Build 9714 (August 6, 2026)
Question asked by Daniele (TDBnet) - 8/7/2026 at 8:15 AM
Unanswered
Any experiences with this new build?


  • Added: [API] User Settings -> UserDisableTwoFactor.
  • Added: [HA] [HUB] 2FA for cluster administrators.
  • Added: [HA] [HUB] An avatar for the cluster administrator user icon.
  • Added: [HA] [HUB] An HTML editor for Send Email and Send Message action types in Routing Rules.
  • Added: [HA] [HUB] Custom webmail login settings.
  • Added: [HA] [HUB] Feature limits for cluster administrators.
  • Added: [HA] [HUB] IP restrictions for cluster administrators.
  • Added: [HA] [HUB] Password requirements for cluster administrators.
  • Added: [HA] [HUB] The ability for a cluster administrator to update their password.
  • Added: [HA] [HUB] The ability to add multiple cluster administrator accounts.
  • Added: Error page when SpamFoo client isn't running.
  • Added: EWS settings to control concurrent subscription limits.
  • Added: MDN/DSN spam weight to Null Sender spam check settings.
  • Changed: [AP] Modified PutDraft API to remove unused input properties sendImmediately and scheduledSend.
  • Changed: [API] API endpoints 'SetUsersMailSettings' and 'SetUsersMailSettingsLegacy' can no longer be used to disable 2FA. Use 'DisableUserTwoFactor' instead.
  • Changed: [API] Modified SendMessage API to change scheduledSend input property from text to date.
  • Changed: [HA] [HUB] Separated out Manage tasks versus Settings tasks and added a top menu bar.
  • Changed: Clicking the SpamFoo dashboard should start the SpamFoo service if it isn't running.
  • Changed: Improved SharePoint Sync authentication checks.
  • Changed: Updated some of the Contacts API documented parameters.
  • Changed: Updated some of the spool message's API documented parameters.
  • Changed: Updated the Calendar's save-time confirmation dialog so it distinguishes an appointment that's started from one that's already finished.
  • Changed: Updated to ClamAV 1.5.3 for Windows. (Linux automatically downloads the latest updates.)
  • Changed: Updated to the latest Froala version 5.3.0.
  • Removed: What's New modals from webmail.
  • Fixed: [HA] [HUB] Error toast for mTLS setup is missing the proper punctuation.
  • Fixed: [HA] [HUB] Logging in as a mailbox user with 2FA enabled causes the login card to switch around.
  • Fixed: [HA] Hub authentication can sometimes return a 400 instead of 401.
  • Fixed: [HA] The Primary Cluster Administrator shouldn't be able to change their display name.
  • Fixed: An issue that can cause an invitee to be unable to use Propose New Time for a meeting invite.
  • Fixed: An issue were international domain names can fail at the a gateway server.
  • Fixed: An issue where authenticated users could search arbitrary mailbox metadata under certain conditions.
  • Fixed: An issue where some reports would show 0 for certain days/times.
  • Fixed: An issue with calendar Scheduling pages that are password protected.
  • Fixed: An issue with XMPP file uploads and the Extension Blacklist.
  • Fixed: Attaching a domain that already exists doesn't show the necessary "Domain Already Attached" error.
  • Fixed: Auto login fails to redirect users to the password update page when their password has expired.
  • Fixed: Auto login URL leads to a blank screen for users who have not logged into webmail previously.
  • Fixed: Changing the read state or flagging a message in the Scheduled folder silently removes it from the Scheduled queue.
  • Fixed: Classification folders can sometimes fail to show associated messages.
  • Fixed: Content Filters and Routing Rules for attachments with specific extensions don’t properly handle wildcards.
  • Fixed: Content Filters that search email headers do not correctly pick up wrapped header values.
  • Fixed: Creating a future recurring event can sometimes marks it as expired in webmail grid view.
