[5818] in Hotline Meeting
bloom-picayne has medium (hard?) errors
daemon@ATHENA.MIT.EDU (Marc Horowitz)
Mon Aug 12 04:22:12 1991
To: hotline@ATHENA.MIT.EDU
Cc: usenet@ATHENA.MIT.EDU, sipb-staff@ATHENA.MIT.EDU, cfyi@ATHENA.MIT.EDU
Reply-To: Marc Horowitz <marc@MIT.EDU>
Date: Mon, 12 Aug 91 04:21:03 EDT
From: Marc Horowitz <marc@ATHENA.MIT.EDU>
--- Short form: there will be no news for awhile. ---
Earlier today, we noticed that reading the news history file on
picayune gave an I/O error. rzdisk flagged many (> 50) bad blocks, so
we tried to reassign them. rzdisk didn't like that:
Reassigning bad block (203292) on device (/dev/rrz1c).
*** ERROR OCCURRED DURING COMMAND (Reassign Block) ***
(/dev/rrz1c) error: sense key (0x3): medium error.
(/dev/rrz1c) error: error code (0x15): seek error.
(/dev/rrz1c) error: sense = f0 0 3 ff ff ff e4 c 0 0 0 0 15 0 0 0 0 0
The kernel never logged a thing, with the disk mounted or not. Before
we could copy critical stuff off and shut down the machine, it froze.
Currently, it won't finish ROM test 0 (^C will bring up the rom
monitor), and if I try to single-user boot, the kernel hangs trying to
config the bad disk.
The machine is currently powered down. (It is the humble opinion of
the author that this disk has gone to meet its maker.)
Hotline info:
bloom-picayune is the decstation in the sipb corner of the w20 machine
room (w20-575a). The bad partition is on an rz55. The serial number
is ab90703sjf. Running news is impossible without this disk, so this
should be fixed as soon as possible. I can already see the olc spike
:-) Thanks.
Marc
