Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 29 Jan 1999 21:44:38 +1030 (CST)
From:      Kris Kennaway <kkennawa@physics.adelaide.edu.au>
To:        Julian Elischer <julian@whistle.com>
Cc:        Matthew Dillon <dillon@apollo.backplane.com>, hackers@FreeBSD.ORG
Subject:   Re: softupdates on /
Message-ID:  <Pine.OSF.4.05.9901291601380.5100-100000@bragg>
In-Reply-To: <Pine.BSF.3.95.990128165737.11856I-100000@current1.whistle.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 28 Jan 1999, Julian Elischer wrote:

> The eventual aim is to never use NON softupdates mode..
> The system is not supposed to bonb with "filesystem full" when there are
> uncommitted delete operations on the softupdates queues.
> if it does, then there is a bug. It's suposed to kick off some faster 
> reconciliation and hang around for a bit.

Unless I'm mistaking you, this is in fact what happens:

[morden|root] 21:34 ~ df
Filesystem  1K-blocks     Used    Avail Capacity  Mounted on
/dev/wd0s2a     32254    21244     8430    72%    /
/dev/wd0s2f    596286   534890    13694    98%    /usr
/dev/wd1s3e    496367   456461      197   100%    /usr2
/dev/wd1s3f    396895   360004     5140    99%    /home
/dev/wd0s2e     29727     7705    19644    28%    /var
/dev/wd0s1     872640   848720    23920    97%    /c
/dev/wd1s1    1023824   942256    81568    92%    /d
/dev/wd1s5     511760   486472    25288    95%    /e
procfs              4        4        0   100%    /proc
mfs:31          20598       16    18936     0%    /tmp
localhost:/         0        0        0   100%    /crypt

[morden|root] 21:34 ~ pwd
/home/kkenn
[morden|root] 21:34 ~ mount
...
/dev/wd1s3f on /home (local, soft-updates, writes: sync 1123 async 11409)
[morden|root] 21:34 ~ dd if=/dev/zero of=bigfile bs=1024k count=20
20+0 records in
20+0 records out
20971520 bytes transferred in 11.454656 secs (1830829 bytes/sec)
[morden|root] 21:34 ~ df

...

/dev/wd1s3f    396895   382092   -16948   105%    /home

[morden|root] 21:34 ~ rm bigfile

[In another window, as a regular user]

[morden|kkenn] 21:35 ~ cp /etc/motd .

/home: write failed, file system is full
cp: ./motd: No space left on device

This persists for 30 seconds exactly, until the clock expires and the
space is reclaimed. Given the state of fullness of my partitions, I run into
this quite frequently. I had always assumed it was just the way things were
with soft-updates.

Kris

-----
(ASP) Microsoft Corporation (MSFT) announced today that the release of its 
productivity suite, Office 2000, will be delayed until the first quarter
of 1901.



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.OSF.4.05.9901291601380.5100-100000>