Multiple Services Failing

[chandro]

im getting email from all services about failing. so i decide to update to RELEASE let see what happen...

now is:

cPanel 11.25.0-R42399 - WHM 11.25.0 - X 3.9
CENTOS 5.2 i686


lets wait...

[chandro]

well, everything looks good. lets hope everything continues like that

[chandro]

all servers failed...

pop
ftpd
imap
spamd
exim
httpd
cpsrvd

mmmm, wich is the problem...

[cPanelDon]

im getting email from all services about failing. so i decide to update to RELEASE let see what happen...

now is:

cPanel 11.25.0-R42399 - WHM 11.25.0 - X 3.9
CENTOS 5.2 i686

lets wait...

well, everything looks good. lets hope everything continues like that

all servers failed...

pop
ftpd
imap
spamd
exim
httpd
cpsrvd

mmmm, wich is the problem...

Was the server rebooted?

Was the notice seen near the same time as when you performed the update (shortly before, during, or shortly after it finished)?

Did the services automatically restart successfully?

If you're experiencing a large number of services frequently failing I recommend escalating the issue via a ticket; there may be any number of causes that can only be determined by investigating the system directly.

[chandro]

hello Don, well, i update cpanel.

Was the server rebooted?

no, i dont restart the server.

Was the notice seen near the same time as when you performed the update (shortly before, during, or shortly after it finished)?

the failing servers start like 10-15 min after update ended.

Did the services automatically restart successfully?

yes, everything now looks good, for example, right now 11 pm, no services are failing. since 6pm everything looks good, let see if tomorrow during the day, services stay up.

thanks

[chandro]

nope, servers failing....

additional, i cant create accounts. everything appears ok, but there is no files inside /home/username

[cPanelDon]

To help better organize and differentiate the new topic I've split the discussion from the original thread.

The original discussion thread is located here: queueprocd on server.xxxxxxxxxx.net failed - cPanel Forums

[cPanelDon]

im getting email from all services about failing. so i decide to update to RELEASE let see what happen...

now is:
cPanel 11.25.0-R42399 - WHM 11.25.0 - X 3.9
CENTOS 5.2 i686

lets wait...

While this may not directly alleviate the entire issue, I would consider ensuring the operating system (OS) is updated to the latest available release. The existing version of CentOS 5.2 could be upgraded to CentOS 5.4 that is currently available. The following command will accomplish this task via root SSH access:

# yum upgrade

To verify the OS version after updating the following command may be used:

# cat /etc/*release

nope, servers failing....

additional, i cant create accounts. everything appears ok, but there is no files inside /home/username

Is the system a VPS or in some form of a virtual environment (e.g., Virtuozzo/OpenVZ, Xen)?

The following command will help determine the above:

# cat /var/cpanel/envtype && echo

If it is Virtuozzo or OpenVZ, please check the following for the "failure count" information (in the last column of data):

# cat /proc/user_beancounters

If the same symptoms persist, due to the nature of the issue, where multiple services are repeatedly failing (or at least being reported as having failed), I recommend escalating the issue via a ticket. I may suggest checking first with your upstream data center as they can physically check and access the machine directly (including for hardware diagnostic testing). If submitting a support request to cPanel, please send me your ticket ID number so that I may follow-up internally and verify the circumstances involved; if appropriate I can then share the solution in your forums thread.

[chandro]

no servers failing, what i do???

modify the vps conf, the QUOTAUGIDLIMIT="380"

was 320, and i change the limit to 380 and the accounts transfers work, and the servers no fail for today.

but here is the paste of the beancounters.



root@xela [~]# cat /proc/user_beancounters
Version: 2.5
uid resource held maxheld barrier limit failcnt
114: kmemsize 6687212 13928722 28745400 29580328 0
lockedpages 0 8 512 512 0
privvmpages 79316 160078 262144 262144 0
shmpages 1125 6565 43008 43008 0
dummy 0 0 0 0 0
numproc 65 129 480 480 0
physpages 36318 88939 0 2147483647 0
vmguarpages 0 0 67584 2147483647 0
oomguarpages 36318 88939 52224 2147483647 0
numtcpsock 58 237 720 720 0
numflock 6 14 376 412 0
numpty 1 2 32 32 0
numsiginfo 1 27 512 512 0
tcpsndbuf 1124708 3534676 3440640 5406720 426030363
tcprcvbuf 429884 1860240 3440640 5406720 0
othersockbuf 41784 2250572 2252160 4194304 83
dgramrcvbuf 0 124836 524288 524288 0
numothersock 35 85 720 720 0
dcachesize 0 0 7249920 7249920 0
numfile 1780 3524 18624 18624 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
numiptent 103 103 256 256 0
root@xela [~]#



what is QUOTAUGIDLIMIT, i search google and in a post i see:

