• The new Python extension is now available. It allows customers to deploy and manage WSGI-based Python applications on their websites directly from Plesk.
  • The Plesk Mobile apps for iOS and Android and the Plesk Mobile Center extension have been deprecated, and are no longer available for new installations. In Plesk Obsidian 18.0.84, the Plesk Mobile Center extension will be removed from servers during the upgrade, and the Plesk Mobile apps will lose their server management features.
  • The Social Login extension has been deprecated and will be removed in Plesk Obsidian 18.0.84. After that, it will no longer be possible to log in to Plesk with a Google, Facebook, GitHub account, or with a passkey. Plesk usernames and passwords, as well as logging in with a WebPros Account, are not affected.

Issue This update addresses a critical security issue. We strongly recommend that you apply it as soon as possible.

Azurel

Silver Pleskian
Server operating system version
AlmaLinux 8.10
Plesk version and microupdate number
18.0.74#2
In Changelog I found this
This update addresses a critical security issue. We strongly recommend that you apply it as soon as possible.
for Plesk Obsidian 18.0.74 Update 2 and Plesk Obsidian 18.0.73 Update 5.

If this is so very important, why is no corresponding notification sent to the admin account by email? I receive an email when an update has been installed automatically, but not when urgent action is required? Or has anyone received such a notice?
 
What is the actual security issue ??

And no for security issues Plesk should do a better job of notifying owners of licenses.

Ladylinux
 
I’ve shared a suggestion regarding this issue in the Plesk Suggestions and Feedback forum for Plesk to improve security notifications:

 
Hi,

I would edit but you only give people a few minutes to edit or delete

Unbelievable this is a ROOT level Vulnerability!!!

This is the actual vulnerability


https://support.plesk.com/hc/en-us/...s-to-execute-arbitrary-code-on-behalf-of-root

And



No where is it even clearly shown that this affects earlier Plesk Versions and those just have clunky workarounds!!

Unacceptable Plesk!!
 
It would be really helpful if security advisories clarified whether an issue can be exploited remotely by an unauthenticated visitor or requires existing access to Plesk (e.g., as a customer or admin).

I don't have any customers or additional admins, yet these critical security warnings always put me on high alert, only to find out weeks later that the vulnerability required authenticated access. A little more transparency about the actual risk would be greatly appreciated!
 
I'm a bit confused by your response. You're referencing articles published approximately 11 months ago.

Are you suggesting that Plesk has only now released a patch for a vulnerability that has been publicly documented on its own website for nearly a year?

If so, could you please clarify why the vulnerability remained unpatched for that length of time, or whether the recently released update addresses a different issue?
 
@nmdpa3 I think there’s just a timing mix-up here. The two articles Ladylinux linked (CVE-2025-66430 and CVE-2025-66431) were published in December 2025, right when this thread started, and they belong to exactly the update Azurel asked about (18.0.74 Update 2 / 18.0.73 Update 5). Both articles list 18.0.73.5 and 18.0.74.2 as the fixed versions, so the vulnerabilities were not left unpatched for a year.

@Azurel you’re right about the exposure. Both are local privilege escalations: an attacker needs a logged-in Plesk user account (for CVE-2025-66431 even with specific permissions like site and domain management). If you run a server without customers or additional admins, that attack path essentially doesn’t exist for you. Updating is still the right call, but no reason for panic.

@Ladylinux fair point that older versions are less convenient: for 18.0.72 and earlier, the articles only offer an upgrade to the next release or a manual patch/workaround.

What does remain a legitimate criticism is the communication. “Critical security issue” without CVE, component or required privileges is hard to assess when the details only show up later on the support site. Maarten’s suggestion for proper advisories deserves an upvote. ;)

Regards
Tessa
 
@tessa67 and @AlL ............ and @Sebahat.hadzhi

This quote

What does remain a legitimate criticism is the communication. “Critical security issue” without CVE, component or required privileges is hard to assess when the details only show up later on the support site. Maarten’s suggestion for proper advisories deserves an upvote

is something that also - again - applies to the latest MU3 (on 18.0.82) and MU8 (18.0.81).


A line like

"This update addresses a critical security issue. We strongly recommend that you apply it as soon as possible."

in the release notes should not be sufficient.


@Sebahat.hadzhi .......... I can imagine that Plesk does not want to share sensitive information online, but there is also a Plesk forum.

For instance, it could also be considered to add "sensitive data" to that part of the Plesk forum that is not accessible to all of Plesk users.

That way, some of the (active) Plesk users can contribute at least to the Plesk forum, by using the up-to-date information to help out ALL Plesk users.....

Nevertheless, it is often better to share all information with everybody.

So, two options : share some information ....... or share all information with everybody.

Please note that some shared information can also lead to some feedback to Plesk team - this can speed up development of security patches.

Stated differently, I do not really understand why Plesk does not take advantage of the latter.


