RE: Strange HPL failure

255 can mean many things, did you check the hpl job/session reject file and log file?, these files will have some details why the hpl job/session failed. usually, when i get a 255 error it is one of the following problems, 1. mis-matched data-type, i mean loading a char to a date/datetime column 2. ipload not getting sufficient virtual memory - bouncing the engine resolves this problem or adjust the onconfig parameter for sharememory 3. corrupt onpload database - may need to run oncheck on the opload database 4. drop and re-create the onpload database 5. hung share memory segment - may need to cleanup the share memory segment can you create a new job and execute it? -----Original Message----- From: owner-informix-list@iiug.org [mailto:owner-informix-list@iiug.org]On Behalf Of Neil Truby Sent: Thursday, June 17, 2004 5:28 PM To: informix-list@iiug.org Subject: Strange HPL failure IDS 9.40FC3W2 on HP-UX 11.11 HPL has "suddenly" stopped working on this database server, although it runs OK on otther databases on the same server, And we did previously have it working on this database server at this version. Job adds, describes etc work OK but when you try to run a job it hangs for a minute or so then barfs with the strange error below. If you then capture the $? return value, it's 255. It's the same on 9.40FC4. Anyone run into this? cheers Neil informix@hamilton[/home/system/informix] onpladm describe device andy_test_ar> BEGIN OBJECT DEVICEARRAY andy_test_arr BEGIN SEQUENCE TYPE FILE FILE /exports/andy/hplscripts/file1 <snip> END OBJECT informix@hamilton[/home/system/informix] onpladm run job prod_andy -fu Connecting to onpload, Please wait... Onpload cannot validate logging state of target database informix@hamilton[/home/system/informix] echo $? 255 sending to informix-list

Thanks Savio In fact, we'd run into a known bug. Here's what UK Tech SUpport told us to do (and it worked): With regards to your problem it seems that you may be hitting bug 167408 Bug: 167408 ONPLOAD RETURNED ERROR ACCESSING TABLE 'SYSDRI', SQL ERROR 100, ISAM ERROR 0 Check to see if the following returns the return codes shown below.. select type from sysmaster:sysdrcb;returns -1 select * from sysdrireturns 'no rows found' ('sql 100'). If they do try the workaround below, you may like to check the sysmaster table first.. Workaround: 9322 echo "INSERT INTO flags_text (tabname, flags, txt) VALUES ('sysdrcb', -1, 'Not Initialized');"| dbaccess sysmaster As a matter of interest, you used to be able to find things like this in TechInfo. Has there something similar replaced it? regards Neil