We were trying to run an export on a table to a file on the system (DB2 V8.1 FP2, AIX 5.2)

db2 +c "export to /archive/proda/DB2/symavc/document_index_search.ixf of ixf select b.* from avc_transaction a, document_index_search b where cast(a.avc_trace_id as varchar(255)) = cast(b.avc_trace_id as varchar(255))and substr(cast(symcor_service_path as varchar(2000)),length(cast(symcor_service_path as varchar(2000))) - 14, 8) >= '20051101'"

and this was the resulting error

ADM6017E The table space "TEMPSPACE1" (ID "1") is full. Detected on container
"/archive/proda/DB2/avcpa1/avc/TEMPSPACE" (ID "0"). The underlying file system
is full or the maximum allowed space usage for the file system has been
reached. It is also possible that there are user limits in place with respect
to maximum file size and these limits have been reached.

2006-03-06-12.34.09.334378 Instance:avcpa1 Node:000
PID:307402(db2pclnr) TID:1 Appid:none
buffer pool services sqlbClnrAsyncWriteCompletion Probe:0

This is the filesystem

/dev/webdb2lv00 5242880 4963744 6% 475 1% /webdb1/hacmpapps/archive/proda/DB2/avcpa1

this is the file it created before it filled up (this file gets written to another filesystem that has plenty of space)

-rw-r----- 1 symavc arcsupp 100651000 Mar 06 12:34 document_index_search.ixf

The funny thing is earlier we were able to create the following files from the export utility without any issue

rw-r--r-- 1 symavc arcsupp 2464792528 Mar 05 14:00 delivery.ixf
-rw-r--r-- 1 symavc arcsupp 322862286 Mar 05 14:04 todolist_entry.ixf
-rw-r--r-- 1 symavc arcsupp 1734053947 Mar 05 14:24 document_index_retrieval.ixf
-rw-r--r-- 1 symavc arcsupp 3046529805 Mar 05 14:51 avc_transaction.ixf

So we are stumped as to what to do next? expand the filesystem that filled up? or do something else?

thanks in advance

Mark

Have you checked ulimit of instanceid?

Also, maybe the select is filling up tempspace.

It looks as though tempspace is filling up but it is not while we are running other exports so yes we do think this select is causing the problem. The instance id and the user creating the file on the filesystem has no limits on file size creation.

So our only recourse currently is increasing the filesystem to increase the amount of TEMPSPACE is available but this is only a stop gap fix and not really a long term solution

Mark

Were there other things that might have used temp space when you ran the export that contributed to the disk filling up?

I don't understand why you are using the +c option, but I would remove it just in case it is causing a problem. You should never use +c on an export or import. You can use commitcount on an import to keep the logs from filling up.