Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations Chris Miller on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

WFICA32.EXE takes 100% CPU

Status
Not open for further replies.

Pritch5000

Technical User
Aug 15, 2000
137
GB
I have a customer who is experiencing SQL database errors. Thru investigating this I have found that they have two issues at the client end.

1. Inconsistent DNS resolution issues, i.e. can ping name, but not IP on some machines and vice-versa on other machines. But there seems to be no link between that and the fact that some machines have different DNS suffixes.

2. WFICA32.EXE runs at 99% CPU but the session seems to function generally OK - that is until it freezes! along with many, if not all, other sessions. Then they get errors like 'ODBC timeout' and miscellaneous SQL errors.

Any help you can offer would be very much appreciated.
I've got loads more investigation info to write if it helps....

cheers
pritch5000

 
Is it possible that they connect to the sql server but that the lines that they are using are being dropped by a router or a hub when idle ? I've seen database apps (clients) that crash when the line is dropped because they can not find the database server fast enough.

wfica32 shouldn't be 99%. However I've seen cases where the hardware was the problem. Have you tried to reïnstall the citrix client ?

DNS issues can become a problem if you setup clients to use DNS names instead of IP numbers.
 
Thanks for your reply, historically, people dont seem to reply to my threads! :(

We have upgraded the client to another one that we use, and get the same thing. What we havent done is remove/reinst which I gather we should do. We are planing to deploy the latest from the server so remove/reinst would be difficult to do.

We arent using names but DNS is being checked out.

The customer did put in new clone hardware around the time that problems started and we are planning to test without those in place.

The customer's networks guys are assuring us that the networks fine, they monitor it daily and have no issues (apparently). It runs at 1-2% util.

Thanks again... any more thoughts?
 
On the DNS issue, if there are resolution issues, then maybe;

- The cache in one or more of the routers is out of sync
- The lookup and reverse lookup tables on the DNS servers are out of sync
- The client DNS settings are incorrect

I'd guess that maybe wfica32 is hogging CPU as a result of a database operation, from the way you describe it - is the database attempting to access client resources in any way?

...I'm sure I've replied to your posts in the past... :) CitrixEngineer@yahoo.co.uk
 
Thanks CE for your response [2thumbsup]

I am going back to site on Monday armed with the latest 1050 client and MDAC 2.6 SP2 - at the moment we are into isolating the different parts of the equation - db, server hardware, wan, lan, their dns, and stuff.

The customer assures me that both the WAN and DNS settings are correct.

Our software uses local receipt and report printers, local barcode swipe readers and we use a fat-client app to take photo that get incorporated using the clipboard!! so yes! - but we have this running all over the UK and abroad.

I'll report back when I know some more....

cheers
aspritch@hotmail.com
 
Hi, as said I am reporting in...

We have found out why the WFICA32.EXE is maxing out! We found that our admin user could log on anywhere and the cpu on that PC no longer ran at 100%. We narrowed this down to the fact that our user does not have a COM port redirected.

ALL of the regular users had the following line in their login script, so that they would be able to use thier local swipe readers:

net use com1: \\client\com1:

We configure our software to use this redirected COM port and then the issue is seen, consistently.

We have disabled ALL COM ports and the users have reported that things are better.

So far so good!

Also we have tested this at other customer sites and found this to be consistently true.

So it seems to be a problem with our code. BUT the only thing is our QA team have tested on this and found that this has been true since 'day 1' !! So why then are we only having trouble with it now? after 3 years!! Bizzarre.

aspritch@hotmail.com
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top