Summary: Help with disquieting error messages and a bad disk block

jensen jensen at slyrf.dkuug.dk
Tue Aug 28 04:13:37 AEST 1990


I have received helpful explanations and suggestions regarding almost all
questions I had listed in my original posting on the above subject.  Below
I include a condensation of the answers to each question.

>We get the following two messages:

>     le0: WARNING: if_snd full

Might be attributable to network problems such as broadcast storms.

>     zs0: silo overflow  [or:] zs1: silo overflow

Nobody commented on this one -- I take that to mean that I shouldn't loose
sleep at nights over it.

>and recently this gem:

>     itrunc: /usr/71999 new size = 0, blocks = -16

Opinions vary here as to the severity of this one.  cs at eng.sun.com takes
the dimmest view:

=     This one's serious.  Are you still running some flavor of 4.0* instead
=of 4.1?  [yes] If so, a patch is available.  [...]
=If you're seeing this on 4.1 (where it's supposed to be fixed), report a high
=severity bug today.  Really.  It's your file system we're talking about here.

I'm getting in contact with my local Sun support organisation.

>Recently one of the Suns began reporting a bad diskblock in the following
>manner:

>       sd0:  warning, abs. block 37586 has failed 14 times 
>       [etc]

>After the kernel told me that the block "needed mapping" I did a man(1) -k
>on `map' and `mapping', but it didn't turn up anything useful as far as I
>could see.

The solution is to use the format(8S) (/usr/etc/format) and select the
Repair-command.  Reboot afterwards as the problem was in the swap area.
Documentation is in the section "Repairing a Defective Sector" in Chapter
10 (Maintaining Disks with format) in the System Administration manual.

I did this, and it seemed painless.  I would like to thank the people that
sent me mail in answer to my questions.  They were:

  cs at eng.sun.com		(Carl Smith)
  del at mlb.semi.harris.com	(Don "Truck" Lewis)
  levy%ihcgf%tellab5 at relay.eu.net
    [ sorry, but that was all that was left of the address ]
  peterg at murphy.com		(Peter Gutmann) 
  trinkle at cs.purdue.edu		(Daniel Trinkle)

Regards,

(jensen at dkuug.dk)



More information about the Comp.sys.sun mailing list