DirectAdmin v1.667

fln said:
@ViT, thanks for reporting the issue. This problem only affects pureftpd (not proftpd).
Click to expand...
Also already reported december last year:
Richard G

Post in thread 'DirectAdmin v1.657'

I maybe have discovered a weird password issue. But I don't know 100% sure if this is this version related or something else. But we never experienced this before.

So we experienced something very weird, newest DA version.
Freshly installed server with new DA admin pass we got after installation.

We restored the old admin account from old server, so we have alle correct settings from the main domain which is in there.
Ofcourse yes, the current admin passwords will be overwritten and then we need to use the old admin password to login again, that is well known.

To get the new passwords...
Your answer:
Password change after FTP and main account password desync issue is not new, but we will look into this, thanks.
Click to expand...

Glad to see it's getting fixed now. ;)
 
Did you read the topic? Post 7?

fln

Post in thread 'DirectAdmin v1.667'

@Dauser2007 and @HageHosting could it be that you have customized nginx configure script? The script was updated in DA 1.667 the official version in custombuild/configure/nginx/configure.nginx adds additional modules with --add-module=static_modules/ngx_cache_purge. Adding module with --add-module=../ngx_cache_purge was used in older DA.

Based on the error message it seems that configure script still uses old location.

Changes to the configure script is made to make sure nginx is being built in temporary directory and does leave temporary files dangling in...
  • Like
 
Hi

Since this update, we cannot change the PHP version anymore via the users' DirectAdmin 'home' page.
When I click the button 'change', it says 'Feature disabled'.
However, it is still possible via Domain Setup > click domain > PHP Version Selecter in the upper right corner.
So it is a minor issue.

Kr
Dries
 
An update is released to fix a regression with password changing in Enhanced and legacy API (to allow changing system and FTP account passwords separately).
 
fln said:
An update is released to fix a regression with password changing in Enhanced and legacy API (to allow changing system and FTP account passwords separately).
Click to expand...

As I believe your license server is on Cloudflare and they are experiencing instability, several Directadmin installations that we manage are down after the update, displaying an error message when validating the license, and pointing to /evo/login/license-expired

Cloudflare Status

Welcome to Cloudflare's home for real-time and historical data on system performance.
www.cloudflarestatus.com www.cloudflarestatus.com

How to solve this? Your license validation system should not take everyone offline due to an availability failure in your own license system.
 
DA services affected by this will retry and start automatically. It seems CF got it under control already. If CF would not be able to react quickly we have the infrastructure in place to bypass CF. Due to small scope of the incident and fast response from CF did not activate it.
 
fln said:
DA services affected by this will retry and start automatically. It seems CF got it under control already. If CF would not be able to react quickly we have the infrastructure in place to bypass CF. Due to small scope of the incident and fast response from CF did not activate it.
Click to expand...

Several servers that we manage had DirectAdmin down for an extended period of time and displayed an expired license message to all customers. This is critical and causes us great concern.

This incident makes it clear that the license validation system puts all customers at risk in the event of network instability, which can occur for a variety of reasons (routing problems, problems with Cloudflare itself, etc.). It is not expected that services will be down due to a problem with your license validation system, and only "some time" after your alternative infrastructure takes over.

It would be important for the license validation system to have a grace period, perhaps 24/48 hours, and during this period the administrator would only receive alert messages, but the panel would remain active.
 
unihostbrasil said:
It would be important for the license validation system to have a grace period, perhaps 24/48 hours, and during this period the administrator would only receive alert messages, but the panel would remain active.
Click to expand...
True, but there will be people abusing this ;)
 
unihostbrasil said:
It would be important for the license validation system to have a grace period, perhaps 24/48 hours, and during this period the administrator would only receive alert messages, but the panel would remain active.
Click to expand...
I've not experienced this myself, but if DirectAdmin suddenly stopped working for reasons outside my control, that would indeed be very bad. If a long grace period is a problem due to abuse, perhaps some mechanism where the grace period is dynamic could work. E.g. a steady license-checkin for x days means a grace period of x hours.
 