  • Fixed: Deleting an email signature causes the Signature page to flicker.
  • Fixed: Emails sometimes do not move to Secondary Paths under certain conditions.
  • Fixed: Impersonation functionality was improved and now works better across various browsers.
  • Fixed: Importing Contacts does not properly refresh in certain browsers.
  • Fixed: Importing Users from a CSV doesn't properly refresh the modal.
  • Fixed: Improved password modal for password protected Scheduling links.
  • Fixed: In Reports, when hovering over Midnight in the graph, no time stamp is displayed.
  • Fixed: It's possible to set Primary and Secondary Storage paths for a domain to the same directory.
  • Fixed: Logins that fail due to network errors can cause webmail to go unresponsive.
  • Fixed: Manually running Content Filters can fail to report completion back to webmail.
  • Fixed: Message Archive showing count for TNEF attachments that are not displayed.
  • Fixed: Non-ASCII characters in a Username can cause SpamFoo Classification to fail.
  • Fixed: POP3 accounts on eM Client fail due to "-ERR NTLM Failure".
  • Fixed: Push and streaming subscriptions aren't terminated when a user changes their password [EWS].
  • Fixed: Push and streaming subscriptions aren't terminated when a user is detached or deleted [EWS].
  • Fixed: Scheduled messages can't be moved out of the Scheduled folder using the context menu.
  • Fixed: Scheduled messages in the Sent Items folder have the Date header set to the original send date instead of the scheduled send date.
  • Fixed: Scheduled messages sent to recipients on the local server aren't using the scheduled send date for the Received header.
  • Fixed: Scrolling a domain's System Settings can hang on a mobile device.
  • Fixed: SmarterMail service shutdown on Linux can sometimes time out.
  • Fixed: SpamFoo dashboards are now available at the user and domain level if email classification is disabled but spam checking is enabled.
  • Fixed: SRS not being honored when sending from an external address to a domain alias.
  • Fixed: Styling issues on the 'Getting Started' page for users logging in to webmail for the first time.
  • Fixed: The Notes section of a contact does not properly wrap text.
  • Fixed: Uploading images for domain mailing list's message can drop the image under certain conditions.
  • Fixed: Using an auto login URL for a mailbox user shows the 2FA page incorrectly.
  • Fixed: Using an auto login URL for a system administrator, after their password was changed, shows a Server Not Responding error.
  • Fixed: Webmail autocomplete doesn't load correctly when a mailing list has a blank entry in Allowed Posters.
  • Fixed: When editing a Port binding, the modal briefly shows an error under Select Certificate.
  • Fixed: When multiple Content Filter Actions for a message modify headers, only the Subject Prefix change gets applied.
  • Fixed: When SmarterMail creates a default Primary System Administrator, the password can get set to null.
  • Fixed: XMPP Connection counts are not always updated correctly.
  • Efficiency: Fixed delay with email reclassification.
  • Efficiency: Improved memory changes related to EWS push event subscriptions.
  • Efficiency: Improved memory changes related to EWS streaming event subscriptions.
  • Efficiency: Webmail UX improvements across various pages.
  • Security: Resolved an issue relating to MailService restarts.
Installed on 1 server. No issues so far
Gabriele Maoret - Head of SysAdmins and CISO at SERSIS
Currently manages 7 SmarterMail installations (1 in the cloud for SERSIS which provides services to a few hundred third-party email domains + 6 on-premise for customers who prefer to have their mail server in-house)
Bruce Replied
No issues so far here either after 12 hours
Total disaster...
Webmail not working and outlo0ok not working... no secure connection despite certificates beeing in place...

