We are facing issue while configuring CA CEM. The Automatic transaction discovery is not working and it is throwing error NO Machine EM collector found for <TIM collector host name>.
I have checked the service assigned to TIM collector in CEM. It is showing status for the Tim collector status UNKNOW but for MOM is showing UP.
The second thing is no events are generating at CEM console.
Amit Singh Bisht
Please check the Reference screen-shot
Please let me know the ips of MOM and collectors. Did you already added TIM or you were not able to add TIM because TIM collection service will be only able to run once you add your TIM in the cluster. Please also share hosts file from MOM, collectors and TIM.
The issue seems to be like MOM and collectors are not able to resolve the host names of each other.
Appreciate your response
Both TIM's are added & I am able to enable/disable both TIM's.I checked the hosts files, there are entries in every host file..
We have 1 MOM, 2 collectors & 2 TIM server's
You have two issues. RTTM won't start amd Unknown EM. Both are documented in knowledge docs
1) RTTM won't start
See if this is an issue. Looking for teh string Svc or RTTM or BT Stats in EM log should say why
[\Manager.com.timestock.tess.services.service.\CheckServices] Could not start service 'TIM.Collection.Svc' on entity 25: the service did not stop
I would look at your EM clusters properties on MOM. Check hostname, ip, and em.x num
2) To fix the unknown EM issue do this
1) Delete all TIMs through CEM UI2) Shutdown all \EMs in the cluster (MOM and all Collectors)3) Execute > delete from ts_entity; > commit;4) Restart all \EMs in this order: 1. MOM 2. Collector which customer is planning to use for TCS (this step will assign TCS to this Collector automatically) 3. All other \EMs5) Re-add and enable only ONE TIM6) Double-check if TCS is assigned to Collector which customer is planning to use, if not assign TCS to this Collector7) Make sure that TCS and RTTM service is working at this point8) Add other 2 \TIMs (do NOT re-assign TCS to different Collector at this point
Thanks for your response,
I had few queries.
1. Is that compulsory to perform 3rd step if we able to delete TIM's though CEM UI.
2. Where can I find EM clusters properties on MOM?
There may be data about inactive TIMs that should be purged
Look in your <EM_Home>/config directory
# Define connection properties for clustered EMs.# Connection properties support hot config.# Each clustered EM requires three properties:# host: hostname or ip address of the collector# port: connecting port of the collector# publickey: the public key to authenticate the MOM to the collector -# either an absolute path or a path relative to the config directory[not commented out)#introscope.enterprisemanager.clustering.login.em1.host=hostname#introscope.enterprisemanager.clustering.login.em1.port=5001#introscope.enterprisemanager.clustering.login.em1.publickey=internal/server/EM.public#introscope.enterprisemanager.clustering.login.em1.weight=
I will follow all the steps according to the suggestions.
If Virtual IPs are being used the "Unknown" status may not be a cause for concern - this KB has details:
"Under the CEM Setup>Services>EM Configuration tab all Collectors show an "EM Running State" of "Unknown" instead of "Up". What is the significance of the "Unknown"" - Under the CEM Setup>Services>EM Configuration tab - CA Knowledge
An "EM Running State" of "Unknown" may or may not indicate that there is a functionality problem.
For example in a APM cluster failover environment with the following configuration:
An "Unknown" running state will be reported instead of "Up" because the VIP does not match any of the NIC IPs. However all functionality is working as expected.
Thank you for updating.
As per the KB article, you suggested the UNKNOWN status will not affect any functionally.
But what could be the issue behind the Automatic Transaction Discovery which is not working?
Meanwhile, I'll try the step's suggested by Hallett.
Amit Singh BSiht
I had tried your suggestion but it doesn't work out. The Automatic transaction discovery is still showing the same error.
The customer had configured 1 VIP by merging the two APM collectors/ servers.
So we have 4 collectors and 2 VIPs
APM Version: 10.5.1
It sounds like you do have a failover environment for the 2 collectors then.
Please create a support case so this problem can be analysed in more detail.
Please also include the link to this Community thread in the case description so the Engineer who takes ownership is aware of this history.
For the record we resolved this problem via a support case:
MOM EM properties file was using Collector VIPs (introscope.enterprisemanager.clustering.login.em*.host=<VIP>.).
The Collector running the Tim Collection Service needed its VIP <-> FQDN mapping to be added to the MOM hosts file.
The MOM was then able to do a successful lookup of the Collector FQDN with the VIP it knew about.
That resolved the Automatic Transaction Discovery (ATD) startup error:"No matching EM collector found for <FQDN of EM running Tim Collection Service>"
It also resolved the problem of the Collector "EM Running State" of "Unknown" which changed to "Up"
I have also updated the above KB article "Under the CEM Setup>Services>EM Configuration tab all Collectors show an "EM Running State" of "Unknown" instead of "Up". What is the significance of the "Unknown"":
Under the CEM Setup>Services>EM Configuration tab - CA Knowledge