ca.portal.admin

Re: question on QUED

Discussion created by ca.portal.admin on Sep 21, 2010
I think your problem is that the QUED task was unable to write an ABRT checkpoint record, so was unable to rollback.

The ""normal"" RHDCRUAL run units should behave similar to any other run unit that takes a COMMIT, locks are freed and journal images are available for offloading.

In the case of QUED, it (evidently) takes no COMMITs, so unless you can hold all the BFOR/AFTR images in your journals, you'll exhaust all your journals.

Just a hypothesis.

Don Casey

----- Original Message -----
From: Chris Hoelscher <choelscher@HUMANA.COM>
To: IDMS-L@LISTSERV.IUASSN.COM
Sent: Tue, 21 Sep 2010 13:47:42 -0000 (UTC)
Subject: question on QUED

i have a very active read-only CV - however it does update Queues (being
r/o journals are very small)

although several applications have been subset - the jobs that write to
these queues are still active- so i execute QUED prompt
I note then when a journal fills and the info message is written

DC205030 V75 T53649733 LID=4179105942 PROG=RHDCRUAL SUBS=IDMSNWK7
BFOR=438KB

rhdcrual is the program - interestingly - even though i exit qued -
condensing does not free up the journal space

is this because a pre-defined (RHDCRUAL) run unit is considered still
active and as such the BFORS cannot be removed?

the DCMT D RUNUNIT shows:

TYPE QUEUE
DRIVER TASK ID 0000000002
SUBSCHEMA IDMSNWK7
NODE
DICTNAME/DBNAME SYSTEM
IDLE INTERVAL OFF
PREDEFINED RUN UNITS 2
RUN UNIT ALLOCATIONS 162
RUN UNIT FREES 162
OVERFLOW RUN UNITS 0
AREA NAME DDLDCRUN
USAGE MODE SHARED UPDATE
TYPE BOUND IN-USE ALLOCS OWNING TASK
PERM YES NO 161
PERM YES NO 1


Chris Hoelscher
IDMS/DB2 System & Database Architect
Humana Inc
502-476-2538
choelscher@humana.com

I refuse to repeat gossip - so listen carefully the first time




The information transmitted is intended only for the person or entity to which it is addressed and may contain CONFIDENTIAL material. If you receive this material/information in error, please contact the sender and delete or destroy the material/information.
"
IDMS Public Discussion Forum
IDMS-L@LISTSERV.IUASSN.COM
SMTP
IDMS-L@LISTSERV.IUASSN.COM
IDMS-L@LISTSERV.IUASSN.COM
SMTP








Normal

Normal
Re: question on QUED
"Each queue record consists of one SROOT-DCS-139 (or something like that, I don't have a dictionary Bachmann in front of me) attached to QUEUE-DCQ-138 record.

Depending on the size of the queue record, there will be 0-n SEXT-DCS-140 records (something like 400-500 bytes each).

As QUED processes the DDLDCRUN AREA, it will remove expired queues. This results in erasing the entire 138/139/140 structure (all the underlying IDMS records). These are journalled like any other records would be, resulting in BFOR/AFTR images being sent to the journal.

Since these are ERASEs, the AFTR image is null, therefore the bulk of the data is in the BFOR record.

Scratch records, although they use the same record types, live in a different AREA where standard journalling is suppressed.

----- Original Message -----
From: Lutz Petzold <PetzoldL@AETNA.COM>
To: IDMS-L@LISTSERV.IUASSN.COM
Sent: Tue, 21 Sep 2010 14:42:29 -0000 (UTC)
Subject: Re: question on QUED

Where do all these BFOR images come from that you're referring to? QUED reads a record, deletes a record, maybe. At most there are two images. At best, only one BFOR image per record read. It's a sequential sweep, so no repetition of records read. So there is nothing to condense. No?

Lutz Petzold
DBSS DB2 LUW/IDMS Support
401-782-2265
Page 860 366 0865 or Telalert

Outcomes