I have seen a couple of dumps occured with the new kernel, the issue seems to be fixed with the latest DBSL.

The DBIF_REPO_SQL_ERROR dump and developer trace refer to such entiries:

C     OCILobRead2(substring) failed with SQL error 12577:

C     ORA-12577: Message 12577 not found;  product=RDBMS; facility=ORA


C  <–OCILobRead2(con=0, BLOBloc=x5dce058, slen=0, offs=1, buf=x9bc2380, bln=65536, ONE_PIECE, amt=0) -> orc=12577

C  used+free LL chunks: num=1+3, cnt=0+0, max=1+3; all locators: Lb.cnt=4, Lb.max=20, Lb.itp=0000NULL

C  —————————————————————————————————

C  Used LOB locators: 0 itp=x5d435c0, locarr=x5d6bca0, cnt=0, max=1

C                       #  0 LOB locator = x5dce058

C  Free LOB locators: 1 itp=x5ec4b80, locarr=x5ec4c00, cnt=0, max=1

C                       #  0 LOB locator = x631dd90

C                     2 itp=x6b2c130, locarr=x6a3ea30, cnt=0, max=1

C                       #  0 LOB locator = x60a1ec0

C                     3 itp=x69f69e0, locarr=x5f67bf0, cnt=0, max=1

C                       #  0 LOB locator = x5fffbf0           >> total memory 96 bytes

C  LOB locators of current transaction:

C  Lb.locarr = x5dd46d0, Lb.cnt = 4

C     Lb.locarr[0]=x5dce058, sch=x5d4ace0, len=7412

C     Lb.locarr[1]=x631dd90, sch=0000NULL, len=0

C     Lb.locarr[2]=x60a1ec0, sch=0000NULL, len=0

C     Lb.locarr[3]=x5fffbf0, sch=0000NULL, len=0

Please refer as well to the KBA 1898521 – “Process list grows after ORA-01017; ORA-00020”.

With the new patches under point 2 several bugs are fixed including ora-12577.

So if you get such dumps first apply the latest DBSL before opening an incident.

To report this post you need to login first.

1 Comment

You must be Logged on to comment or reply to a post.

Leave a Reply