I fucking hate beeing a guinea pig...

Upgraded the fucking server and downgraded and still not working....

What a freaking nightmare. Emergency ticket here we come.... me paying for shit.
IMAP is working and the server exchanges emails. MAPI and webinterface is down  for all clients.


 

135-3209CDC7-007C Emergency ticket number.
3 servers upgraded: no issues here
Gabriele Maoret - Head of SysAdmins and CISO at SERSIS
Currently manages 7 SmarterMail installations (1 in the cloud for SERSIS which provides services to a few hundred third-party email domains + 6 on-premise for customers who prefer to have their mail server in-house)
Server fucked and SM cant resolve it so far. MAPI and HTTPS not working.

Fuck my life....
Answer from support


Taken together, these results show that SmarterMail is accepting and responding to the local requests normally, while certain external connections are being terminated during the Windows TLS/SSL handshake. Since the issue remains unchanged across both SmarterMail builds, there is no indication that the SmarterMail application or the 9714 update is responsible for the connection resets.

At this point, we have exhausted the SmarterMail-side troubleshooting avenues available for this particular behavior. The remaining evidence points to the environment rather than the SmarterMail application.

Server is fucked and MAPI is unusable.

Just a coincidence that thius happened after the 9714 ugrade... cause everything is working flawlessly.

Except that HTTPS and MAPI not working....
Roger Replied
Hi Brian,

Just a suggestion:
- Have you tried setting up a completely new virtual server with the latest version to see if it works?
- After reinstalling the server, could you migrate the existing accounts and then completely shut down the old server that’s having these issues?
- Have you used Fiddler Everywhere to analyze your site and the server side to see exactly what’s happening and where the problem lies? For example, with regard to cipher suites, compatibility, .NET Framework version, etc.?
- Does this affect only certain client systems making requests, or all of them, regardless of location, etc.? You can easily check this using webpagetest.org

What tests did you run during those 8 hours in your messages yesterday, and what results did you get? Were you able to narrow down the cause any further or identify any patterns?

Is there anything I can do to help you? Would it help if I set up a server for you, created email accounts on one of our systems, or anything else?
J. LaDow Replied
I don't know if this helps you, but at some point during one of our upgrades, we had to reset all the certificates that SM was trying use and re-save the certificate settings even though nothing changed with the certificates themselves...


MailEnable survivor / convert --
Douglas Foster Replied
Let go of the panic and start working systematically.
Can you log onto webmail on localhost:17017?   That will confirm whether or not the core system is fine.   If so, you have a connectivity issue.
Can you log onto webmail using localhost or actual name on port 443?   If so, then IIS is fine.  If you get a certificate warning, you should be able to click through the warning to confirm that certificates are the only problem?
While logged on locally, can you connect to Internet sites?   Check the network adapter settings to see if something is wrong there.   I had one system lose its gateway address at the seemingly same as I upgraded to build 9693.  I could not explain it and three other upgrades were fine, so I wrote it off, but maybe you have hit something similar.
Hopefully these tests will narrow your search. 
Sébastien Riccio Replied
It looks a bit like something that is happening between the browser and the reverse proxy (IIS if you're on Windows).

Could be a network issue (MTU size wrong for example).

Have you tried to access the webmail locally from a browser on the server to bypass network, this both directly on SM built in webserver localhost:17017 and the IIS endpoint (port 443) ?

Maybe also check the Event viewer:
  1. Press Win + R
  2. Type eventvwr and press Enter.
  3. Check these locations:
    • Windows Logs → Application — IIS/application errors are commonly recorded here.
    • Windows Logs → System — especially useful for WAS (Windows Process Activation Service) and IIS application-pool problems.
    • Applications and Services Logs → Microsoft → Windows → IIS-Configuration → Operational — useful for IIS configuration changes.

For an IIS application-pool crash

Go to Windows Logs → System, then filter for:

  • Source: WAS
  • Event ID: 5011

Microsoft identifies WAS Event ID 5011 as an indicator of an IIS worker-process crash. 

Sébastien Riccio
System & Network Admin

The server is fine according to SM.

HTTP works fine and the server exchanges ermail.

Its certificate or HTTPS related. Only login via https and the MAPI connectivity are broken. IMAP and everything else works fine.

NOthing changed in regards to Cipher suites and certificates before the upgrade.

Getting this in the logs

The key material used to protect TLS Session Tickets was not found at C:\ProgramData\Microsoft\Crypto\TlsSessionTicketKeys\S-1-5-18\SessionTicketKey.key.

SM did a Wireshark to the server and the OS terminates the TLS session before it hits SM software. Which seems logical in regards to the logs.

Havent got a clue how to fix that.
Sébastien Riccio Replied
For what it's worth, here is what claude is saying about your log entry:


What the message actually means

This isn't a certificate or binding failure — HTTPS keeps working. Schannel is telling you that an administrator-managed TLS session ticket key was configured for the LocalSystem account (S-1-5-18, which is what IIS/HTTP.sys runs as), but the key file itself is gone. The result is that TLS session resumption via tickets is disabled: every client does a full handshake, so you get slightly higher CPU and latency on TLS, nothing more.

That path is created by the Enable-TlsSessionTicketKey cmdlet. It configures a TLS server with a key created by New-TlsSessionTicketKey and sets up the rule that uses the key for the service account — and since IIS runs under the System account, the -ServiceAccountName parameter is System. So someone (or a hardening script / SSL Labs remediation) ran that cmdlet on this box at some point, and the .key file has since disappeared.

Common causes

  • Server cloned from a template / restored from image where the config existed but the key didn't survive (DPAPI-protected, tied to the machine).
  • Antivirus, disk cleanup, or a cleanup GPO deleting files under C:\ProgramData\Microsoft\Crypto\.
  • Broken NTFS permissions on TlsSessionTicketKeys — SYSTEM needs full control.
  • Partial Disable-TlsSessionTicketKey that removed the key but left the registry rule.

Diagnose


powershell
Get-ChildItem 'C:\ProgramData\Microsoft\Crypto\TlsSessionTicketKeys' -Recurse -Force
icacls 'C:\ProgramData\Microsoft\Crypto\TlsSessionTicketKeys'

If the S-1-5-18 folder exists but is empty, that confirms the diagnosis.

Two ways to fix it

Option A — drop the admin-managed key (simplest, single server). Windows then falls back to its own internally generated ticket keys.


powershell
Disable-TlsSessionTicketKey -ServiceAccountName System

Then reboot. This deletes the key and the corresponding rule that uses it. Note the trade-off: without an admin-managed key the server can't reuse existing session tickets, so it renegotiates session information more often.

Option B — regenerate the key (required if this is a load-balanced farm). If several nodes are meant to share the same ticket key so a client can resume against any node, you must recreate and redeploy it:


powershell
$Password = Read-Host -AsSecureString
New-TlsSessionTicketKey -Password $Password -Path 'C:\KeyConfig\TlsSessionTicketKey.config'
Enable-TlsSessionTicketKey -Password $Password -Path 'C:\KeyConfig\TlsSessionTicketKey.config' -ServiceAccountName System

Run the Enable- command with the same .config file and password on every node in the farm, then reboot each one (the key is loaded by LSASS at boot). Store the .config file and password somewhere safe — you'll need them again for any new node.

Then prevent the recurrence

Exclude C:\ProgramData\Microsoft\Crypto\ from AV scanning/quarantine and from any cleanup scripts, and verify SYSTEM still has full control on the folder. If you don't fix the underlying deletion, the event will simply come back after the next reboot.

If you want to verify the outcome, an SSL Labs scan (or openssl s_client -connect host:443 -reconnect) will show whether ticket-based resumption is working afterwards.

-- 
WITHOUT any warranty that this will help you with the issue you have

Sébastien Riccio
System & Network Admin

Sébastien Riccio Replied
And some diag suggestions about the error returned by the browser:


First, what the error narrows down

PR_CONNECT_RESET_ERROR is Firefox's way of saying it received a TCP RST, not a TLS alert. That distinction is the most useful clue you have:

  • A plain protocol or cipher mismatch normally produces a clean TLS handshake_failure alert, and Firefox reports SSL_ERROR_NO_CYPHER_OVERLAP or SSL_ERROR_PROTOCOL_VERSION_ALERT.
  • An RST means something tore the socket down instead of answering — HTTP.sys aborting before it could build a credential, a middlebox interfering, or the server killing the connection after the ClientHello.

So look first at things that make the listener fail rather than negotiate.

Most likely causes, roughly in order

Broken Schannel credential / private key access. HTTP.sys can't build the server credential, so it resets. Classic symptoms are Schannel event 36870 (0x8009030D) or 36871 in the System log. Causes: private key ACLs on C:\ProgramData\Microsoft\Crypto\RSA\MachineKeys, a certificate imported without its key, or a cert whose keyset broke after a renewal or restore.

Missing or orphaned HTTP.sys SSL binding. The IIS site binding exists in the UI but netsh http show sslcert shows no entry for that IP:port, or shows one with a stale certhash/appid. Very common after a certificate renewal done outside IIS Manager, or after a site was recreated.

SNI mismatch. A binding set to Require Server Name Indication with a client or health-check that doesn't send SNI, or two bindings on the same IP:port where only one has SNI. Also relevant if you use Centralized Certificate Store and the .pfx filename doesn't match the requested hostname — CCS resets rather than falls back.

Over-aggressive hardening. IIS Crypto templates, CIS baselines, or the SSL Cipher Suite Order GPO. Two specific traps here: the GPO string is limited to 1023 characters, so a long cipher list gets silently truncated and can leave an unusable set; and restricting ECC Curve Order (e.g. to P-384 only) will break clients that only offer P-256. Enabling FIPS mode (HKLM\SYSTEM\CurrentControlSet\Control\Lsa\FipsAlgorithmPolicy) has a similar effect.

Client certificate configuration. Require client certificate combined with TLS 1.3, or with renegotiation disabled, is a known source of resets on Server 2022. Try setting the SSL negotiation to Accept rather than Require, or pin the site to TLS 1.2, as a test.

Something in the path. Firewall/IPS with TLS inspection, a load balancer with its own cipher policy, or endpoint antivirus doing HTTPS scanning. Test from a machine on the same subnet as the server to rule this out.

How to pin it down

Enable verbose Schannel logging, then reproduce:


reg add "HKLM\SYSTEM\CurrentControlSet\Control\SecurityProviders\SCHANNEL" /v EventLogging /t REG_DWORD /d 7 /f

That requires a reboot to take full effect. Watch the System log for 36870, 36871, 36874, 36887 and 36888 — the internal error state number in those events is what actually identifies the failure.

Then check the binding layer:


powershell
netsh http show sslcert
Get-WebBinding | Format-List

And test the handshake directly from a client, which tells you whether the server answers at all:


openssl s_client -connect yourhost:443 -servername yourhost -tls1_2
openssl s_client -connect yourhost:443 -servername yourhost -tls1_3

If TLS 1.2 works and 1.3 resets (or vice versa), you have your answer immediately. If the ClientHello gets an RST with no ServerHello at any version, it's a credential or binding problem, not a cipher problem.

Finally, check C:\Windows\System32\LogFiles\HTTPERR\ — HTTP.sys logs connection-level aborts there with reason codes, and the IIS site logs will show nothing at all for these requests (which itself confirms the connection died before HTTP.sys handed it to IIS).

One note: this is unrelated to the session ticket key warning from your earlier question — that one degrades performance but never resets connections.

Sébastien Riccio
System & Network Admin

It works in inkognito mode. But MAPI still not working.
Disabled HTTP2 on the server and MAPI and https started working again.

It was running http2 before the update and somehow something changed.

Reverting back to http1.1 solved the issues.

Can anyone verify that handshakes is handled in http2 or http1.1 on your servers??
Sébastien Riccio Replied
[root@prism:~] # curl -I -v https://mail01.xxx.com/interface/root

