Tidal master 22.214.171.1246 windows
Agent added windows 3.1.0.05
Just added new Agent and agent is dropping every 11 minutes
Re-installed agent and rebooted no luck
Checked agent log and I am seeing this:
_20141217 13:36:53 <err.TIDAL_AGENT_1.ES_INVJOBSCHED01_1> Resource Config - Settings : Active = 0 Master = ES_INVJOBSCHED01_1 CpuPoll = -1 VmemPoll = -1
_20141217 13:47:56 <inf.TIDAL_AGENT_1.ES_INVJOBSCHED01_1> Master Agent ID changed from 265 to 94
_20141217 13:47:56 <inf.TIDAL_AGENT_1.ES_INVJOBSCHED01_1> Connecting to master port 5591@INVJOBSCHED01
_20141217 13:47:56 <inf.TIDAL_AGENT_1.ES_INVJOBSCHED01_1> Suspend request received from master ES_INVJOBSCHED01_1, Unregistering master ES_INVJOBSCHED01_1
_20141217 13:48:01 <inf.TIDAL_AGENT_1.ES_INVJOBSCHED01_1> Master Agent ID changed from 94 to 265
In props file on agent ID=265
Checked nodmst table I see ID 265 and is assigned to right node, this node, there is no node 94.
Solved! Go to Solution.
Thank you for the response. I had been hoping that upgrading would resolve our issue. Seems like that is not the case.
Were you ever able to get a work around for this? I am concerned about jobs going to orphaned status and hanging up production.
I've had this problem start for a single windows agent since installing a new license.
It was an extension key to add some more agents but since then this one host has been bleating about dropping its connection. Its got the latest agent and we're patched up on 6.2.1 just shy of SP3.
I'm going to wait until we load SP3 before I start playing about with the license files since its always such fun getting new ones.
We were on version 126.96.36.1993 (Windows Masters), with 3.1.0.05 agents which kept restarting every 11 mins.... turns out for us it was a bad license on the master. Once we received and installed the new license, the problem was resolved.
Hope this is the answer to your problem.
I have this issue on a 6.1 install as well, only my disconnects are every 55 minutes. Do you have a Fault monitor setup in your environment? Reason I ask is that when the Backup master takes over job execution in my environment I do not see the disconnects, and the license file is only installed on the primary master.
thanks, Bill S
We have the same issue in our Production and Sandbox environment. We have loaded multiple licenses in each environment and still are experiencing this issue. Currently we are running on a Demo license as a work around. Please see CSCuv74153.
Has anyone seen this before? If so, has anyone found a solution?
Senior Tidal Administrator
John Laird suggested this fix:
02/26 11:25:41:130[1:main]: (mem=3551847384/4116709376) Scheduler: No license [Demo or Permanent] to process
Now you can start the Client Manager.
Since we went through this process (over a month ago), we have not experienced the dropped connections.
how much memory do you have allocated to your agent?
how many other windows agents are in your environment...and do they exhibit the same behavior?
have you begun scheduling any jobs yet on this newest agent?
We have floating licenses for all of our agents.
9 of our 12 Windows agents have this behavior every 55 minutes.
11 of 12 are at 188.8.131.52,1 of the offending 9 is 184.108.40.206, We have jobs scheduled on all of them.
When you say "how much memory do you have allocated to your agent? " are you referring to on the VM's or within Tidal?