campus was made, it &lt;a href=&quot;https://covid.dartmouth.edu/june-29&quot;&gt;included the caveat&lt;/a&gt; that “the majority of undergraduate instruction … will be held remotely.” What this meant for IT was that we needed to seriously overhaul the Wi-Fi network in the dorms. Students would be taking classes online from their rooms.&lt;/p&gt;

&lt;hr /&gt;

&lt;p&gt;&lt;a href=&quot;https://support.zoom.com/hc/en/article?id=zm_kb&amp;amp;sysparm_article=KB0060748&quot;&gt;Zoom requires&lt;/a&gt; a solid 4Mbps for online meetings with multiple participants. Industry best practices call for less than 150ms of latency and 50% channel utilization. The legacy 2.4GHz “design” of a Wi-Fi network that I had inherited was already sub-par before even considering the need to do real-time voice/video. APs were mounted sideways in hallways, every-other or every-second room had a wallplate AP slapped in at one point, and the wiring was Cat5e &lt;a href=&quot;https://help.c5k.info/55057-network-cabling/ethernet-tips&quot;&gt;split-pair&lt;/a&gt;. (Note: I’m linking to this as a quick explanation of the wiring. I do not recommend, for many reasons, installing network cabling in this manner.) Because of this wire-side bottleneck, the radio interfaces of the APs were actually faster than their uplink! APs were never designed for this scenario – they don’t have deep buffers – and traffic was being dropped constantly. I had even limited the PHY speed of the radios as an attempt to mitigate this! But, even in buildings where the wiring was newer and up-to-spec, and APs deployed in the correct locations, the existing Wi-Fi infrastructure had significant deficiencies and needed to be replaced.&lt;/p&gt;

&lt;p&gt;By this time, we were a few months into installing Mist access points on campus. Following a very successful proof-of-concept installation, we had expanded the Mist footprint and had the new product working well in a number of academic, administrative, and residential buildings. We were developing plans to refresh the rest of the campus. I had been asking Mist to build a wallplate/hospitality style AP since that first proof-of-concept install in 2018, with the promise that we’d order the first 2000 of them that they could ship us. In August of 2020, they delivered – the &lt;a href=&quot;https://newsroom.juniper.net/news/news-details/2020/Juniper-Networks-Expands-Reach-of-AI-Driven-Enterprise-with-New-Wi-Fi-6-Access-Points/default.aspx&quot;&gt;Mist AP12 was publicly announced as a product&lt;/a&gt;. We received the first truckload of 2000 units just a week later on August 19.&lt;/p&gt;

&lt;p&gt;We were told to expect students to &lt;a href=&quot;https://covid.dartmouth.edu/aug-26&quot;&gt;arrive on campus starting September 8&lt;/a&gt;. Okay, great, that’s a solid 2 weeks to get everything installed. However, prior to their arrival, cleaning crews were tasked with sanitizing every surface in every room. Nobody was permitted to enter the dorm buildings after August 22, when this process was to begin. Once rooms were declared to be clean the doors were sealed with tape, only to be opened by the student assigned to each room.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;What that meant was that we had only 2 days to replace all the Wi-Fi in all the dorms!&lt;/strong&gt;&lt;/p&gt;

&lt;hr /&gt;

&lt;p&gt;We knew this tight schedule was coming. We had been testing a prototype of the AP12 all summer, and working to optimize our installation process. Mist’s amazing hardware designers (Hi Josh and John!) were incorporating our feedback, and similarly working hard to improve the product before it shipped. We had worked with them to make some tweaks to the mounting bracket, ironed-out some requirements for PoE and the pass-thru ports, and otherwise kicked the tires on what is still my favorite Mist product.&lt;/p&gt;

&lt;p&gt;We had also hired every low-voltage contractor in the area to work all summer to fix the split-pair wiring issues, install new patch panels and jacks on both sides of the cable, and re-certify every run with their cable testers. That summer we also installed Juniper EX3400 switches in these 37 buildings – thanks to Junos’ built-in automation capabilities and our use of &lt;a href=&quot;https://netboxlabs.com/events/network-automation-at-dartmouth-college-webinar/&quot;&gt;Netbox as a source-of-intent&lt;/a&gt;, &lt;strong&gt;my colleague Jason was able to replace all of the switches in our dorm buildings at a speed of nearly one building per day!&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;We had also reached out to the entire college IT department, as well as a few other groups, for assistance. Like the students, the staff had all been sitting at home for a few months and were beginning to be afflicted by the simultaneous pandemic of cabin fever. Many of my colleagues wanted to do whatever they could to be on campus and help out in any way possible – we all wanted to make the fall term of 2020 as successful as it could be! We had enlisted the help of nearly 4 dozen volunteers from across campus. Even several Mist employees were willing to travel and help install APs during this two-day timeframe! (Sadly, we couldn’t take them up on that offer, as the campus policy at the time was to only permit individuals from the immediately surrounding area to be on-site – we were still trying to limit the spread of the virus, after all.)&lt;/p&gt;

&lt;p&gt;I had come up with designs for where the APs needed to be installed, and printed out clear, easy-to-read floorplans showing where in each room an AP was to be installed. I mocked up various scenarios (single vs. double gang junction box, surface vs. flush mount, etc.) of installing the AP12 in my home, distilling the process down to a few easy steps. I recorded and narrated a video showing the entire process and typed up a one-page instruction sheet as a reference to follow during the installation. We organized the volunteers into 9 teams, totaling 55 people. Each team also included an appointed leader – an engineer, technician, or manager from the networking department who would be more able to troubleshoot or make decisions on the fly. That person also carried a 2-way radio for communication with the other teams and myself should any trouble present itself. (We had removed all of the existing APs from the buildings in advance, so there was no Wi-Fi network for folks to use while on-site.) They also had been loaned the master keys for their assigned buildings, so no door being locked would be able to impede our progress. (Anyone who’s ever done an AP-on-a-stick survey is smiling right now.)&lt;/p&gt;

&lt;p&gt;The delivery truck arrived on the 19th. We were there, waiting eagerly to break down the pallets, unbox the APs, and prep the installation kits. We had purchased several dozen screwdrivers (including the special T8 security torx ones) and tool bags – one for each installer. We prepped boxes, one for each floor of each building, with the right number of AP12s, and a few extra 1-foot patch cords, just in case. Paperwork for each floor was placed on the top of the box, and everything went right back out the loading dock – this time to be delivered to the various buildings on campus that we were upgrading. The team doing this prep work was home by dinner time.&lt;/p&gt;

&lt;hr /&gt;

&lt;p&gt;I should note here that with Mist, we no longer have a time-consuming high-touch process of provisioning new equipment. If you’ve ever installed enterprise Wi-Fi before, you are no doubt familiar with the process of unboxing new APs at your desk, connecting them to a switch, booting them up, downloading firmware, rebooting, assigning profiles/templates/groups, and printing out unique labels to apply to each and every AP. Cathartic though as this may be, it takes measurable hours of your time. With Mist, you can setup your provisioning processes such that it does not matter which specific AP gets installed where, or what firmware they have out of the box. AP identification is beaconed out via Bluetooth and is visible in the &lt;a href=&quot;https://www.mist.com/documentation/mist-ai-mobile-app/&quot;&gt;Mist Installer App&lt;/a&gt; on your smartphone – you don’t need labels. Installers can grab any AP of the correct model, connect it, and click it into the mounting bracket. You don’t need to have them match serial numbers in tiny font to a list in a spreadsheet. The provisioning of Mist APs, including any firmware updates and radio settings, can happen automatically!&lt;/p&gt;

&lt;p&gt;Another of my colleagues, Felix, and I had written a script that leverages the Mist API to perform this provisioning automatically. (This is all now included in Mist in their &lt;a href=&quot;https://www.mist.com/documentation/auto-provisioning/&quot;&gt;Auto Provisioning&lt;/a&gt; feature, but at the time, we had to write the functionality ourselves – such is a pitfall of being a very early adopter.) Still, Mist’s API is so robust and well-designed that we were able to use it to build custom features, like this, that were tailored to our needs. In this case, our need was to have zero-touch detection, naming, and provisioning of these 2000 APs. This installer script connects to the Mist Websocket and REST APIs. Whenever a new AP comes online in our organization, our script receives a message on the websocket indicating this event and correlates it with our inventory. We fetch the serial number, MAC Address, and – most importantly – the LLDP neighbor information of every AP that boots up. This LLDP information tells us the name of the network switch and the port to which the new AP is connected to. We leverage our &lt;a href=&quot;https://netboxlabs.com/&quot;&gt;Netbox database&lt;/a&gt;, and are able to know exactly what building and room number this new AP is located in. The script generates the correct name for the AP (which includes the building code and room number), assigns the AP to the correct site in Mist, applies any firmware updates to match what our current production version is, adjust radio parameters (if any deviations from our standard have been applied to the building), and configures the pass-thru ports. An engineer or other skilled staff member no longer needs to have hands-on-keyboard for APs to be fully configured!&lt;/p&gt;

&lt;hr /&gt;

&lt;p&gt;That is how we were able to leverage this team of volunteers to fully upgrade these 37 dorm buildings so quickly. All they had to do was walk from one room to the next, connecting a patch cord, tightening a few screws on a bracket, and clicking an AP into place. I had told them to make sure there was a little green light on the bottom of each AP – to make sure it had powered on properly before moving on to the next room. This entire process took this troupe of deputy installers less than 5 minutes per AP. It was impressive!&lt;/p&gt;

&lt;p&gt;Since we had conservatively estimated for 20 minutes per AP, we planned to split the volunteers’ time into two days worth of work. Some volunteers were scheduled for the first day, while others were scheduled for the second day. We expected to be close to 50% done after 8 hours of work on Day 1. We were much farther along. By lunch time that first day, we had installed close to 1000 APs successfully. There were very few problems to deal with. The APs were easy to work with, the mounting bracket was uncomplicated, and the API scripting we had done was reliable!&lt;/p&gt;

