Link to home
Start Free TrialLog in
Avatar of andy0789
andy0789

asked on

Slow/Incomplete Print Jobs Over 56k Frame Relay

Hey,

We are running Windows 2k3 with Citrix Metaframe XP. We have remote locations around Indiana. Some, not all, of these locations have problems printing. Our printers are a mix of HP Laserjet 4's, 5's, 2300's, and 4050's. They are all networked with JetDirect boxes and put onto a print server in the corporate office. So when someone prints from citrix it goes to the printer server (in the same room) then out to the printer at the remote location.

These locations are connected to us by AT&T 56k Frame Relay with 35k Guarenteed Rate. Now 4 of these locations are having printing problems. They are not the largest locations. The print job goes into the queue and gets upto however many bytes then stops and eventually errors out. Then the queue will try again to print the document and MAYBE after awhile (15min) the job will go through.

VNC seems to work fine over these links to the locations. Citrix sometimes they have problems connecting to. At the locations they either have an HP 10/100mbps Switch or a 10mbps Hub with a Cisco 1700 Router which seems to make no difference.

I am thinking is has something to do with the connection but ping rates seem to be resonable (60 - 90ms) and steady.

If anyone has any ideas on howto troubleshoot this please help!

Thanks,
Andy
Avatar of Big5250
Big5250

Small world, I'm also in Indiana and a Large Citrix Farm user.  We use the HP Print Spoolers to offload our Print jobs.  We have many remote facilities with 56k frame.  Quick question, how many users at the remote sites?  How large are the print jobs on average?  Are your remote switches manageable?  If so, reset the JetDirects to do Full 100 through the Web GUI, and change the settings on the Switch as well.

When you can print, at the same time try to Telnet to the LPR port (515) on the remote JetDirect, see if you get a response.
Avatar of andy0789

ASKER

Hey, at the facilties that are having the problem there is no more than 7 nodes. Most are WYSE terminals and few are desktops. The print jobs are usually around 56kb. Some do range upto 3mb though but those are not to common. Yes the remote switches are managable.

I will try what you said

Thanks!
Andy
Too funny, my name is Andy Too!
Hey Andy, I forgot the JetDirect Boxes are only 10mbps not 100. I dont think that really is the problem. I mostly think its the 56k Frame Relay connection. Seems to be in one area of the state that is having the problem (Around Batesville/Dillsboro). Have you experienced any problems like that in your experience in Indiana? The facilties that are having the problem have at most 4 printers.

At times while watching the print queue on one of them it goes through about 4 40kb documents fine then half way through the 5th it just stops and we get "Error - Printing"

The printers are setup to go over port 9100 (JetDirect). Is there a different port we could try or just a different way of trying it with what we have?

Let me know,
Thanks!!
We use LPR printing from our Print Servers, which utilizes port 515.  This seems to work for us.  No real issues to speak of.  What kind of appliances are you using to offload your printer processing?
Not sure exactly what you mean by offloading our printer processing But we have all 170x JetDirect Boxes. Those printers are setup on a windows 2003 server. Then of course we have 3 metaframe xp servers.
The print rendering process on the servers can be quite a load, so instead of having this process done by the Citrix boxes, we offloaded the job to print appliances.  It actually does the print processing and takes the load off of the Citrix servers.

As well, you may want to look into some packet shaping products like Packeteer:

http://www.packeteer.com/prod-sol/products/packetshaper.cfm

This gives you the ability to not overload your 56k lines with print traffic which will slow down your ICA sessions.
It is still odd though because we have larger facilities with more nodes but on the same line speed and they have no problem with printing. I know the facilities that are having the problems are not all on the same at&t circuit because I have asked them thinking something might be wrong with just the location. I have noticed though that when someone does print and it does decide to go through the ping response time shoots upto 1600+ms while the document is being spooled just for those facilities. I think AT&T is playing with us and doesnt want to admit that the line speed at those facilities is a lot lower than the ones at the larger facilities that work fine.
Again, the packet shaping may help alleviate some of your problems.  But, verify that your CIR is what they are saying.  Try running MRTG to baseline your line usage, this may be a starting point.
I have solved the problem!!! It was the IOS on the routers believe it or not. We noticed that the facilites that were having the problems have a cisco 1720 router and the facilities that were not have a 1750 router. So what we did was upgrade the IOS on the 1720 routers to the same IOS that are on the 1750 routers and everything seems to be working smoothly now..

FYI:
Cisco 1720 IOS - 12.0(4)T
Cisco 1750 IOS - 12.0(5)XQ

Never would have guessed that one.

Thanks for you help!
ASKER CERTIFIED SOLUTION
Avatar of Big5250
Big5250

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial