cphulkd broken after WHM 54.0 (build 20) update

[PPNSteve]

Server just updated and it can't seen to restart cphulkd:

[B]Service Name[/B]

cphulkd

[B]Service Status[/B]

failed ⛔

[B]Notification[/B]

The service “cphulkd” appears to be down.

[B]Service Check Method[/B]

The system’s command to check or to restart this service failed.

[B]Number of Restart Attempts[/B]

8

[B]Startup Log[/B]

Starting PID 10355: cPhulkd - processor - dormant mode - accepting connections

[B]Memory Information[/B]

Used

1.87 GB

Available

5.79 GB

Installed

7.67 GB

[B]Load Information[/B]

0.41 0.45 0.43

[B]Uptime[/B]

33 days, 2 hours, 15 minutes, and 29 seconds

[B]IOStat Information[/B]


avg-cpu:   %user %nice %system %iowait %steal %idle 3.74 0.01 1.02 0.71 0.00 94.52   Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn sda 15.90 134.03 291.46   383235522 833388920 sdb 0.85 13.51 15.63 38635032 44693168

[B]Top Processes[/B]

[B]PID[/B]

[B]Owner[/B]

[B]CPU %[/B]

[B]Memory %[/B]

[B]Command[/B]

14070

root

17.19

1.11

  
whostmgr     - servup ./servup

14084

root

7.69

0.09

  
/usr/local/cpanel/3rdparty/perl/514/bin/perl     -T -w /usr/local/cpanel/3rdparty/perl/514/bin/spamassassin --version

13988

root

5.75

0.31

  
tailwatchd     - chkservd - cpanellogd check

6347

mysql

5.73

3.34

  
/usr/sbin/mysqld     --basedir=/usr --datadir=/var/lib/mysql     --plugin-dir=/usr/lib64/mysql/plugin --user=mysql     --log-error=/var/lib/mysql/svr4.1-gb.net.err --open-files-limit=15528     --pid-file=/var/lib/mysql/svr4.1-gb.net.pid

14086

root

4.00

0.29

  
/usr/local/cpanel/scripts/restartsrv_cpanellogd     --check --notconfigured-ok


and so on...

A little help here please.. what now?

 

[PPNSteve]

Just had another server update and it's also failing cphuld as well.. looks like there was something in the latest update that borked it..

/me looks at cPanel people..

 

[TapanB]

Hi,

I am also facing same issue, though I don't have any real accounts on the server yet and I just got it 4 days back, but I guess after the update my inbox was filled today with cphulkd crashing and recovering:

---
Server: xyz.host.com
Primary IP Address: xxx.xxx.xxx.xxx
Service Name:    cphulkd
Service Status:    failed ⛔
Notification: The service “cphulkd” appears to be down.
Service Check Method:    The system’s command to check or to restart this service failed.
Number of Restart Attempts:    1
Startup Log: Mar 22 10:35:04 xyz.host.com systemd[1]: Starting cPanel brute force detector services...
Mar 22 10:35:04 xyz.host.com systemd[1]: PID file /var/run/cphulkd_processor.pid not readable (yet?) after start.
Mar 22 10:35:04 xyz.host.com restartsrv_cphulkd[4918]: Starting PID 4927: cPhulkd - processor - dormant mode - accepting connections
Mar 22 10:35:04 xyz.host.com systemd[1]: cphulkd.service: Supervising process 4927 which is not our child. We'll most likely not notice when it exits.
Mar 22 10:35:04 xyz.host.com systemd[1]: Started cPanel brute force detector services.

Memory Information  
Used: 1.15 GB
Available: 45.85 GB
Installed: 47 GB
Load Information: 0.00 0.01 0.05
Uptime: 42 minutes and 49 seconds
IOStat Information: avg-cpu: %user %nice %system %iowait %steal %idle 0.26 0.01 0.10 0.17 0.00 99.46 Device: tps kB_read/s kB_wrtn/s kB_read kB_wrtn sda 10.67 545.05 32.10 1400141 82470 sdb 0.17 1.17 0.80 3008 2048

Top Processes  
PID    Owner    CPU %    Memory %    Command
5629    root    15.00    0.04    /usr/local/cpanel/scripts/restartsrv_cpanellogd --check --notconfigured-ok
5511    root    6.89    0.05    tailwatchd - chkservd - cpanellogd check
5446    root    1.17    0.02    /usr/local/cpanel/bin/dcpumon
5510    root    0.60    0.00    top -n 2 -b -c
2759    root    0.40    0.70    /usr/local/cpanel/3rdparty/bin/clamd
---

How to fix ???

