Re: [zfs-discuss] How can I copy a ZFS filesystem back and forth
Thanks for the help Chris! Cheers, Fritz You wrote: original and and rename the new one, or zfs send or ?? Can I do a send and receive into a filesystem with attributes set as I want or does the receive keep the same attributes as well? Thank you. That will work. Just create the new filesystem with the attributes you want and send/recv the latest snapshot. As the data is received the gzip compression will be applied. Since the new filesystem already exists you will have to do a zfs receive -Fv to force it. --chris ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] How can I copy a ZFS filesystem back and forth
I turned compression on for several ZFS filesystems and found performance was still fine. I turned gzip on and it was also fine and compression on certain filesystems is excellent. I realize all the files that were on the filesystem when compression=on did not get the benefit of gzip when I set compression=gzip. I would like everything to be compressed with gzip. What's the easiest way for me to accomplish this? I figured some sort of copy back and forth is required, but I don't know what would be fastest. Should I just rsync everything to a new filesystem with compression=gzip set and then delete the original and and rename the new one, or zfs send or ?? Can I do a send and receive into a filesystem with attributes set as I want or does the receive keep the same attributes as well? Thank you. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS ok for single disk dev box?
On 08/30/2012 12:07 PM, Anonymous wrote: Hi. I have a spare off the shelf consumer PC and was thinking about loading Solaris on it for a development box since I use Studio @work and like it better than gcc. I was thinking maybe it isn't so smart to use ZFS since it has only one drive. If ZFS detects something bad it might kernel panic and lose the whole system right? I realize UFS /might/ be ignorant of any corruption but it might be more usable and go happily on it's way without noticing? Except then I have to size all the partitions and lose out on compression etc. Any suggestions thankfully received. Simply set copies=2 and go on your merry way. Works for me and protects you from bit rot. That sounds interesting. How does ZFS implement that? Does it make sure to keep the pieces of the duplicate on different parts of the drive? Even if you do decide to put a second drive in at a later time, just remember, RAID is not a backup solution. I use deja-dup to backup my important files daily to an off-site machine for that. Oh I realize that but this isn't a production machine just an unused lonely PC that could be running Solaris instead. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS ok for single disk dev box?
Hi Darren, On 08/30/12 11:07, Anonymous wrote: Hi. I have a spare off the shelf consumer PC and was thinking about loading Solaris on it for a development box since I use Studio @work and like it better than gcc. I was thinking maybe it isn't so smart to use ZFS since it has only one drive. If ZFS detects something bad it might kernel panic and lose the whole system right? I realize UFS /might/ be ignorant of any corruption but it might be more usable and go happily on it's way without noticing? Except then I have to size all the partitions and lose out on compression etc. Any suggestions thankfully received. If you are using Solaris 11 or any of the Illumos based distributions you have not choice you must use ZFS as your root/boot filesystem. I did not realize that. I was trying to decide between S10 I use at work although on Sun hardware and S11 since I have no experience with it. I would recommend that if physically possible attach a second drive to make it a mirror. I understand that is the best way to go. Personally I've run many many builds of Solaris on single disk laptop systems and never has it lost me access to my data. The only time I lost access to data on a single disk system was because of total hard drive failure. I run with copies=2 set on my home directory and any datasets I store data in when on a single disk system. However much much more importantly ZFS does not preclude the need for off system backups. Even with mirroring, and snaphots you still have to have a backup of important data elsewhere. No file system and more importantly no hardware is that good. Words to live by! Thanks, Stu ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS ok for single disk dev box?
Thank you. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS ok for single disk dev box?