We already have all this. When there is no license hopping the grace period is being extended. DA starts by checking the license immediately and then exponentially increases the check period.

This is why CloudFlare problem was barely visible in our licensing system metrics (number or live servers). It looks like a big deal from the server admin perspective if it affected his server. However looking at the big picture it was barely noticeable. Please keep in mind that DA does not bring down any other services (like websites, emails, databases, etc.). So if there is a transient DA problem you might not be able to add a new user or domain but all your existing services are working.
 
Hi,

I got this while wanting to update:
adding module in ../modsecurity-nginx-v1.0.3
./configure: error: no ../modsecurity-nginx-v1.0.3/config was found
doNginx: failed to compile '/usr/local/directadmin/custombuild/cache/nginx-1.27.1.tar.gz' inside '/usr/local/directadmin/custombuild/tmp/tmp.fWAaECfP1d.nginx-1.27.1.tar.gz'
failed to build nginx
Click to expand...
 
@ericosman. Most likely you have customized nginx configure script. Please check original configure script bundled with DA.
 
We updated DirectAdmin on several servers to version 1.667 and identified two issues:

1- We use the "Icons Grid" layout, and when logged in as a "user," the account statistics information now displays a horizontal scrollbar, cutting off a large portion of the text.

print1.jpg

At the reseller level, only a small portion is cut off, while at the "admin" level, the information is displayed correctly without cutting off any text or showing a scrollbar. This issue didn’t occur in version 1.666, for example.

2- The other issue is related to the LiteSpeed version. We’ve tried updating it through CustomBuild in every way possible, as it indicates that an update is available. However, it always installs either an older version or a newer one, but never the expected version 6.3.1-0. Because of this, the update message remains visible, even though we are on a newer version (6.3.1-1).

print2.jpg
 
paulonichio said:
2- The other issue is related to the LiteSpeed version. We’ve tried updating it through CustomBuild in every way possible, as it indicates that an update is available. However, it always installs either an older version or a newer one, but never the expected version 6.3.1-0. Because of this, the update message remains visible, even though we are on a newer version (6.3.1-1).
Click to expand...
We have this same problem, and contacted DA about it. Short story is that Litespeed is reporting its version wrong, they have release a hotfix for it, and it will be available in 1.668. You can also force the new version by putting litespeed:6.3.1-1: into custom_versions.txt.
 
unihostbrasil said:
Several servers that we manage had DirectAdmin down for an extended period of time and displayed an expired license message to all customers. This is critical and causes us great concern.

This incident makes it clear that the license validation system puts all customers at risk in the event of network instability, which can occur for a variety of reasons (routing problems, problems with Cloudflare itself, etc.). It is not expected that services will be down due to a problem with your license validation system, and only "some time" after your alternative infrastructure takes over.

It would be important for the license validation system to have a grace period, perhaps 24/48 hours, and during this period the administrator would only receive alert messages, but the panel would remain active.
Click to expand...

this happens to me on restart on some of the servers it seems further away server is from Canada more ofthen is happens

Seems timing is not enough on initial start to verify the license

It fixable by just restarting direct admin
 
Could there be other possible causes like Cloudflare or Cloudlinux?
Because we have servers in Sweden and Germany, which is quite a distance from Canada and we almost never experience this issue.
 
kristian said:
We have this same problem, and contacted DA about it. Short story is that Litespeed is reporting its version wrong, they have release a hotfix for it, and it will be available in 1.668. You can also force the new version by putting litespeed:6.3.1-1: into custom_versions.txt.
Click to expand...

The issue is that on some of our servers, we're still running CloudLinux 7, which means DirectAdmin will no longer be updated. In theory, we won't receive version 1.668 on these servers. So, essentially, this error will remain in CustomBuild indefinitely unless we manually change the version in custom_versions.txt.

@fln Will we continue receiving updates via CustomBuild on these servers, even though DirectAdmin itself will no longer be updated?
 
Can't set user email limit more then 500 also user can't sent more then 200
 

Attachments

  • Screenshot 2024-09-20 100557.png
    Screenshot 2024-09-20 100557.png
    14.2 KB · Views: 6
You must log in or register to reply here.
Back
Top