Date: Tue, 6 Feb 2001 19:26:56 +0200 From: Panagiotis Astithas <past@netmode.ntua.gr> To: stable@freebsd.org Subject: df output incosistency Message-ID: <20010206192656.B6939@netmode.ece.ntua.gr>
next in thread | raw e-mail | index | archive | help
Running df on my laptop with softupdates enabled, produces the following output: Filesystem 1K-blocks Used Avail Capacity Mounted on /dev/ad0s1a 99183 43709 47540 48% / /dev/ad0s1e 2750654 2112356 418246 83% /usr procfs 4 4 0 100% /proc As you can see there is a difference between the total number of blocks and the sum of used+available blocks. Particularly in the case of /usr there are more than 200MB "lost". Now, this came up while I was building a large port and erasing a very large file, about the size of the lost space. In the process the filesystem became full, and I had to stop the rm process, stop the build, reboot and fsck the disks (which did find errors). After that df always reported the mismatch you can see above. At that time the system was 4.2-STABLE as of early December IIRC, so I suspected a bug in df or ufs or softupdates and I upgraded the system, to no avail. I have since used the system as usual, with the only remarkable exception that a few times when I exceeded the available disk space (as indicated by df), the operations continued without a problem, although df reported a negative available space. My question is, can I reclaim the apparently lost space without using newfs? Is df lying and I can safely ignore the available space reading, or doing so will corrupt my filesystem even worse? Any help on this would be highly appreciated, since I will not be able to newfs my drive for a few months. Regards, -past To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010206192656.B6939>