In conclusion, NOT an option : sharing no information at all with nobody.

It simple does not make sense.

Kind regards....
 
@tessa67 and @AlL ............ and @Sebahat.hadzhi

This quote



is something that also - again - applies to the latest MU3 (on 18.0.82) and MU8 (18.0.81).


A line like

"This update addresses a critical security issue. We strongly recommend that you apply it as soon as possible."

in the release notes should not be sufficient.


@Sebahat.hadzhi .......... I can imagine that Plesk does not want to share sensitive information online, but there is also a Plesk forum.

For instance, it could also be considered to add "sensitive data" to that part of the Plesk forum that is not accessible to all of Plesk users.

That way, some of the (active) Plesk users can contribute at least to the Plesk forum, by using the up-to-date information to help out ALL Plesk users.....

Nevertheless, it is often better to share all information with everybody.

So, two options : share some information ....... or share all information with everybody.

Please note that some shared information can also lead to some feedback to Plesk team - this can speed up development of security patches.

Stated differently, I do not really understand why Plesk does not take advantage of the latter.


In conclusion, NOT an option : sharing no information at all with nobody.

It simple does not make sense.

Kind regards....
I first felt the same, but now I'd actually prefer that Plesk does not publish what exactly was improved. Because, if they did, they'd foremost give hackers a hint how to attack. Maybe a compromise, first publish the message that we must absolutely update, then wait another two to four weeks, then explain why the update was important.
 
I first felt the same, but now I'd actually prefer that Plesk does not publish what exactly was improved. Because, if they did, they'd foremost give hackers a hint how to attack. Maybe a compromise, first publish the message that we must absolutely update, then wait another two to four weeks, then explain why the update was important.

@Bitpalast

As for me, the evolution of opinions is - exactly - the other way around.

At first, I would have agreed with you : careful release of vulnerability information would have been preferred.

Now, this has completely changed ....... and I prefer to have full disclosure upfront.


There are three reasons for this change :

1 - some vulnerabilities are quite easy to mitigate or annihilate completely : information upfront allows Plesk admins to take immediate action

2 - Plesk updates are not released instantly and/or can be buggy (with some of them even increasing the vulnerability that was intended to be patched) and some Plesk updates do not patch all known vulnerabilities (which is normal) : information upfront allows Plesk admins to take appropriate action after updates

3 - the attack surface and the number of vulnerabilities will increase with the continuous release of more advanced AI : information upfront (to humans, Plesk admins) is stating "the obvious" that has already been disclosed (by humans, security researchers and hackers alike) as more or less "public" information that has been "learned" (by AI) and that still has to be "learned" by most Plesk admins (humans, needing information upfront to know something)


However, the main reason for this change and evolution towards "full disclosure upfront is required" is totally different from the 3 aforementioned reasons.

In essence, people are inclined to become passive and to wait for Plesk updates ........ if those updates "promise" to resolve issues.

The difference between "active" and "passive" is the time to react upon a certain situation and this time is positive, not zero.

The time to react upon a certain situation, if and when relying on Plesk updates, is exactly equal to SUM of the (given) time to develop the Plesk update, still to be augmented with the (variable) time to implement the Plesk update - which time of implementation can vary across Plesk admins.

The time to react upon a certain situation, if and when relying on "upfront information", can - by definition - be shorter and - probably - will be shorter.

In fact, the time always starts at the moment that the aforementioned "certain situation" is known.

In the case of relying on Plesk updates, there is the starting time (the moment where the situation is known), followed by the (given) time to develop Plesk, again followed by a (variable) time to implement the Plesk update.

In the case of relying on upfront information, there is the starting time (again the moment where the situation is known), followed by an opportunity to patch and/or mitigate the issue caused by the situation in question, hence providing time to safely wait for the Plesk update.


The aforementioned opportunity is real and valuable, since in most cases, mitigation or annihilation of specific situations can be fast.

Consider an issue that simply requires a kernel update ........ no solution in Plesk updates, but solutions are to be found in kernel updates (which are fast).

Consider an issue that simply requires one line of config code or an update of PHP ......... simply update PHP or adjust the config.


In short, it is highly valuable and desirable to reduce the time to react ......... and that can only be done when "information upfront" is provided.


Kind regards...

PS I do agree with you that, at least for some rare scenario's, full disclosure of information should never be considered. Nevertheless, these scenario's are really rare and then still it is valuable and desirable to share parts of the relevant information with respect to symptoms (hence not causes).
 
@trialotto @Bitpalast

I think there is an interesting inconsistency in the current changelog itself: Site Import 1.13.2 gets a proper security note with a CVE:

This update addresses a critical security issue. We strongly recommend that you apply it as soon as possible. (CVE-2026-107246)

But the core updates 18.0.82 Update 3 and 18.0.81 Update 8, released on the same day, only contain the generic sentence with no CVE, affected component, affected versions, or indication whether the issue requires an authenticated Plesk user.

