undef error - open(/var/run/system_needs_reboot.cache): Too many levels of symbolic links at /usr/local/cpanel/Cpanel/Serverinfo/CachedRebootStatus.pm line 47.
Answered
Hello,
I started experiencing an issue while accessing WHM/cPanel and wanted to ask if anyone has encountered this before or knows the proper fix.
When logging into WHM, I receive the following error:
undef error - open(/var/run/system_needs_reboot.cache): Too many levels of symbolic links at /usr/local/cpanel/Cpanel/Serverinfo/CachedRebootStatus.pm line 47.
Then, while accessing the cPanel user interface, I get this Template::Exception error:
Template::Exception:
[TYPE]=[file]
[INFO]=[libraries/cjt2/cjt2_header_include.tt: not found]
[TEXT]=[
]
at cpanel.pl line 1127.
cpanel::cpanel::cptt_exectag("/usr/local/cpanel/base/frontend/jupiter/index.auto.tmpl", 1) called at cpanel.pl line 4649
cpanel::cpanel::run_standard_mode() called at cpanel.pl line 931
cpanel::cpanel::script("cpanel::cpanel", "./frontend/jupiter/index.html") called at cpanel.pl line 324
Environment:
-
cPanel/WHM Version: 134.0.23
-
OS: CloudLinux
-
Theme: Jupiter
What I suspect:
-
A symlink loop related to
/var/run/system_needs_reboot.cache -
Missing or corrupted cPanel template files
-
File corruption after a cPanel update
-
Permission or symbolic link issues after EasyApache/cPanel updates
Commands already attempted:
/scripts/upcp --force
/usr/local/cpanel/scripts/check_cpanel_pkgs --fix
/usr/local/cpanel/scripts/rebuildhttpdconf
/usr/local/cpanel/scripts/restartsrv_cpsrvd
However, the issue still persists.
Has anyone experienced this before?
Especially the following errors:
-
Too many levels of symbolic links -
cjt2_header_include.tt: not found
Any advice or confirmed fix would be greatly appreciated.
Thank you.
-
Hey there! Unfortunately no, I'm not finding any record of this exact error in our system. Would you be able to create a ticket so we can check that out?
0 -
Hi!
We face the exact same issue today after a cPanel update that produced errors in the error log.
Have you got a fix for that by cPanel support?
0 -
We do have one ticket on this today. Our team is looking into this and I'll update as soon as I know more.
0 -
Hi!
This should most probably be our ticket... It was opened about 8 hours ago, but there still no resolution and besides the fact that we cannot access WHM, and our customers cannot access cPanel, webmail, etc... we also have some sites producing "error 500".
Could you please do something to speed-up the process? We would be greatful.
0 -
We are experiencing the same problem after updating the kernel to patch the dirty flag vulnerability.
0 -
ANTONIOS 20VT - our team is still investigating as we don't know the root cause at this time. khalid smaili - if you're able to submit a ticket that would help us look into this.
0 -
Hello,
We are still experiencing the same critical issue.
Additional information from our side:
-
We completely removed and reinstalled Apache, but the issue still persists.
-
The problem may be related to the recent test kernel update.
-
The server environment was running AlmaLinux 8 with CloudLinux 8.
-
Kernel version in use was
136.0.23.
After the kernel/security update, the following problems started appearing simultaneously:
-
WHM instability or inaccessibility
-
cPanel/Webmail access failures
-
Random 500 Internal Server Errors on websites
-
Service instability across the entire hosting environment
This does not appear to be an Apache configuration issue anymore, as even a full reinstall did not resolve the problem.
We strongly believe there may be a compatibility or stability issue related to the latest kernel/test kernel deployment on AlmaLinux 8 + CloudLinux 8 environments.
Could the cPanel team please escalate this urgently and confirm whether there are any known issues, rollback recommendations, or emergency fixes available?
Thank you.
0 -
-
Hello,
I’m seeing the same issue after updating CloudLinux/kernel/cPanel.
Environment:
- CloudLinux 8.10
- cPanel & WHM 11.134.0.25
- Kernel: 4.18.0-553.123.2.lve.el8.x86_64WHM shows this error:
undef error - open(/var/run/system_needs_reboot.cache): Too many levels of symbolic links at /usr/local/cpanel/Cpanel/Serverinfo/CachedRebootStatus.pm line 47.