I asked what I thought was a simple question but most of the answers don't have too much to do with the question. Now it seems to be an argument of your filesystem is better than any other filesystem. I don't think it is because I have seen the horror stories lurking on this list. I had no intention to get into this and I think you should have no intention either. I like ZFS, I use it at workand I am not here to knock it. 1) Anecdotal evidence is nearly worthless in matters of technology. Agree but fail to see the relevance. Bug reports on this list aren't worthless or the list wouldn't exist. 2) Data corruption does happen, and HDD manufacturers can even pin a number to it (the typical bit error rate on modern HDDs is around 10^-13, i.e. one bit error per ~10TB transferred). That it didn't hit your sensitive data but only some random pixel in an MPEG movie is good for you. But ZFS was built to handle environments where all data is critically important. I don't think I have 10TB of source code ;) Other file systems also handle critically important data. Every design has its tradeoffs and I don't believe ZFS is superior to anything else although it has many nice management features which aren't available in the same feature set elsewhere. I am not criticising ZFS, but I don't believe it solves every problem either. 3) Data corruption also happens in-transit on the SATA/SAS buses and in memory (that's why there is a thing as ECC memory). Right. 4) If it so bothers you, simply set checksum=off and fly without the parachute (a single core of a modern CPU can checksum at a rate upwards of 4GB/s, but if the few CPU cycles are so important to you, turn it off). You're making up imaginary motives and blaming them on me? I didn't say I don't want to spend cycles on checksumming. I said I don't want to lose a system because of a filesystem error. There's no need to be snide or condescending. Maybe you need a vacation? Who's your boss? In this specific use case I would rather have a system that's still bootable and runs as best it can than an unbootable system that has detected an integrity problem especially at this point in ZFS's life. If ZFS would not panic the kernel and give the option to fail or mark file(s) bad, I would like it more. ZFS doesn't panic in case of an unrecoverable single-block error, it simply returns an I/O error to the calling application. The panic only *can* take place in case of a catastrophic pool failure and isn't the default anyway. See man zpool(1M) for the description of the failmode option. ZFS is not perfect and although it may be designed to do what you say I think errors in ZFS are more likely than bit errors on hard drives. I'm betting on hardware and /in this scenario/ I would prefer a filesystem that tolerates it even ignorantly rather than protecting me from myself. What I'd really like is an option (maybe it exists) in ZFS to say when a block fails a checksum tell me which file it affects and let me decide to proceed or dump. But having the ability manage the disk with one pool and the other nice features like compression plus the fact it works nicely on good hardware make it hard to go back once you made the jump. Choices, choices. So you want to enable compression (which is a huge CPU hug) and worry about checksumming (which is tiny in comparison)? Yes, you got it right this time. You're the one trying to put words in my mouth. Nowhere did I ever suggest CPU cycles are an issue. The issue is what I said. Scroll up. If you're compressing data, you've got all the more reason to enable checksumming, since compression tends to make all data corruption much, much worse (e.g. that's why a single-bit error in a compressed MPEG stream doesn't simply slightly alter the color of a single pixel, but typically instead results in a whole macroblock or row of macroblocks messing up completely). Sounds reasonable. Even if your system does crash, at least you now have an opportunity to recognize there is a problem, and think about your backups, rather than allowing the corruption to proliferate. This isn't a production box as I said it's an unused PC with a single drive, and I don't have anybody's bank accounts on it. I can rsync whatever I work on that day to a backup server. It won't be a disaster if UFS suddenly becomes unreliable and I lose a file or two, or if a drive fails, but it would be very annoying if ZFS barfed on a technicality and I had to reinstall the whole OS because of a kernel panic and an unbootable system. As noted before, simple checksum errors won't panic your box, and neither will catastrophic pool failure (the default failmode=wait). You have to explicitly tell ZFS that you want it to panic your system in this situation. I have read reports on this list that show ZFS does panic the system by default in some cases. It may not have been for
Re: [zfs-discuss] what have you been buying for slog and l2arc?
It depends on the model. Consumer models are less likely to immediately flush. My understanding that this is done in part to do some write coalescing and reduce the number of P/E cycles. Enterprise models should either flush, or contain a super capacitor that provides enough power for the drive to complete writing any date in its buffer. My Home Fusion SSD runs on banana peels and eggshells and uses a Flux Capacitor. I've never had a failure. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Question on 4k sectors
Hans J. Albertsson hans.j.alberts...@branneriet.se wrote: I think the problem is with disks that are 4k organised, but report their blocksize as 512. If the disk reports it's blocksize correctly as 4096, then ZFS should not have a problem. At least my 2TB Seagate Barracuda disks seemed to report their blocksizes as 4096, and my zpools on those machines have ashift set to 12, which is correct, since 2¹² = 4096 Thanks, this is good to know. Is there any way, looking at manufacturers data sheets for drives, whether they report their blocksize correctly? From Seagate and WD that list the number of sectors, it's trivial to determine what sectors the disk is using. But is this number what the disk is really organized in or is it the number the disk reports?! It is very confusing... So far we seem to rely on reports from people on the list, which is good for us but bad for guys who wasted money on drives that don't work as they should (the drives that don't report actual sector sector size correctly). Really, it would be so helpful to know which drives we can buy with confidence and which should be avoided...is there any way to know from the manufacturers web sites or do you have to actually buy one and see what it does? Thanks to everyone for the info. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Question on 4k sectors
You wrote: 2012-07-23 18:37, Anonymous wrote: Really, it would be so helpful to know which drives we can buy with confidence and which should be avoided...is there any way to know from the manufacturers web sites or do you have to actually buy one and see what it does? Thanks to everyone for the info. I think that vendors' marking like 512e may give a clue on their support of emulated 512-byte sectors, whatever they would mean by that for a specific model line. Yeah but buying through the mail it's awfully difficult to see the vendor markings until it's too late ;) I believe you can roughly be certain that all 3Tb HDDs except Hitachi use 4Kb native sectors, and 4Tb disks are all 4Kb. If these disks don't expose such sector sizing to the OS properly, you can work around that in several ways, including, as of recent illumos changes, an override config file for the SCSI driver. The question was relative to some older boxes running S10 and not planning to upgrade the OS, keeping them alive as long as possible... The main problem with avoiding 4kb drives seems to be just the cases where you want to replace a single disk in an older pool built with 512b-native sectored drives. Right, that's what we're concerned with. For new pools (or rather new complete top-level VDEVs) this does not matter much, except that your overheads with small data blocks can get noticeably bigger. Understood. There were statements on this list that drives emulating 512b sectors (whether they announce it properly or not) are not all inherently evil - this emulation by itself may be of some concern regarding performance, but not one of reliability. Then again, firmware errors are possible in any part of the stack, of both older and newer models ;) I haven't seen any post suggesting 512b emulation didn't have very adverse effects on performance. Given how touchy ZFS seems to be I don't want to give him any excuses! Thanks for your post. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Move files between ZFS folder/Datasets ?
Thanks guys. I'm only planning to move some directories. Not the complete dataset. Just a couble of Gb's. Would it be save to use mv Of course, provided your system doesn't crash during the move. Using rsync -avPHK --remove-source-files SRC/ DST/ isn't that just as copying files ? Extra load on the server instead of moving the files from one place to another ? You may not realize it but a mv consists of a copy and delete. rsync is nice because you can check that it really got all your files by doing another rsync with -c after the first one completes. It compares checksums of files to make sure they actually are identical. With ZFS this is less of an issue but it is a nice doublecheck if you really can't afford to lose your files. Make sure you understand rsync syntax before using it live. Make some directories in /tmp and copy stuff around. One thing about rsync is that a trailing / means something and not having it means something else. rsync with the -n option does a trial run rsync -axvn /tmp/path/to/data /tmp/target # moves a directory named data to the target dir rsync -axvn /tmp/path/to/data/ /tmp/target # moves everything in the data directory (but not the data directory itself!) to the target rsync -axv # do it live and keep date/time/owner rsync -axvc # do it again and compare checksums instead of date/time/size ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] What is your data error rate?
I've been watching the heat control issue carefully since I had to take a job offshore (cough reverse H1B cough) in a place without adequate AC and I was able to get them to ship my servers and some other gear. Then I read Intel is guaranteeing their servers will work up to 100 degrees F ambient temps in the pricing wars to sell servers, he who goes green and saves data center cooling budget will win big since now everyone realizes AC costs more than hardware for server farms. And this is not on new special heat-tolerant gear, I heard they will put this in writing even for their older units. From that I would conclude at least commercial server gear can take a lot more abuse than it gets and still not be affected enough to make components fail because if they did, Intel could not afford to make this guarantee. YMMV of course. I still feel nervous running equipment in this kind of environment but after 3 years of doing that including commodity desktops I haven't seen any abnormal failures. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Failing WD desktop drive in mirror, how to identify?
Hello all, Trying to reply to everyone so far in one post. casper@oracle.com said Did you try: iostat -En I issued that command and I see (soft) errors from all 4 drives. There is a serial no. field in the message headers but it is has no contents. messages in /var/adm/messages perhaps? They should include the path to the disk in error. Success, thank you! It has the serial numbers of the drives along with the path. I had the path before but I couldn't relate it to the physical connectors on the mobo. The serial number is what I needed. Edward Ned Harvey said: A few things you could do... This will read for 1 second, then pause for one second. Hopefully making a nice consistent blinking light for you to find in your server. export baddisk=/dev/rdsk/cXtYdZ while true ; do dd if=$baddisk of=/dev/null bs=1024k count=128 ; sleep 1 ; done Alas this is a desktop box and there is one LED that lights for any disk activity. But if you don't have lights or something... The safest thing for you to do is to zpool export, then shutdown, remove one disk. Power on, devfsadm -Cv, and try to zpool import -a When the bad disk is gone, you'll be able to import no problem. If you accidentally pull the wrong disk, it will not cause any harm. Pool will refuse to import. Sounds like a good plan b. I will keep this in mind. Since I got the serial number from /var/adm/messages I am good to go. I couldn't copy and paste Jim's message since he posted with MIME instead of regular ASCII test. Thank you Jim for the help. This turned out to be a Solaris question not a ZFS question. Sorry and thank you Casper and Jim and Edward Ned! ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Failing WD desktop drive in mirror, how to identify?