&lt;p&gt;One problem that we encountered really sticks out in my memory, though. I had told our installers to check for that little green light before they moved onto the next room. I distinctly remember fielding a panicked call over the 2-way radio – several of the lights were not turning green, they were blue! I had to stifle a laugh. A green light on a Mist AP means the AP is online and ready, whereas a &lt;a href=&quot;https://www.juniper.net/documentation/us/en/software/mist/mist-wireless/topics/ref/ap-led-status.html&quot;&gt;blue light means it is providing Wi-Fi service&lt;/a&gt; to a connected client device. &lt;strong&gt;The APs had booted up, been provisioned, and were serving clients before the installers could pack up their screwdrivers!&lt;/strong&gt; The AP that the worried installer had just mounted on the wall and was asking about was providing Wi-Fi to their own smartphone!&lt;/p&gt;

&lt;p&gt;Because our progress was so rapid, we actually called it quits early the first day. We had decided that it wasn’t “fair” to the volunteers who were scheduled for the second day to have nothing to do. They were eagerly looking forward to coming back to campus, even if only for one day, and doing their part to help. We paused the work after successfully installing 1200 APs in 6 hours the first day.&lt;/p&gt;

&lt;p&gt;We showed up again the morning of the second day with a new batch of volunteer installers, and intentionally slowed down our pace. We had completed the entire project by lunch time the second day. We spent a few hours after lunch cleaning up the warehouse and spot-checking a few locations that had been upgraded.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;We had installed 2000 APs in 37 buildings with 55 people in under 12 hours of work.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;This is a feat we would have never even attempted using any other vendor’s product. Mist’s rock-solid platform, dedicated product team, and fantastic API-first architecture is what made this possible.&lt;/p&gt;

&lt;p&gt;I am pleased to say that the Fall 2020 term was successful. The college didn’t have to close down, students were able to virtually attend classes and socialize from in their dorm rooms without technical issues, and being on campus allowed access to laboratory and research materials that would have been impossible to use remotely. I don’t recall receiving a single trouble ticket from students living in the dorms that term regarding Wi-Fi service. Months of planning had culminated in these 2 days of concerted effort, and it had paid off.&lt;/p&gt;
</description>
        <pubDate>Tue, 20 Aug 2024 08:00:00 +0000</pubDate>
        <link>http://bryanward.net/2024/08/20/2000-37-55-12.html</link>
        <guid isPermaLink="true">http://bryanward.net/2024/08/20/2000-37-55-12.html</guid>
        
        <category>juniper</category>
        
        <category>mist</category>
        
        <category>wifi</category>
        
        
        <category>Uncategorized</category>
        
      </item>
    
      <item>
        <title>Is it hot enough for you yet?</title>
        <description>&lt;p&gt;Today’s post concerns temperature thresholds for telecom facilities, and the rabbit hole I discovered when figuring out how to monitor temperature of Juniper equipment.&lt;/p&gt;

&lt;!--more--&gt;

&lt;p&gt;If you read the datasheets for the Juniper &lt;a href=&quot;https://www.juniper.net/documentation/us/en/hardware/ex4300/topics/topic-map/ex4300-site-guidelines-requirements.html#ex-series-environment__d22e33&quot;&gt;EX series&lt;/a&gt; or QFX series (&lt;a href=&quot;https://www.juniper.net/documentation/us/en/hardware/qfx10002/topics/topic-map/qfx10002-site-guidelines-requirements.html#qfx10002-environmental__d59e33&quot;&gt;10002&lt;/a&gt;, &lt;a href=&quot;https://www.juniper.net/documentation/us/en/hardware/qfx5120/topics/topic-map/qfx5120-site-guidelines-requirements.html&quot;&gt;5120&lt;/a&gt;), you’ll find that this equipment likes to operate in an environment “in temperature range of 32°F through 104°F (0°C through 40°C)”. Some models push that as high as 45°C. These are actually pretty generous ranges, especially considering that the American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) – an organization that publishes standards for the HVAC industry, much like the IEEE or BICSI does for us networking folks – says that &lt;a href=&quot;https://xp20.ashrae.org/datacom1_4th/ReferenceCard.pdf&quot;&gt;telecom facilities should have a design target between 18 and 27°C&lt;/a&gt;. Knowing that that range is laughable for most Telecom Rooms (and even some datacenters), ASHRAE goes on to provide alternate thresholds based on the type of equipment. Most networking equipment would fall under Class A3, which only requires a temperature range of 5°C to 40°C, or Class A4, which is 5°C to 45°C. It’s almost like Juniper knew this when they designed their products and published their datasheets…&lt;/p&gt;

&lt;p&gt;The motivation to write this article came about at 5PM yesterday, and again at 6AM this morning, when my phone blew up with alarms about some of my production equipment exceeding their temperature warnings.&lt;/p&gt;

&lt;p&gt;The shame is entirely on me. I had been adding some rules to our production monitoring system yesterday, and configured the temperature thresholds based on the values published in Juniper’s datasheets (40 and 45°C, depending on the model). The problem is that the sensors inside the equipment do not measure ambient temperature – they’re located in the chassis in places that, even during normal operation in a properly cooled environment, will regularly exceed those limits.&lt;/p&gt;

