Constant "Alarm level changed" emails

[Stuff That Spins]

Bump... This problem is annoying. 12 cores and 1 core over % and boom!

 

[Dave W]

/etc/init.d/psa-health-monitor-notificationd stop
chkconfig psa-health-monitor-notificationd off

 

[BestServedCold]

Bump... This problem is annoying. 12 cores and 1 core over % and boom!

Same here, I've set the parameter to 900 percent and I'm still getting alert emails when one core goes high, very frustrating, can anyone advise?

 

[Kingsley]

THis mails are kinda annoying too

 

[UFHH01]

Complaining about a custom definition is a quite strange behaviour, I think. If you don't like these eMail - notifications, just turn them off, please... it's YOUR choice!

Parallels Plesk Panel for Linux services logs and configuration files ( KB - article: 111 283 )​

Please have a CLOSER look at: "Health Monitor manager" - "Configuration" - "/usr/local/psa/var/custom-health-config.xml"​

 

[Kingsley]

Complaining about a custom definition is a quite strange behaviour, I think. If you don't like these eMail - notifications, just turn them off, please... it's YOUR choice!

Parallels Plesk Panel for Linux services logs and configuration files ( KB - article: 111 283 )​

Please have a CLOSER look at: "Health Monitor manager" - "Configuration" - "/usr/local/psa/var/custom-health-config.xml"​

Cant even find "Health Monitor manager" in my admin area

 

[UFHH01]

Hi Kingsley,

the part "Health Monitor manager" is quoted from the KB - article, which I mentioned. The KB - article lists all needed paths for your Plesk installation.

 

[Kingsley]

Hi Kingsley,

the part "Health Monitor manager" is quoted from the KB - article, which I mentioned. The KB - article lists all needed paths for your Plesk installation.

what i mean is when i install it , it will be active but i cant find "Health Monitor" from my server management list

 

[UFHH01]

The link for the Health Monitor can be viewed in "Service Provider view" and it is then located in the "Server Management"

​

 

[Kingsley]

The link for the Health Monitor can be viewed in "Service Provider view" and it is then located in the "Server Management"

View attachment 9473​

Its not there, so i have to unistall. were i using different view i guess

 

[UFHH01]

HI Kingsley,

Did you try to switch to the server providers view at "Login > Tools and Settings > Panel Appearance > Interface Management"
( Direct - link after a login with the admin - account: "https://YOUR_SERVER_IP:8443/plesk/server/interface/uimodes@/" )

 

[Kingsley]

HI Kingsley,

Did you try to switch to the server providers view at "Login > Tools and Settings > Panel Appearance > Interface Management"
( Direct - link after a login with the admin - account: "https://YOUR_SERVER_IP:8443/plesk/server/interface/uimodes@/" )

Thanks

 

[Neo1]

Hi,

Can anyone explain what could be causing this usage of CPU (97.8%id ?) - I think it would be better to resolve the problem that causes such high cpu usage, than stop email alerts.

Server health parameter "Services > Apache CPU usage" changed its status from "green" to "red".

top - 10:20:14 up 169 days, 1:27, 0 users, load average: 0.11, 0.20, 0.27
Tasks: 241 total, 1 running, 240 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.3%us, 0.4%sy, 0.7%ni, 97.8%id, 0.7%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 32871892k total, 4921084k used, 27950808k free, 670200k buffers
Swap: 16777080k total, 269104k used, 16507976k free, 2654656k cached

 

[Nelson_Silva]

I have the same problem on plesk 11.5 and now on Plesk 12.

I have multi-core processors (4) and I believe that may be a cause.
In Service tab, I have problems on Apache CPU, but on CPU tab (instead services) it's all ok.

This is a bug (on services=>apache CPU) or a misconfiguration?

Hi,

Can anyone explain what could be causing this usage of CPU (97.8%id ?) - I think it would be better to resolve the problem that causes such high cpu usage, than stop email alerts.

Server health parameter "Services > Apache CPU usage" changed its status from "green" to "red".

top - 10:20:14 up 169 days, 1:27, 0 users, load average: 0.11, 0.20, 0.27
Tasks: 241 total, 1 running, 240 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.3%us, 0.4%sy, 0.7%ni, 97.8%id, 0.7%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 32871892k total, 4921084k used, 27950808k free, 670200k buffers
Swap: 16777080k total, 269104k used, 16507976k free, 2654656k cached

 

[Niko H.]

The really problem is, that obviously nowhere is descibed why all these manuals are unsuccessful and I wonder why Parallels does not gives the solution here, and leaves their paying community alone with it. I do not understand that.

In most these cases the custom-health-config.xml is in a state without knowing what hardware is working.

1. download the custom-health-config.xml and rename it to s. th like custom-health-config-beforehardwarechanges.xml
2. click -Detect Hardware Changes-
3. download the custom-health-config.xml again!
4 compare both custom-health-config.xml-Files and in most cases you will see that the later downloaded configuration now has parameters that were not present in the inital custom-health-config.xml. Often a OS for example of a virtual machine is installed in a minimal hardware configuration and the admin forgets to tell Plesk that s. th. has changed
5. now make changes to that custom-health-config.xml. In my case it was RAM of Apache that caused those annoying emails
I changed the values vom 40 to 60 and from 30 to 50 for red and yellow alarm.

<MemoryUsage>
<service name="Web" monitor="true">
<alarm level="red" threshold="60" type="percent"/>
<alarm level="yellow" threshold="50" type="percent"/>
</service>
.
.
.
6. After uploading the modified file the alarm state immediately changed to green.
That was the whole secret of my problem.

This is also descibed in german language on
ASPprofessionell ASP FAQ: Health Monitor Plesk Onyx, nervende mail messages red yellow green status