"quotaugidlimit is used to track the numer of group and users with quotas. As you add more users and groups it gets near the limit"

sounds reazonable for me, that was my problem and that cause the server been restarting the services...


thanks Don

[cPanelDon]

no servers failing, what i do???

modify the vps conf, the QUOTAUGIDLIMIT="380"

was 320, and i change the limit to 380 and the accounts transfers work, and the servers no fail for today.

...

what is QUOTAUGIDLIMIT, i search google and in a post i see:

"quotaugidlimit is used to track the numer of group and users with quotas. As you add more users and groups it gets near the limit"

sounds reazonable for me, that was my problem and that cause the server been restarting the services...

thanks Don

Thank you for the detailed update and information. The specified parameter is used by Virtuozzo and OpenVZ; for more verbose detail please reference the following documentation:
KB Parallels: What is the difference between 'disk quota' and 'hard disk quota'?
KB Parallels: How do I solve this problem: edquota: "Can't write quota for /dev/vzfs"?
User Guide/Managing Resources - OpenVZ Wiki

Number of user/group IDs allowed for the CT internal disk quota. If set to 0, the UID/GID quota will not be enabled.

...

The parameter that controls the second-level disk quotas is QUOTAUGIDLIMIT in the Container configuration file. By default, the value of this parameter is zero and this corresponds to disabled per-user/group quotas.

[chandro]

Hi Don, today services fail and restart automatically, but only in the morning. let me read more about my VPS`s config. something is bad there i think.

[chandro]

all the day services are good but in the morning, services fail and restart, and then in the day continue good.

[cPanelDon]

no servers failing, what i do???

modify the vps conf, the QUOTAUGIDLIMIT="380"

was 320, and i change the limit to 380 and the accounts transfers work, and the servers no fail for today.

but here is the paste of the beancounters.

root@xela [~]# cat /proc/user_beancounters
Version: 2.5
       uid  resource           held    maxheld    barrier      limit    failcnt
      114:  kmemsize        6687212   13928722   28745400   29580328          0
            lockedpages           0          8        512        512          0
            privvmpages       79316     160078     262144     262144          0
            shmpages           1125       6565      43008      43008          0
            dummy                 0          0          0          0          0
            numproc              65        129        480        480          0
            physpages         36318      88939          0 2147483647          0
            vmguarpages           0          0      67584 2147483647          0
            oomguarpages      36318      88939      52224 2147483647          0
            numtcpsock           58        237        720        720          0
            numflock              6         14        376        412          0
            numpty                1          2         32         32          0
            numsiginfo            1         27        512        512          0
            tcpsndbuf       1124708    3534676    3440640    5406720  426030363
            tcprcvbuf        429884    1860240    3440640    5406720          0
            othersockbuf      41784    2250572    2252160    4194304         83
            dgramrcvbuf           0     124836     524288     524288          0
            numothersock         35         85        720        720          0
            dcachesize            0          0    7249920    7249920          0
            numfile            1780       3524      18624      18624          0
            dummy                 0          0          0          0          0
            dummy                 0          0          0          0          0
            dummy                 0          0          0          0          0
            numiptent           103        103        256        256          0
root@xela [~]#

what is QUOTAUGIDLIMIT, i search google and in a post i see:

"quotaugidlimit is used to track the numer of group and users with quotas. As you add more users and groups it gets near the limit"

sounds reazonable for me, that was my problem and that cause the server been restarting the services...

thanks Don

Hi Don, today services fail and restart automatically, but only in the morning. let me read more about my VPS`s config. something is bad there i think.

all the day services are good but in the morning, services fail and restart, and then in the day continue good.

Does the file "user_beancounters" (for Virtuozzo/OpenVZ) still show a failure count for any of the listed parameters?

Are there any errors in the log files for the affected services?