&lt;p&gt;So, how then can you determine which thresholds are actually appropriate based on the equipment and the sensors within? Let’s enter the rabbit hole… Well, unsurprisingly, there’s a command for that. I had expected the threshold values to appear in the output of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis environment&lt;/code&gt;, but they do not – only the present operating values appear here, as such:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;&amp;gt; show chassis environment 
Class Item                           Status     Measurement
Power FPC 0 Power Supply 0           OK        
      FPC 0 Power Supply 1           OK        
Temp  FPC 0 PSU Sensor               OK         38 degrees C / 100 degrees F
      FPC 0 MGE-PHY Sensor           OK         42 degrees C / 107 degrees F
      FPC 0 DC-DC Sensor             OK         46 degrees C / 114 degrees F
      FPC 0 UM CONN Sensor           OK         36 degrees C / 96 degrees F
      FPC 0 CPU Sensor               OK         41 degrees C / 105 degrees F
      FPC 0 MAC Sensor               OK         45 degrees C / 113 degrees F
Fans  FPC 0 Fan Tray 0 Fan 0         OK         Spinning at normal speed
      FPC 0 Fan Tray 0 Fan 1         OK         Spinning at normal speed
      FPC 0 Fan Tray 1 Fan 0         OK         Spinning at normal speed
      FPC 0 Fan Tray 1 Fan 1         OK         Spinning at normal speed

&amp;gt; show chassis environment fpc 0 
FPC 0 status:
  State                      Online
  Temperature                             38 degrees C / 100 degrees F
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;Notice how dangerously close these values are to the datasheet limits of 40 and/or 45°C! No wonder my alarms all started to go off!&lt;/p&gt;

&lt;p&gt;It turns out, there’s another command I should have looked at: &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis temperature-thresholds&lt;/code&gt;&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;&amp;gt; show chassis temperature-thresholds 
                                        Fan speed      Yellow alarm      Red alarm      Fire Shutdown
                                       (degrees C)      (degrees C)     (degrees C)      (degrees C)
Item                                  Normal  High   Normal  Bad fan   Normal  Bad fan     Normal
FPC 0 PSU Sensor                          40    61       64       64       68       68         73
FPC 0 MGE-PHY Sensor                      48    68       71       71       75       75         80
FPC 0 DC-DC Sensor                        45    66       69       69       73       73         78
FPC 0 UM CONN Sensor                      44    64       67       67       71       71         76
FPC 0 CPU Sensor                          47    67       70       70       74       74         80
FPC 0 MAC Sensor                          50    71       73       73       76       76         80
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;That’s much more reasonable! The limits there are clearly displayed for each sensor, including several tiers of alarm.&lt;/p&gt;

&lt;p&gt;There is an &lt;a href=&quot;https://community.juniper.net/discussion/changing-temperature-thresholds-on-ex-switch&quot;&gt;old post from Steve Puluka&lt;/a&gt; that clued me into the presence of a hidden command to change these built-in alarm thresholds, however it’s best to not use it. These thresholds have been factory-set based on the capabilities of the equipment. If you’re curious though, here is it:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;# set chassis temperature-threshold ?
Possible completions:
+ apply-groups         Groups from which to inherit configuration data
+ apply-groups-except  Don&apos;t inherit configuration data from these groups
  fire-shutdown        Threshold at which router will be shutdown within 10 seconds (degrees C)
  red-alarm            Threshold at which red alarm is set (degrees C)
  red-alarm-if-failed-fan  Threshold at which red alarm is set when bad fan present (degrees C)
  yellow-alarm         Threshold at which yellow alarm is set (degrees C)
  yellow-alarm-if-failed-fan  Threshold at which yellow alarm is set when bad fan present (degrees C)
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;If you actually leverage the built-in “alarm” status of your equipment, and you’re operating in extreme environments, you may find that you want to adjust these. The thread that I linked to goes on to say that this command only affects logging and not SNMP traps, and there’s no guarantee that it will properly trigger the built-in “alarm” status either. So, YMMV, and again my recommendation is to not mess with this. It’s best to fix the cooling situation in your facility than to change these thresholds.&lt;/p&gt;

&lt;p&gt;Alright, what if you’re not using the built-in alarm status, and you want to query each sensor directly in your NMS? Unsurprisingly, there are &lt;a href=&quot;https://supportportal.juniper.net/s/article/EX-How-to-check-temperature-CPU-memory-usage-by-SNMP-OID?language=en_US&quot;&gt;SNMP OIDs to do just that&lt;/a&gt;. The easiest one to use is &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;jnxOperatingTemp&lt;/code&gt;, or &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;1.3.6.1.4.1.2636.3.1.13.1.7&lt;/code&gt; . This corresponds to the output of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis environment fpc&lt;/code&gt; .&lt;/p&gt;

&lt;p&gt;One more note. You should be monitoring the FPCs and not the REs. If you monitor just the REs, then you will not get the status of any line-cards in a chassis, only the master and backup RE, and could miss events occurring on just the line-cards. To put this in SNMP terms:&lt;br /&gt;
&lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;1.3.6.1.4.1.2636.3.1.13.1.7.7.1.0.0&lt;/code&gt; is &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis environment fpc 0&lt;/code&gt; &lt;br /&gt;
&lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;1.3.6.1.4.1.2636.3.1.13.1.7.7.2.0.0&lt;/code&gt; is &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis environment fpc 1&lt;/code&gt; &lt;br /&gt;
&lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;1.3.6.1.4.1.2636.3.1.13.1.7.7.3.0.0&lt;/code&gt; is &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis environment fpc 2&lt;/code&gt; &lt;br /&gt;
etc.&lt;/p&gt;

