Disabled Temp DBspace -- Revisited

This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------ =_NextPart_000_01BCD4CA.3A5E4470 Content-Type: text/plain; charset="iso-8859-1" Hello All, We're running ODS 7.20.UC3 on a DEC 8200, O/S version 3.2. Summary: Following an assert failure / segmentation violation that brought down the engine, we rebooted the box. Recovery failed on a chunk in a temp dbspace. Now the chunk / dbspace is marked DOWN. For a workaround, we re-assigned the DBSPACETEMP to another dbspace/chunk, then bounced the instance. The interesting part: Browsing thru dejanews, I found that someone else had this problem over a YEAR ago on an AIX 4.1 platform w/ ODS 7.12.UC2. There was speculation then that the problem would be fixed in 7.2x. Obviously, it has not been fixed, at least not on the DEC platform. Question: Has anyone else had this problem?? Any cures?? For those interested, here are the gory details: ... 14:28:03 Checkpoint Completed: duration was 0 seconds. 14:33:03 Checkpoint Completed: duration was 0 seconds. 14:34:42 Assert Failed: Internal Error - Segmentation Violation 14:34:42 Who: Session(1319, userid@<UNKNOWN-HOST>, -1, 118598248) Thread(1355, sqlexec, 5d5b1f8, 3) 14:34:42 Results: OnLine must abort 14:34:42 Action: Reinitialize shared memory 14:34:42 See Also: /tmp/af.54b5971 14:34:42 rsinit.c, line 9640, thread 1355, proc id 25164, Segmentation Violation. 14:34:42 PANIC: Attempting to bring system down Mon Oct 6 15:22:48 1997 15:22:48 Event alarms enabled. ALARMPROG = '/opt/informix/prod/log_full_dq1.sh' 15:22:52 DR: DRAUTO is 0 (Off) 15:22:54 INFORMIX-OnLine Initialized -- Shared Memory Initialized. 15:22:54 Physical Recovery Started. 15:22:55 Physical Recovery Complete: 64 Pages Restored. 15:22:55 Logical Recovery Started. 15:22:56 Assert Failed: WARNING! pthdrpage:ptalloc:bad partn page 15:22:56 Who: Session(24, informix@infx_srvr, 0, 97903240) Thread(56, xchg_1.4, 5d32218, 3) 15:22:56 Results: Cannot use TBLSpace page for TBLSpace 11534374 15:22:56 Action: Run 'oncheck -pt 11534374' <<< NOTE: Here's where the real problem began. <<< This "oncheck" just gave us: <<< Error opening TBLspace b00026. ?????? 15:22:56 See Also: /tmp/af.3864bf 15:22:56 Rollforward of log record failed. iserrno = 101 15:22:56 Log Record: log = 429, pos = 4337e0, type = 91[PTCOLUMN],trans = 157 15:22:56 Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/prod/log_full_dq1.sh 3 1 "Table failure: 'dq1p:"aboivin"._temptable'." "pthdrpage:ptallo 15:22:57 Assert Failed: WARNING! Chunk 11 is being taken OFFLINE. 15:22:57 Who: Session(24, informix@infx_srvr, 0, 97900040) Thread(56, xchg_1.4, 5d32218, 3) 15:22:57 Results: DBspace dq1tmp02s is disabled. 15:22:57 Action: Restore DBspace dq1tmp02s <<< NOTE: Tried to do this restore of dq1tmp02s (a temp dbspace) -- didn't work!! 15:22:57 See Also: /tmp/af.3864bf 15:22:57 Logical Recovery Complete. 20 Committed, 0 Rolled Back, 0 Open, 0 Bad Locks 15:22:57 Assert Failed: Memory freed at bad boundary. 2432 bytes at2062a5680 15:22:57 Who: Session(24, informix@infx_srvr, 0, 0) Thread(49, fast_rec, 0, 5) 15:22:57 Results: Memory lost 15:22:57 Action: Action: %s 15:22:57 See Also: /tmp/af.3164c1 15:22:57 Dataskip is now ON for all dbspaces 15:22:58 On-Line Mode 15:22:58 Checkpoint Completed: duration was 0 seconds. 