I use Timesten 184.108.40.206, and have a read-only cache-group of a table, state is on.
- When I execute statement to refresh data from Oracle to TimesTen: call ttcacheautorefresh('<user_name>','<cache_group_name>', 1);
I met error:
*5039: An error occurred while refreshing <cache_group_name>: ftSchedulerGetRefreshStats: cannot find cache group with cgID 7284*
- I searched in Error Messages and SNMP traps, this is Internal Error :(
- P/s: When I alter cachegroup's state from to off, then on, and run 'call ttcacheautorefresh()' it's ???
Can anyone help?
As the TT instance admin user, can you do
select cgname, cgowner, cgid from sys.cache_group where cgid = #####
where cgid is the number returned in the error e.g 7284. Do this before you alter the cachegroup state (ie when it should return a valid row) and then after the error is reported.
The CGID should not change because you change the state of the cachegroup and its entry should not be removed from sys.cache_group. The REFRESH_STATE column will show Y or N depending whether it is on or off.
I met another error when call ttcacheautorefresh as following:
6003: Lock request denied because of time-out
Details: Tran 6.32826333 (pid 9264) wants IS lock on table SYS.CACHE_GROUP. But tran 1.18571354 (pid 29574) has it in X (request was X). Holder SQL (alter cache group <cachegroup_name> set autorefresh state on)
8227: Cache group <user_name>.<cachegroup_name> not found
Are you pasting in the actual error text? I need to know whether the error is actually displaying the string
or whether you are substituting this in to hide the actual username and cachegroup name for security purposes. That's fine but we need to be certain.