• Plesk Uservoice will be deprecated by October. Moving forward, all product feature requests and improvement suggestions will be managed through our new platform Plesk Productboard.
    To continue sharing your ideas and feedback, please visit features.plesk.com

Resolved 421 Misdirected Request, Activesync no longer working

C

Compy

Basic Pleskian
Server operating system version
os_Ubuntu 22.04
Plesk version and microupdate number
Obsidian v18.0.71_build1800250704.07
So after the Misdirected Request, the sites are up but Outlook cant connect with Activesync.
Imap works.

Using Plesk Premium Email, Postfix, Dovecot
 
The error started at the same time as Misdirected Request, the server has since been updated to v18.0.71
 
scsa20 said:
It's probably related to the apache change that the plesk team is working on releasing an actual patch but for now there is a work on.

Refer to this post:

Issue - 421 Misdirected Request

for fix CVE-2025-23048 the SSLStrictSNIVHostCheck directive is now enabled by default and proxy must send host to apache
talk.plesk.com talk.plesk.com

Click to expand...

Thanks @scsa20! You are a hero, my head hurts like hell after this day. :p

Also found that a lot of updates for kolab and roundcubemail were stuck. Nothing you could see in the control panel.
 
Has anyone managed to get Activesync working on their server after the Apache bug? Issue - 421 Misdirected Request patch does not resolve the error.
Or if anyone knows which log to look at
 
Hello, @Compy . Have you updated Plesk to 18.0.71 Update 1 already or you only applied the patch? Can you please also provide the exact error message you encounter?
 
Compy said:
Has anyone managed to get Activesync working on their server after the Apache bug? Issue - 421 Misdirected Request patch does not resolve the error.
Or if anyone knows which log to look at
Click to expand...

Yes, after applying the fix or installing 18.0.71 Update 1, ActiveSync should be fine.
 
Still disconnected after update to Plesk Obsidian 18.0.71 Update #1 Web Pro Edition

have tried to renew ssl, lets encrypt.
Disable Reverse Proxy Server (nginx) will result in 404

web
https://**domain**/Autodiscover/Autodiscover.xml

Misdirected Request​

The client needs a new connection for thisrequest as the requested host name does not matchthe Server Name Indication (SNI) in use for thisconnection.
Apache Server at webmail.**domain** Port 443
Click to expand...

Outlook
2025.07.18 23:25:49 **email** Synch operation started (flags = 00000031)
2025.07.18 23:25:49 **email** StartImport(flags = 00000000, max msg = ffffffff): full items
2025.07.18 23:25:49 **email** EnsureProtoInfo: Error getting remote URL (hr = 0x80070057)
2025.07.18 23:25:49 **email** Unable to parse URL for protocol handler:
Click to expand...
Mozilla Thunderbird
Communications error (HTTP status 421)
Click to expand...
 
Guys, it was confirmed that the workaround previously suggested for the Apache update is indeed causing an issue of ActiveSync and IMAP connectivity with Plesk Premium Email. The workaround we can suggest is:

  1. Connect to the server via SSH.
  2. Rename the file deployed by the hotfix:
    Bash: 
    mv /etc/nginx/conf.d/fixssl.conf /etc/nginx/conf.d/fixssl.conf-old
  3. Regenerate the web server configuration:
    Bash: 
    plesk repair web -y

@Compy can you please confirm if that solves the issue for you? Thank you in advance.
 
& if that doesn't work, check the error log for ActiveSync here:

/var/log/kolab-syncroton/errors.log

I had my own dramas on a 20.04 LTS server after updating kolab/roundcube resulting in missing columns in the roundcube SQL database (was causing 500 errors in the domain's NGINX log for both sync and ping AS commands from the client).

It was all resolved simply by adding back the two missing columns that were reported missing in the errors.log.

Even if your issue isn't the same, it's the errors.log file that will most likely reveal the reason your AS isn't working.
 
Sebahat.hadzhi said:
Guys, it was confirmed that the workaround previously suggested for the Apache update is indeed causing an issue of ActiveSync and IMAP connectivity with Plesk Premium Email. The workaround we can suggest is:

  1. Connect to the server via SSH.
  2. Rename the file deployed by the hotfix:
    Bash: 
    mv /etc/nginx/conf.d/fixssl.conf /etc/nginx/conf.d/fixssl.conf-old
  3. Regenerate the web server configuration:
    Bash: 
    plesk repair web -y

@Compy can you please confirm if that solves the issue for you? Thank you in advance.
Click to expand...
No, no change.
 
BuzzLightBurger said:
& if that doesn't work, check the error log for ActiveSync here:

/var/log/kolab-syncroton/errors.log

I had my own dramas on a 20.04 LTS server after updating kolab/roundcube resulting in missing columns in the roundcube SQL database (was causing 500 errors in the domain's NGINX log for both sync and ping AS commands from the client).

It was all resolved simply by adding back the two missing columns that were reported missing in the errors.log.