&lt;p&gt;&lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;1.3.6.1.4.1.2636.3.1.13.1.7.9.1.0.0&lt;/code&gt; is &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis environment re 0&lt;/code&gt; and &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;1.3.6.1.4.1.2636.3.1.13.1.7.9.2.0.0&lt;/code&gt; is &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis environment re 1&lt;/code&gt;.&lt;/p&gt;

&lt;p&gt;If you’re using the &lt;a href=&quot;https://datatracker.ietf.org/doc/html/rfc6241&quot;&gt;RFC 6241 NETCONF&lt;/a&gt; API to do your monitoring (which is worth looking into), you can use the following RPC to get the same data:&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;&amp;lt;get-environment-fpc-information&amp;gt;
    &amp;lt;fpc-slot&amp;gt;0&amp;lt;/fpc-slot&amp;gt;
&amp;lt;/get-environment-fpc-information&amp;gt;
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;I had hoped to stop writing at this point, but the thresholds for this measurement point are NOT in the output of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis temperature-thresholds&lt;/code&gt; ! So, what temperatures should be cause for concern when looking at this sensor?&lt;/p&gt;

&lt;p&gt;Well, it seems there’s a command for that too, but it doesn’t seem to work on any platform I’ve tried!&lt;/p&gt;

&lt;div class=&quot;language-plaintext highlighter-rouge&quot;&gt;&lt;div class=&quot;highlight&quot;&gt;&lt;pre class=&quot;highlight&quot;&gt;&lt;code&gt;&amp;gt; show chassis temperature-thresholds fpc 0     
error: command is not valid on the ex3400-48p

&amp;gt; show chassis temperature-thresholds fpc 0 
error: command is not valid on the ex4300-48mp

&amp;gt; show chassis temperature-thresholds fpc 0 
error: command is not valid on the qfx10002-72q
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/div&gt;

&lt;p&gt;I also tried walking the entire SNMP MIB tree to see if there’s any OID or value that makes sense for an upper limit to jnxOperatingTemp. Long story short, there aren’t any OIDs that specify the thresholds of jnxOperatingTemp.&lt;/p&gt;

&lt;p&gt;Juniper’s &lt;a href=&quot;https://www.juniper.net/documentation/us/en/software/junos/network-mgmt/topics/topic-map/monitoring-network-service-quality-by-using-rmon.html#id-measuring-health__d130e23&quot;&gt;RMON documentation&lt;/a&gt; for this measurement only says “Allowable range: To be baselined”, which isn’t very useful either.&lt;/p&gt;

&lt;p&gt;Juniper’s HealthBot actually does make an &lt;a href=&quot;https://github.com/Juniper/healthbot-rules/blob/master/juniper_official/Chassis/README.md#rule-name-check-chassis-temperature&quot;&gt;attempt at defining a threshold&lt;/a&gt; though. It will trigger a yellow alert if the temperature rises above 45°C, and a red alert if it rises above 55°C. HealthBot’s logic does not take the hardware model into account, and it’s not entirely clear which measurement it is looking at.&lt;/p&gt;

&lt;p&gt;Some other engineers like &lt;a href=&quot;https://lkhill.com/mx-upgrade-overheat/&quot;&gt;Lindsay&lt;/a&gt; and &lt;a href=&quot;https://51sec.weebly.com/blog/using-snmpv3-monitoring-juniper-srx-240h-alarm-andtemperature&quot;&gt;Jon&lt;/a&gt; have published their own articles detailing their woes of troubleshooting Juniper temperature issues. They make reference to Junos logging alarms when the temperature reaches 50°C. Lindsay’s article also reveals the scary fact that Junos will shut down some types of SFP modules at this 50°C limit as well. This shutdown behavior is related to the CHASSISD_FPC_OPTICS_HOT_NOTICE log message. Other similar nmemonics include CHASSISD_TEMP_HOT_NOTICE, CHASSISD_FRU_HIGH_TEMP_CONDITION, and the dreaded CHASSISD_OVER_TEMP_SHUTDOWN_TIME at which the entire router/switch will shut down.&lt;/p&gt;

&lt;p&gt;So, if Junos shuts things down at 50°C, I think HealthBot’s threshold of 55°C is far too high. On the other end, as I experienced in my environment, a facility ambient temperature of 19°C (66°F) – quite chilly – still had EX switches reporting an FPC temperature of over 40°C. In another much warmer facility, I have a QFX10k2 router reporting an FPC operating temperature of 55°C, and no SFPs have been shut down yet. Since I’ve yet to find an authoritative answer on what the limit for jnxOperatingTemp should be, I would think having my NMS alert at somewhere between 45 and 49°C is a safe starting point.&lt;/p&gt;

