Recommended Posts

I'm having a problem with my internet connectivity with pfSense, my WAN interface keeps going offline after a set of actions that regularly appear in the logs.

 

My WAN interface remains up at all times, it has a public IP via DHCP from my ISP, via a bridged modem. I think this repeating offline/online cycle is affecting my line speed and I can't reach anywhere near my lines capability. Description of problem below, can anyone advise what is going on or how to troubleshoot this?

 

This is a screenshot of the log showing the actions that appear to coincide with the WAN gateway going offline. I'm not saying this is the cause, just that this happens repeatedly at the same time, not sure if cause or effect, or I suppose, if even unrelated to the gateway going offline.

Screen Shot 2016-08-21 at 09.39.15 redacted.png

 

And this is what the gatway log shows for the WAN going offline. Note the times of the instances match those of the set of actions in the system log.

Screen Shot 2016-08-21 at 09.38.20 redacted.png

 

Any help appreciated, this is doing my head in! :yes:

 

rancid

if your wan goes offline, or the monitoring system thinks it does because of high latency then yeah it can reset the states and disconnect you..  This can be a problem if your starting to load up your pipe and your buffer bloat causes your latency to go way up.. So monitor thinks your gateway is offline and resets all the states.

 

You can turn that function off here

 

System / Advanced / Miscellaneous

flushstates.jpg

 

Other option is to disable gateway monitoring.. But that is not good idea.  Other option is work with traffic shaping to prevent buffer bloat increasing the latency to your gateway IP your monitoring, etc..  Another option is to change the monitor values to really really high latency doesn't count as your gateway being offline so the states flush.. The quick easy fix is to just uncheck that box ;)

 

That function is really for when you have more than 1 wan, and you want to use say your 2nd wan when 1st goes down then yeah you would want to flush all the states on that 1st gateway so your clients create new connections via the 2nd failover gateway, etc..  But in a 1 wan configuration that option really has no use, not really a reason to reset your states..  And as you can see if monitoring thinks your connection is down because your latency exceeds a specific threshold wack go all your sessions ;)

 

 

  • Like 1

That sounds logical, only problem is that box is unchecked already! :s

 

Also, not sure if it is related to loading the pipe up, it seems to be more time based, as in every half hour it does this, regardless of what I am doing.

 

Any thoughts?!

clearly dpinger is giving you a warning about your connectivity - showing 20% packet loss, etc.  So maybe your wan connection is just sucks?  Did you contact your isp?

 

Just because the interface is up doesn't mean your not having line issues..

 

What does your quality graph look like.. Also you have ipv6 on your wan - are you using it?  If not you could try turning off ipv6 on pfsense wan..

packetloss.jpg

 

see last night I had a bit of a problem for a bit.. Connection just went belly up for a bit.. Interface never went down.. But pinging to gateway that dpinger monitors wasn't answering.. See normally I get about 10ms response time to my gateway..

 

Here is same graph for my ipv6 tunnel.

 

tunnel.jpg

 

you can see same exact time having a problem with it.. This rides on top of my ipv4 connection, so not just gateway wasn't answering ping, etc..

 

Here is my connection to my vps via a vpn connection..

vpnconnection.jpg

 

You can see same exact time some sort of issue!!  It has some minor connection issues later while you don't see those on the other graphs for normal ipv4 wan, my ipv6 tunnel, etc..  So what does your quality graph look like when you say you have these problems?

 

if I zoom in

zoomin.jpg

 

You can see I was offline for a few minutes.. I was not actually on then so didn't notice it..

 

From log you can see showed same alarms as you, and then they cleared once the packetloss dropped below threshold..

logclear.jpg

 

Normally my connection is rock solid stable - but as of late they have been having some minor hiccups now and then.. It think its prob related to the gig rollout that is coming to chicagoland from comcast.. I have my name on list.. Going to be freaking sweet!!! ;)

 

This topic is now closed to further replies.
  • Posts

    • BrowserOS 0.46.0 by Razvan Serea BrowserOS is a free, open-source Chromium-based browser that runs AI agents natively, offering a smarter, more productive browsing experience. It supports Chrome extensions and integrates AI agents to automate tasks, fill forms, and streamline workflows. Your data stays on your computer: you can use your own API keys or run local models via Ollama, making it a privacy-first alternative to tools like Perplexity, Comet, or Dia. With built-in productivity tools and app integrations, BrowserOS boosts efficiency while keeping control firmly in your hands. Being Chromium-based, BrowserOS lets you effortlessly import your bookmarks, passwords, and Chrome extensions in just a few clicks. BrowserOS works with OpenAI GPT models, Anthropic Claude, Google Gemini, and local AI models via Ollama or LMStudio. You can use your own API keys and effortlessly switch between providers. BrowserOS Agent Your AI productivity assistant that organizes and manages your browsing effortlessly Quickly list, group, or close tabs Save and resume browsing sessions Search your history and organize bookmarks Switch instantly to the tab you need BrowserOS Navigator – Automate web tasks with ease Navigate websites and search automatically Interact with pages without manual effort Handle repetitive tasks in seconds What makes BrowserOS special Feels like home - same familiar interface as Google Chrome, works with all your extensions AI agents that run on YOUR browser, not in the cloud Privacy first - bring your own keys or use local models with Ollama. Your browsing history stays on your computer Open source and community driven - see exactly what's happening under the hood MCP store to one-click install popular MCPs and use them directly in the browser bar (coming soon) Built-in AI ad blocker that works across more scenarios! BrowserOS 0.46.0 changelog: Run Claude Code & Codex right in your browser — We've extended the agent harness to bring full coding agents into BrowserOS. Claude Code and Codex now come bundled and plug straight into the assistant, so you can drive your browser with the agent — and the subscription — you already use. A brand new experience — A redesigned new tab, a calmer composer, and a rebuilt command center for switching between agents. The whole assistant is cleaner, faster to reach, and easier to live in. New MCP tools — We rebuilt the browser tool surface from the ground up — a tighter, more reliable set of tools for agents to drive the browser. Plus one-click install of BrowserOS as an MCP server into the agents you already run, with automatic URL sync. Chromium 148 — Updated to the latest Chromium base with all recent upstream fixes and security patches. Streamlined — We've pulled back a few features that weren't getting much use — Skills, Soul, and Memory — so we can focus and ship better versions of them soon. Download: BrowserOS 0.46.0 | 181.0 MB (Open Source) Download: BrowserOS for macOS | 485.0 MB Links: BrowserOS Homepage | Github | Screenshot Get alerted to all of our Software updates on Twitter at @NeowinSoftware
    • Microsoft finally admits its default Windows 11 25H2, 24H2 action broke key legacy component by Sayan Sen Microsoft last week released Windows 11 KB5094126 and KB5093998 as the latest Patch Tuesday updates. Following that the company also published the accompanying dynamic updates under KB5094149, KB5095971, and KB5094156. So far the company has acknowledged two known issues that have popped up after the release which include bugged-out Office apps as well as the Recycle Bin; though there could be more at play too. Speaking of bugs and issues, Microsoft seems to have finally acknowledged a problem that probably has been around for close to a year. That's because back in July of 2025 the company made a default change to the latest Windows 11 versions, wherein it switched to JScript9Legacy on Windows 11 24H2 and later releases. Hence following the release of version 25H2 in October 2025, JScript9Legacy also remained default-enabled. As a result there has been a compatibility issue ever since then. For those wondering, by switching to JScript9Legacy Microsoft intended to improve the security of modern Windows PCs by reducing vulnerabilities tied to legacy scripting like cross-site scripting (XSS), among others. XSS exploits can allow cyber-attackers to attach malicious code onto legitimate websites and use them to execute the code when a potential victim loads such a website. Hence the new JScript9Legacy engine enforced stricter execution policies and improved object handling, which should help mitigate such attacks. Microsoft today has published a new support article detailing the problem. Neowin spotted it while browsing. The company says that JScript global definitions and execution context may fail to persist across scripts, potentially breaking older dependent apps and web-based components that relied on this legacy behavior. In the article Microsoft has confirmed that the issue stems from its move away from the older jscript9.dll engine in favor of jscript9legacy.dll. As mentioned above, while the newer engine was designed to address vulnerabilities and strengthen security it also changes how JScript handles execution context. As a result functions and definitions loaded by one script could no longer remain available to subsequent scripts once execution ended. The company notes that some applications worked correctly on earlier Windows versions because the older JScript engine automatically retained global definitions and execution state between scripts. Under the newer model though that behavior is disabled by default causing certain legacy workloads and polyfill-dependent scripts to fail. Microsoft says it addressed the problem via the KB5077241 update though the fix had not been enabled automatically in the following updates. As such admins must explicitly turn on persistent JScript execution context using a Registry setting that the tech giant shared today. The configuration can be applied to individual processes or system-wide through the FEATURE_ENABLE_PERSISTENCE registry key. The steps have been outlined below: Run the following command to create the feature control registry key: reg add "HKLM\Software\Policies\Microsoft\Internet Explorer\Main\FeatureControl\FEATURE_ENABLE_PERSISTENCE" Under this key, create a new DWORD (32-bit) value. Configure the value as follows: To enable persistence for specific processes only: Set the value to 1 for each target process name. To enable persistence for all processes: Add * as the key name and set its value to 1. You can find the official support article here on Microsoft's website.
    • The possibility that milk gathers back into a glass implies that gravity can be 'reversed'.
  • Recent Achievements

    • Week One Done
      Jordan Smith earned a badge
      Week One Done
    • Reacting Well
      BizSAR earned a badge
      Reacting Well
    • First Post
      AndreaB earned a badge
      First Post
    • Week One Done
      Huge Trailer earned a badge
      Week One Done
    • Week One Done
      Classifyskilleducation earned a badge
      Week One Done
  • Popular Contributors

    1. 1
      +primortal
      590
    2. 2
      +Edouard
      186
    3. 3
      PsYcHoKiLLa
      76
    4. 4
      Michael Scrip
      73
    5. 5
      Steven P.
      66
  • Tell a friend

    Love Neowin? Tell a friend!