- From my POV there are 2 opportunities why this happens:
1. You have a second XYmon server which has no checks defined for that
host and intermittently sends this erroneous data to your primary server.
2. The client message is too big and does not contain ALL data.
This can happen e.g. if there is a very long process listing sent with
the client data.
When the checks turn to clear just click on the "Client data" link
available on each check page of a server OR on the "clientlog" check for
that server and check if the message is complete.
A complete message would start with the line:
[collector:]
and end with 4 lines with time information like:
[clock]
epoch: 1580142177.579044
local: 2020-01-27 17:22:57 CET
UTC: 2020-01-27 16:22:57 GMT
If those 4 lines are missing then your message is too big to be
transferred and processed and you'll need to tweak some settings in
xymonserver.cfg like MAXMSG or MAXLINE.
Also, if you have the check "xymond" for your XYmon server check if that
check changed the color when the other checks were flapping to clear.
Regards,
Torsten
On 27.01.2020 13:41, user-ba12ef048dd6@xymon.invalid wrote:
Le Fri, 24 Jan 2020 12:48:06 +0000,
John Horne <user-e95f1ec2f147@xymon.invalid> a ?crit :
If Xymon is processing the client data netstat output, then I suspect
the '$' in your regex above is causing the problem. The netstat
output includes other info after the port number, but your regex is
saying it should be the end of the line.
I've already try a simple check, but nothing better, this one not more :
PORT LOCAL=%[\.:]3306 STATE=LISTEN min=1 TEXT=mysql3306
By the way, i'll try procs also, and it's the same trouble :
PROC mysqld 1 1
WARNING: Flapping status
"clear" No process checks defined
strange !