On Mon, 5 Nov 2007, Timo Zimmermann wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi, > I'm new to Solaris and this maillinglist so let me introduce myself. > My name is Timo Zimmermann. A video made by "CSI:Munich" (on youtube) > about ZFS caught my attention and I decided that I move my private > fileserver to Solaris. > Because a lack of time (and coding skills) the only way I see to > contribute is with testing what you guys release and maybe some > documentation. > > After installing OpenSolaris Developer Preview, which was (btw.) really > easy and worked without problems, I got some errors and gdm didn't > start. > >> svc.startd[7]: svc:/application/font/fc-cache:default: Method >> "/usr/bin/fc-cache" failed due to signal BUS. >> svc.startd[7]: svc:/application/font/fc-cache:default: failed: >> transitioned to maintenance (see 'svcs -xv' for details) > > fc-cache only returned "Bus Error (core dumped)". > > zpool status -v showed me: >> name state read write cksum >> zpl_slim degraded 0 0 18 >> c7d0s0 degraded 0 0 18 too many errors >> >> errors: Permanent errors have been detected in the following files: >> //usr/X11/lib/X11/fonts/TTF/luximb.ttf > > replacing /usr/X11/lib/X11/fonts/TTF/luximb.ttf with luximb.ttf from the > livecd solved this and fc-cache works without problems and gdm started. > > Some time later I had another "permanent error" in a lib under > /lib/amd64. Because I don't need amd64-libs I just deleted the file. > But somehow I'm curious how this can happen. > I don't believe in a hardware error, Debian Etch ran on this system > since the first release, FreeBSD 7 beta and Solaris Express Developer > Edition didn't come up with one error. > memtest86 shows no error and smartmontools tells me my harddrive is > okay. So everything as expected. But because this looks okay to me > I don't know how this could happen. > >
Since the problem seems to have occurred multiple times, I would be very worried about the disk or the controller. The fact that you didn't notice problems with other operating systems on the same hardware doesn't mean silent corruption wasn't occurring; unless you had ZFS or a RAID configuration you would not detect most errors. format(1M) has an "analyze" option that might be useful for further testing. -Albert _______________________________________________ indiana-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/indiana-discuss
