CVE-2026-126253 FortiOS patch versions — 7.6.8 and 7.4.13 shipped 19 days before the advisory
CVE-2026-126253 in Fortinet FortiOS is fixed in FortiOS 8.0.2, 7.6.8, 7.4.13 and 7.2.14 or later; the 7.0 branch gets no fix and must move to a fixed branch. The flaw is an authentication bypass (CVSS 9.3) in the OSPF routing daemon of FortiOS and FortiSASE that lets an attacker on the same network segment broadcast forged routing information (LSAs) without logging in. Devices that do not run OSPF are not exposed to this path. Fortinet published advisory FG-IR-26-267 on November 10, 2026 (US time), but it had shipped the fixed builds 19 days earlier, on October 22. The advisory marks the flaw as not known to be exploited, and it is absent from the November 13, 2026 version of the US Cybersecurity and Infrastructure Security Agency's Known Exploited Vulnerabilities (KEV) catalog.
The three lines
- Versions — FortiOS 8.0.2, 7.6.8, 7.4.13, 7.2.14 or later. 7.0 has no fix and must migrate. Builds shipped October 22; the advisory followed on November 10
- Conditions — only devices running OSPF, and the attacker must sit on the same network segment. No login is needed. At 9.3 it is not the highest score in the batch
- Status — not exploited per Fortinet, not in CISA KEV as of the November 13 version. FortiMail and FortiManager carry separate 9.8-rated flaws from the same day
Key questions
- CVE-2026-126253 FortiOS patch version
- **FortiOS 8.0.2, 7.6.8, 7.4.13 and 7.2.14 or later. The 7.0 branch has no fixed build.** | Branch | Affected | Fixed | |---|---|---| | FortiOS 8.0 | 8.0.0–8.0.1 | 8.0.2 or later | | FortiOS 7.6 | 7.6.0–7.6.7 | 7.6.8 or later | | FortiOS 7.4 | 7.4.0–7.4.12 | 7.4.13 or later | | FortiOS 7.2 | 7.2.0–7.2.13 | 7.2.14 or later | | FortiOS 7.0 | 7.0.1–7.0.19 | No fix; migrate to a fixed branch | | FortiSASE 26.3 | 26.3.1, 26.3.1_a | Remediated by Fortinet in 26.3.2; no customer action |
- Does CVE-2026-126253 require authentication
- **No login is required. The device must be running OSPF, and the attacker must be on the same network segment.** | Item | Detail | |---|---| | Type | Authentication bypass by primary weakness (CWE-305) | | Location | OSPF daemon in FortiOS and FortiSASE | | CVSS 3 | 9.3 (Critical) | | Attack vector | Adjacent network (AV:A); not directly reachable from the internet | | Authentication | None | | Impact | Forged LSAs broadcast; integrity and availability high, confidentiality none | | Workaround in the advisory | None |
- Fortinet November 2026 security updates exploited in the wild
- **Fortinet marks them as not known to be exploited, and none of the batch appears in the November 13, 2026 version of CISA's KEV catalog.** | Date (US time) | Event | |---|---| | October 21 | Fortinet announces a security hardening release; FortiMail and FortiAuthenticator builds ship | | October 22 | FortiOS, FortiManager and FortiAnalyzer builds ship | | November 10 | 89 advisories published (nearly 100 vulnerabilities, about 17 critical) | | November 13 | Fortinet adds virtual-patch details to the BGP advisory; KEV version of this date lists none of the batch |
CVE-2026-126253, a routing flaw in Fortinet's firewall operating system FortiOS, is fixed in FortiOS 8.0.2, 7.6.8, 7.4.13 and 7.2.14. Fortinet disclosed it in advisory FG-IR-26-267 on November 10, 2026 (US time, the second Tuesday of the month), but the fixed builds had been available since October 22, nineteen days earlier. In October the company told customers to upgrade quickly without saying what had been fixed; on November 10 it published 89 advisories at once and the details came out.
