Mozilla Foundation reports :

CVE-2017-7843: Web worker in Private Browsing mode can write IndexedDB data

CVE-2017-7844: Visited history information leak through SVG image

The version of Firefox installed on the remote Windows host is prior to 57.0.1. It is, therefore, affected by multiple vulnerabilities as referenced in the mfsa2017-27 advisory.

  - When Private Browsing mode is used, it is possible for a web worker to write persistent data to IndexedDB     and fingerprint a user uniquely. IndexedDB should not be available in Private Browsing mode and this     stored data will persist across multiple private browsing mode sessions because it is not cleared when     exiting. (CVE-2017-7843)

  - A combination of an external SVG image referenced on a page and the coloring of anchor links stored within     this image can be used to determine which pages a user has in their history. This can allow a malicious     website to query user history.  Note: This issue only affects Firefox 57. Earlier releases are not     affected. (CVE-2017-7844)

Note that Nessus has not tested for these issues but has instead relied only on the application's self-reported version number.

The version of Mozilla Firefox installed on the remote macOS or Mac OS X host is prior to 57.0.1. It is, therefore, affected by multiple vulnerabilities.

Note: CVE-2017-7844 only affects version 57. Earlier releases are not affected.

The version of Firefox installed on the remote macOS or Mac OS X host is prior to 57.0.1. It is, therefore, affected by multiple vulnerabilities as referenced in the mfsa2017-27 advisory.

  - When Private Browsing mode is used, it is possible for a web worker to write persistent data to IndexedDB     and fingerprint a user uniquely. IndexedDB should not be available in Private Browsing mode and this     stored data will persist across multiple private browsing mode sessions because it is not cleared when     exiting. (CVE-2017-7843)

  - A combination of an external SVG image referenced on a page and the coloring of anchor links stored within     this image can be used to determine which pages a user has in their history. This can allow a malicious     website to query user history.  Note: This issue only affects Firefox 57. Earlier releases are not     affected. (CVE-2017-7844)

Note that Nessus has not tested for these issues but has instead relied only on the application's self-reported version number.

The remote host is affected by the vulnerability described in GLSA-201802-03 (Mozilla Firefox: Multiple vulnerabilities)

    Multiple vulnerabilities have been discovered in Mozilla Firefox. Please       review the referenced CVE identifiers for details.
  Impact :

    A remote attacker could entice a user to view a specially crafted web       page, possibly resulting in the execution of arbitrary code with the       privileges of the process or a Denial of Service condition. Furthermore,       a remote attacker may be able to perform Man-in-the-Middle attacks,       obtain sensitive information, spoof the address bar, conduct clickjacking       attacks, bypass security restrictions and protection mechanisms, or have       other unspecified impact.
  Workaround :

    There is no known workaround at this time.

Versions of Mozilla Firefox earlier than 57.0.1 are unpatched for the following vulnerabilities :

 - A flaw exists in the 'PerDocumentStyleDataImpl::visited_styles_enabled()' function in 'servo/components/style/gecko/data.rs'. The issue is triggered when handling the CSS ':visited' selector for a document being used as an SVG image. With a specially crafted web page, a context-dependent attacker disclose visited history information. (CVE-2017-7844)
 - A flaw exists in the 'FactoryOp::CheckPermission()' function in 'dom/indexedDB/ActorsParent.cpp' that is triggered as a web worker in Private Browsing mode can write to IndexedDB. With a specially crafted web page, a context-dependent attacker can uniquely fingerprint a user even when browsing in Private Browsing mode. (CVE-2017-7843)

Versions of Mozilla Firefox earlier than 57 are unpatched for the following vulnerabilities :