So Plesk clearly can publish structured security information in the changelog. It just does not apply the same standard consistently to core updates.

I am not necessarily asking for full technical details immediately. I understand the argument about not handing attackers a roadmap while patches are still rolling out. But a minimum set of metadata for every security-relevant core update would already make a huge difference:
  • CVE number, or at least a stable Plesk advisory ID until a CVE is assigned
  • First fixed Plesk version and affected versions
  • The affected component or feature area
  • Whether exploitation requires an unauthenticated remote attacker, an authenticated Plesk user, a specific permission, or local access
  • A severity classification, ideally CVSS or SSVC
  • A safe, non-exploit-enabling mitigation, if one exists
  • A permanent link to the advisory
This could be published in the changelog RSS and ideally also in the inf3 metadata. It would let admins distinguish between “patch immediately regardless of server usage” and “the practical risk depends on whether I have customers, resellers, additional admins, or a particular feature enabled.”

That would also directly address the original concern in this thread: a solo admin without additional accounts could quickly assess the real exposure instead of waiting for details to appear later in support articles. Until then, assessing the actual risk is mostly guesswork.

Regards
Tessa
 
@Sebahat.hadzhi .......... I can imagine that Plesk does not want to share sensitive information online, but there is also a Plesk forum.

@trialotto , we have discussed this matter many times before. I can't share any details on my behalf until the Plesk team decides the information is ready to go public. Once that time comes, I can highlight the vulnerability details on the forum, so at least regular forum users won't need to go through the support articles.

I’m not keen on the idea of sharing information with only specific group of users. To me, that leaves impression that some users are treated as more privileged than others. I believe everyone should be treated equally and have access to the same information.
 
@Sebahat.hadzhi

This

I’m not keen on the idea of sharing information with only specific group of users. To me, that leaves impression that some users are treated as more privileged than others. I believe everyone should be treated equally and have access to the same information.

is a valid point and I certainly agree with and/or believe in equality.

However, your point is missing the point : Plesk already does treat Plesk forum users differently for many many many years, by assigning them different roles and names (from Regular to Golden Pleskian) ........ and, more importantly, by granting some people access to restricted areas of the Plesk forum.

In essence, Plesk could make use of the already existing Plesk Guru system.

Otherwise, Plesk could choose for "equality" and, taking into account

I can't share any details on my behalf until the Plesk team decides the information is ready to go public. Once that time comes, I can highlight the vulnerability details on the forum, so at least regular forum users won't need to go through the support articles.

stick to the principle of : not sharing information with nobody, unless information can be made public and shared afterwards.


The simple fact is that Plesk can make it's own choices.

The important fact here is that there are no good or bad choices.

To be honest, in the world of security and vulnerability, there are no optimal choices, it is always a trade-off.


However, Plesk does need to realize that "equality" cannot be obtained with the current principles and policies used by Plesk.

Some people (mainly Plesk Guru's) already have identified and - sometimes - have even patched security / vulnerability issues before Plesk identifies them.

Equality - apparently - is obtained by

a) for all people, the possibility to report a bug to Plesk and/or to mention an issue on the Plesk forum,

b) for some people, the possibility to post "workarounds" and "manual patches" on the Plesk forum,

c) for all people, to implement "workarounds" and "manual patches" if and only if they know how,

d) for all people, to wait for a Plesk macro or micro update that includes a fix or patch.


This - simply - means that "equality" is not present at once, it is a gradual process over time .......... that can take some considerable time.

This - also - means that everybody has to wait for Plesk updates if nobody is posting identified security / vulnerability issues and/or solutions on the forum.


In conclusion, the current policy of Plesk might be debatable or justifiable, but it certainly takes out any incentive to take action and help on the forum.

Not having information is simply equivalent to not having any possibility to do something about the thing that is unknown.


@Sebahat.hadzhi ......... I do not have the intention to enter into a debate and there is no need to.

I simply assume that Plesk is mitigating / resolving issues sooner or later (after minutes, hours or even after many many years), knowing that Plesk is not very different from other software : there is no guarantee, there is only a responsibility of a sysadmin to check and double-check everything and, if necessary, apply patches or fixes or improvements before they are released.

However, if Plesk forum members raise questions about sharing information

........ and that information is already widely spread online as a factual security / vulnerability issue

........ then it should not come as a surprise that these Plesk forum members ask these questions

........ and it certainly is not strange that some Plesk forum members find it odd or even undesirable that Plesk does not provide information about a security or vulnerability issue (that is publicly known) and the way that issue affects Plesk.


To be honest, I am still writing because I am to a minor extent flabergasted.

Stated differently, wouldn't it be better if Plesk simply could say : Plesk is not vulnerable for CVE such-and-so ....... or ........ patch being released soon ... or ... patch already included in the OS / kernel update such-and-so ????


Kind regards..........
 
Last edited:
Back
Top