持续使用外部数据更新的生产数据库报告了完整性检查 1719 错误。
这些可以分为:
116×
<Dbcc ID="1" Error="8909" Severity="16" State="1">Table error: Object ID 0, index ID -1, partition ID 0, alloc un
it ID 14918418583781376 (type Unknown), page ID (1:358049) contains an incorrect page ID in its page header. The
PageId in the page header = (45:3473461).</Dbcc>
总结:
<Dbcc ID="118" Error="8954" Severity="10" State="1">CHECKDB found 0 allocation errors and 118 consistency errors
not associated with any single object.</Dbcc>
544×
<Dbcc ID="119" Error="8928" Severity="16" State="1">Object ID 854294103, index ID 6, partition ID 720575941117214
72, alloc unit ID 72057594157203456 (type In-row data): Page (1:358048) could not be processed. See other errors
for details.</Dbcc>
总结:
<Dbcc ID="663" Error="8990" Severity="10" State="1">CHECKDB found 0 allocation errors and 544 consistency errors
in table 'Rpm.TrainRecord' (object ID 854294103).</Dbcc>
1021×
<Dbcc ID="1684" Error="8909" Severity="16" State="1">Table error: Object ID 1327343793, index ID 4, partition ID
72057594114670592, alloc unit ID 72057594160611328 (type In-row data), page ID (1:7031943) contains an incorrect
page ID in its page header. The PageId in the page header = (1:4795983).</Dbcc>
总结:
<Dbcc ID="1685" Error="8990" Severity="10" State="1">CHECKDB found 0 allocation errors and 1021 consistency error
s in table 'Audit.AuditLog' (object ID 1327343793).</Dbcc>
31×
<Dbcc ID="1716" Error="8928" Severity="16" State="1">Object ID 1349579846, index ID 7, partition ID 7205759410936
2176, alloc unit ID 72057594154647552 (type In-row data): Page (3:600863) could not be processed. See other erro
rs for details.</Dbcc>
总结:
<Dbcc ID="1717" Error="8990" Severity="10" State="1">CHECKDB found 0 allocation errors and 31 consistency errors
in table 'Rpm.AlarmEventRecord' (object ID 1349579846).</Dbcc>
最终输出是
<Dbcc ID="1720" Error="8958" Severity="10" State="1">repair_allow_data_loss is the minimum repair level for the e
rrors found by DBCC CHECKDB ([databasename]).</Dbcc>
我们有备份,问题似乎是在将数据库从一台服务器移动到另一台服务器(上周末)之后出现的。检查旧服务器,根本没有一致性错误。
数据库功能正常,可以备份/恢复。
我猜我需要修复数据丢失 - 但我能做比这更好的事情吗?
对于面临这种特定情况的任何其他人来说,解决方法是调整 Azure 中 VM 的大小。
很奇怪,但只是调整 VM 的大小,然后将其恢复到解决问题的大小。
我只能假设它将磁盘移动到另一个集群,而原始集群有某种问题,作为 SQL IO/一致性错误出现。