The score is CVSS 9.3, in the critical band. Ordering work by that number alone would be a mistake. The flaw applies only to devices running the OSPF routing protocol, and the attacker has to be inside the same network segment. The same batch contains 9.8-rated flaws in FortiMail and FortiManager that are reachable without a login and, in FortiMail's case, without any special configuration. This piece sets out which version to move to, how to tell whether a device is in scope, and what exposure can be reduced when an upgrade has to wait, based on the text of Fortinet's advisories.
1. Which versions are fixed — 8.0.2, 7.6.8, 7.4.13, 7.2.14, and nothing for 7.0
The version table in FG-IR-26-267 reads as follows. Red Tiger Security, Rescana and Penligent give the same numbers (nFlo gives the same affected ranges only), and they match the version table Fortinet posted on its blog on October 21.
| Branch | Affected | Action |
|---|---|---|
| FortiOS 8.0 | 8.0.0–8.0.1 | Upgrade to 8.0.2 or later |
| FortiOS 7.6 | 7.6.0–7.6.7 | Upgrade to 7.6.8 or later |
| FortiOS 7.4 | 7.4.0–7.4.12 | Upgrade to 7.4.13 or later |
| FortiOS 7.2 | 7.2.0–7.2.13 | Upgrade to 7.2.14 or later |
| FortiOS 7.0 | 7.0.1–7.0.19 | No fix; migrate to a fixed release |
| FortiSASE 26.3 | 26.3.1, 26.3.1_a | Remediated by Fortinet in 26.3.2; no customer action |
| FortiSASE 25.1–26.2 | All versions of each branch | Migrate to a fixed release |
| FortiSASE 24.4 and earlier | Not affected | — |
Three places in this table trip people up.
First, the 7.0 branch is not being fixed. The advisory lists 7.0.1 through 7.0.19 as affected and offers only "migrate to a fixed release". A device still on 7.0 has to change branch, to 7.2.14 or later. An alert from Rwanda's National Cyber Security Authority (NCSA) dated November 13 lists 7.0.18 or above among fixed versions, but no such version appears in Fortinet's advisory, and the advisory is the one to follow. Why an aging branch stops receiving patches is covered in our explainer on end of life (EOL).
Second, 7.4.12 means opposite things on different products. On FortiOS, 7.4.12 is still vulnerable and 7.4.13 is the fix. On FortiManager and FortiAnalyzer, which received advisories the same day, 7.4.12 is the fixed build for several flaws. A team running firewalls and management appliances together can easily leave a firewall on 7.4.12 and consider the job done.
Third, the fixed builds have been out since October 22.
| Date (US time) | Event | Basis |
|---|---|---|
| October 21, 2026 | Fortinet chief information security officer Carl Windsor announces a "security hardening release" on the company blog. FortiMail and FortiAuthenticator builds ship the same day | Fortinet blog |
| October 22 | FortiOS 8.0.2, 7.6.8, 7.4.13 and 7.2.14 ship, along with FortiManager and FortiAnalyzer builds | Version table in the Fortinet blog |
| October 26 | Help Net Security reports that Fortinet is withholding details of the fixes until November 10 | Help Net Security |
| November 10 | 89 advisories published. CVE-2026-126253 is FG-IR-26-267 | FortiGuard PSIRT |
| November 13 | Fortinet adds virtual-patch information to the BGP advisory (FG-IR-26-268). Rwanda NCSA issues its alert | FortiGuard PSIRT, Rwanda NCSA |
October 22 to November 10 is 19 days. An organization that upgraded FortiOS in a routine maintenance window during that period was already clear of this flaw before the advisory existed. One that deferred "an update that doesn't say what it fixes" met November 10 with the description and the fix public at the same moment. Windsor wrote in the October post that once updates are public, the fixes themselves reveal where changes were made. That is the stated reason for the gap between build and advisory.
2. What breaks — forged routes without a login, but only where OSPF is running
OSPF is the protocol routers and firewalls use to tell each other which networks lie in which direction and to choose paths accordingly. One unit of that information is a link-state advertisement, or LSA. Devices build their routing tables from the LSAs their neighbors send, which is why OSPF has an authentication step to check that a neighbor is genuine.
