Re: [zfs-discuss] ZFS root compressed ?
Bill Sommerfeld wrote: On Thu, 2008-06-05 at 23:04 +0300, Cyril Plisko wrote: 2. How can I do it ? (I think I can run zfs set compression=on rpool/ROOT/snv_90 in the other window, right after the installation begins, but I would like less hacky way.) what I did was to migrate via live upgrade, creating the pool and the pool/ROOT filesystem myself, tweaking both copies and compression on pool/ROOT before using lucreate. I haven't tried this on a fresh install yet. I tried this and the lucreate failed right at the end, unable to find grub on the root device. Because the grub copy failed, ludelete is unable to remove the broken BE. Ian ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] ZFS root compressed ?
Hi ! As, probably, many of us I am playing with snv_90 and ZFS root. During the installation I didn't see any possibility to set the properties of the datasets. In particular compression. I see that default installation leaves the compression property untouched for all the created datasets, with rpool/dump being the only exception. For it the compression property is explicitly set to off. venus:/home/imp zfs get compression NAME PROPERTY VALUE SOURCE rpool compression offdefault rpool/ROOT compression offdefault rpool/ROOT/snv_90 compression offdefault rpool/dump compression offlocal rpool/export compression offdefault rpool/export/home compression offdefault rpool/snv_90_gzip compression gzip local rpool/snv_90_lzjb compression lzjb local rpool/swap compression offdefault I checked what would be the disk saving if compression would be turned on and here are the results for the nearly full SXCE install. rpool/snv_90_gzip and rpool/snv_90_lzjb are copies of the rpool/ROOT/snv_90 ith respective compression turned on. venus:/home/imp zfs list | grep snv_90 rpool/ROOT/snv_90 4.79G 136G 4.79G / rpool/snv_90_gzip 1.82G 136G 1.82G /rpool/snv_90_gzip rpool/snv_90_lzjb 2.80G 136G 2.80G /rpool/snv_90_lzjb venus:/home/imp zfs get compressratio | grep snv_90 rpool/ROOT/snv_90 compressratio 1.00x - rpool/snv_90_gzip compressratio 2.63x - rpool/snv_90_lzjb compressratio 1.70x - venus:/home/imp With lzjb the compression ratio is significant and with gzip is even more so. I believe that having compression enabled on root file system can be a good thing. System files are usually read-only, so the CPU penalty on write for higher compression should be negligible. And decompression on read supposed to be pretty fast. So the questions are: 1. Are there any reasons to *not* enable compression by default ? 2. How can I do it ? (I think I can run zfs set compression=on rpool/ROOT/snv_90 in the other window, right after the installation begins, but I would like less hacky way.) -- Regards, Cyril ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS root compressed ?
On Thu, 2008-06-05 at 23:04 +0300, Cyril Plisko wrote: 1. Are there any reasons to *not* enable compression by default ? Not exactly an answer: Most of the systems I'm running today on ZFS root have compression=on and copies=2 for rpool/ROOT 2. How can I do it ? (I think I can run zfs set compression=on rpool/ROOT/snv_90 in the other window, right after the installation begins, but I would like less hacky way.) what I did was to migrate via live upgrade, creating the pool and the pool/ROOT filesystem myself, tweaking both copies and compression on pool/ROOT before using lucreate. I haven't tried this on a fresh install yet. after install, I'd think you could play games with zfs send | zfs receive on an inactive BE to rewrite everything with the desired attributes (more important for copies than compression). - Bill ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS root compressed ?
Bill Sommerfeld wrote: On Thu, 2008-06-05 at 23:04 +0300, Cyril Plisko wrote: 1. Are there any reasons to *not* enable compression by default ? Not exactly an answer: Most of the systems I'm running today on ZFS root have compression=on and copies=2 for rpool/ROOT I've been experimenting with this, too, and will likely settle on the same: copies=2, compression=on IIRC, grub ZFS does lzjb, but not gzip 2. How can I do it ? (I think I can run zfs set compression=on rpool/ROOT/snv_90 in the other window, right after the installation begins, but I would like less hacky way.) what I did was to migrate via live upgrade, creating the pool and the pool/ROOT filesystem myself, tweaking both copies and compression on pool/ROOT before using lucreate. I haven't tried this on a fresh install yet. Two ways to do this from DVD/LiveCD: 1. change the parameters after the rpool is created, but before packages are being installed. This can be a little tricky to get the timing right, especially for OpenSolaris -- it is easier with SXCE because there is a fair amount of time required to build the dump and swap volumes prior to building the root file system. Anyway, I just open a terminal window and type the appropriate zfs commands. 2. replace /sbin/zpool with a script that does what you want prior to beginning the installation -- ugly but possible. JumpStart may allow you more flexibility, if you do JumpStarts you should check the docs to see if there is an easier way. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS root compressed ?
On Jun 5, 2008, at 4:43 PM, Bill Sommerfeld wrote: after install, I'd think you could play games with zfs send | zfs receive on an inactive BE to rewrite everything with the desired attributes (more important for copies than compression). I blogged about something similar about a year ago: http://blogs.sun.com/mmusante/entry/zfs_compression_and_you Regards, markm ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS root compressed ?
Bill Sommerfeld writes: 2. How can I do it ? (I think I can run zfs set compression=on rpool/ROOT/snv_90 in the other window, right after the installation begins, but I would like less hacky way.) what I did was to migrate via live upgrade, creating the pool and the pool/ROOT filesystem myself, tweaking both copies and compression on pool/ROOT before using lucreate. I haven't tried this on a fresh install yet. after install, I'd think you could play games with zfs send | zfs receive on an inactive BE to rewrite everything with the desired attributes (more important for copies than compression). Would it be possible to create a new BE on a compressed filesystem and activate it? Is snap upgrade implemted yet? If so this should be quick. Ian. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS root compressed ?
On Fri, Jun 6, 2008 at 2:58 AM, [EMAIL PROTECTED] wrote: Bill Sommerfeld writes: 2. How can I do it ? (I think I can run zfs set compression=on rpool/ROOT/snv_90 in the other window, right after the installation begins, but I would like less hacky way.) what I did was to migrate via live upgrade, creating the pool and the pool/ROOT filesystem myself, tweaking both copies and compression on pool/ROOT before using lucreate. I haven't tried this on a fresh install yet. after install, I'd think you could play games with zfs send | zfs receive on an inactive BE to rewrite everything with the desired attributes (more important for copies than compression). Would it be possible to create a new BE on a compressed filesystem and activate it? Is snap upgrade implemted yet? If so this should be quick. Wouldn't snapupgrade clone the original BE ? In this case no data would be rewritten. Ian. -- Regards, Cyril ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS root compressed ?
Cyril Plisko wrote: On Fri, Jun 6, 2008 at 2:58 AM, [EMAIL PROTECTED] wrote: Bill Sommerfeld writes: 2. How can I do it ? (I think I can run zfs set compression=on rpool/ROOT/snv_90 in the other window, right after the installation begins, but I would like less hacky way.) what I did was to migrate via live upgrade, creating the pool and the pool/ROOT filesystem myself, tweaking both copies and compression on pool/ROOT before using lucreate. I haven't tried this on a fresh install yet. after install, I'd think you could play games with zfs send | zfs receive on an inactive BE to rewrite everything with the desired attributes (more important for copies than compression). Would it be possible to create a new BE on a compressed filesystem and activate it? Is snap upgrade implemted yet? If so this should be quick. Wouldn't snapupgrade clone the original BE ? In this case no data would be rewritten. Ah, good point! Ian ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS root compressed ?
On Fri, 2008-06-06 at 06:47 +0300, Cyril Plisko wrote: On Fri, Jun 6, 2008 at 2:58 AM, [EMAIL PROTECTED] wrote: Bill Sommerfeld writes: 2. How can I do it ? (I think I can run zfs set compression=on rpool/ROOT/snv_90 in the other window, right after the installation begins, but I would like less hacky way.) what I did was to migrate via live upgrade, creating the pool and the pool/ROOT filesystem myself, tweaking both copies and compression on pool/ROOT before using lucreate. I haven't tried this on a fresh install yet. after install, I'd think you could play games with zfs send | zfs receive on an inactive BE to rewrite everything with the desired attributes (more important for copies than compression). Would it be possible to create a new BE on a compressed filesystem and activate it? Is snap upgrade implemted yet? If so this should be quick. Wouldn't snapupgrade clone the original BE ? In this case no data would be rewritten. Correct, and Live Upgrade also clones the active BE when you do lucreate. Unless you copy all the data manually, it's going to inherit the uncompressed blocks from the current filesystem. -Albert ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss