Problems with Morning Cashflow process

Sally and Brent,
The write-up below is based upon discussions Jonathan Le and Kevin Lee had
with Steve Stock and Jeremy Wong.
Jonathan and Kevin are working on a new cash flow process for us which should
help to insulate us from the current intra-night process for the ERMS and
EnPower cash flows.
Sheila
---------------------- Forwarded by Sheila Glover/HOU/ECT on 09/15/2000 08:32
AM ---------------------------

Jonathan Le
09/14/2000 04:40 PM
To: Sheila Glover/HOU/ECT@ECT
cc:
Subject: Problems with Morning Cashflow process

fyi
---------------------- Forwarded by Jonathan Le/HOU/ECT on 09/14/2000 04:39
PM ---------------------------

Kevin Lee@ENRON
09/14/2000 11:27 AM
To: Jonathan Le/HOU/ECT@ECT
cc:
Subject: Problems with Morning Cashflow process

Hi Jonathan,

Listed below are the details of each days problems:

Fri, Sep 8,2000
AutoRho failed at 3:15am, uploaded only 8 books into Infinity.
It failed on the step in AutoRho where it runs the query to get all the books
for the Infinity USD portfolio. Jeremy does not know exactly what caused it
but rerunning AutoRho seemed to work.
Recovery used the backup file and was on time.

Tue, Sep 12,2000
Similar problem to Friday, but ENPOWER also failed. The manenpower.txt file
had TNS errors in it. AutoRho was not able to kick off the ENPOWER script to
get the cashflows. Rerunning fixed the problem.
AutoRho failed at 3:15am, uploaded only 9 books into Infinity.
It failed on the step in AutoRho where it runs the query to get all the books
for the Infinity USD portfolio. Jeremy does not know exactly what caused it
but rerunning AutoRho seemed to work.
Recovery used the file that Jeremy provided in the rerun, they were
approximately 10 minutes late.

Wed, Sep 13,2000
ENPOWER failed creating a header ONLY manenpower.txt file. There was no
cashflows in the file.
Jeremy confirmed the Autorho process was fine.
The problem was the ENPOWER databse failed (hardrive failure) and they
immediately went to a backup. But on the backup database, NOT all the tables
are replicated. The cashlow table with the ENPOWER data we needed was not in
the backup database resulting in a file with zero flows!
This problem has been fixed by the enpower people.
Recovery included using the previous day's enpower flows. They were on time.

Thurs, Sep 14,2000
AutoRho at 3:15 am and 11:00 pm was hung. Enpower also was not run since
AutoRho failed.
Reason was because the UNIX NFS file system "solar" was full As a result,
our directory /worksheets2, and also /prod/egs-home is mounted on solar but
could not write anything to it. Supposily the SA are suppose to have been
paged when ever a file system gets close to 100% (can you please confirm
this?). The Unix NFS was fixed by 5:00 am. AutoRho was rerun and was fine.
Recovery included Jeremy's file. They were 1 hour late (partly due to other
reasons on MO side, not system related).

Thanks,
Kevin Lee