 - A race condition exists in 'dom/media/systemservices/MediaParent.cpp' that is triggered when getting deviceId keys. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A flaw exists in 'js/src/proxy/DeadObjectProxy.cpp' that is triggered when handling object proxies. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - An unspecified flaw exists in 'xpcom/threads/SystemGroup.cpp' that is triggered as certain input is not properly validated. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - An overflow condition exists in the 'nr_transport_addr_fmt_ifname_addr_string()' function in 'media/mtransport/third_party/nICEr/src/net/transport_addr.c' that is triggered as certain input is not properly validated. This may allow a context-dependent attacker to cause a stack-based buffer overflow and potentially execute arbitrary code.
 - A flaw exists in the 'nsNodeUtils::CloneAndAdopt()' function in 'dom/base/nsNodeUtils.cpp' that is triggered as certain input is not properly validated. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A flaw exists in the 'GPUProcessHost::OnChannelClosed()' function in 'gfx/ipc/GPUProcessHost.cpp' that is triggered as certain input is not properly validated. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - An unspecified flaw exists in 'gfx/layers/AtomicRefCountedWithFinalize.h' that is triggered as certain input is not properly validated. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A flaw exists in the 'NewReactionRecord()' function in 'js/src/builtin/Promise.cpp' that is triggered as certain input is not properly validated. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A flaw exists in the 'VerifyCMSDetachedSignatureIncludingCertificate()' function in 'security/manager/ssl/nsDataSignatureVerifier.cpp' that is triggered when handling PKCS#7 signedData content. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A flaw exists in the 'ContainerLayerComposite::mPrepared()' function in 'gfx/layers/composite/ContainerLayerComposite.cpp' that is triggered when handling layers. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A flaw exists in the 'MArgumentsLength::computeRange()' function in 'js/src/jit/RangeAnalysis.cpp' that is triggered as certain input is not properly validated. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A use-after-free error exists in 'widget/windows/AudioSession.cpp' that is triggered when handling AudioSession objects. This may allow a context-dependent attacker to dereference already freed memory and potentially execute arbitrary code.
 - An unspecified flaw exists that is triggered when handling WebGL texture images. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A use-after-free flaw exists in the 'nsDocShell::~nsDocShell()' function in 'docshell/base/nsDocShell.cpp' that is triggered when notifying observers. This may allow a context-dependent attacker to dereference already freed memory and potentially execute arbitrary code.
 - A flaw exists in the 'CleanupOSFileConstants()' function in 'dom/system/OSFileConstants.cpp' related to use of uninitialized memory. This may allow a context-dependent attacker to potentially execute arbitrary code.
 - A flaw exists in the 'ApplicationReputationService::~ApplicationReputationService()' function in 'toolkit/components/downloads/ApplicationReputation.cpp' that is triggered as certain pointers are not properly cleared. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A flaw exists in the 'EnumerableOwnProperties()' function in 'js/src/builtin/Object.cpp' that is triggered when rooting objects. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A use-after-free flaw exists in the 'CompositorBridgeChild::RecvDidComposite()' function in 'gfx/layers/ipc/CompositorBridgeChild.cpp' that is triggered when handling texture pools. This may allow a context-dependent attacker to dereference already freed memory and potentially execute arbitrary code.
 - A flaw exists in the 'nsViewManager::~nsViewManager()' function in 'view/nsViewManager.cpp' that is triggered as the PresShell object is not properly handled. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A flaw exists in the 'TLSFilterTransaction::Close()' function in 'netwerk/protocol/http/TunnelUtils.cpp' that is triggered as timers are not properly handled when a transaction is canceled. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A flaw exists in the 'EventStateManager::DispatchCrossProcessEvent()' function in 'dom/events/EventStateManager.cpp' that is triggered when handling drag events. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A flaw exists in the 'nsNodeUtils::CloneAndAdopt()' function in 'dom/base/nsNodeUtils.cpp' that is triggered when handling properties of adopted nodes. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A type confusion flaw exists in the 'PropertyReadNeedsTypeBarrier()' function in 'jit/MIR.cpp' that is triggered as certain input is not properly validated. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A logic flaw exists in the 'IsMarkedBlack()' function in 'js/src/gc/Barrier.cpp' that is triggered during gray marking asserts. This may allow a context-dependent attacker to corrupt memory and potentially execute arbitrary code.
 - A use-after-free error exists that triggered when flushing and resizing the layout, which may cause the PressShell object to be freed while still in use. This may allow a context-dependent attacker to potentially execute arbitrary code.
 - A flaw exists in the 'HttpBaseChannel::GetPerformance()' function in 'netwerk/protocol/http/HttpBaseChannel.cpp' that is triggered as navigations in cross-origin iframes are revealed. Using the Resource Timing API, a context-dependent attacker to gain access to cross-origin URL information.
 - A flaw exists that is triggered as the security wrapper fails to deny access to certain exposed properties using the deprecated exposedProps mechanism on proxy objects. This may allow a context-dependent attacker to gain access to potentially sensitive information.
 - A flaw exists in the 'nsIDNService::isLabelSafe()' function in 'netwerk/dns/nsIDNService.cpp' that is triggered during the handling of the combined, single character, version of the letter 'i' with any of the potential accents in unicode, such as acute or grave. This combinations do not display properly in punycode. With a specially crafted domain including a dotless version of  'i', a context-dependent attacker can spoof a domain.
 - A flaw exists in the 'nsIDNService::isLabelSafe()' function in 'netwerk/dns/nsIDNService.cpp' that is triggered during the handling of Arabic and Indic vowel marker characters that are combined with Latin characters in a domain name, which can cause the non-latin characters to be eclipsed in the address bar. This may allow a context-dependent attacker to spoof a domain.
 - A flaw exists that is triggered as data: URLs loaded into new tabs do not properly inherit the Content Security Policy (CSP) of the original page. This may allow a context-dependent attacker to bypass protection mechanisms.
 - A flaw exists in the 'nsMixedContentBlocker::AsyncOnChannelRedirect()' function in 'dom/security/nsMixedContentBlocker.cpp' that is triggered as mixed content blocking was not properly applied to resources when being redirected from an HTTPS environment to an HTTP environment. This may allow a context-dependent attacker to bypass restricts and load blocked content on pages.
 - A flaw exists in the 'CurlWrapper::Init()' function in 'toolkit/components/telemetry/pingsender/pingsender_unix_common.cpp' that is triggered as it fails to validate the libcurl library before loading it. This may allow a local attacker to replace libcurl files and gain elevated privileges.
 - A flaw exists in the 'nsContentSink::ProcessMETATag()' function in 'dom/base/nsContentSink.cpp' that is triggered during the handling of <meta> tags in SVG passed via <img> tags. This may allow a context-dependent attacker to set cookies for a page.
 - A flaw exists in 'netwerk/dns/nsIDNService.cpp' that is triggered as Punycode format text is not properly displayed in international domain names when triggered by a sub-domain. This may allow a context-dependent attacker to conduct a limited spoofing attack.
 - A flaw exists in the 'stripUnsafeProtocolOnPaste()' function in 'base/content/browser.js' that is triggered during the handling of control characters before javascript: URLs. This may allow a context-dependent attacker to bypass self-XSS protection mechanisms.
 - A flaw exists in the '_writeItem()' function in 'toolkit/components/places/BookmarkHTMLUtils.jsm' that allows a cross-site scripting (self-XSS) attack. This flaw exists exists because the program does not validate input to exported bookmarks before returning it to users. This may allow a context-dependent attacker to create a specially crafted request that would execute arbitrary script code in their own browser.
 - A flaw exists that is triggered as the referrer policy is not always honored when handling <link> elements. This may allow a context-dependent attacker to bypass the referrer policy.
 - A flaw exists in the 'PerDocumentStyleDataImpl::visited_styles_enabled()' function in 'servo/components/style/gecko/data.rs'. The issue is triggered when handling the CSS ':visited' selector for a document being used as an SVG image. With a specially crafted web page, a context-dependent attacker disclose visited history information.
 - A flaw exists in the 'FactoryOp::CheckPermission()' function in 'dom/indexedDB/ActorsParent.cpp' that is triggered as a web worker in Private Browsing mode can write to IndexedDB. With a specially crafted web page, a context-dependent attacker can uniquely fingerprint a user even when browsing in Private Browsing mode.
 - An overflow condition exists that is triggered as certain input is not properly validated when drawing and validating elements using Direct 3D 9 with the ANGLE graphics library. This may allow a context-dependent attacker to cause a buffer overflow and potentially execute arbitrary code.
 - A flaw exists related to speculative execution, which is used as a performance feature to speed up operations. This optimization can result in memory being cached during conditional branches handling out-of-bounds checks. Using a vulnerable code pattern, or a JIT engine or interpreter to generate such a pattern, an attacker can perform a Flush+Reload or Evict+Reload side-channel attack on the cache and disclose parts of the privileged kernel memory.
 - A flaw exists in the fundamental design related to out-of-order process execution, which is used as a performance feature to speed up operations. This optimization can result in memory being cached before exceptions are raised for restricted memory access. Using transient instructions in combination with a Flush+Reload side-channel attack a local attacker can disclose parts of the privileged kernel memory.
 - A flaw exists related to speculative execution, which is used as a performance feature to speed up operations. This optimization can result in memory being cached during indirect branch prediction. This may allow a local attacker to train the Branch Target Buffer (BTB) to trigger a false prediction to a specially crafted memory location, causing a speculative execution of a crafted gadget and the caching of arbitrary memory. Using a side-channel attack on the cache the attacker can disclose parts of the privileged kernel memory.

