当前位置:Oracle研究中心 > 故障案例 >

【案例】Oracle报错ORA-00600 4193 分析产生原因和解决办法

时间:2016-09-10 14:00   来源:Oracle研究中心   作者:惜分飞   点击:

天萃荷净 分享一篇关于Oracle数据库报错ORA-00600 4193错误说明和解决办法。

联系:手机(+86 13429648788) QQ(107644445)QQ咨询惜分飞
标题:ORA-600 4193 错误说明和解决

分析Ora-00600 [5193]报错


  Format: ORA-600 [4193] [a] [b]

  versions 6.0 to 12.1


  A mismatch has been detected between Redo records and Rollback (Undo)

  We are validating the Undo block sequence number in the undo block against
  the Redo block sequence number relating to the change being applied.

  This error is reported when this validation fails.

  Arg [a] Undo record seq number
  Arg [b] Redo record seq number



ORA-600 [4193] [a] [b] [ ] [ ]  [ ]       
Versions: 7.2.2  - 9.2.0                              Source: ktuc.c
Meaning: seq# mismatch while adding an undo record to an undo block. This
         is done by the application of redo.
Argument Description:

    a. (ktubhseq): undo record seq# - this is the seq# of the block that
                                      this undo record WILL BE APPLIED TO.
                                      This is from the Undo Block. It is
                                      NOT the seq# of the undo block itself.

    b. (ktudbseq): redo RECORD seq# - this is the seq# number in the block
                                      that this redo WILL BE APPLIED TO.
                                      This is from the Redo Record. 


    This error is raised in kturdb which handles the adding of undo records
    by the application of redo. 

    When we try to apply redo to an undo block (forward changes are made by
    the application of redo to a block) we check that the seq# in the undo
    record matches the seq# in the redo record. These seq# should be the
    same because when we apply a redo record we must apply it to the
    correct version of the block. We can only apply a redo record to a
    block that contains the same seq# as in the redo record. 

    If the seq# do not match then this error is raised. This implies some
    kind of block corruption in either the redo or the undo block. 

7.3.x - 8.1.7.x
ASSERT2(ubh->ktubhseq == db->ktudbseq, OERI(4193), KSESVSGN,
            ubh->ktubhseq, db->ktudbseq);
ksesic2(OERI(4193), ksenrg(ubh->ktubhseq), ksenrg(db->ktudbseq));

struct ktubh
  kxid  ktubhxid;      /* txid of tx currently using or last used this block */
  ub2   ktubhseq;                              /* undo block sequence number */
  ub1   ktubhcnt;    /* high water mark record index, number of undo entries */
  ub1   ktubhirb;  /* rollback record index, rec index to start the rollback */
  ub1   ktubhicl;  /* collecting record index, rec index to start retrieving col info */
  ub1   ktubhflg;                                                 /* dummy */
  ub2   ktubhidx[1];     /* byte offset of record in block, grows at runtime */

struct ktudb   Kernel Transaction Undo Data operation Block (redo)
  ub2    ktudbsiz;                                          /* size of entry */
  ub2    ktudbspc;                 /* verification: space left in undo block */
  ub2    ktudbflg;            /* flag to indicate the kind of redo operation */
  kxid   ktudbxid;                                          /* current tx id */
  ub2    ktudbseq;                                  /* block sequence number */
  ub1    ktudbrec;                       /* new record index for this change */


最权威、专业的Oracle案例资源汇总之【案例】Oracle报错ORA-00600 4193 分析产生原因和解决办法





ORA-00600[4193] 分析产生原因

Ora-600 4193解决办法