• If you are still using CentOS 7.9, it's time to convert to Alma 8 with the free centos2alma tool by Plesk or Plesk Migrator. Please let us know your experiences or concerns in this thread:
    CentOS2Alma discussion

Upgraded and now PSA service will no longer start

Smasher456

New Pleskian
I upgraded today from 11 to 12, however since the upgrade I am now unable to start the PSA Service, it comes up with the below error.

Starting psa... Starting sw-engine-fpm: Thu Jun 19 13:22:29 2014 (7885): Fatal Error Unable to allocate shared memory segment of 104857600 bytes: mmap: Cannot allocate memory (12)

I have checked resources and everything should be fine, I cant see a reason why it cant allocate the memory, as there is plenty free for it to take from.

I have tried a Bootstrapper repair for 12 however this has not found any issues other than its not able to stop the PSA Service (As its not running)


Suggestions are welcome, everything else is up and running its just the Parallels backend which im guessing is what the PSA Service is!

Cheers

Smoosh
 
Same here!

xxx.com:8443 = 502 Bad Gateway nginx

# service sw-engine restart
Stopping sw-engine-fpm: [FAILED]
Starting sw-engine-fpm: Fri Jun 20 00:06:11 2014 (4350): Fatal Error Unable to allocate shared memory segment of 104857600 bytes: mmap: Cannot allocate memory (12)

How to fix that???????

Thanks.
 
Please post here output of "free -m" command?
Also try to use following workaround:

1. Run
# plesk conf php.ini

2. Disable cache with option
opcache.enabled=1
 
Same Error

Please post here output of "free -m" command?
Also try to use following workaround:

1. Run
# plesk conf php.ini

2. Disable cache with option
opcache.enabled=1



Disable opcache dont fix the Problem

free -m

total used free shared buffers cached
Mem: 2048 90 1957 0 0 35
-/+ buffers/cache: 54 1993
Swap: 0 0 0
 
Thanks for your answer.

total used free shared buffers cached
Mem: 2048 301 1746 0 0 110
-/+ buffers/cache: 190 1857
Swap: 0 0 0

I have set opcache.enabled to 1, restarted the server but nothing.
 
Sorry, cache should be disabled with

opcache.enabled=0
 
Before I set it to 1 it was = 0 ...

In my case it's a virtual server from 1&1, maybe the problem is caused due to 1&1 ?
 
Please post here output of "free -m" command?
Also try to use following workaround:

1. Run
# plesk conf php.ini

2. Disable cache with option
opcache.enabled=1

Free -m

[root@BlaBla ~]# free -m
total used free shared buffers cached
Mem: 2048 251 1796 0 0 98
-/+ buffers/cache: 153 1894
Swap: 0 0 0

As others my Opcache.enabled = 0 already

Also mine is a 1&1 container as well
 
Please show me output of

# cat /proc/user_beancounters


Version: 2.5
uid resource held maxheld barrier limit failcnt
57655048: kmemsize 14204767 14925824 20971520 23068672 0
lockedpages 0 0 256 256 0
privvmpages 83140 84728 524288 576716 0
shmpages 1190 1206 20480 20480 6
dummy 0 0 9223372036854775807 9223372036854775807 0
numproc 65 75 96 96 0
physpages 25068 26368 524288 524288 0
vmguarpages 0 0 131072 2147483647 0
oomguarpages 11425 11425 131072 2147483647 0
numtcpsock 28 28 360 360 0
numflock 13 14 188 206 0
numpty 1 1 16 16 0
numsiginfo 0 12 256 256 0
tcpsndbuf 213904 218528 1720320 2703360 0
tcprcvbuf 212992 265312 1720320 2703360 0
othersockbuf 161840 187720 1126080 2097152 0
dgramrcvbuf 0 4360 262144 288358 0
numothersock 139 143 360 360 0
dcachesize 3440199 3443102 3145728 3460300 0
numfile 1443 1480 4096 4096 0
dummy 0 0 9223372036854775807 9223372036854775807 0
dummy 0 0 9223372036854775807 9223372036854775807 0
dummy 0 0 9223372036854775807 9223372036854775807 0
numiptent 18 18 100 105 0
 
[root@BlaBla ~]# cat /proc/user_beancounters
Version: 2.5
uid resource held maxheld barrier limit failcnt
76792265: kmemsize 23247794 24113152 33554432 41943040 0
lockedpages 0 0 256 256 0
privvmpages 152045 154131 524288 576716 0
shmpages 1833 1849 20480 20480 18
dummy 0 0 9223372036854775807 92233726854775807 0
numproc 112 115 128 128 19
physpages 65628 68391 524288 524288 0
vmguarpages 0 0 262144 262144 0
oomguarpages 34452 34452 262144 262144 0
numtcpsock 35 35 360 360 0
numflock 13 16 188 206 0
numpty 2 2 16 16 0
numsiginfo 0 15 256 256 0
tcpsndbuf 196200 196200 1720320 2703360 0
tcprcvbuf 212992 212992 1720320 2703360 0
othersockbuf 184184 210064 1126080 2097152 0
dgramrcvbuf 0 8720 262144 288358 0
numothersock 151 155 360 360 0
dcachesize 6221784 6241478 5242880 6291456 0
numfile 2565 2615 4096 4096 0
dummy 0 0 9223372036854775807 92233726854775807 0
dummy 0 0 9223372036854775807 92233726854775807 0
dummy 0 0 9223372036854775807 92233726854775807 0
numiptent 18 18 100 105 0
 
First of all try to reboot your VPS. If this doesn't help you, ask your provider for assistance. Physical server may have no enough RAM and you can't detect that fact by checking of /proc/user_beancounters inside container.
 
Have tried several reboots, and currently have a ticket in with my supplier so will wait and see what they come back with
 
I think I have this exact same error, upgraded to Plesk 12 from 11 and the server went berserk, all kinds of limits reached. Now I get permanent "502 Bad Gateway" when trying to access Plesk.

I'm sure I had something similar happen to me in the past and all i had to do was delete a file that was stuck open or something?

Any help would be appreciated! :)
 
I had to roll-back the container to a back-up before the error occurred in order to get it working again!
 
I put it down to not having enough resources available on my VPS, only way to combat this would be to upgrade my VPS package...
 
I put it down to not having enough resources available on my VPS, only way to combat this would be to upgrade my VPS package...

how much ram u have? an how much is in use?
I have 2gb ram and 300mb are used. plesk used 512mb ram up to 768mb than must be 2gb more than enough
 
Back
Top