CVE-2026-126253 is a way around that check. Fortinet's advisory describes an "authentication bypass by primary weakness" (CWE-305) in the OSPF daemon of FortiOS and FortiSASE, through which an unauthenticated attacker may broadcast LSAs using a crafted request.
| Item | Detail | What it means |
|---|---|---|
| Advisory | FG-IR-26-267 | Published November 10, 2026; no revisions since |
| Type | Authentication bypass (CWE-305) | The lock exists but there is a way around it |
| Location | OSPF daemon | No OSPF, no path |
| CVSS 3 score | 9.3 (Critical) | Fortinet's own rating |
| Attack vector | Adjacent network (AV:A) | The attacker must be on the same network segment; not directly reachable across the internet |
| Privileges | None (PR:N) | No account, no password |
| Scope | Changed (S:C) | The effect spreads beyond the one device attacked |
| Impact | Confidentiality none, integrity high, availability high | It does not read data directly; it changes or cuts routes |
| Exploitation | Marked "No" in the advisory | As of publication on November 10 |
| Discovery | Internal | Loic Pantano of Fortinet's product security team, using the AI model Claude Mythos |
The changed scope is what drives the 9.3. An LSA is information a device passes on to its own neighbors. If one device accepts a forged LSA, the other routers in the same OSPF area learn that route too. nFlo notes that injected LSAs can redirect or block traffic; traffic steered through a segment the attacker watches could become the starting point for a further attack. The impact the advisory itself states goes no further than improper access control, and no concrete attack scenario has been published.
Two conditions pull in the other direction.
| Condition | If yes | If no |
|---|---|---|
| Does the device run OSPF? | It is in scope | This path is closed. The same builds fix other flaws, so the upgrade is still needed |
| Can an attacker get onto a segment where OSPF runs? | Attack is possible without a login | Not directly reachable from outside; still open to an attacker already inside |
Rescana points out that Fortinet does not describe a remote internet path, and Panelica classes the flaw as configuration-dependent, behind five findings in the batch that are reachable unconditionally. A small office with a single firewall and static routes probably has OSPF off. An organization that ties headquarters, data centers and branches together with dynamic routing on its firewalls, or a carrier or plant network where the firewall doubles as a router, is squarely in scope.
And 9.3 is not the top score in this batch. Within FortiOS alone, CVE-2026-126254 in the BGP daemon is rated 9.6, and FortiMail, FortiManager and FortiAnalyzer carry 9.8s (section 4). A score describes how bad the worst case is, not whether the worst case holds in a given environment; our CVSS explainer goes through how the number is built.
3. What to do — check the version, follow the upgrade path, reduce exposure where an upgrade must wait
The order is: establish whether a device is in scope, upgrade, then reduce exposure on anything that cannot be upgraded yet.
| Step | Task | How | What to look for |
|---|---|---|---|
| 1 | Check the version | The Version line of get system status in the CLI, or the firmware field on the dashboard | 8.0.2, 7.6.8, 7.4.13 or 7.2.14 and later closes this flaw |
| 2 | Check whether OSPF is in use | show router ospf for configuration, get router info ospf neighbor for neighbors | Empty configuration and no neighbors means this path does not apply |
| 3 | Build the inventory | Every FortiGate at headquarters, branches and in the cloud, including units run by integrators or managed service providers | One table with branch, version and OSPF use per device |
| 4 | Back up the configuration | Download the configuration file before upgrading | Recommended in Fortinet's October post |
| 5 | Check the upgrade path | Enter current and target versions in Fortinet's upgrade path tool (docs.fortinet.com/upgrade-tool) | Do not skip intermediate versions |
| 6 | Upgrade | OSPF devices first; every member of a high-availability pair | Rerun the step 1 command afterward |
| 7 | Verify | Neighbors show FULL in get router info ospf neighbor; nothing unfamiliar in get router info ospf database brief | Compare the routing table with the one from before the upgrade |
Step 5 matters because FortiOS defines the intermediate versions a device must pass through when it crosses branches. Moving from 7.0 to 7.2.14, or from 7.2 to 7.4.13, triggers configuration conversion, so the sequence the tool gives should be followed. Penligent makes the same point: use the supported upgrade path and do not skip steps.
