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 Mike Lewis on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Userenv errors....

Status
Not open for further replies.

Highland

MIS
Jul 4, 2001
451
GB
We've recently been getting logon problems in one of our farms....the farm has four Metaframe 1.8 servers load balanced, basically we start to see the userenv error (event id 1000) appearing in the application log when the user count is at around 58 per server (LB is set to use user count as its primary factor)....useenv errors are apparently caused by "lack of server resources - such as registry size, pagefile size (or lack of) etc...however we have spent several weeks "tuning" the servers and none of the above are anywhere near their limit....before we load balanced the servers we did have user counts of 80+, any ideas....????
 
- How often do you reboot?
- Check licensing, both MS Terminal Services and Citrix
- Look into the Userenv events on
Other things you can do include: cleaning old .TMP files, running a defrag with DiskKeeper, and removing obsolete user profiles from "My Computer" properties.

hope this helps,
-gbiello
 
Reboot every night, Licensing is 65+ per server, checked eventid.net, and other suggestions are carried out regularly.....(we are set-up to logon with Roaming Profiles...)...policy set to clear these...

Userenv seems to be a resource issue, but our registry size is approx. 110-120 MB (max - 156Mb), pagefile is set-up to correspond to registry size, System partition has plenty of free-space......

If some of you could maybe respond with how many concurrent users you have logging onto NT/Metaframe 1.8, just wondering if we have an "average" number of users on our servers or if some of you are getting much more per server...????

Thanks.....
 
At the last site I worked at that implemented 1.8, we had 80 users per server on 6 x DELL PowerEdge 6300's. These servers had 4 x Xeon 450 processors, 4Gb RAM and 2 x 9.1 Gb Fast SCSI disks in a mirrored configuration on a hardware PERC controller.

If you do a query server /load, do the resulting figures give any clues? Load Balancing is a bit of a black art, IME, and never quite responds the way you think it's going to.

Your Registry's a bit on the large size - are your users saving data into their profiles?

Also, have you checked the "Advanced Concepts" pdf on Citrix's site? That contains some good tuning tips.

Hope this helps CitrixEngineer@yahoo.co.uk
 
No there should be nothing saved with the profile...??!!

Checked the Query Server a few times and it's been within the "normal" range, the load balancing remains consistent across each server in each farm...

I didn't think the registry size was excessive (I was comparing 110Mb with the max permitted of 156Mb...????)

Yeah, we have had 80ish users on our servers too (pre LB), the only change was the introduction of Load Balancing, were the servers you mentioned running LB or supporting direct connections...???

thanks.....
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top