Even if your issue isn't the same, it's the errors.log file that will most likely reveal the reason your AS isn't working.
Click to expand...
The log is empty. console.log empty, only the sendmail.log is used.
 
Check the domain's nginx log for any entries related to a client connecting to AS. Filter by "ActiveSync".
You should at least see some POST requests reaching the server.

If not, then the client might need inspection (a firewall on the client computer or possibly on the server)
 
When running Diagnose & Repair, Mail Server. No issues detected, It could be that the server cannot check Autodiscover when Plesk Premium Email is installed. The script should find the 421 error right away.
 
BuzzLightBurger said:
Check the domain's nginx log for any entries related to a client connecting to AS. Filter by "ActiveSync".
You should at least see some POST requests reaching the server.

If not, then the client might need inspection (a firewall on the client computer or possibly on the server)
Click to expand...
web
**IP** - - [22/Jul/2025:14:37:17 +0200] "GET /Autodiscover/Autodiscover.xml HTTP/2.0" 421 389 "-" "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:140.0) Gecko/20100101 Firefox/140.0"
Click to expand...
proxy_access_ssl_log
**IP** - **mail** [22/Jul/2025:06:47:32 +0200] "POST /Microsoft-Server-ActiveSync?Cmd=FolderSync&User=**mail**&DeviceId=**ID**&DeviceType=WindowsOutlook15 HTTP/1.1" 421 389 "-" "Outlook/15.0 (15.0.5337.1000; MSI; x64)"
Click to expand...
Mozilla Thunderbird TbSync
**IP** - **mail** [22/Jul/2025:14:38:59 +0200] "POST /Microsoft-Server-ActiveSync?Cmd=Settings&User=**mail**&DeviceType=TbSync&DeviceId=**id** HTTP/2.0" 421 389 "-" "Thunderbird ActiveSync"
Click to expand...
 
proxy_error_log
2025/07/22 09:38:06 [error] 38537#0: *4085 open() "/var/www/vhosts/**domain**/httpdocs/favicon.ico" failed (2: No such file or directory), client: **ip**, server: **domain**, request: "GET /favicon.ico HTTP/2.0", host: "**domain**", referrer: "https://**domain**/Autodiscover/Autodiscover.xml"
Click to expand...
 
Yup, your entire problem is that 421 that hasn't been resolved by applying either the patch or the Plesk update :(

Are there any additional directives in the Apache/NGINX settings for the domain?

The domain I repaired overnight relies on both Apache and NGINX, running both the original patch (and I applied the Plesk update) and was sketchy with iOS 18 on ActiveSync (hence the reason I updated kolab/roundcube last night) which, after fixing the SQL issue, has at least fixed everything so it truly is a mystery why you're getting this 421, when you're not getting it for your domain(s).

If you updated kolab/roundcube, did you retain your original config files or replace with new?
If you haven't updated kolab/roundcueb (due to the GPG key issue recently) try updating them.

I'm hoping someone from Plesk will dip into this and look at your logs to work out what is happening here. I know what it's like to Google like crazy to find someone with a similar issue (did it last night) but conceded mine was an isolated issue and required time to diagnose. It would appear you're in the same boat :(

Keep looking for clues and post them here. It's probably something really stupidly simple that'll take hours to find :/
 
Sebahat.hadzhi said:
Guys, it was confirmed that the workaround previously suggested for the Apache update is indeed causing an issue of ActiveSync and IMAP connectivity with Plesk Premium Email. The workaround we can suggest is:

  1. Connect to the server via SSH.
  2. Rename the file deployed by the hotfix:
    Bash: 
    mv /etc/nginx/conf.d/fixssl.conf /etc/nginx/conf.d/fixssl.conf-old
  3. Regenerate the web server configuration:
    Bash: 
    plesk repair web -y

@Compy can you please confirm if that solves the issue for you? Thank you in advance.
Click to expand...

After this now all sites has the 421 again, and some css errors.

This will happen when the server is running
Checking web server configuration. Please wait
Click to expand...

running
echo -e "proxy_ssl_server_name on;\nproxy_ssl_name \$host;\nproxy_ssl_session_reuse off;" > /etc/nginx/conf.d/fixssl.conf && systemctl restart nginx
Click to expand...
will fix the sites again, not the ActiveSync error.
 
You must log in or register to reply here.

Similar threads

P
Resolved 421 Misdirected Request
2 3 4
Replies
74
Views
13K
VNick
V
W
Issue 421 Misdirected Request on IP address after Apache update
Replies
2
Views
373
BuzzLightBurger
B
B
Question Spam Folder Location
Replies
5
Views
558
trialotto
T
C
Issue ActiveSync SOGoMail doesn't work
Replies
1
Views
3K
Peter Debik
P
M
Issue Postfix - warning: connect to transport private/plesk-domain.org.uk-123.145.167.189-: Connection refused
Replies
8
Views
1K
MHC_1
M
Back
Top