*   Trying 94.x.96.x:443...
* Connected to mail01.xxx.com (94.x.96.x) port 443 (#0)
* ALPN: offers h2,http/1.1
* TLSv1.3 (OUT), TLS handshake, Client hello (1):
*  CAfile: /etc/ssl/certs/ca-certificates.crt
*  CApath: /etc/ssl/certs
* TLSv1.3 (IN), TLS handshake, Server hello (2):
* TLSv1.2 (IN), TLS handshake, Certificate (11):
* TLSv1.2 (IN), TLS handshake, Server key exchange (12):
* TLSv1.2 (IN), TLS handshake, Server finished (14):
* TLSv1.2 (OUT), TLS handshake, Client key exchange (16):
* TLSv1.2 (OUT), TLS change cipher, Change cipher spec (1):
* TLSv1.2 (OUT), TLS handshake, Finished (20):
* TLSv1.2 (IN), TLS handshake, Finished (20):
* SSL connection using TLSv1.2 / ECDHE-RSA-AES256-GCM-SHA384
* ALPN: server accepted h2
* Server certificate:
*  subject: CN=mail01.xxx.com
*  start date: Jul  7 07:03:12 2026 GMT
*  expire date: Oct  5 07:03:11 2026 GMT
*  subjectAltName: host "mail01.xxx.com" matched cert's "mail01.xxx.com"
*  issuer: C=US; O=Let's Encrypt; CN=YR1
*  SSL certificate verify ok.
* using HTTP/2
* h2h3 [:method: HEAD]
* h2h3 [:path: /interface/root]
* h2h3 [:scheme: https]
* h2h3 [:authority: mail01.xxx.com]
* h2h3 [user-agent: curl/7.88.1]
* h2h3 [accept: */*]
* Using Stream ID: 1 (easy handle 0x55b91406c040)
> HEAD /interface/root HTTP/2
> Host: mail01.xxx.com
> user-agent: curl/7.88.1
> accept: */*
> 
< HTTP/2 200 
HTTP/2 200 
< content-length: 0
content-length: 0
< content-type: text/html; charset=utf-8
content-type: text/html; charset=utf-8
< content-security-policy: default-src 'self';frame-src 'self' *.youtube.com youtu.be *.smartertools.com docs.google.com;script-src * 'unsafe-inline';font-src * 'unsafe-inline' data:;img-src * 'unsafe-inline' data: blob:;style-src * 'unsafe-inline';media-src *;frame-ancestors 'self';connect-src *;
content-security-policy: default-src 'self';frame-src 'self' *.youtube.com youtu.be *.smartertools.com docs.google.com;script-src * 'unsafe-inline';font-src * 'unsafe-inline' data:;img-src * 'unsafe-inline' data: blob:;style-src * 'unsafe-inline';media-src *;frame-ancestors 'self';connect-src *;
< x-frame-options: SAMEORIGIN
x-frame-options: SAMEORIGIN
< x-xss-protection: 1; mode=block
x-xss-protection: 1; mode=block
< x-content-type-options: nosniff
x-content-type-options: nosniff
< x-robots-tag: noindex
x-robots-tag: noindex
< x-powered-by: ARR/3.0
x-powered-by: ARR/3.0
< x-powered-by: ASP.NET
x-powered-by: ASP.NET
< strict-transport-security: max-age=0
strict-transport-security: max-age=0
< date: Sun, 09 Aug 2026 13:55:14 GMT
date: Sun, 09 Aug 2026 13:55:14 GMT
Ours answer correctly to http/2 requests, but I'm on the previous build, not upgraded yet due to changes in the API I need to reflect on our control panel.
Sébastien Riccio
System & Network Admin

Douglas Foster Replied
Thank you, Sebastien, for a reminder of how to use A.I. to systematically investigate obscure problems.
Douglas Foster Replied
Brian, I am posting an abuse report about your handling of this crisis.

First, I will remind you that the f-bomb is a prayer to the demon world to cause the object of your curse to be raped.   This is never an appropriate response to a crisis, and it is wise to assume that the Faust story contains an element of truth:   the demon world never answers your prayers without taking your soul in exchange.

Second, it is unwise to curse the people whom you desperately need to help you out.

Third, ST provides this forum, so we are guests in a house they provide.   When a guest, it is necessary to act like a guest, even when we are unhappy with the host.

Fourth, the "Build NNNN experience" topics are for summary information about your experience.    You can post once to say, "I have an unexplained problem about ...", and again to say, "my problem was resolved by...."   If you want to talk in detail about your particular problem, you need to start a separate topic.

Fifth, some problems are just plain mysterious and hard for anyone to find.   Life is like that.   In Downton Abbey, the dowager said, "Life is solving one problem after another, and then you die."   Not a reason to give up, just a reason to keep perspective.




Reply to Thread

Enter the verification text