IBM E027SLL-H Troubleshooting Guide - Page 213
Error list appears in warehouse logs
![]() |
View all IBM E027SLL-H manuals
Add to My Manuals
Save this manual to your list of manuals |
Page 213 highlights
the persistent datastore detects that there are no empty datasets left, it finds the one with the oldest data and maintains it. If the BACKUP or EXPORT options are not specified, maintenance is done within the persistent datastore to initialize the dataset so that its status changes from full to empty. If BACKUP or EXPORT are specified, a job runs to save the data, then the dataset is initialized and marked as empty. If the BACKUP or EXPORT was specified, but the maintenance jobs fail to do their job, the recording would stop in the persistent datastore. In this case, datasets are taken off-line until there are no more available datasets for reading or writing. Note: If you allocate persistent datastores that fit more than 24 hours worth. The agent initializes and writes as much data as it can fit in the persistent datastores, 24 hours or more. The Tivoli Enterprise Portal, for short term history display, only pulls up 24 hours worth of data. The warehouse can archive ALL the data in the persistent datastores, regardless if it has been 24 hours or more worth of data. If you create a Tivoli Enterprise Portal query that is over 24 hours, then the warehouse fulfills that request regardless if the data is in the online persistent datastores. Also, because the agent or Tivoli Enterprise Monitoring Server reads the entire persistent datastores at initialization time, you should not allocate very large persistent datastores to potentially store more than 24 hours. That increases the Tivoli Enterprise Monitoring Server and agent startup time. As mentioned above, the agent writes to it, but the Tivoli Enterprise Portal only displays 24 hours from it. The warehouse processing reads all the data in the persistent datastores (24 hours or more), but there is a trade-off in Tivoli Enterprise Monitoring Server and agent startup time. It is always best to calculate space for 24 hours as best as possible. As far as the potential of losing historical data, if the warehouse is down over 24 hours, that is a potential problem, assuming the persistent datastore's backup processing is functioning and, therefore, the agent does not stop writing to the persistent datastores. Since you have the choice of collecting history data at the Tivoli Enterprise Monitoring Server or the Tivoli Enterprise Monitoring Agent, the persistent datastore should be defined in both places. If you are 100% sure that you will always collect at the Tivoli Enterprise Monitoring Agent or always collect at the Tivoli Enterprise Monitoring Server then you can optionally define the persistent datastore in only one location. Note that many configuration issues occur because the person installing the product selects one location for the persistent datastore and sometime later someone else enables history collection for the other location. Error list appears in warehouse logs The following error list appears in the warehouse logs: == 25 t=Thread-1 com.ibm.db2.jcc.c.DisconnectException: A communication error has been detected. Communication protocol being used: T4Agent.sendRequest(). Communication API being used: OutputStream.flush(). Location where the error was detected: There is no process to read data written to a pipe. Communication function detecting the error: *. Protocol specific error codes(s) TCP/IP SOCKETS DB2ConnectionCorrelator: G92A17E8.C3D2.071018074633 at com.ibm.db2.jcc.b.a.a(a.java:373) at com.ibm.db2.jcc.b.a.y(a.java:346) at com.ibm.db2.jcc.b.a.l(a.java:298) at com.ibm.db2.jcc.c.j.c(j.java:234) Chapter 10. Monitoring agent troubleshooting 195
![](/manual_guide/products/ibm-e027sllh-troubleshooting-guide-7c8d102/213.png)