G
George Potemkin
Guest
> what does the error saying ? At some point in time one process has read the block number 7205141 from disk (from the prdcrn.d1 file). The block has passed the sanity checks - it was not corrupted. Later (at 23:31:35) user 43 decided to evict the block from buffer pool and to replace it by another block from disk. At this moment the sanity checks have found the corruption in the block's header. The corruption was left by some unknown "bad" process who updated the shared memory using a wrong offset.We don't know when it happened - there were no errors. We known only a footprint left by the process. Block dump contains the table's IDs that the "bad" process was updating but you posted only a part of block dump (only 336 bytes of 8192). The real contents of the block reported in the errors do not matter - the block was just an "incidental victim" of the "bad" process. It's really hard to envistigate a root case of the errors 49 and 4232. Contact Progress technical support.
Continue reading...
Continue reading...