Devices that cannot be upgraded right away are the hard part. Fortinet's advisory offers no workaround for this flaw and lists no virtual patch. The measures below come from the InfoSec Monkey blog, Red Tiger Security and nFlo. They are not mitigations confirmed by Fortinet, and InfoSec Monkey itself says such countermeasures reduce exposure and are not fixes.
| Measure | What it does | Where | Limit |
|---|---|---|---|
| Passive interfaces | Stops OSPF packets on interfaces with no real OSPF neighbor | set passive-interface under config router ospf | Interfaces with real neighbors stay open |
| Known neighbors only | Accepts OSPF (IP protocol 89) to the device only from known neighbor addresses | Allow and deny rules in config firewall local-in-policy | Untested against an attacker on the same segment spoofing a neighbor address |
| Route filter | Installs only expected prefixes learned from OSPF; rejects a default route (0.0.0.0/0) | config router access-list with set distribute-list-in | A forged route inside the permitted ranges passes |
| Segment review | Confirms OSPF is not enabled on user, guest or internet-facing interfaces | get router info ospf interface | Finds configuration mistakes |
| Monitoring | Records unexpected changes in the routing table and LSA database | Query and compare on a schedule | Detection after the fact |
One caution. InfoSec Monkey warns not to rely on OSPF MD5 authentication alone, since the flaw is precisely a bypass of authentication; keep authentication on and add the controls above. Local-in policies and route filters can also break adjacency with legitimate neighbors if entered wrongly and destabilize the whole network. Apply them in a maintenance window while watching that neighbors stay in FULL state.
FortiSASE needs little separate attention. According to the advisory, 26.3.1 and 26.3.1_a were remediated by Fortinet in 26.3.2 and customers need do nothing. Branches 25.1 through 26.2 are listed as affected in all versions with the instruction to migrate to a fixed release.
4. The other critical flaws published the same day — the 9.8s are in FortiMail and the management appliances
SecurityWeek reported that Fortinet's 89 advisories of November 10 cover nearly 100 vulnerabilities, about 17 of them critical and nearly 80 high. The SANS Internet Storm Center gave the same 89 and 17 in its November 12 Stormcast and noted that FortiMail received the most critical fixes. SecurityWeek's ranking of products with critical fixes runs FortiMail, FortiAuthenticator, FortiOS, FortiManager, FortiAnalyzer.
Scores below are Fortinet's. Fixed versions are given only where the advisory was opened directly or two or more sources agree.
| Product | CVE (advisory) | Score | What it is | Condition | Fixed in |
|---|---|---|---|---|---|
| FortiMail | CVE-2026-126147 (FG-IR-26-199) | 9.8 | Path traversal; code or command execution without a login | None | 8.0.2, 7.6.7, 7.4.9; 7.2 must migrate |
| FortiMail | CVE-2026-126148 (FG-IR-26-200) | 9.8 | Command injection through HA join-request packets | 8.0 and 7.6 branches only | 8.0.2, 7.6.7 |
| FortiMail | CVE-2026-126164 (FG-IR-26-216) | 9.4 | SQL injection | None | 8.0.2, 7.6.7, 7.4.9 |
| FortiMail | CVE-2026-126155, CVE-2026-126156 | 9.1 | Session or account takeover of encrypted-mail (IBE) recipients | IBE in use | 8.0.2, 7.6.7, 7.4.9 |
| FortiManager, FortiAnalyzer | CVE-2026-126170 (FG-IR-26-219) | 9.8 | JSON injection in the JSON-RPC API; code or command execution without a login | JSON API enabled (off by default); 7.4 and 7.2 not affected | 8.0.3, 7.6.8, or the workaround |
