|
On Thu, 14 Sep 2000, Keith Barrett wrote:
> Unfortunately, whether it was obvious on the mailing list or not, there was
> a lot of other issues involved. For example; he was offered a refund but
> refused it, then complained here that he was out $2k (which is for the
Kieth, that is bull crap. The refund was offered in the beginning with
**NO TROUBLSHOOTING** done with the problem. And the refund was offered in
response to my question of "If we're not going to recieve assistance on
the problem then why include technical support for the product!"
** Of course we're not going to just take the refund right off the
bat! This is a business! We have deadlines to meet, We have a technology
model that needs to be put in place. We purchased a rather expensive piece
of software to accomplish that mission. IT comes with a technical support
contract to assist us in the matter. OF COURSE we're not going to take the
refund right yet! We haven't even recieved the barest of technical support
which we just paid $2000 for! ***
Not once was I asked OK what set up do you have? How are you building the
config file? What equipment is this running on? These are some pretty
standard questions to ask from a technical support stand point. Why were
they never asked? No one attempted to contact us back regarding the
problem. All contact was done because *I* institued the calls,
because *I* pushed the emails. You guys were content to just let the
problem hang there. Of COURSE we refused the refund inthe
beginning! We wanted the goddamn troubleshooting to take place that
you are so avid about pointing out. "It comes with a one year 24x7
telephone support" We got NONE of this!
OK let's get the goddamn facts straight.
First, we purchased this product from you after speaking with Nathan
Thomas about some concerns. Things like caching of dead routes even when
restarting, having to stop the secondary node in order to correctly get
the primary node online and lvs working, having to reapply ipchains rules
after restarting lvs, lvs installing multiple default routes on start up,
lvs not correctly adding or removing backend nodes from/to the routing
paths. All this was spoken about BEFORE we purchased.
We were told that all these issues worked in the product. We installed the
product according to your documentation and found that either nanny was
failing to pass the lvs.cf file information to pulse or the lvs daemon was
failing to correctly parse it's own configuration file which in turn kept
the ipvsadm rules from being properly created and/or maintained. Unless
lvs creates the rules itself just manually adding the rules does not get
them monitored or controlled.
We spoke quite extensively with technical support with me being told that
they were still too new to the product and that they would ahve to speak
with those responsible for the product (in this case that being you and
members of your HA team). Not to mentionthe fact that you guys lost the
original registration of the product and a member of the technical
support team had to go back and REREGISTER the product! Then we we are
told that the technical support only covers configuration and
administration of the product. When we told you that it's a bug, were
told that we need to get a development contract since bugs are not
covered inthe technical support. OK, THEN we're told by development that
it's a CONFIGURATION issue! OK, then THAT is convered under the original
technical support contract. Nope. It belongs under development, nope it
belongs under technical support. Not once did anyone try to actually
RESOLVE the issue. Over 75% of my conversations with various members at
Red hat revolved around you guys trying to get MORE MONEY out of us.
We never recieved a call back until I called up about the issue. Finally
after posting my config files AS YOU REQUESTED to
bugzilla, I never ONCE recieved an email regarding the problem from you
Kieth WHO ORIGINALLY REQUESTED the cofig file. I had to spend another day
calling Red Hat repeatedly to get ANY type of information on where things
stood. Next thing I know the problem is closed as WORKSFORME. That was the
ONLY indication i had that you guys were even working on the
problem. Finally I call back to technical support looking for
assistance in this and the answers are "Everything is in tha manual" or
"Read the manual" Not a single one of you listened to the fact that I
told you that I had actually yanked the system out of the co-location
facility to work on them here and that we had done a STEP BY STEP set
up of the product. We actually sat here and set up a testbed network
with the actual settings from the book. We gave the systems the IPs
specifically mentioned in the manual. We did everything THREE TIMES to
make sure we weren't goofing the problem. NOT A SINGLE TECH SUPPORT
AGENT LISTENED TO THAT FACT! The same response was given, "It's in the
manual". The entire team back here is sitting back going "No, really?? Are
you sure it's in the manual?"
Not once did YOU call or email saying OK let's walk through this and see
what's wrong. I called and gave technical support a number of steps we had
taken to debug the problem all with the response "We'll pass it on and see
what they say". We told you guys we were not able to pass the -v
switch the man page says to use to enable verbose mode for nanny but no
one paid any attention to that. LITERALLY, mind you. NONE of the
troublshooting steps were taken. This was a core piece of assistance
that we requested since without it, there was no way to tell exactly
where the product was dropping the ball. THEN, You say that I was more
interested in taking a stand than getting the issues resolved. Show me
where we HAVEN'T worked with you or tried to get things answered! You have
a lot of nerve to say that!
> support, not the software). If you read the bugzilla entries, you can also
> see that I tried to work with him,
And anyone reading the entries will see that I GAVE YOU what you
requested.
> There were additional interactions (many not with me) that it is not
> my place to discuss, so I have to accept that it may appear unbalanced.
Yeah the interactions were my phone calls to you guys saying "Where's the
support that comes with this product?" Goddamn it!
GET FOCUSED!! ADDRESS THE DAMN ISSUES I'M PRESENTING! You have walked
around every damn issue I have broached! You go around pointing out "He
has personal issues with redhat". "He was offered a refund but he
refused" "you hate redhat but promote and rhce"
Screw that! Answer the damn questions. How does one pass the -v switch to
nanny to get more verbosity so we can see where things are failing. Is
nanny being called hardcoded from inside pulse or some other spot or a
script? How do we pass the verbose switch to lvs itself so we can seee
what IT'S doing? We can see that the lvs daeemon is parsing the file to
some degree since the NAT IP and the VIP are coming online. It's dropping
the ball on the virtual declaration(s). How do we follow along with it to
find out exactly WHERE it's dropping the ball?
We've been asking these questions all the way along with no reply. Either
we get technical support agents that don't knwo the product and say they
need to ask the HA development team to find this stuff out or tell us we
need to buy a development contract in order to get the problem
resolved? WHY? This is functionality that should be in the base product
and isn't. If it's supposed to work then troubleshoot WHY it's not
working. Quit referring folks back to a manual. Actually TALK to folks
about the problem. Taking a configuration file and running it in a lab
and it working would normally mean you call or email the client back
and say OK it works here, let's see where the differences are so we
can fix this. Who at Red hat did that? None!
> Ignoring the legal possibilities, many of his statements
> false, and he never gave specifics or stating it's just his opinion or
> experience.
Please PLEASE PLEASE take this legal! PLEASE take this to court. Oh
please! please please!
I have tried throughout this whole thing to NOT give the exact details
BECAUSE I didn't want this to get out of hand. But when you sit there and
say the product works fine, that the person involved has personal issues
that extend beyond the scope of the product, that Red Hat has tried to
work with the client and the CLIENT refused assistance, AND state that you
personally tried to assist (ALL OF WHICH IS NOT TRUE) I HAVE to come
forward and bring in the extra information. I'm not going to just sit back
and let Red Hat claim that it's incompetency on the client's part and that
they have no fault or blame in this.
--
David D.W. Downey
Systems Administrator
RHCE, CUA/CLA
Internet Security Specialist
QIXO.com
|
