Date: Sat, 4 Sep 2004 08:02:52 -0700 (PDT) From: David Wolfskill <david@catwhisker.org> To: current@freebsd.org, PeterJeremy@optushome.com.au Subject: Re: 5.3-BETA2 df(1) reports incorrect values for UFS1 FS Message-ID: <200409041502.i84F2qjS024533@bunrab.catwhisker.org> In-Reply-To: <20040904080058.GV423@cirb503493.alcatel.com.au>
next in thread | previous in thread | raw e-mail | index | archive | help
>Date: Sat, 4 Sep 2004 18:01:11 +1000 >From: Peter Jeremy <PeterJeremy@optushome.com.au> >To: current@freebsd.org >Subject: 5.3-BETA2 df(1) reports incorrect values for UFS1 FS >Sender: owner-freebsd-current@freebsd.org >On a 4.10 system, my /home reports as: >Filesystem 1K-blocks Used Avail Capacity iused ifree %iused Mounted on >/dev/ad0s3g 86710002 71758104 8015098 90% 1803049 9038549 17% /home >When I mount it on a 5.3-BETA2 system, I get: >Filesystem 1K-blocks Used Avail Capacity iused ifree %iused Mounted on >/dev/ad0s3g 86710002 1929924 77843278 2% 70983 10770615 1% /home Odd.... >This is somewhat disconcerting. As far as I can tell, all the files are >there but 5.3 doesn't realise it. >fsck on 5.3 reports no errors and: >1803048 files, 35879051 used, 42390039 free (1903 frags, 5298517 blocks, 0.0% fragmentation) >fsck on 4.10 reports no errors and: >1803048 files, 35879051 used, 7475950 free (390462 frags, 885656 blocks, 0.9% fragmentation) Hmmm... >Any ideas what is wrong with 5.3? More critically, is it safe to write >to a UFS1 filesystem with 5.3? Although it might be considered risky, each system where I run 5.x is set up so that: * Each file system backed by its own partitin (within a slice) is UFS1. This way, I can mount each of them while running 4.x and do repair work, for example. * The system multi-boots FreeBSD, and I do most of my work on 4.x (at this time). (My "production" systems run snapshots of 4.x that I build every couple of weeks.) * I track each of RELENG_4 and RELENG_5 (was HEAD before RELENG_5 was tagged) on a daily basis (unless "cvs update" shows no changes to the sources in question since last time). I am not seeing the problems you report. In particular, here's my build machine (the laptop is still doing this morning's "make buildworld" for RELENG_5, so it won't be done for a while): First, RELENG_5, freshly built: freebeast(5.3)[1] uname -a FreeBSD freebeast.catwhisker.org 5.3-BETA3 FreeBSD 5.3-BETA3 #14: Sat Sep 4 07:26:33 PDT 2004 root@freebeast.catwhisker.org:/common/S3/obj/usr/src/sys/FREEBEAST i386 freebeast(5.3)[2] df -ki Filesystem 1K-blocks Used Avail Capacity iused ifree %iused Mounted on /dev/ad0s3a 158767 108172 37894 74% 3909 36025 10% / devfs 1 1 0 100% 0 0 100% /dev /dev/ad0s3e 1873113 913919 809345 53% 87517 383073 19% /usr /dev/ad0s4h 27728233 14294775 11215200 56% 862175 6087263 12% /common /dev/ad0s4g 2032839 425583 1444629 23% 47282 461068 9% /var procfs 4 4 0 100% 1 0 100% /proc /dev/md0 507630 8 467012 0% 5 65785 0% /tmp freebeast(5.3)[3] And here's RELENG_4: freebeast(4.10-S)[1] uname -a FreeBSD freebeast.catwhisker.org 4.10-STABLE FreeBSD 4.10-STABLE #987: Fri Sep 3 05:26:57 PDT 2004 root@freebeast.catwhisker.org:/common/S1/obj/usr/src/sys/FREEBEAST i386 freebeast(4.10-S)[2] df -ki Filesystem 1K-blocks Used Avail Capacity iused ifree %iused Mounted on /dev/ad0s1a 158767 47187 98879 32% 1902 38032 5% / mfs:25 515606 6 474352 0% 4 64506 0% /tmp /dev/ad0s1e 1873082 1009417 713819 59% 90564 380026 19% /usr /dev/ad0s2a 158767 47187 98879 32% 1902 38032 5% /S2 /dev/ad0s2e 1873113 946906 776358 55% 87586 383004 19% /S2/usr /dev/ad0s3a 158767 108172 37894 74% 3909 36025 10% /S3 /dev/ad0s3e 1873113 913919 809345 53% 87517 383073 19% /S3/usr /dev/ad0s4a 158767 108249 37817 74% 3909 36025 10% /S4 /dev/ad0s4e 1872759 910011 812928 53% 84059 386531 18% /S4/usr /dev/ad0s4h 27728233 14294777 11215198 56% 862176 6087262 12% /common /dev/ad0s4g 2032839 425642 1444570 23% 47278 461072 9% /var procfs 4 4 0 100% 50 4034 1% /proc freebeast(4.10-S)[3] (You may see that more file systems show up under RELENG_4. I do not mount the RELENG_4 file systems necessary for running the system by default when booting 5.x or higher.) As noted, I have not had the problems you are reporting. The last issue I recall was something that was committed to fsck (iirC) by Matt Dillon (to avoid a 4.x fsck seeing a superblock updated by a 5.x fsck as broken). As implied by the committer, that wa a while back. Peace, david -- David H. Wolfskill david@catwhisker.org Evidence of curmudgeonliness: becoming irritated with the usage of the word "speed" in contexts referring to quantification of network performance, as opposed to "bandwidth" or "latency."
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200409041502.i84F2qjS024533>