Technical Blog Post
Abstract
75 ways to demystify DB2: #25: Techtip: Connections being rejected on a V9.8 pureScale environment
Body
Connections being rejected on a V9.8 pureScale environment
Any connections being made to the database that is on a v9.8 pureScale instance is experiencing a connection hang and instance may experience a 'hang' like situation.
If we take a few iteration of stacks (db2pd -stacks all -rep 10 5), and the stack analysis may have the following output:
0x0900000000533990 _p_nsleep + 0x10
0x0900000000038BE4 nsleep + 0xE4
0x090000000015B290 nanosleep + 0x190
0x0900000007058B60 ossSleep + 0x80
0x090000001209DAA8 shouldRejectIncoming__8sqeDBMgrFP8sqeAgentP14sqeDbStatusKeyP18sqeConnectDecisionP5sqlcaPbT5 + 0x420
0x090000001209E688 StartUsingLocalDatabase__8sqeDBMgrFP8SQLE_BWAP8sqeAgentRccP8sqlo_gmtPb + 0x318
0x0900000011FAB64C AppStartUsing__14sqeApplicationFP8SQLE_BWAP8sqeAgentcT3P5sqlcaPc + 0x464
0x0900000011FAD08C AppLocalStart__14sqeApplicationFP14db2UCinterface + 0x514
0x0900000011CE1424 sqlelostWrp__FP14db2UCinterface + 0x38
0x0900000011CE14D8 sqleUCengnInit__FP14db2UCinterfaceUs + 0x64
0x0900000011B6B940 sqleUCagentConnect + 0x2C4
0x0900000011BC5790 sqljsConnectAttach__FP13sqljsDrdaAsCbP14db2UCinterface + 0x100
0x0900000011BC5500 sqljs_ddm_accsec__FP14db2UCinterfaceP13sqljDDMObject + 0x330
0x0900000011BC5144 sqljsParseConnect__FP13sqljsDrdaAsCbP13sqljDDMObjectP14db2UCinterface + 0x4C
0x0900000011BC508C .sqljsParse__FP13sqljsDrdaAsCbP14db2UCinterfaceP8sqeAgentb.fdpr.clone.95 + 0x2C
0x0900000011D16F2C @69@sqljsSqlam__FP14db2UCinterfaceP8sqeAgentb + 0x22C
0x0900000011C3E064 @69@sqljsDriveRequests__FP8sqeAgentP14db2UCconHandle + 0xA8
0x0900000011C3DCD8 @69@sqljsDrdaAsInnerDriver__FP18SQLCC_INITSTRUCT_Tb + 0x38C
0x0900000011C3D888 sqljsDrdaAsDriver__FP18SQLCC_INITSTRUCT_T + 0xE8
0x0900000011F8364C RunEDU__8sqeAgentFv + 0x14C
0x0900000011F86C48 EDUDriver__9sqzEDUObjFv + 0x2EC
0x0900000011F90020 sqloEDUEntry + 0x2E4
Based on this, first check we can do is to ensure DB2_DATABASE_CF_MEMORY AND NUMDB variables are set properly.
For the problem I faced the work around is to set DB2_DATABASE_CF_MEMORY to -1 and NUMDB to something larger than the actual number of databases you have.
If the above workaround is not applicable or doesn't work, then more analysis needs to be done and DB2 Support can be engaged.
[{"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SSEPGG","label":"Db2 for Linux, UNIX and Windows"},"Component":"","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"","Edition":"","Line of Business":{"code":"LOB10","label":"Data and AI"}}]
UID
ibm11141156