jazzman on 10 Dec 2007 01:46:43 -0000


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: [PLUG] More SCSI Weirdness


As far as I know it's terminated at both ends. The card reports auto 
termination and I have an active terminator on the external drive case.

On Sun, 9 Dec 2007, Eric wrote:

> Active termination is better than passive but both exist.
> 
> Here's what I know is a nutshell:
> 
> Both ends of the bus must be terminated but no other device in the bus should be
> terminated.
> 
> Example 1:
> 
>    T---------+---------+----------+-------+-------T
>   SCSI     drive     drive      drive           cable end
>  adapter
> 
> Alternatively, last drive has termination set on:
> 
>    T---------+---------+--------+-------+-------T
>   SCSI     drive              drive          drive
>  adapter                                   on cable end
> 
> (+) is connector
> (T) indicates termination point.
> 
> Incorrect or missing termination causes "ringing" on the bus and might cause
> this kind of error.
> 
> Eric
> 
> jazzman@exdomain.org wrote:
> > Both the card and case claim auto termination, but I also have a 
> > terminator (marked active terminator... is there a "passive" kind?) that 
> > I've tried... same behavior...
> > 
> >  On Sun, 9 Dec 2007, Eric wrote:
> > 
> >> Hey, I'm no SCSI expert but this has a certain "oddness" that might be explained
> >> by improper SCSI bus termination.
> >>
> >> Have you checked that?
> >>
> >>
> >> Eric
> >>
> >>
> >>
> >> jazzman@exdomain.org wrote:
> >>> Ok, so I think I may have some more info on the weird drive behavior I was 
> >>> seeing with my scsi array.
> >>>
> >>> I tried the drive array on a second machine, different OS (Fedora 5) but 
> >>> the machine behaves similarly (doesn't always detect ALL of the drives on 
> >>> each boot... the scsi drives randomly pop in and out of existence on each 
> >>> reboot). The only thing I did notice is that each machine has the SAME 
> >>> scsi card (tekram DC395x) and that when the drives disappear on the FC5 
> >>> machine I get the following error message flooding the logs:
> >>>
> >>> dc395x: QUEUE_FULL for dev (0X-0)
> >>>
> >>> where X is the scsi ID of the first "missing" drive.
> >>>
> >>> Does anyone know anything good/bad about this card? It was cheap when I 
> >>> bought it, but I'm hesitant to go buying new cards/cables and such without 
> >>> knowing for sure. If this is a "could be anything" type of problem I have 
> >>> a feeling the scsi case and it's drives end up on ebay befoer long...
> >>>
> >>> any thoughts?
> >>>
> >>
> > 
> > ___________________________________________________________________________
> > Philadelphia Linux Users Group         --        http://www.phillylinux.org
> > Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce
> > General Discussion  --   http://lists.phillylinux.org/mailman/listinfo/plug
> 
> 
> 

___________________________________________________________________________
Philadelphia Linux Users Group         --        http://www.phillylinux.org
Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce
General Discussion  --   http://lists.phillylinux.org/mailman/listinfo/plug