ID	Name	Product	Family	Published	Updated	Severity
105026	FreeBSD : mozilla -- multiple vulnerabilities (b7e23050-2d5d-4e61-9b48-62e89db222ca)	Nessus	FreeBSD Local Security Checks	12/6/2017	11/12/2025	high
105040	Mozilla Firefox < 57.0.1	Nessus	Windows	12/6/2017	11/18/2025	high
105039	Mozilla Firefox < 57.0.1 Multiple Vulnerabilities (macOS)	Nessus	MacOS X Local Security Checks	12/6/2017	11/12/2025	high
275696	Mozilla Firefox < 57.0.1	Nessus	MacOS X Local Security Checks	11/18/2025	11/18/2025	high
106884	GLSA-201802-03 : Mozilla Firefox: Multiple vulnerabilities	Nessus	Gentoo Local Security Checks	2/20/2018	10/29/2025	critical
700323	Mozilla Firefox < 57.0.1 Multiple Vulnerabilities	Nessus Network Monitor	Web Clients	8/21/2018	3/6/2019	low
700322	Mozilla Firefox < 57 Multiple Vulnerabilities	Nessus Network Monitor	Web Clients	8/21/2018	3/6/2019	critical

Detections

Analytics

Plugins Search

high

high

high

high

critical

low

critical