| FortiManager | CVE-2026-126261 (FG-IR-26-275) | 9.0 | Elevation in the device-registration channel (FGFM) before the registration password is verified | Reach to the FGFM port (TCP 541) | 8.0.1, 7.6.8, 7.4.6, 7.2.9 |
| FortiManager, FortiAnalyzer | CVE-2026-126229 (FG-IR-26-244) | 9.1 | An administrator reaches virtual domains (VDOMs) outside their permissions | Authenticated administrator | 8.0.2, 7.6.8, 7.4.12; 7.2 must migrate |
| FortiAnalyzer, FortiManager | CVE-2026-126262 (FG-IR-26-276), CVE-2026-126260 (FG-IR-26-274) | 9.3, 9.0 | SQL injection and server-side request forgery in the log channel (OFTP) | Reach to the OFTP port (TCP 514) | Check the advisory |
| FortiOS, FortiSASE | CVE-2026-126254 (FG-IR-26-268) | 9.6 | Numeric truncation in the BGP daemon; malformed attributes pushed to all connected routers | Must be a configured iBGP neighbor (authenticated) | 8.0.2, 7.6.8, 7.4.13, 7.2.14 |
| FortiOS | CVE-2026-126259 (FG-IR-26-273) | 9.0 | Super-admin command execution through HA packets | Only with FGSP plus standalone-config-sync, or autoscale | 8.0.1, 7.6.8, 7.4.13, 7.2.14; 7.0 must migrate |
| FortiAuthenticator | CVE-2026-126251, CVE-2026-126257 | 9.1, 9.0 | Incorrect authorization; hard-coded password | High attack complexity for the second | 8.0.4, 6.6.11 |
Three things stand out.
Unconditional reach comes first. CVE-2026-126147 in FortiMail needs no login and no particular configuration, and a mail appliance faces the internet by design. For an organization that does not run OSPF, FortiMail's 9.8 is far more urgent than FortiOS's 9.3. Fortinet's advisory says FortiMail Cloud was remediated on October 5, 2026 and its customers need not act.
FortiManager 8.0.2 leaves one flaw open. The FortiManager and FortiAnalyzer 8.0.2 builds from the October 22 hardening release fix most of the batch, but the advisory for the JSON-RPC flaw (CVE-2026-126170) lists 8.0.0 through 8.0.2 as affected and 8.0.3 as the fix. SecurityOnline and GBHackers report 8.0.3 as marked forthcoming. Until it ships, 8.0-branch management appliances depend on the advisory's workaround: set the JSON API permission (rpc-permit) to none on administrator profiles. Fortinet says the API is disabled by default.
Management ports do not belong on the internet. FGFM (TCP 541) and OFTP (TCP 514) are the channels firewalls use to register with a management appliance and send it logs. GBHackers notes that protecting the browser login page does nothing for these services. One FortiManager pushes policy to hundreds of firewalls, so a compromise there reaches everything below it.
The batch also includes a flaw in the FortiClient endpoint software. SecurityWeek reports a bug nicknamed "Fortinet2Shell" (CVE-2026-121529) by Kevin Joensen of Baldur Security, which he says lets an attacker run code when a user with FortiPAM enabled visits a malicious site. This rests on a single report.
5. Exploitation status, and why Fortinet keeps being targeted
The current state first.
| Checked | Result | As of |
|---|---|---|
| "Known Exploited" field in Fortinet's advisories | "No" in all six advisories opened directly | November 10 publication; November 13 revision for the BGP advisory |
| CISA Known Exploited Vulnerabilities catalog | None of the batch is listed | Catalog version 2026.11.13, 1,781 entries |
| SecurityWeek | Fortinet makes no mention of exploitation | November 11 |
| Indicators of compromise | None published | Rescana, checked November 11 |
"Not exploited yet" is a dated judgment. Penligent observes that the field is no proof that a particular deployment was never targeted. For Fortinet equipment the statement has had a short shelf life more than once.
