SUMMARY: INN / fclose / IO Error writing history file

From: News Master <news_at_althea.ucs.uwplatt.edu>
Date: Mon, 4 Aug 1997 21:04:11 -0500 (CDT)

Well thanks for everybody's help, but we were all wrong this time. This
was a tricky one!

Original Question:
I have been running INN 1.5.1 for some time now with no problems. The
only thing that happened recently is that we upgrade to DU 4.0B about 2
months ago.

However, this week I have been getting this error sequence:

Jul 29 02:33:33 althea innd: SERVER cant write history I/O error
Jul 29 02:33:33 althea innd: SERVER cant fclose history I/O error
Jul 29 02:33:33 althea innd: SERVER throttle I/O error writing history
file
-- throttling
Jul 29 02:33:33 althea innd: SERVER cant write history
<cancel.33DA3DC2.1DBE_at_dialup.zaporizhzhe.ua> I/O error
Jul 29 02:33:33 althea innd: SERVER cant write history I/O error
Jul 29 02:33:33 althea innd: SERVER cant fclose history I/O error
Jul 29 02:33:33 althea innd: SERVER throttled I/O error writing history
file
--throttling
Jul 29 02:33:33 althea innd: SERVER throttle I/O error writing history
file --
throttling
Jul 29 02:33:33 althea innd: SERVER cant write history
<cancel.33DA3DC2.1DBE_at_dialup.zaporizhzhe.ua> I/O error

Basically it looks like there is something wrong with the history file.
So I did a makehistory on it, to make a new one. Started things up
again, and it puked just like this..

So, anybody have any clues?

--Brett M. Thorson
  News Boy

============================================================================
      _-THE SCOOP-_

When Digital issues their new patches and versions I guess they don't
clean up what their patches were supposed to prevent. I don't keep up
with the fixes (And what they do, I barely have enough time to install
the darn things). I am guessing that 4.0B made some fixes to ADVFS. But
it didn't clean up the errors that the previous version made. And I
guess when you install new versions, things start to get messy.

The errors I had indeed did look like a full disk from the logs above.
But I knew that wasn't the problem with over 3.5 gig free. So I ran
verify on the drives. That puked trying to do something with the mcells
while on the 3 drive in the fileset. So then I decided that the drive
was probably bad. I removed it from the fileset and ran verify again.

Well like is usually the case, you kill one bug and 10 more come out to
defend it. Verify ran further but gave some really weird errors with
something to do with headers in the wrong place.

Time to call digital again. I guess when these things happen the best
thing to do is to back up all the data, blow away the fileset and start
over. Another option (the one I will be trying) is to remove and install
each drive fromn the file set. I guess this is suppose to rewrite the
bitmap file on each drive and make the world a happier place.

So in the end, I still don't know what caused it, and I am still praying
that this will work.

--Brett M. Thorson
  Probably the only person on the list making $7.00/hour for this.
  Unix Systems Admin _at_ University Wisconsin Platteville & Student Pilot
Received on Tue Aug 05 1997 - 04:20:59 NZST

This archive was generated by hypermail 2.4.0 : Wed Nov 08 2023 - 11:53:36 NZDT