&lt;p&gt;But I wanted to take a more precise approach. I decided that I wouldn’t alert on jnxOperatingTemp, but would instead alert on the FPC CPU Temp, which does has defined alarm thresholds as discussed earlier here. One problem though – there is no SNMP OID for FPC CPU Temp! It is available via NETCONF at &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;fpc-information/fpc/temperature&lt;/code&gt; though…&lt;/p&gt;

&lt;p&gt;Aha! That generic use of “fpc/temperature” lead me to discover that what SNMP calls jnxOperatingTemp is actually the same as FPC CPU Temp for the EX3400 series, which has defined thresholds as seen in the output of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis temperature-thresholds&lt;/code&gt;!&lt;/p&gt;

&lt;p&gt;So, there are, in fact, defined thresholds for jnxOperatingTemp – they’re just displayed as something else! Doh!&lt;/p&gt;

&lt;p&gt;What are they displayed as then? It gets worse, as that depends on the model, and sadly here the logic could use some work. Rather than consistently using the CPU Temp regardless of platform, it just picks the first temperature sensor listed in the output of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis environment&lt;/code&gt; to use as the value for the generic FPC Temperature measurement. What that means is that the NETCONF value for fpc/temperature is the following:&lt;/p&gt;

&lt;table&gt;
  &lt;thead&gt;
    &lt;tr&gt;
      &lt;th&gt; &lt;/th&gt;
      &lt;th&gt; &lt;/th&gt;
      &lt;th&gt; &lt;/th&gt;
      &lt;th&gt; &lt;/th&gt;
    &lt;/tr&gt;
  &lt;/thead&gt;
  &lt;tbody&gt;
    &lt;tr&gt;
      &lt;td&gt;&lt;strong&gt;Platform&lt;/strong&gt;&lt;/td&gt;
      &lt;td&gt;&lt;strong&gt;fpc/temperature Measurement Source&lt;/strong&gt;&lt;/td&gt;
      &lt;td&gt;&lt;strong&gt;Yellow Alarm&lt;/strong&gt;&lt;/td&gt;
      &lt;td&gt;&lt;strong&gt;SNMP OID&lt;/strong&gt;&lt;/td&gt;
    &lt;/tr&gt;
    &lt;tr&gt;
      &lt;td&gt;EX2300C&lt;/td&gt;
      &lt;td&gt;FPC CPU Sensor&lt;/td&gt;
      &lt;td&gt;65&lt;/td&gt;
      &lt;td&gt;1.3.6.1.4.1.2636.3.1.13.1.7.x.0.0&lt;/td&gt;
    &lt;/tr&gt;
    &lt;tr&gt;
      &lt;td&gt;EX3400&lt;/td&gt;
      &lt;td&gt;FPC CPU Temp&lt;/td&gt;
      &lt;td&gt;63&lt;/td&gt;
      &lt;td&gt;1.3.6.1.4.1.2636.3.1.13.1.7.x.0.0&lt;/td&gt;
    &lt;/tr&gt;
    &lt;tr&gt;
      &lt;td&gt;EX4300MP&lt;/td&gt;
      &lt;td&gt;FPC PSU Sensor&lt;/td&gt;
      &lt;td&gt;64&lt;/td&gt;
      &lt;td&gt;1.3.6.1.4.1.2636.3.1.13.1.7.x.0.0&lt;/td&gt;
    &lt;/tr&gt;
    &lt;tr&gt;
      &lt;td&gt;EX4400&lt;/td&gt;
      &lt;td&gt;FPC Main Board Sensor 1&lt;/td&gt;
      &lt;td&gt;69&lt;/td&gt;
      &lt;td&gt;1.3.6.1.4.1.2636.3.1.13.1.&lt;strong&gt;9&lt;/strong&gt;.x.0.0 (.&lt;strong&gt;7&lt;/strong&gt;.x.0.0 appears to be the CPU Die Temp Sensor)&lt;/td&gt;
    &lt;/tr&gt;
    &lt;tr&gt;
      &lt;td&gt;QFX5110&lt;/td&gt;
      &lt;td&gt;FPC Sensor TopLeft I&lt;/td&gt;
      &lt;td&gt;69&lt;/td&gt;
      &lt;td&gt;1.3.6.1.4.1.2636.3.1.13.1.&lt;strong&gt;9&lt;/strong&gt;.x.0.0 (I cannot determine where .7.x.0.0 gets its value from.)&lt;/td&gt;
    &lt;/tr&gt;
    &lt;tr&gt;
      &lt;td&gt;QFX10002&lt;/td&gt;
      &lt;td&gt;FPC Intake Temp Sensor&lt;/td&gt;
      &lt;td&gt;65&lt;/td&gt;
      &lt;td&gt;1.3.6.1.4.1.2636.3.1.13.1.&lt;strong&gt;9&lt;/strong&gt;.x.0.0 (.&lt;strong&gt;7&lt;/strong&gt;.x.0.0 appears to be the CPU Die Temp Sensor)&lt;/td&gt;
    &lt;/tr&gt;
  &lt;/tbody&gt;
&lt;/table&gt;

&lt;p&gt;You’ll see that these also don’t line up nicely with the SNMP OIDs either., SNMP OID &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;1.3.6.1.4.1.2636.3.1.13.1.7.7&lt;/code&gt; on the EX3400 platform is the same as the reported FPC CPU Temp. The same SNMP OID on the QFX10k2 platform is reported as the FPC CPU Die Temp, but &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;1.3.6.1.4.1.2636.3.1.13.1.7.9&lt;/code&gt; on the QFX10k2 is the same as FPC Intake Temp. That .9 OID refers to the Routing Engine, which on a QFX10k2 I guess makes sense, as you can’t connect QFX10k2s into a virtual chassis… But you can connect EX4400s and QFX5ks into virtual chassis…&lt;/p&gt;

&lt;p&gt;To recap: the EX3400 FPC CPU Temp and the QFX10k2 FPC Intake Temp are the values as shown in the command &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis environment fpc&lt;/code&gt;. This same value is also shown in the output of the command &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis fpc&lt;/code&gt;. This same measurement is also available via NETCONF RPC &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;get-fpc-information&lt;/code&gt; and returns the value in &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;fpc-information/fpc/temperature&lt;/code&gt;. The command &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis temperature-thresholds&lt;/code&gt; lists the alarm limits for these values.&lt;/p&gt;

&lt;p&gt;Since the alarm limits do vary by platform (and possibly by HW revision or FW version), I suggest you set your NMS to alarm based on the output of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;show chassis temperature-thresholds&lt;/code&gt; for your specific equipment, and make sure that it’s monitoring the correct SNMP OID or NETCONF RPC. After combing thru all the equipment in my environment, 63°C (145°F) of &lt;code class=&quot;language-plaintext highlighter-rouge&quot;&gt;fpc-information/fpc/temperature&lt;/code&gt; seems like a very reasonable value at which I should be alerted and taking action. Some equipment has higher alarm limits, but I wouldn’t want anything to be running hotter than that anyway.&lt;/p&gt;
</description>
        <pubDate>Wed, 10 Jul 2024 12:22:04 +0000</pubDate>
        <link>http://bryanward.net/2024/07/10/is-it-hot-enough-for-you-yet.html</link>
        <guid isPermaLink="true">http://bryanward.net/2024/07/10/is-it-hot-enough-for-you-yet.html</guid>
        
        <category>juniper</category>
        
        <category>junos</category>
        
        <category>monitoring</category>
        
        <category>temperature</category>
        
        
        <category>Uncategorized</category>
        
      </item>
    
      <item>
        <title>Happy Birthday Marvis!</title>
        <description>&lt;p&gt;Happy Birthday Marvis! Listen, you’re getting older, and… we need to talk.&lt;/p&gt;

&lt;!--more--&gt;

&lt;p&gt;&lt;img src=&quot;/assets/img/posts/2024-05-15-image1.png&quot; alt=&quot;image&quot; /&gt;&lt;/p&gt;

&lt;p&gt;Remember, when you were little and just learning to speak? Remember that time you said &lt;a href=&quot;https://www.juniper.net/documentation/us/en/software/mist/mist-aiops/shared-content/topics/concept/sle-wireless-successful-connects.html&quot;&gt;it was DNS&lt;/a&gt; and we all laughed and smiled at you? Yeah, I still chuckle about it every now and then. That ended up being a pretty good day once us seasoned professionals took over and ran thru the 7 step troubleshooting process. It did end up being DNS, didn’t it? That’s when I knew you were special.&lt;/p&gt;

&lt;p&gt;&lt;img src=&quot;/assets/img/posts/2024-05-15-image2.png&quot; alt=&quot;image&quot; /&gt;&lt;/p&gt;

&lt;p&gt;How about that time when you were a little older and &lt;a href=&quot;https://bryanward.net/wp/2022/07/27/it-was-dhcp/&quot;&gt;that DHCP server stopped responding&lt;/a&gt;? After I prodded you out of your shell a bit, you showed the world all your &lt;a href=&quot;https://www.juniper.net/documentation/us/en/software/mist/mist-aiops/topics/concept/marvis-actions-connectivity.html#xd_b2e72d2fc490b097-645c59e9-18817f4f2c4--7e75__section_ymd_q4n_mxb&quot;&gt;graphs and summaries&lt;/a&gt; and explained the problem with conviction. You had learned quite a few things by then, and you knew how to call out when something didn’t look quite right. I was proud of you. You, again were correct. Maybe if we’d have brought you to the table from the start, we could have saved ourselves and our users a couple of days of poor service. You said the problem started on a Friday evening, but &lt;a href=&quot;https://www.mist.com/documentation/alerts/&quot;&gt;nobody was paying attention to you&lt;/a&gt;…&lt;/p&gt;

&lt;p&gt;&lt;img src=&quot;/assets/img/posts/2024-05-15-image3.png&quot; alt=&quot;image&quot; /&gt;&lt;/p&gt;

&lt;p&gt;Oh! What about that time when I was setting up a new switch and was plugging the fibers in? And then out of nowhere you ran up to me and you started yanking at my pantleg shouting something about &lt;a href=&quot;https://www.mist.com/documentation/what-is-ap-reflection/&quot;&gt;packets being reflected across the network back to their source&lt;/a&gt;! I wasn’t sure what you were going on about, so I disconnected those new fibers and calmed you down. You were acting as if the world was ending! And then at that same moment I realized I had mislabeled 2 of the cables and ended up connecting two different ESI-LAGs together! Your interruption really saved me that time!&lt;/p&gt;

