Troubleshooting
Problem
In pureScale, starting CF with db2start failed with error SQL1677N
Symptom
db2start fails with following error:
128 0 SQL1677N DB2START or DB2STOP processing failed due to a DB2 cluster services error.
Following error can be found in db2diag.log:
<timestamp> LEVEL: Severe
PID : 1234567 TID : 1 PROC : ca-wdog 128 [db2in1]
INSTANCE: db2in1 NODE : 128
HOSTNAME: HOST0001
EDUID : 1 EDUNAME: ca-wdog 128 [db2in1]
FUNCTION: DB2 UDB, high avail services, rocmGetHCARSCTHandles,
probe:1198
MESSAGE : ZRC=0x827300AC=-2106392404=HA_ZRC_CONFIGURATION_ERROR
"HA is configured incorrectly"
DATA #1 : String, 135 bytes
No RSCT handles were found for any HCA. Make sure the host for the CF
or member is configured correctly. (db2nodes.cfg and /etc/hosts)
DATA #2 : Codepath, 8 bytes
9:18:21:24
DATA #3 : Database Partition Number, PD_TYPE_NODE, 2 bytes
128
Cause
This may due to the configuration issues on InfiniBand or RoCE card. First verify the configuration is correct for InfiniBand or RoCE.
It may also be due to the configuration of the following parameter in database manager configuration:
Transport method to CF (CF_TRANSPORT_METHOD) = TCP
Resolving The Problem
Update the parameter to RDMA and try again:
db2 update dbm cfg using CF_TRANSPORT_METHOD RDMA
Was this topic helpful?
Document Information
Modified date:
16 June 2018
UID
swg21982540