| Flaw | What it was | Published | Added to CISA KEV | Course |
|---|---|---|---|---|
| CVE-2022-40684 (FG-IR-22-377) | Authentication bypass in the administrative interface of FortiOS, FortiProxy and FortiSwitchManager, 9.8 | October 10, 2022 | October 11, 2022 | Fortinet's advisory says it was aware of exploitation. KEV marks it as used in ransomware campaigns |
| CVE-2024-21762 (FG-IR-24-015) | Out-of-bounds write in the FortiOS SSL VPN daemon, 9.8 | February 8, 2024 | February 9, 2024 | The advisory said it was potentially being exploited; KEV listing followed a day later with a seven-day federal deadline |
| CVE-2024-47575 (FG-IR-24-423) | Missing authentication in the FortiManager registration daemon (fgfmd), 9.8 | October 23, 2024 | October 23, 2024 | Already exploited at disclosure; the advisory says so explicitly |
In all three, publication and KEV listing were at most a day apart. The 2024 FortiManager flaw sat in the same channel, FGFM, as CVE-2026-126261 in the current batch, which is why Red Tiger Security sets the two side by side.
The November 13 KEV version holds 32 Fortinet entries. The most recent are CVE-2025-61624, a path traversal in FortiOS and other products added on October 16, 2026, and CVE-2026-104286, a path traversal in FortiMail added on October 1, 2026. Both carried three-day deadlines for US federal civilian agencies; the deadlines bind only those agencies, but they are a useful signal of urgency for anyone. In other words, a FortiMail path traversal was being exploited six weeks before November 10, the day another FortiMail path traversal, CVE-2026-126147, was published at 9.8.
Why Fortinet? The answer lies in where its products sit.
| Reason | Explanation |
|---|---|
| They stand at the entrance | Firewalls, VPN gateways and mail gateways sit between the internet and the internal network. Take one and you can see what is behind it |
| The management appliance is a lever | One FortiManager pushes policy to hundreds of firewalls |
| Monitoring tools are hard to install on them | Unlike a general-purpose server, an appliance rarely runs endpoint detection software, so compromise is noticed late |
| Patching is slow | A firewall upgrade interrupts the network and waits for a maintenance window |
| The patch becomes a map | Once a fix ships, attackers compare builds to work back to the flaw |
The last row explains the shape of this release. In the October 21 post Fortinet said some of the vulnerabilities were found through its use of frontier AI models in product security testing, and thanked Anthropic, OpenAI, Google and SpaceX for their collaboration. Windsor wrote that the window between disclosure and exploitation keeps shrinking and that the company expects it to keep shrinking. Fortinet also warned that AI can link seemingly unrelated lower-severity vulnerabilities into something more severe. Hence the sequence: ship the fix, give customers time to upgrade, publish the details 19 days later.
Five of the six advisories opened for this article credit a named Fortinet product security researcher together with Claude Mythos; the sixth, for the JSON-RPC flaw, credits an AI model without naming it. Magica cautions that Fortinet has not said how many of the roughly 100 vulnerabilities AI found, and that an "internal discovery" label does not by itself establish AI involvement. This is a case of a defender finding and fixing first with AI, but Fortinet's own premise is that the same capability shortens the time an attacker needs to read a patch. For a recent measurement of that interval, our piece on the Barracuda ESG flaws documents a one-week gap between patch and confirmed exploitation.
6. What remains open and what could not be confirmed
| Item | Confirmed | Not confirmed |
|---|---|---|
| Fixed versions | FortiOS 8.0.2, 7.6.8, 7.4.13, 7.2.14 (advisory and multiple sources agree) | The 7.0.18 listed by Rwanda NCSA is not in the advisory |
| Attack method | One sentence: a crafted request broadcasts LSAs | The form of the request, differences by OSPF authentication type, whether exploit code exists |
| Exploitation | "No" per Fortinet; absent from the November 13 KEV version | Any change since; whether anyone analyzed the builds after October 22 |
| Workaround | None in the advisory | Whether passive interfaces and local-in policies actually stop the attack |
| Size of the release | 89 advisories, nearly 100 vulnerabilities, about 17 critical (as reported) | Totals stated by Fortinet itself; a definitive list of the 17 |
| FortiManager 8.0.3 | Named in the advisory as the fix for the JSON-RPC flaw | Its release date |
| AI's share | Several advisories credit Claude Mythos | How many of the roughly 100 AI found |
| Belgium's CCB alert | Title and address only | The page refused access and was not read |
| South Korea | — | Number of affected devices in the country; any separate advisory from Korean authorities |
The order, then: FortiMail first if you run it; management appliances first if ports 541 or 514 face the internet; then FortiGates running OSPF or BGP, to 8.0.2, 7.6.8, 7.4.13 or 7.2.14. For 7.0, plan the branch migration. Conditions set the order, not the score.