Thanks.

 

[polariz]

Exact Same issue here =(

Hope for a quick solution / answer from anyone here....

 

[Adeel Rizvi]

Same for me, its keeping restarting, failed to start and then it says recovered and so on

 

[marcochiesi]

Hi there,
I am experiencing the very same problem in some of my servers. I also tried soft and hard service restart from command line (see docs), but still no luck.

 

[TapanB]

Are all effected by any chance on mod_ruid + dso ??? I am asking this becuase only server in which i have mod_ruid + dso is facing the issue. Other which have suphp does'nt have this issue. Weird!!!

 

[tylerjr92]

cPHulkd is down for me as well. Getting an odd error when I attempt to reboot the damon |
The system did not expect the .exe file for the 'cphulkd' service's PID file '/var/run/cphulkd_processor.pid'. The system will not kill the PID: /usr/local/cpanel/libexec/cphulkd-dormant at /usr/local/cpanel/Cpanel/ServiceManager/Manager/Initd.pm line 174, <$fh> line 1.

This started after the update to WHM 54, Build 20. No command line attempts to restart are successful. This is the error I find in the log for cphuld at /usr/local/cpanel/logs/cphulkd_errors.log

[2016-03-22 08:19:08 -0400] warn [cphulkd] 4412 Hulkd processor is already running [PID 4356] at /usr/local/cpanel/Cpanel/Hulkd.pm line 429.

        Cpanel::Hulkd::_write_log(Cpanel::Hulkd(0xfd0180), "errlog", "warn", "Hulkd processor is already running [PID 4356]") called at /usr/local/cpanel/Cpanel/Hulkd.pm line 460

        Cpanel::Hulkd::warn(Cpanel::Hulkd(0xfd0180), "Hulkd processor is already running [PID 4356]") called at /usr/local/cpanel/Cpanel/Hulkd/Proc.pm line 23

        Cpanel::Hulkd::Proc::get_apps_to_start(Cpanel::Hulkd(0xfd0180)) called at /usr/local/cpanel/Cpanel/Hulkd.pm line 174

        Cpanel::Hulkd::launcher(Cpanel::Hulkd(0xfd0180)) called at /usr/local/cpanel/Cpanel/Hulkd.pm line 167

        Cpanel::Hulkd::start_daemon(Cpanel::Hulkd(0xfd0180)) called at /usr/local/cpanel/Cpanel/Hulkd.pm line 89

        Cpanel::Hulkd::run_daemon(Cpanel::Hulkd(0xfd0180)) called at /usr/local/cpanel/libexec/cphulkd line 47

 

[cPanelMichael]

Hello

Internal case CPANEL-5023 is open to address an issue where cphulkd continuously restarts after updating to cPanel version 54.0.20. Our developers have determined the cause of the issue, and made changes to correct it. A new build of cPanel that addresses this issue should be released soon, but I have no specific time frame to provide right now. I'll update this thread as soon the new build is published. In the meantime, the temporary workaround is to disable "cphulkd" from the "Dormant services" option under the "Software" tab in "WHM >> Tweak Settings".

Thank you.

 

[storminternet]

cPanel 54 is going to be release for stable build in couple of days. I hope it should be sorted out before that otherwise
it will create security threat on shared servers. Disabling cphulk on shared servers will invite spammers and hackers.

 

[cPanelMichael]

cPanel 54 is going to be release for stable build in couple of days. I hope it should be sorted out before that otherwise
it will create security threat on shared servers. Disabling cphulk on shared servers will invite spammers and hackers.

Disabling "cphulkd" from the the "Dormant services" option will not disable cPHulk, but instead it disables the dormant service functionality for cPhulk. Here's a snippet from our documentation that explains the dormant service feature:

This setting configures the system to unload idle services from memory after two minutes of inactivity, and then loads listening devices that correspond to it. This reduces memory usage, but delays responses from dormant services.

Thus, the temporary workaround will allow cPhulk to continue to function.

Thank you.

 

[PPNSteve]

Good to hear it was located and addressed. looking forward to the update.
Will try the workaround for now and see if that helps keep it up.

 

[cPanelMichael]

New builds have been published to the affected build tiers to address this issue:

Fixed case CPANEL-5023: Fix redaemonization of cphulk when resuming from dormant mode.

You can update cPanel via the "/scripts/upcp" command or wait for cPanel to update automatically if you have configured automatic updates. Remember to adjust any settings changed to manually workaround the issue back to their original values.

Thank you.

 

[storminternet]

That's great news.