Translating with DrWatson… this can take a few seconds the first time.
This is a genuine, complex translation. DrWatson protects commands, error codes, and log output while naturally translating the surrounding text. It’s translated once and saved.
Dennis Melnikov got error 9810/12014 (cannot find entry in lofd table) from UPDATE STATISTICS HIGH on a table with no LOB columns on IDS 11.70.FC5XE. Art Kagel asked about large LVARCHARs and suggested oncheck. Mike Walker found the table was fragmented by expression and said fragment-level statistics use smart blobs. He advised checking sbspace free space, noted APAR IT27566, and said switching to table-level statistics was a fallback.
Auto-generated by Claude from the posts below — may be imperfect; read the full thread.
Dennis Melnikov — source: IBM Community (ConnectedCommunity.org) Informix forum
Hi,
IBM Informix Dynamic Server Version 11.70.FC5XE
We have a rather weird error on UPDATE STATISTICS,
UPDATE STATISTICS HIGH FOR TABLE ord_l_2 ( c_deliv,deliv_printed,notif_id,ofd_id,ord_id,ord_l_id,pp_status,status_req,time_deliv1) ;9810: Smart-large-object error.
12014: Smart Large Objects: Cannot find entry in lofd table.
Error in line 3Near character position 129
A fact is that the table has no LOB.
------------------------------
Sincerely,
Dennis
------------------------------
↪ replying to Dennis Melnikov
Art Kagel — source: IBM Community (ConnectedCommunity.org) Informix forum
Dennis:
Does the table contain any LVARCHAR columns defined as longer than 4K? Those strings are stored out-of-row in smart blob objects when the particular string is longer than 4K (lvarchar string shorter than 4K, even if the definition is longer, are stored in-row).
Art
------------------------------
Art S. Kagel, President and Principal Consultant
ASK Database Management Corp.
www.askdbmgt.com
------------------------------
↪ replying to Art Kagel
Dennis Melnikov — source: IBM Community (ConnectedCommunity.org) Informix forum
Art,
No column longer than CHAR(75).
------------------------------
Sincerely,
Dennis
------------------------------
↪ replying to Dennis Melnikov
Art Kagel — source: IBM Community (ConnectedCommunity.org) Informix forum
That is odd! Can you run two onchecks to make sure there is no corruption:
oncheck -cc <database>
oncheck -cDI <database>.<tablename>Let's get that out of the way.
Art
------------------------------
Art S. Kagel, President and Principal Consultant
ASK Database Management Corp.
www.askdbmgt.com
------------------------------
↪ replying to Dennis Melnikov
Mike Walker — source: IBM Community (ConnectedCommunity.org) Informix forum
Hi Dennis,
Does the table have any fragmentation, especially RANGE fragmentation?
------------------------------
Mike Walker
xDB Systems, Inc
www.xdbsystems.com
------------------------------
↪ replying to Mike Walker
Dennis Melnikov — source: IBM Community (ConnectedCommunity.org) Informix forum
Mike,
Yes, the table as well as some indices has fragmentation, by expression.
------------------------------
Sincerely,
Dennis
------------------------------
↪ replying to Dennis Melnikov
Mike Walker — source: IBM Community (ConnectedCommunity.org) Informix forum
You are likely running into an issue with fragment level statistics. These use blobs.
First make sure that your sbspace in SYSSBSPACENAME has free space.
You may be running into this problem, even though you are not using range fragmentation:
IT27566: RANGE FRAGMENTED TABLE WITH FRAGMENT LEVEL STATISTICS CAN HAVE NEGATIVE COUNT IN BIN AFTER UPDATE STATISTICS MEDIUM (ibm.com)
If you really can't find a way around this, you can try changing the stats for the table to table-level instead of fragment level. See
Fragment-level statistics - IBM Documentation
------------------------------
Mike Walker
xDB Systems, Inc
www.xdbsystems.com
------------------------------
We use strictly necessary cookies to make this site work. With your
consent we’d also use optional cookies for analytics and marketing. You can accept all,
reject all, or choose. Read our Cookie Policy.