Index update command in NSE does a disk space check before it
proceed to update the index to avoid index corruption due to
insufficient disk space.
However there is rare scenario where there might be enough disk
space and NSE reports CTE0310 error, like the one below:
2012-05-15-11.37.47.000000+000 I433A421 LEVEL: Error
PID : 12345678 TID : 1 PROC : cteprupx
INSTANCE: instance NODE : 000
APPID : *LOCAL.instance.120515160017
EDUID : 1
FUNCTION: DB2 UDB, trace services, sqlt_logerr_string (secondary
logging fu, probe:0
MESSAGE : Message
DATA #1 : String, 67 bytes
CTE0310 Index update is not allowed due to insufficient disk
space.
In such case additional traces needs to be gathered.
db2trc on -m *.*.96.*.* -f db2nse.trc
db2text update index INDEX_NAME for text commitcount 500 connect
to DB_NAME
db2trc stop
db2trc off
db2trc fmt db2nse.trc db2nse.fmt
Note that above index update command is an example. Adjust it
according to your need.
In db2nse.fmt file find checkDiskSpace(probe 94) entry like the
one below:
2181 data DB2 UDB Net Search Extender
CteUpdateIndexUtilities::checkDiskSpace(CteCLIConnection
&pConnection,CteTextIndexExtendedInformation &
pIndexInformation,string databaseName,string
instanceName,CteError & pError) fnc (3.3.96.1007.0.94)
pid 29688032 tid 1 cpid 19289 node 0 sec 0 nsec 419335041
probe 94
bytes 12
Data1 (PD_TYPE_DEFAULT,4) Hexdump:
0000 0001 ....
Returned hexadecimal value 0000 0001 refers to number of
documents indexed within current index. If above value doesn't
match the number of row indexed it means that problem's criteria
is met. |