What I checked:
1. The file is not a symlink. It is a regular file:
/var/run -> ../run
/run/system_needs_reboot.cache = regular file
/var/run/system_needs_reboot.cache = regular file2. Perl can open the file successfully:
perl -e 'open my $fh, "<", "/var/run/system_needs_reboot.cache" or die "OPEN_FAILED: $!\n"; print "OPEN_OK\n"; print while <$fh>;'
Output:
OPEN_OK
{"needs_reboot":0,"details":{}}3. cPanel Perl can also open it:
/usr/local/cpanel/3rdparty/bin/perl -e 'open my $fh, "<", "/var/run/system_needs_reboot.cache" or die "OPEN_FAILED: $!\n"; print "CPANEL_PERL_OPEN_OK\n"; print while <$fh>;'
Output:
CPANEL_PERL_OPEN_OK
{"needs_reboot":0,"details":{}}4. WHM API works fine:
whmapi1 system_needs_reboot
Output:
data:
details: {}
needs_reboot: 0
metadata:
command: system_needs_reboot
reason: OK
result: 15. I also tested from the cPanel/WHM process namespace using nsenter, and the file is still visible as a regular file and opens correctly.
6. Webmail is working on port 2096:
curl -kI https://127.0.0.1:2096
Output:
HTTP/1.1 200 OK
7. I tried disabling Feature Showcase with:
mkdir -p /var/cpanel/activate/features
touch /var/cpanel/activate/features/disable_feature_showcase
/usr/local/cpanel/scripts/restartsrv_cpsrvd --hardBut the WHM error still appears.
8. I also noticed these related log entries:
- WP Toolkit Feature Showcase warnings / JSONParseError
- “Failed to import Cpanel::Config::ConfigObj::Driver::WpToolkitACL::META”
- “Can’t exec /usr/local/cpanel/etc/init/kill_apps_on_ports”However, I confirmed that kill_apps_on_ports exists, has correct permissions, has a valid shebang, no CRLF, and runs manually with exit code 0.
9. I temporarily modified CachedRebootStatus.pm to bypass the die statement at line 47, but the same WHM error still appeared, so I reverted the file from backup.
At this point, the system_needs_reboot.cache file itself appears healthy, the API works, and the file opens correctly from both normal shell and cPanel context, but WHM still throws the symlink loop error.
Can cPanel confirm if this is a known bug in 11.134.0.25 on CloudLinux 8 after the latest kernel/security updates? Is there an internal case or ETA for a fix?
Thank you.
0 -
Hi!
It is extremely disappointing that there is still no resolution, fix or workaround to this issue, provided by cPanel. We are paying extremely expensive and ever increasing cPanel license and support fees and this is 100% unacceptable. Please, provide a solution ASAP.
Regarding the issue itself.
- For the random 500 Internal Server Errors we found that changing PHP handler from suPHP to PHP-FPM mitigated the issue.
- Was any of the affected servers running the Bitninja agent? I have a suspicion (not based on facts) that Bitninja might have played some role on the issue.
0 -
ANTONIOS 20VT I saw this posted earlier. Perhaps this will help?
0 -
Hello,
Thank you Vladimir Ivanov for the solution. It appears to be resolved now on our side as well.
Also thanks to Allen Daniels for sharing the related article and helping point us in the right direction.
Best regards,
0 -
Thank you very much @Allen Daniels!
I have informed our admin team about this fix since Saturday, but we have an open ticket with cPanel and we still wait for their response @cPRex on how we should procced...
I am glad it worked for @DoÄŸucan ugurlu
I will post again when we have an update on the issue.
Have a nice day!
0 -
It looks like all is well here now, correct? If I'm still needed just let me know!
0 -
Tried this workound: https://support.cpanel.net/hc/en-us/articles/40347469257111-kmod-lve-causing-WHM-login-to-fail-with-Too-many-levels-of-symbolic-links-error
But failed:
====================
[root@shared ~]# dnf remove kmod-lve-2.1-60.el8.x86_64
This system is receiving updates from CloudLinux Network server.
Error:
Problem: The operation would result in removing the following protected packages: kernel-core
(try to add '--skip-broken' to skip uninstallable packages or '--nobest' to use not only best candidate packages)
[root@shared ~]# yum remove kmod-lve-2.1-60.el8.x86_64
This system is receiving updates from CloudLinux Network server.
Error:
Problem: The operation would result in removing the following protected packages: kernel-core
(try to add '--skip-broken' to skip uninstallable packages or '--nobest' to use not only best candidate packages)0 -
Just as an update, the above fix worked for us.
0 -
Glad to hear it!
0
Please sign in to leave a comment.
Comments
16 comments