FUJ00059070
FUJ00059070
HORIZON KEL rcoleman1253j (Version 2)
Type: Information
Status: Authorised
Title: "The device, \Device\ScsiPort? did not respond within the timeout period.’
Raised: by Richard Coleman on 09/11/1999
Last updated: by Aprime on 29/06/2006
Release: B13
System product: Counter
Keywords: scsi, port, disc, disk scsiport”
Status: Authorised
Visi Medium
Peak: PC32983
Tis: 9911090001
Symptoms
Events seen on a counter:<br><br>Sub_source=atapi<br>Sub_origin=15<br>Event text: <br>'The device, \Device\ScsiPort? did not respond
within the timeout period.’<br>and/or<br>'The device, \Device\ScsiPort1, is not ready for access
yet.'<br><br>and/or<br><br>Sub_source=Disk<br>Sub_Origin=11<br>Event Text: <br>'The driver detected a controller error on
\Device\Harddisk<N>\Partition <X>'<br>(where <N> is 0 or 1 corresponding to the ScsiPort event and as described below and <X> is the
dartition of the drive
Problem
NT has detected a fault on a disc drive or IDE controller.<br>ScsiPort0 = Primary IDE controller = Main hard disc (Harddisk0)<br>ScsiPort =
Secondard IDE controller = Mirror disc for SCOs (Harddisk1)(including mobiles)
Solution - Helpdesk
First check for other events which may have lead to this error, such as bad block or corrupt storage unit events. If such events exist follow
the appropriate KELs for those.<br><br>To test if the error is simply a result of processor delay due to intensive processing by the counter at
the time of the event, reboot counter.<br>If the error recurrs this is indication of a more serious fault either with the computers motherboard
or the affected drive. <br>Therefore, if the message reappears then send an engineer to replace the part at fault.<br><br>For a B-base or C-
base unit, the base unit should be swapped out. The engineer should contact SMC for setting of the KMS (gateway) or clearance of LAN
(slave). SMC should check events to ensure there are no messagestore corruptions and copies should be deleted from all counters if these
exist.<br><br>For a SCO the counter should be synchronised before the engineer begins the swapout and the PM should be requested to print
a balance or office snapshot before an after the work is carried out, to ensure that transactions are not lost.<br>On a SCO. <br>If the
problem is <kel><b>only ScsiPort0</b> <html>, the existing mirrordisk should be used in a new base unit.<br>If the problem is <htmi><b>only
ScsiPort1 </b> <html>, a new mirrordisk in the old baseunit should be tried first. If the problem is not resolved then the old mirrordisk should be
put in a new base unit and this tried.<br>If problems exist for <html><b>both ScsiPort0 and ScsiPort1 </b> <htmi>, swap the whole
unit. <br> <br><html><b>BEFORE</b> the base unit is replaced (at a SCO) the engineer <b>MUST</b> contact the SMC to synchronize the
messagestores.<br>See KEL PCarroll919Z for further details. </kel>
Evidence
For the engineer calls, be careful to check for SCOs whether or not the site is a mobile. If so the call MUST include the storage address of the
mobile where the engineer will need to attend.<br>Access times for the appropriate location required.