This brings back memories pre IBMLINK.
I vaguely remember an IBM product (INFO-MVS) which was (if memory
serves me) was a single (??) reel 6250 tape sent monthly by IBM.
I was the one designated to load the tape into an VSAM dataset. The
team was 8 (or was it 9) sysprogs and all of us disliked calling into
IBM to look up problems so there was a lot of push to get the tape
loaded ASAP.
One day the delivery of the tape was delayed because our in house
delivery system broke down. ll we knew after some frantic phone calls
it was somewhere between the 3rd and 18th floor. The group split up
and went to every floor in the the system trying to find the tape. We
found out that it was still on the internal lift system it was
unavailable. We all ended up on the 3rd floor pestering the
technician to get the system back up and running. After 15 minutes
we left the trainee to pester the technician. The building was rather
new and I know things break down but the tape was deemed rather
critical. The rest of us had to go back and share one microfiche
reader and the infamous 1-800 (it could have been a local number but
its immaterial) number and we opened so many problems with IBM that
day that I think IBM sort of wondered if the building burned down.
The INFO MVS dataset was really critical to our group and having to
to sit there at a fiche reader was not exactly appealing to any of us.
I am really shocked that IBM is apparently treating the outages like
they are not worth talking about. Although with IPL failures you
would not have been able to us INFO/MVS you still have had a fall
back of the 1-800 number or groan even fiche. To me the SIS and the
other (maybe not all) features of IBMLINK would be similar to a
airline reservation system. Does anyone have another comparison they
would prefer to use?
Ed
On May 21, 2012, at 9:09 PM, Clark Morris wrote:
On 21 May 2012 13:27:19 -0700, in bit.listserv.ibm-main Sam Knutson
wrote:
My consideration is many applications in IBMLink don't seem to
merit the cost of true continuous operations. I think it is
unreasonable to ask for 24/7/365(6), for all of IBMLink.
As someone who has been away from the actual trouble shooting for over
20 years (I still remember my site's customer number) so I don't know
what the critical functions are or what all is included in IBMLINK.
From your posting below, I can see that IBMLINK covers a number of
areas. I wonder how Amazon or LL Bean define what has to have
virtually total availability. Similar questions arise for Apple and
Microsoft. I would definitely think that problem databases require it
and that frozen ones should be available for reading even when they
are unavailable for update due to maintenance. I'm wondering if this
is sort of like EREP was back when I was dealing with it, a necessary
stepchild that seemed to not have any one group responsible for it.
Possibly it would be worthwhile to identify all of the functions in
IBMLINK and identify which ones need the availability. It sounds like
something SHARE is well suited for. Also we need to identify the
customer and IBM costs incurred by lack of availability both soft and
hard.
Have any of the people on this list found that the problems with
IBMLINK affect the perception of the z series within their
organizations and of IBM's ability to come up with appropriate
products for business needs?
Clark Morris
We need high availability for applications that enable us to
effectively support our own high availability systems. This seems
to be operational support for our System z hardware and software,
reporting problems electronically, researching problem databases
for known solutions and open APARs, and retrieving service.
ResourceLink which we use for hardware support IBM moved to a
higher availability infrastructure some time ago and has made
process changes to make the infrequent outages less problematic
for customers.
SR on the current highly available infrastructure seems to be
filling the need to submit problems electronically. I didn't like
SR as much when I first saw it. The community at SHARE and
customers directly gave IBM very critical feedback. IBM's leaders
listened and delayed the replacement of ETR by SR until many
significant issues had been addressed. There are remaining bugs
and opportunities to improve i.e. the double sign-in requirement
for some use cases. SR is any many ways better than ETR and I now
prefer it i.e. attach small files, multiple users updates on
status of PMR, etc. IBM is doing OK here.
IBMLink SIS is critical to us. This weekend it was fortunate that
IBMLink was up early as we had a planned infrastructure outage and
ran into a problem that we found in five minutes using SIS
probably would have taken hours using phone support at 0400 on a
Sunday. IBMLink SIS or some other tool that is highly available
and lets me search in one place all System z products solutions,
problems, and information is important to us. IBMLink availability
for SIS has not always met our expectations. This weekend for
instance it was scheduled down while we had a major quarterly
infrastructure upgrade. We hit a problem that caused our daily
backups to fail after deploying RSU1203 (OA38632). We were lucky
and IBMLink was up early from its scheduled down time so we were
able to very quickly in a few minutes determine that a fixing PTF
and a workaround was available. I think IBM needs to make
moving the search function improving completeness, usability, and
transition to a hig!
h!
availability infrastructure a priority.
ShopzSeries or an equivalent facility to order and retrieve
service and hold data electronically is the third leg on the stool
for production support. This is an area where IBM has not met
expectations and outages have occurred on a recurring basis with
little or no explanation.
The applications in IBMLink outside that core are conveniences and
should have good availability but probably don't have a valid
business case to be highly available.
IBM is not meeting my expectations in terms of announcing downtime
with sufficient lead time, providing access to a help desk with
current availability information, and minimizing outages.
I think IBM has an opportunity to showcase its own technologies in
its customer faced systems that it seems to be missing.
Best Regards,
Sam Knutson, GEICO
System z Team Leader
mailto:sknut...@geico.com
(office) 301.986.3574
(cell) 301.996.1318
"Think big, act bold, start simple, grow fast..."
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu]
On Behalf Of Thomas Conley
Sent: Sunday, May 20, 2012 9:28 PM
To: IBM-MAIN@bama.ua.edu
Subject: Re: IBMLink outages in 2012
For our purposes, an outage is an outage, scheduled or not. Our
requirement to IBM is clearly 24/7/365(6), for all of IBMLink, SR,
ShopZ, Internet Service Retrieval, etc. Ed's right in that IBM
exacerbates the problem by having the outages during most z/OS
installation's prime time for maintenance. They should move these
outages to 0dark30 on Thursday morning or something.
Regards,
Tom Conley
====================
This email/fax message is for the sole use of the intended
recipient(s) and may contain confidential and privileged information.
Any unauthorized review, use, disclosure or distribution of this
email/fax is prohibited. If you are not the intended recipient,
please
destroy all paper and electronic copies of the original message.
---------------------------------------------------------------------
-
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