15:22:58 Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/prod/log_full_dq1.sh 4 5 "DBspace is off-line:'dq1tmp02s'." "Chunk 11 is being taken OF <<< NOTE: Chunk 11 is the only one in that temp dbspace. 15:28:26 Checkpoint Completed: duration was 0 seconds. 15:33:26 Checkpoint Completed: duration was 0 seconds. [SNIP] 16:08:26 Checkpoint Completed: duration was 0 seconds. 16:13:26 Checkpoint Completed: duration was 0 seconds. 16:18:07 Skipped existing audit trail /onaudit/infx_srvr1_shm.0. 16:18:07 Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/prod/log_full_dq1.sh 2 6 "Internal Subsystem failure:'Audit'" "Skipped existing audit t 16:18:07 Skipped existing audit trail /onaudit/infx_srvr1_shm.1. 16:18:07 Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/prod/log_full_dq1.sh 2 6 "Internal Subsystem failure:'Audit'" "Skipped existing audit t 16:18:07 Skipped existing audit trail /onaudit/infx_srvr1_shm.2. 16:18:07 Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/prod/log_full_dq1.sh 2 6 "Internal Subsystem failure:'Audit'" "Skipped existing audit t 16:18:07 Skipped existing audit trail /onaudit/infx_srvr1_shm.3. 16:18:07 Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/prod/log_full_dq1.sh 2 6 "Internal Subsystem failure:'Audit'" "Skipped existing audit t 16:18:07 Skipped existing audit trail /onaudit/infx_srvr1_shm.4. 16:18:07 Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/prod/log_full_dq1.sh 2 6 "Internal Subsystem failure:'Audit'" "Skipped existing audit t 16:18:07 Skipped existing audit trail /onaudit/infx_srvr1_shm.5. 16:18:07 Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/prod/log_full_dq1.sh 2 6 "Internal Subsystem failure:'Audit'" "Skipped existing audit t 16:18:07 Skipped existing audit trail /onaudit/infx_srvr1_shm.6. 16:18:07 Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/prod/log_full_dq1.sh 2 6 "Internal Subsystem failure:'Audit'" "Skipped existing audit t 16:18:26 Checkpoint Completed: duration was 0 seconds. 16:23:26 Checkpoint Completed: duration was 0 seconds. 16:28:26 Checkpoint Completed: duration was 0 seconds. 16:43:26 Checkpoint Completed: duration was 0 seconds. 16:48:26 Checkpoint Completed: duration was 0 seconds. 16:49:30 Warning: Invalid dbspace 'dq1tmp02s' listed in DBSPACETEMP. 16:49:31 Checkpoint Completed: duration was 0 seconds. 16:50:24 Checkpoint Completed: duration was 0 seconds. 16:50:26 INFORMIX-OnLine Stopped. Mon Oct 6 16:50:38 1997 16:50:38 Event alarms enabled. ALARMPROG = '/opt/informix/prod/log_full_dq1.sh' 16:50:41 DR: DRAUTO is 0 (Off) 16:50:42 INFORMIX-OnLine Initialized -- Shared Memory Initialized. 16:50:42 Physical Recovery Started. 16:50:42 Physical Recovery Complete: 0 Pages Restored. 16:50:42 Logical Recovery Started. 16:50:45 Logical Recovery Complete. 0 Committed, 0 Rolled Back, 0 Open, 0 Bad Locks 16:50:45 Onconfig parameter DBSPACETEMP modified from dq1tmp02s to dq1tmp01s. 16:50:45 Dataskip is now ON for all dbspaces 16:50:45 Checkpoint Completed: duration was 0 seconds. 16:50:45 Dropping temporary TBLspace 100067, recovering 8 pages. 16:50:45 Dropping temporary TBLspace 100068, recovering 8 pages. 16:50:45 On-Line Mode 16:56:14 Checkpoint Completed: duration was 0 seconds. 17:01:14 Checkpoint Completed: duration