09-12-2013 06:54 AM
Hello Forum,
Recently this error message showed up on my system log,
Sep 11 08:17:37 dbsp1 Had[2008]: [ID 702911 daemon.notice] VCS ERROR V-16-1-13027 (dbsp1) Resource(MNIC) - monitor procedure did not complete
I am guessing that this is because the monitor procedure failed to completed within the time window.
Probably because the CPU was busy. Someone mentioned it was because of swap space.
I checked the swap but there is only about 5% capacity utlization of swap space.
Can anyone guide me as to what I need to look for?
The cluster shows no faults or errors.
Thank you very much in advance.
Solved! Go to Solution.
09-12-2013 08:07 AM
If this is just a one off message, then it is not very serious as by default a resource has to timeout 4 times in a row before it fails. If you are seeing this message a lot, then you need to identify what is causing monitor to hang which can be a resource issue (CPU, memory, swap) or a problem with the resource itself.
Issue with the resource itself, could be network timeout or if the resource type of MNIC is MultiNICB and the UseMpathd attribute is set to 1, then there could be an issue with mpathd.
Mike
09-12-2013 08:07 AM
If this is just a one off message, then it is not very serious as by default a resource has to timeout 4 times in a row before it fails. If you are seeing this message a lot, then you need to identify what is causing monitor to hang which can be a resource issue (CPU, memory, swap) or a problem with the resource itself.
Issue with the resource itself, could be network timeout or if the resource type of MNIC is MultiNICB and the UseMpathd attribute is set to 1, then there could be an issue with mpathd.
Mike
09-14-2013 11:50 PM
Hi,
Is this error occurring in regular intervals or at particular point of time In a day?
If error is happening in regular intervals most probably due to high network traffic on interface and congestion, then you need to check out speed of interface and duplex settings.
If error is happening in a particular point of time, then probably some intended heavy data transfers happening via interfaces may be due to batch jobs from cron or backups
As Mike Said there is a chance of issue with mapthd particularly when it is configured as probe based
-Girish Puppala
10-22-2013 09:49 AM
Thank you,
This was most likely due to heavy network traffic.