"ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU> wrote on 2005-02-26
04:35:49:
> Andy,
> I just had a quick look at the 1st few pages of the v4.0 of the WinPE
> document and found the following:
>
>
> 1) Doesn't state that the machine you're taking a backup of must be
> already registered as a node in TSM.
The field guide is not intended as a primer on the product, but is an
adjunct to the formal product documentation (in which the basics are
already covered). In particular, the first page after the table of
contents states the following regarding the target audience:
"This field guide is intended for storage administrators and skilled
technical support personnel who need to recover Windows 2000, Windows XP
and Windows 2003 systems from a catastrophic system or hardware failure.
The guide assumes technical proficiency with installation, configuration,
and use of Windows systems with the Tivoli Storage Manager (TSM)
Backup-Archive client."
It a reasonable expectation for such an audience to understand the need to
have a registered node; or, if they forgot to register one, to know what
to do when the error message appears telling them that the node is not
registered.
> 2) States that you have to have two volumes available to carryout a
> "online" backup.
As you subsequently noted, this document was released before availability
of 5.3. It is certainly reasonable that any subsequent update to this
document should include a modification of this statement to account for
the lifting of the restriction in TSM client version 5.3.
> 3) diskpart: When you go to the link in the doc about diskpart
> (http://www.microsoft.com/library/errorpages/smarterror.aspx?404;http://
> www.microsoft.com/resources/documentation/windows/xp/all/prodocs/en-us/d
> iskpart.mspx)
> you get the following msg:
>
> We're sorry, but there is no Microsoft.com Web page that matches your
> entry. It is possible you typed the address incorrectly, or the page may
> no longer exist. You may wish to try another entry or choose from the
> links below, which we hope will help you find what you're looking for.
It is an unfortunate fact that URLs are not necessarily static. If they
change over time, then despite our best intentions to provide direct
links, there will almost always be a lag from the time the URL changes to
when we can identify the change and make the corresponding correction.
Having said that, I do not think such is the case here. The rev 4 guide
contains the following link, which works for me:
http://www.microsoft.com/resources/documentation/windows/xp/all/proddocs/en-us/diskpart.mspx
Noting the differences between this link and the one you cited, you will
see that the one I cite has two letter 'd's in the "proddocs" component of
the URL.
> It might be better if in the document there was an explanation relating
> to the fact that not all versions of Diskpart work on all versions of
> Windows servers.
I will feed this back to the authors of the field guide.
> My understanding is that documentation should take into consideration
> that not everyone is an expert, and indeed there are a lot of people
> using TSM that aren't, including myself.
In order for the guide to not grow to an unwieldy size (knowing that you
need a registered node is only one of many potential knowledge gaps), we
have to start somewhere with an assumption about the base knowledge of the
user. I realize that the field guide has not met your expectations, but to
be fair to the field guide in this regard, it does take this into
consideration by disclosing the expected technical expertise of the target
audience, as I mentioned above.
> Now if someone picking up this
> document and following it to the letter, is going to run into problem
> after problem. Presumably people aren't going to try this for the first
> time on a production server, so it may not already be a node in TSM. A
> small thing but the 1st of many issues you'd run into following this
> documentation.
Even assuming a high level of proficiency on the part of the user, it is
not possible to anticipate every possible pitfall. Testing of *any* new
system should occur before putting that system into production. And
[periodic] testing of disaster recovery procedures should be a given.
As I said in an earlier post, the procedures in the field guide have been
tested by the authors, so they are not entirely hypothetical. But that
does not mean problems cannot be discovered after publication, be they
errors we should have caught in advance or unanticipated problems related
to the particular environment in which the procedures are used.
Notwithstanding the problems you have encountered in trying to get your
feedback through, we welcome any feedback that helps us correct glaring
errors or omissions in the documentation, or otherwise improves the
quality of the documentation.
Regards,
Andy
|
