[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference netcad::hub_mgnt

Title:DEChub/HUBwatch/PROBEwatch CONFERENCE
Notice:Firmware -2, Doc -3, Power -4, HW kits -5, firm load -6&7
Moderator:NETCAD::COLELLADT
Created:Wed Nov 13 1991
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:4455
Total number of notes:16761

2418.0. "DB90FL DR90FL jabber errors" by GIDDAY::MORETTI (Networks Integration, Oz) Fri Jun 23 1995 02:17

    
    
    I have been fault-finding jabber errors on a small hub-based network
    for the last 2 weeks and have had a continually moving target.
    
    Following is the topology :
    
    hubA : dechub90
    	   DR90FL
    	   DR90T+
    
    hubB : dechub90
    	   DB90FL
    	   DR90T+
    
    hubC : dechub90
    	   DB90FL
    	   DR90T+
    
    At hubA we have one XL server which is the only file server on the net.
    There are approx 20 PCs at each of the other hubs all accessing the
    fileserver at hubA.
    
    The fibre runs from A-B and A-C are in the order of 100-200metres.
    
    I can see no reason for the DB90FLs as all PCs are accessing the
    Fileserver at hubA but here is the problem :
    
    I run a batch copy file procedure on a PC at hubB to the fileserver at
    hubA and we get 300 jabber errors per minute at the hubA end, none are
    appearing at the hubB end due to the bridge separating the segments.
    
    The same happened when I did this on hubC to hubA.
    
    Therefore I put a bridge in hubA and ran the copy test again from hubA
    and hubB (DB90FL - DB90FL) which gave me no jabber errors so I
    concluded that the DR90FL in hubA was causing the errors.
    
    I changed the DEFMR-MA repeater twice and still have the same result
    and I have noticed in TIMA that these repeaters have problems with
    large packets (the header said this was 18th NOV 94 and I am wondering
    if we have fixed this problem?
    
    The revision level of the DEFMR that I have used are all E01 and the OS
    of the file server is NT using a DE435 NIC card version 2.0 drivers.
    
    I have changed the DE435 card as this was also a possibility at one
    stage but have not installed the interim driver version 2.1C as the
    jabbers are normally generated at the lower stack levels.
    
    I am currently in the process of changing the DB90FLs for DR90FAs and
    will see what this will do but still need to know if the long packets
    problem has been fixed .
    
    john moretti
    nis bbo
T.RTitleUserPersonal
Name
DateLines
2418.1Problem locatedGIDDAY::MORETTINetworks Integration, OzSun Jul 02 1995 23:1117
    
    
    Have found the root of all the jabbers :
    
    	DECrepeater90T+
    
    Whenever you have a fileserver on one 90T+ port and access the server
    from a PC on other 90T+s you will generate large amounts of jabbers
    under heavy load ie network utilisation >20%
    
    I changed the 90T+ for a 90T and this problem cleared up, it was
    exacerbated by the SMC cards in the PCs but was still evident when
    DE202 cards were substituted.
    
    This problem has been escalated to engineering.
    
    John
2418.290T-16 do not have this problem.GIDDAY::MORETTIRSSG Brisbane 617-3223-3741Wed Dec 20 1995 01:4910
    
    Have finally come got a solution to the jabber problem which is to
    replace the 90T+ digital phase loop lock orientated modules with the
    new 90T-16 repeaters which are running analogue circuitry.
    
    The nnew repeaters are also cheaper and provide more ports per dollar.
    
    A nice win/win situation.
    
    John