Date: Mon, 4 Feb 2008 16:38:39 +0100 (CET) From: Wojciech Puchar <wojtek@wojtek.tensor.gdynia.pl> To: Christian Baer <christian.baer@uni-dortmund.de> Cc: freebsd-questions@freebsd.org Subject: Re: Looking for a Text on ZFS Message-ID: <20080204162233.M8365@wojtek.tensor.gdynia.pl> In-Reply-To: <fo78an$1vo5$3@nermal.rz1.convenimus.net> References: <fo2f26$1l2q$1@nermal.rz1.convenimus.net> <200802022111.21862.fbsd.questions@rachie.is-a-geek.net> <fo48j6$1p19$1@nermal.rz1.convenimus.net> <20080203173245.U1631@wojtek.tensor.gdynia.pl> <fo6vjo$1vo5$2@nermal.rz1.convenimus.net> <20080204133351.P7781@wojtek.tensor.gdynia.pl> <fo78an$1vo5$3@nermal.rz1.convenimus.net>
next in thread | previous in thread | raw e-mail | index | archive | help
>> did you ever got your UFS filesystem broken not because your drive failed? > > That is not the point here. I have been using FreeBSD sind version 3.3, > which was released in 1999. Before that I used Linux. So I can't even look while i was using linux - crashed filesystem was quite common without any disk failure. that's one of the reason i started using *BSD, another was performance. > To answer the question: Yes, it did happen and not only once. This was in > the time when I was setting up a new computer with 6.0-RELEASE and a new > S-ATA controller. There was a bug in the driver which the developer > managed to fix after we exhanged a few eMails. Before the error was fixed, > my machine crashed several times with a kernel panic. There were something of course - with buggy driver it may be big problems. > connected to the new controller and I was using this machine as a > plattform to debug the driver, no real damage was caused. you of course will not use new/untested hardware/software combination for production machine? me too, so what a problem? > That's ok to believe if you want to. UFS is designed to minimize errors. > There is no guarantee that there will be none. there is never guarantee. if ZFS will calculate block checksum from memory and miscalculate it because of hardware/software problem, it will write it, and then write good data to say 2 disks. then it will take data from both disk and report uncorrectable error - without any error at all. just an example. there is no "ultimate" solution. > systems still work. :-) my works too with one partition. so what a problem - except i had less work? > >> still - making all in / is much easier and works fine. > > Maybe I'm just too conservative for that. in EVERY unix book it's repeated countless times that partitioning is good, and make it more secure, more prone to errors etc. it was - on original unix FS, but not on UFS, which automatically "partitions" your drive to cylinder groups. >> making all in / and /lessused, where / is at first part on disk, and >> /lessused on second - make big performance improvements (shorter seeks!). > > There are about 10 things I can think of that I'd do before I tried > something like that. I'm a little surprised about a suggestion like this > coming from you because you seem to be a great advocacy of dynamic what you mean "dynamic system"? > systems. And here you have to decides what is used often and what not. > This is an estimate that you could also mess up - I'm sure I probably > would. :-) And chaninge a file from the seldom to the often area isn't > that trivial either. i prefer / for everything. but sometimes i need this to speed things up, and definitely need it when using gmirror - to not waste lot of space by mirroring everything. i just mirror what's have to be mirrored. > But ok, noone will judge either of us for working with our systems the way > we please. :-) Anyone with Unix knowledge will find his way around my > boxes and the same should be true for you. The rest are just details. :-) yes - but it's bad repeating "truth"s because it's said. ZFS advocacy first creates problems then solving it. i STRONGLY state most of these problems are artifical for most administrators and users. so ZFS may be useful for someone, but for most of as it's just waste of CPU and memory and... ours time (contrary to what ZFS stated - saving ours time). >> mirror. that's all. > > Which doesn't really address the issue of what happens if a drive that is > part of a big ZFS is removed (because it's broken). it will say "read error" on all files and directories that happened to be placed on that disk! >> you understand quota as workaround of problems creating 1000 partitions. >> or simply - looks like you don't understand it at all, because it is not >> workaround. it's excellent tool. > > Maybe you just don't understand my English? :-) maybe..but you stated that quota is needed because partitions can't me easily created by mass. and that's exactly wrong. > Quota does not address the different needs of certain applications. With > quota you can limit the amount of inodes a user may grab but you cannot > create areas with more inodes and others with less. Quota solves many in most cases average inode count is important with a bit excess. with rare cases really lots of files planned in some directory (like my squid spools) i make separate partitions. having this spools on separate partitions i can greatly reduce seeking as it's all in narrow part of disk plate. with ZFS it's no problem to have lots of small files, but they will be mixed up with other - without any control of placement. actually - i think it could be done automatically quite good, like few other things. maybe (i'm too lazy) i will write UFS3 ;) - but for sure it will be UFS-style filesystem, with some improvements. there is no need to revolution > problems and is a great tool, no doubt in that, but it doesn't make your > computer fast, you less thirsty and it doesn't improve your sexlife either - > at least that didn't happen here. :-) i think windows vista may be better in it :). but not for me.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080204162233.M8365>