Discussions - Recent Topics

Are passkeys really phishing resistant if users still get phished?

wan_wrangler — Tue, 22 Sep 2026 00:15:40 +0000

We are rolling out passkeys and management thinks phishing is basically solved. Then Microsoft published that Storm-2372 writeup and recent attacks seem to use fake enrollment pages and device code flows instead of stealing the passkey itself. How are you explaining this without making the rollout sound pointless?

wifi 7 upgrade for office networks in 2026... is it actually worth it

packetlost — Mon, 21 Sep 2026 19:24:59 +0000

We need to refresh 70 access points next year and the Aruba quote (its HPE with Juniper Mist now, still getting used to that) is the obvious Wi Fi 7 option. Most clients are still Wi Fi 6 or 6E though. For people who already rolled it out, did users notice anything beyond speed tests? I'm more interested in crowded meeting rooms, roaming and latency than a big number next to one laptop.

IPv6 only office network with NAT64 and DNS64, is anyone running this in production?

blue_team_bob — Mon, 21 Sep 2026 14:03:30 +0000

We are opening a small office and somebody suggested making the client VLAN IPv6 only with NAT64 and DNS64. I like the idea of not carrying dual stack forever, but I also do not want to discover that the finance app or a random VPN client only speaks IPv4. Has anyone done this for normal users rather than a lab?

GCP vs AWS vs Azure in 2026, why did you choose Google Cloud?

cloud_newbie_404 — Mon, 21 Sep 2026 10:08:40 +0000

We are greenfield and not tied to Microsoft. AWS has the largest ecosystem, Azure is what most local partners sell, but the developers keep asking for GCP because of data services and a cleaner console. For people who chose GCP recently, what was the real reason and what did you regret later?

Immutable backup and ransomware recovery, what failed in your restore test

blue_team_bob — Mon, 21 Sep 2026 06:55:11 +0000

We ran a ransomware tabletop and learned that our Veeam repository is immutable but the backup console, identity provider and DNS all depend on the same environment we would be recovering. The data is safe, but the team could not start a clean restore without production credentials. How are people designing around this?

SASE vs branch firewall, did costs and operations actually get simpler?

route_table — Mon, 21 Sep 2026 02:15:07 +0000

We moved ten branches to Palo Alto Prisma Access. Connectivity is easier and remote users finally have the same policy, but licensing, log retention and traffic charges are harder to predict than the old firewalls. For teams a year or two into this, did the operating model get simpler or did the complexity just move to the vendor portal?

ai SOC analyst tools, useful automation or expensive autocomplete?

packetlost — Sun, 20 Sep 2026 18:45:59 +0000

Our Microsoft Sentinel rep is pushing Security Copilot as an add on. The demo summarizes alerts, writes KQL queries and suggests next steps. It looks nice, but our problem is noisy data and weak playbooks. Has this actually reduced investigation time for anyone, or does it just write confident summaries of bad alerts??

multi region active active architecture, when is the extra cost worth it?

sre_gus — Sun, 20 Sep 2026 14:16:53 +0000

A recent auth outage took down our application even though compute was healthy in two regions. We now have pressure to go active active everywhere. The app can run in two regions, but identity, queues and the database still have single region assumptions. Where do you draw the line before resilience becomes a second product?

Best VMware alternative in 2026, Proxmox vs Hyper V vs Nutanix

hyperv_hannah — Sun, 20 Sep 2026 13:49:40 +0000

Our old agreement expires this year and the new VMware quote finally landed. We run three clusters, around 180 VMs, vSAN and Veeam. Management wants alternatives now, not a strategy slide. For teams that finished a migration, where did you land and what was harder than expected?

Cloud repatriation cost savings... real FinOps win or another migration?

pve_migrator — Sun, 20 Sep 2026 13:16:15 +0000

our steady workloads cost far more in cloud than the original estimate, mostly RDS instances, egress and an always on EKS cluster we never scale down. Leadership now wants to move everything back on premises. Has anyone done the math honestly, including staff, hardware refresh and disaster recovery?