"Javier Martinez" <> wrote in message
news:%
....
> Hi
>
> When I try donwload critical updates I get this error 0x80070057 the other
> components: drivers and Windows 64 elements gone well.
>
> The log is the folowing:
....
> 2005-09-01 18:31:53 16:31:53 Success IUENGINE Querying software
> update catalog from
> https://v4.windowsupdate.microsoft.c...etmanifest.asp
> 2005-09-01 18:32:02 16:32:02 Error IUENGINE loadXML: line 1,
> pos 3566, A semi colon character was expected.
> (Error 0xC00CE50D)
This error seems more important to me than the one you are noting.
If the file is normally valid I would suspect a problem with your link.
However if the protocol is https: (which would imply that the data
is encrypted) that would make that case less likely. E.g. then if there
was a problem with the link you would probably being seeing a different
symptom.
I don't think that the error is necesarily occurring in the previously listed URL.
Hence we can't say what protocol is being used to download the problem file.
The best way to tell would be to capture a packet trace of this
(difficult to do if the protocol is https

but at least a regular packet trace
would tell us that much.
In any case you could try to find the file which contains that error
for more clues. E.g. try finding all files which have a date modified
on the day of the update attempt, sort them by that field (thus ordering
them by timestamp and thus hopefully grouping all files involved in the
update so they can be opened and inspected. You might be lucky
and be able as well to restrict your search to files which contained
that string loadXML but don't be surprised if that search does not work.
For example, none of the standard file search tools can show you matches
which involve doublebyte characters (such as Unicode) and some can't
even search within container files such as .cab or .zip files.
For a more precise refinement of the above you could use the freeware
utility FileMon. E.g. it would show you the files actually being used,
not just what is left after the update attempt is over. Another advantage
to using FileMon would be that it can show you more exact timestamps
for the writes of the above log messages. (The writes have to be inferred
from the coarse timestamps used in the log and the length of the record
being written but it is a very useful way of providing a clear context to the
rest of the FileMon trace.)
HTH
Robert Aldwinckle
---