Kbase P125224: Error 1152 with database crash using RAID 0.
| Autor |
Progress Software Corporation - Progress |
| Acesso |
Público |
| Publicação |
01/08/2007 |
|
Status: Unverified
FACT(s) (Environment):
Windows
Progress/OpenEdge Versions
SYMPTOM(s):
Error 1152 with database crash using RAID 0.
The RAID level being used is RAID 0.
Hardware disk failure caused was corrected with a new disk. RAID then recovered the data and the database continued to work normally.
Some time after the disk had been replaced and everything had been working normally, the database crashes with multiple 1152 and 1124 errors.
SYSTEM ERROR: read wrong dbkey at offset <offset> in file <file>
found <dbkey>, expected <dbkey>, retrying. (1152)
SYSTEM ERROR: wrong dbkey in block. Found <dbkey>, should be <dbkey2> (1124)
After rebooting the server machine the users can no longer log in. When they try this they get error 604:
Extent <file-name> has the wrong creation date. (604)
There are large gaps in the database log file since the original hardware failure, and some of the logging messages have been truncated or cut off while writing to the log file. For example:
11:34:32 BROKER 0: BI File Threshold Stall (-bistall): Disabled. (6552)
11:34:32 BROKER 0: Number of Before-Image Buffers (-bibufs): 25. (4252)
11:34:32 BROKER 0: After-Image File Name (-a): Not Enabled. (4253)
11:34:32 BROKER 0: After-Image Stall (
: Userid is now ATMSCYL. (708)
11:47:36 SRV 43: Previous message sent on behalf of user 134. (5512)
11:47:42 SRV 44: Started on port 3043 using TCP, pid 2632. (5646)
11:47:43 SRV 44: Login usernum 133, userid pascualb, on alziraw031.
(742)
CAUSE:
RAID 0 is disk striping and does not provide data redundancy. Subsequently when a physical disk fails, its data is lost and there are holes in the database similar to those in the database log file. This eventually results in the 1152 and 1124 errors.
FIX:
Restore the database from a reliable backup taken before the original hardware failure.