Sadly, I have seen this. Several times, in fact. Today Ill get no less than 7
disks replaced in the last array to suffer similar problems.
The root cause is old firmware in the disk array that allows disks to fail
silently. We did not notice until the raid5 array underlying ASM became
'punctured' and lost a database.
My advice would be to check the OS logs for IO errors and if there are any,
contact your hatdware vendor.
Cheers,
Tony
On 13 December 2016 8:40:20 AM AEST, Mladen Gogala <gogala.mladen@xxxxxxxxx>
wrote:
On 12/12/2016 03:23 PM, GG wrote:
Corrupt block 70167 found during reading backup piece, file=network,
corr_type=3
Continuing reading piece network, no other copies available.
Fri Dec 09 13:59:07 2016
Datafile rdba reconstruction error, expected block greater than
140983, got 140862 for datafile 3
Corrupt block 125161 found during reading backup piece, file=network,
corr_type=4
Continuing reading piece network, no other copies available.
This looks like a "call support now" case. I have never seen this
before, but it doesn't look good.
--
Mladen Gogala
Oracle DBA
Tel: (347) 321-1217
--
//www.freelists.org/webpage/oracle-l
