Data Corruption Error
Posted in 1995
Topics: Storage & Space Management, SQL Development & Query Writing, Error Codes & Troubleshooting, Connectivity: ESQL/C, 4GL & Embedded SQL
HELP!
We have come across a problem with our database and would appreciate
any information about similar problems and/or fixes you may have
experienced.
-------------------------------------------------
04:32:30 bfcheck: bad page: pg_addr 23152e != bp->bf_pagenum 231525, userp = 80
402084, pid = 7555, uid = 270
04:32:30 buffer header:
04:32:30 8047e3f4: 00000000 00000000 ec264980 5c8f4780 ........ .&I.\\.G.
04:32:30 8047e404: 30024080 0c614980 02000000 02040000 0.@..aI. ........
04:32:30 8047e414: 25152300 00104f80 00000000 00000000 %.#...O. ........
04:32:30 8047e424: 01000000 3ca24980 00010000 ....<.I. ....
04:32:30 page header:
04:32:30 804f1000: 2e152300 a3ff90a9 65000100 68060000 ..#..... e...h...
04:32:30 804f1010: 00000000 00000000 ........
04:32:30 slot table and stamp:
04:32:30 804f17fc: a3ff90a9 ....
04:32:30 -- Fail Consistency Check -- isread:bad data page -- pid=7555 user=270
us=80402084Program stopped at "PLupdate.4gl", line number 429.
SQL statement error number -246.
Could not do an indexed read to get the next row.
SYSTEM error number -105.
ISAM error: bad isam file format.
------------------------------------------------- We have seen this corruption error when joining a large permanent
table (over 100,000 rows) with a temp table in the root dbspace
(created with 'INTO TEMP ourtemp WITH NO LOG', and indexed). We
regularly perform consistency checks on all database using tbcheck,
and have had no signs of problems. We have not been able to
consistently re-produce this error. We have no evidence that we
are having hardware failures with memory or hard disks and according
to the informix manual, our database would be down if our root
dbspace was corrupt.
This problem has occurred twice since we increased our BUFFERS and
TBLSPACES paramaters in tbconfig. We believe this may be part
of our problem, so we will be resetting our paramaters in an attempt
to count this out.
Informix technical support could find no related reports in their
bugs database, and suggested that we reset our buffers to see if
this solves our problem.
We are using the Informix Online engine version 5.03.UC1 on a SCO Unix
platform version 3.2.4.3 with MPX 3.0. These errors occured using
the RDS 4GL version 4.11.UC1.
If you have experienced similar problems with your databases we would
greatly appreciate your insight. Please respond via email as our
news feed is behind.
Spencer Hanson
spencer@group1.com