Microsoft's side of the same month is in our "November 2026 Patch Tuesday — why a 6.5-rated Outlook zero-day, CVE-2026-124194, comes first among 1,094 fixes"; its conclusion, that exploitation and exposure outrank the score, applies equally to a 9.3-rated OSPF flaw. A recent case of a mail security appliance under attack is in "Barracuda ESG vulnerabilities patch versions"; handling a branch that gets no fix is covered in "What end of life (EOL) means"; and reading scores is covered in "What CVSS is".
Sources
- FortiGuard PSIRT — FG-IR-26-267 Authentication bypass by primary weakness (CVE-2026-126253, 2026-11-10)
- FortiGuard PSIRT — FG-IR-26-268 Numeric truncation error (CVE-2026-126254, 2026-11-10, updated 11-13)
- FortiGuard PSIRT — FG-IR-26-273 Unauthenticated adjacent command execution via HA (CVE-2026-126259, 2026-11-10)
- FortiGuard PSIRT — FG-IR-26-199 FortiMail path traversal (CVE-2026-126147, 2026-11-10)
- FortiGuard PSIRT — FG-IR-26-219 JSON Injection in JRPC API (CVE-2026-126170, 2026-11-10)
- FortiGuard PSIRT — FG-IR-26-275 FGFM elevation before registration password is verified (CVE-2026-126261, 2026-11-10)
- FortiGuard PSIRT — advisory index
- Fortinet blog — Fortinet Releases Security Hardening Release for Multiple Products (Carl Windsor, 2026-10-21, updated 11-10)
- CISA — Known Exploited Vulnerabilities Catalog JSON (catalog version 2026.11.13)
- SecurityWeek — Fortinet Patches Nearly 100 Vulnerabilities (2026-11-11)
- SANS Internet Storm Center — Stormcast 10176 (2026-11-12)
- Rwanda NCSA — Security Alert: Fortinet Security Updates, November 2026 (2026-11-13)
- Red Tiger Security — Fortinet November 2026: Critical FortiOS Routing Flaw and FortiManager Console Fixes
- Rescana — Fortinet November 2026 PSIRT: FortiMail and FortiManager Unauthenticated RCE (2026-11-11)
- InfoSec Monkey — Fortinet PSIRT Countermeasures for Pre-Auth Bugs
- SecurityOnline — Fortinet Patches Eight Critical Flaws in FortiOS, FortiManager and FortiAnalyzer (2026-11-10)
- GBHackers — Critical Fortinet Vulnerabilities (2026-11-11)
- Panelica — Fortinet November 2026 Advisories: Which of the 17 CVEs to Patch First (2026-11-12)
- nFlo — CVE-2026-126253: Authentication Bypass in FortiOS (2026-11-12)
- Penligent — Fortinet November 2026: FortiOS and Manager Fixes
- Help Net Security — Fortinet patches critical FortiOS flaws found with AI help (2026-10-26)
- Magica — Fortinet discloses nearly 100 flaws after AI-assisted testing (2026-11-11)
- FortiGuard PSIRT — FG-IR-24-423 Missing authentication in fgfmsd (CVE-2024-47575, 2024-10-23)
- FortiGuard PSIRT — FG-IR-24-015 Out-of-bound Write in sslvpnd (CVE-2024-21762, 2024-02-08)
- FortiGuard PSIRT — FG-IR-22-377 Authentication bypass in administrative interface (CVE-2022-40684, 2022-10-10)