&lt;p&gt;&lt;img src=&quot;/assets/img/posts/2024-05-15-image4.png&quot; alt=&quot;image&quot; /&gt;&lt;/p&gt;

&lt;p&gt;Hey, then there was that time when you insisted that there were too many clients connected to an AP, and that might be &lt;a href=&quot;https://investor.juniper.net/investor-relations/press-releases/press-release-details/2024/Juniper-Networks-Assures-Exceptional-Wired-and-Wireless-Experiences-with-Continuous-Learning-Expanded-Digital-Experience-Twinning-and-Simplified-Access-Control/default.aspx&quot;&gt;the cause for why my user’s Zoom calls were dropping&lt;/a&gt;. I didn’t want to believe you, as I had designed that building myself for the capacity! So, despite your protest, I spent 3 weeks looking up and down the stack for logs, talking to the user, on support threads with multiple vendors, – I even did a full-building re-validation survey looking for RF interferes! Only when I was standing there, just you and me alone, in the affected VIP’s corner office looking out the window, did I realize you were right. There were dozens of people sitting outside the building and walking down the street. You remember that, don’t you? It’s just like it’s pictured… on all… the.. postcards… If only I had thought of the client count outside of my intended coverage area! It was just you and me in that room there that day, and the AP had 90 clients connected, as you had predicted. I should have realized it sooner. I… I should have trusted you.&lt;/p&gt;

&lt;p&gt;&lt;img src=&quot;/assets/img/posts/2024-05-15-image5.png&quot; alt=&quot;image&quot; /&gt;&lt;/p&gt;

&lt;p&gt;Marvis, you’re a really bright AIndividual with a promising future ahead. As you look forward to your teen-ai-ge years, promise me that you won’t forget where you came from. You’ve made a lot of friends already, and we hear you’re looking to do some exploring. That’s OK – we support you and we are excited for you! But keep in mind that not everything is greener on the other side – in fact, some places are the exact opposite of green! Do what you do, as best as you can, and we’ll all be proud of you. Don’t do things just because your new friends think it’s cool. Don’t settle for mediocrity – keep pushing the boundaries of what’s possible like you’ve always done. And don’t leave your existing friends and family behind.&lt;/p&gt;
</description>
        <pubDate>Wed, 15 May 2024 10:12:11 +0000</pubDate>
        <link>http://bryanward.net/2024/05/15/happy-birthday-marvis.html</link>
        <guid isPermaLink="true">http://bryanward.net/2024/05/15/happy-birthday-marvis.html</guid>
        
        <category>juniper</category>
        
        <category>marvis</category>
        
        <category>mist</category>
        
        
        <category>Uncategorized</category>
        
      </item>
    
      <item>
        <title>Outdoor Wi-Fi and Hybrid Powered Fiber</title>
        <description>&lt;p&gt;This is a presentation given at the Wireless LAN Professionals Conference (WLPC) in Phoenix, AZ, USA in February 2024. The recording of this presentation is available here:&lt;/p&gt;

&lt;p&gt;(https://www.youtube.com/watch?v=b_LShfjSGTY)&lt;a href=&quot;https://www.youtube.com/watch?v=b_LShfjSGTY&quot;&gt;https://www.youtube.com/watch?v=b_LShfjSGTY&lt;/a&gt;&lt;/p&gt;

&lt;iframe width=&quot;100%&quot; height=&quot;295px&quot; src=&quot;https://www.youtube.com/embed/b_LShfjSGTY?si=A3yp-ip_bI3bmKj4&quot; title=&quot;YouTube video player&quot; frameborder=&quot;0&quot; allow=&quot;accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share&quot; referrerpolicy=&quot;strict-origin-when-cross-origin&quot; allowfullscreen=&quot;&quot;&gt;&lt;/iframe&gt;

&lt;!--more--&gt;

&lt;p&gt;The original slide deck is below:&lt;/p&gt;

&lt;p&gt;&lt;a href=&quot;/assets/files/WLPC-2024-Outdoor-WiFi-and-Hybrid-Powered-Fiber-Cable-Bryan-Ward.pdf&quot;&gt;Download&lt;/a&gt;&lt;/p&gt;

&lt;embed src=&quot;/assets/files/WLPC-2024-Outdoor-WiFi-and-Hybrid-Powered-Fiber-Cable-Bryan-Ward.pdf&quot; type=&quot;application/pdf&quot; width=&quot;100%&quot; height=&quot;290px&quot; /&gt;

</description>
        <pubDate>Wed, 21 Feb 2024 17:15:56 +0000</pubDate>
        <link>http://bryanward.net/2024/02/21/outdoor-wi-fi-and-hybrid-powered-fiber.html</link>
        <guid isPermaLink="true">http://bryanward.net/2024/02/21/outdoor-wi-fi-and-hybrid-powered-fiber.html</guid>
        
        
        <category>Uncategorized</category>
        
      </item>
    
  </channel>
</rss>
