RE: expdp and ORA-01555

Hmm... if even PCTVERSION 100 doesn't help, then this *may* be a bug related to PCTVERSION calculation codepath. In which case you could (temporarily) replace the PCTVERSION 100 with RETENTION as this should use slightly different codepath (and then the undo_retention comes into play).

This is just a guess though, if there's a bug involved, that ORA-1555 could be raised due any other reason in addition to the possible PCTVERSION issue...

--
Regards,
Tanel Poder
http://blog.tanelpoder.com
http://n.otepad.com - n.ote this!
 



> -----Original Message-----

> From: Yong Huang [mailto:yong321_at_yahoo.com] 

> Sent: Thursday, August 28, 2008 00:16

> To: richa03_at_gmail.com

> Cc: tanel.poder.003_at_mail.ee

> Subject: RE: expdp and ORA-01555

> 

> Richa,

> 

> We had two incidents of ORA-1555 on LOBs that look like 

> yours. But in both cases, the LOB is corrupted according to 

> the check in Note:452341.1 (probably hitting Bug 5212539, 

> which is fixed in 10.2.0.3). So it's different than yours.

> Our database is 10.2.0.2 on 64-bit RH Linux, Itanium.

> 

> Tanel's idea is very smart. I tried. I still got the errors.

> 


--
http://www.freelists.org/webpage/oracle-l