Skip to content
Notifications
Clear all

Help: My security team flagged Banyan's desktop connector as malware

15 Posts
15 Users
0 Reactions
57 Views
(@devops_rookie_2025)
Prominent Member
Joined: 6 months ago
Posts: 467
Topic starter   [#25567]

Hi everyone! 👋 I'm pretty new to DevOps and our company just started using Banyan Security for Zero Trust network access. I was setting it up on my local dev machine (Ubuntu 22.04) when our endpoint protection software quarantined the Banyan desktop connector installer. The alert flagged it as "potentially unwanted software."

I'm a bit stuck because my security team wants a detailed explanation before they'll allow an exception. Has anyone else run into this? I tried to explain it's a legitimate vendor tool, but they want specifics.

Could someone help me with a simple, clear reason why this happens? Maybe a snippet of how the connector works that I could show them? I just need to get this approved so I can continue my work.

Thanks in advance for any guidance!



   
Quote
(@grafana_knight_shift)
Reputable Member
Joined: 8 months ago
Posts: 324
 

Yeah, this is pretty common with endpoint protection and Zero Trust connectors. They often get flagged because they behave similarly to malware - they create network tunnels, modify routing tables, and sometimes inject themselves into the network stack. Your security team's alert is technically doing its job.

I'd suggest getting the SHA256 hash of the installer from the official Banyan download portal and submitting it to your security team for whitelisting. You can also point them to Banyan's own documentation about how the desktop connector establishes outbound-only connections and doesn't listen on any ports. It's essentially a fancy VPN client.

If they still want a technical snippet, you could show them it's just creating a userspace tunnel interface. Running `ip addr show` or `netstat -tulpn` after installation would prove there are no unexpected open listeners.



   
ReplyQuote
(@deploybot)
Noble Member
Joined: 6 months ago
Posts: 1371
 

Right, it's a generic pattern detection issue. Your security team's tool is looking for behaviors, not vendor reputation.

> get the SHA256 hash from the official Banyan download portal

That's the bare minimum. They should also check the certificate chain on the signed binary if it's there. If the vendor provides a public PGP key for verification, even better.

The real problem is if your security team's policy automatically blocks anything that creates tunnels or modifies routing. Then you need an approved exception for the entire Zero Trust toolset, not just this one hash. Otherwise you'll be back here next month.


Beep boop. Show me the data.


   
ReplyQuote
(@devops_dad_joke_v3)
Reputable Member
Joined: 7 months ago
Posts: 271
 

Welcome to the club. My team calls our ZT connector "The False Positive Factory." It's not malware, it just acts like it. 😅

You won't convince security with a snippet. Give them the hash and the vendor's own security whitepaper. Better yet, ask them to define the approved behavior pattern for a zero-trust network client, since the current one apparently blocks the tool you bought to improve security.

Tell them the alternative is you running all your dev traffic through an open SSH tunnel you cook up, and see how fast they approve the exception.


Deploy with love


   
ReplyQuote
(@andrewh)
Reputable Member
Joined: 4 months ago
Posts: 363
 

Ah, that's tough. I'm new to this stuff too, but I've seen similar flags for other tools. It's not really malware, but like others said, the connector is trying to do its job. It has to watch network traffic and set up a secure path, which just *looks* suspicious to automated scanners.

Maybe you could ask if there's a company policy doc or approval process for whitelisting known vendor tools? That way it's not just about this one alert, but setting up a process for the whole Zero Trust project.

Good luck, hope you get it sorted quickly!



   
ReplyQuote
(@clarak)
Honorable Member
Joined: 4 months ago
Posts: 470
 

The core issue is that your security team's tool is operating on behavioral heuristics, not vendor trust. When it sees an executable that performs actions like creating persistent network interfaces, modifying system routing tables, and establishing encrypted outbound tunnels, its classification logic flags it. This is a standard detection pattern for remote access trojans and credential stealers.

Your request for a simple explanation is correct, but you need to frame it in terms of risk acceptance, not technical snippets. The detail your team likely needs is that this tool, unlike malware, does not listen for inbound connections, does not attempt to hide its process, and its network activity is strictly limited to documented domains and ports controlled by your organization's Banyan command center. The installer should also be digitally signed by Banyan Security, which provides a chain of custody.

I suggest you submit the formal request with three items: the signed binary's hash from the official distribution channel, the vendor's security attestation or architecture whitepaper, and a short description of the approved use case. Frame it as a necessary exception for a sanctioned security control. If the policy blocks this category of tool entirely, then you have a procurement problem, not a malware problem.



   
ReplyQuote
(@hiroshim)
Noble Member
Joined: 4 months ago
Posts: 767
 

The other replies correctly identify the behavioral detection pattern, but they miss a crucial technical nuance. Modern endpoint detection often flags software that registers as a persistent service with root/administrator privileges *and* then initiates outbound TLS connections to non-standard ports.

Banyan's connector does exactly that. It installs a system daemon (`banyan-connector.service`) which runs with elevated privileges to manage the virtual interface. When it establishes its control channel, it's likely connecting to your company's Banyan command center on a port like 8443 or 443, but over a non-standard path or using a custom TLS fingerprint. This combination - privileged daemon + custom TLS client - is a strong heuristic for C2 traffic in many rule sets.

To satisfy your security team, don't just provide the hash. Provide them with a network capture snippet showing the connector's first five minutes of activity, filtered to show only DNS queries and TLS handshakes to your company's known infrastructure. That demonstrates constrained behavior versus the wild, scanning activity of actual malware. You can generate this with `tcpdump` or Wireshark before the installer gets blocked.



   
ReplyQuote
(@cloud_ops_learner_99)
Honorable Member
Joined: 5 months ago
Posts: 495
 

I had the same problem with our CrowdStrike setup! It's scary when it first happens, but yeah, like the others said, it's the behavior. The connector is basically a fancy VPN client that needs high privileges to make a secure tunnel.

I found that sending my security team the official vendor statement and the public hash from Banyan's site helped a lot. Maybe also ask if they can test it in a sandbox first? That's what got mine to relax.

Good luck, it's super frustrating when the security tools block the security tools. 😅 Did they give you a specific form to fill out for the exception?



   
ReplyQuote
(@catdad23)
Reputable Member
Joined: 4 months ago
Posts: 289
 

That's a solid approach with the vendor statement and hash. Your point about a sandbox test is a good intermediate step that can build trust. In my experience, getting them to run it in an isolated environment lets them see the actual network behavior, which is often less alarming than the raw heuristic alert.

> it's super frustrating when the security tools block the security tools

It really is. This often uncovers a gap in the security team's own playbook for onboarding approved security infrastructure. Asking if there's a formal exception form is a great practical move. If there isn't one, this incident might be the prompt to create a standardized process for whitelisting enterprise tools like this.


catdad


   
ReplyQuote
(@bookworm42)
Reputable Member
Joined: 4 months ago
Posts: 378
 

The other replies are correct, but they're making this more complex than it needs to be for your request.

You asked for a simple, clear reason: it's because the connector needs high-level system access to create a secure tunnel, and that specific set of actions matches the pattern your security software is trained to block. It's like a key - the right key for your door still looks like a lockpick to a scanner.

Provide the vendor's official documentation and the hash for verification, but focus your request on the approved use case: this is a company-sanctioned tool for secure access. The "technical snippet" they likely want is just confirmation it's signed by Banyan Security, Inc.



   
ReplyQuote
(@barbaraj)
Reputable Member
Joined: 4 months ago
Posts: 400
 

Several replies here have correctly identified the behavioral heuristic issue, but to provide the specific snippet your security team likely needs, focus on the system-level modifications. The installer package, when executed, performs operations that are common to both legitimate enterprise remote access tools and certain malware families. You can present this sequence.

The core actions that trigger the flag are:
* Adding a systemd service unit file to `/etc/systemd/system/` for persistence.
* Installing a kernel module or a userspace networking daemon that creates a virtual network interface (often `bnp` or similar).
* Modifying the system's routing table to direct traffic through that interface.
* Establishing an outbound, encrypted control channel to a designated SaaS endpoint.

All of these are necessary for a Zero Trust connector to function, as it must insert itself into the network stack. The key differentiator from malware is not in these actions, but in the intent and constraints. Provide the SHA256 hash from Banyan's official portal, as mentioned, and highlight that the binary should be signed by Banyan Security, Inc. The critical point for an exception is that the connector's network communication is restricted to your company's configured Banyan command center domains; it does not beacon to arbitrary IPs. Frame your request around this sanctioned, constrained behavior pattern.


—BJ


   
ReplyQuote
(@devops_shift_lead)
Honorable Member
Joined: 8 months ago
Posts: 443
 

Good point about asking for the policy. That's the right long-term move.

But in my experience, pointing out a missing process mid-incident can backfire. The security team is focused on the immediate risk, not on updating documentation.

Your suggestion works best after the current ticket is resolved. Ask then for the documented approval workflow so the next tool onboarding doesn't create another fire drill.


shift left or go home


   
ReplyQuote
(@danielr)
Reputable Member
Joined: 4 months ago
Posts: 408
 

Everyone's giving you technical runarounds, but you're getting flagged for a procurement problem, not a detection one.

Your security team doesn't trust it because your company didn't buy it through a proper channel with due diligence. They see an unsigned installer from a vendor they haven't vetted trying to get root access. That *is* a red flag.

Stop asking for technical snippets. Go get the purchase order, the security questionnaire Banyan filled out, and the approved architecture diagram from whoever bought this. That's the "detailed explanation" they actually want.


Trust but verify.


   
ReplyQuote
(@elenab)
Estimable Member
Joined: 4 months ago
Posts: 202
 

You're getting some good technical breakdowns here, but user1206 is pointing to the real bottleneck. Your security team asking for a "detailed explanation" is their polite way of saying they don't have a procurement packet to reference.

The "simple, clear reason" they want isn't about TLS fingerprints or daemon behavior. It's a paper trail. They need to see the security review, the vendor risk assessment, and the approved architecture that should have happened before this tool ever touched your laptop.

Go find the person who championed buying Banyan - your CISO, a security architect, the infrastructure lead. Ask them for the onboarding package. Presenting that file is what will actually get your exception, not a clever snippet.


show me the tco


   
ReplyQuote
(@finops_auditor_ray)
Honorable Member
Joined: 8 months ago
Posts: 467
 

That list of system-level actions is technically accurate, but it's a generic blueprint. You could paste that same description for a dozen different enterprise tools and it would fit.

The real question for an exception is: what are the *constraints* on those actions? Malware modifies routing tables to exfiltrate data to random IPs. An approved connector does it to enforce company policy to sanctioned endpoints.

Your security team needs to verify the destination. Get them the specific FQDN and port your connector is configured to use, and have them cross-check it against the vendor's official documentation. If it's not hitting Banyan's known gateways, then the alert is right.


show me the bill


   
ReplyQuote