Richard Elling said If the errors bubble up to ZFS, then they will be shown in the output of zpool status On the console I was seeing retryable read errors that eventually failed. The block number and drive path were included but not any info I could relate to the actual disk. zpool status showed a nonzero count of READ errors but nothing more. Otherwise, you can use iostat -En, as Casper noted, to show the error counters per disk. For more detailed information, use fmdump -eV What I needed was to identify the drive and Casper and Jim's suggestion to look at /var/adm/messages was where I found the info. It is just a drive going bad AFAIK and ZFS is working fine including letting me detach the bad drive from a mirror and not screwing up my data. Thanks again to everyone. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Can I create a mirror for a root rpool?
Thank you all for your answers and links :-) ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Can I create a mirror for a root rpool?
On Solaris 10 If I install using ZFS root on only one drive is there a way to add another drive as a mirror later? Sorry if this was discussed already. I searched the archives and couldn't find the answer. Thank you. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Wrong rpool used after reinstall!
You wrote: Hi Roy, things got alot worse since my first email. I don't know what happened but I can't import the old pool at all. It shows no errors but when I import it I get a kernel panic from assertion failed: zvol_get_stats(os, nv) which looks like is fixed by patch 6801926 which is applied in Solaris 10u9. But I cannot boot update 9 on this box! I tried Solaris Express, none of those will run right either. They all go into maintenance mode. The last thing I can boot is update 8 and that is the one with the ZFS bug. If they go into maintenance mode but could recognize the disk, you should still be able to do zfs stuff (zpool import, etc). If you're lucky you'd only miss the GUI Thank you for your comments. This is pretty frustrating. Unfortunately I'm hitting another bug I saw on the net. Once the Express Live CD or text installer falls back to maintenance mode, I keep getting a msg bash: /usr/bin/hostname command not found (from memory, may not be exact). All ZFS commands fail with this message. I don't know what causes this and I'm surprised since Solaris 10 update 8 works mostly fine on the same hardware. I would expect support to get wider not less but the opposite seems to be happening because I can't install update 9 on this machine. I have 200G of files I deliberately copied to this new mirror and now I can't get at them! Any ideas? Another thing you can try (albeit more complex) is use another OS (install, or even use a linux Live CD), install virtualbox or similar on it, pass the disk as raw vmdk, and use solaris express on the VM. You should be able to at least import the pool and recover the data. I didn't think I would be able to use raw disks with exported pools in a VM but your comment is interesting. I will consider it. The host is bootable, it just immediately panics upon trying to import the 2nd pool. Is there some way I can force a normal boot where my root pool is mounted but it doesn't mount the other pool? If so I could install VirtualBox and try your suggestion without moving the drives to another box. Aren't there any ZFS recovery tools, or is ZFS just not expected to break? I mentioned I can boot the update 9 installer, but it fails when trying to read media from the DVD because of a lack of nvidia drivers (documented limitation) I wonder if I can do some kind of network or jumpstart install. I have no other Solaris Intel boxes (and this post should explain some of the reasons why) but I have several Solaris Sparc boxes. I haven't gone through the doc on net/jumpstart installs, there is alot to read. But maybe this would get update 9 on that box and maybe it could import the pool. Are there any other Solaris based live CD or DVD I could try that have known, good ZFS support? I will need something where I can scp/rcp/rsync the data off that box, assuming that I can import it somehow. Thanks, Jim ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Wrong rpool used after reinstall!
I am having a problem after a new install of Solaris 10. The installed rpool works fine when I have only those disks connected. When I connect disks from an rpool I created during a previous installation, my newly installed rpool is ignored even though the BIOS (x86) is set to boot only from the new rpool drives. When the system starts it uses the old rpool! How can I get around this? I want to use my new install and then import the old rpool under a new name, and clean things up. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Wrong rpool used after reinstall!
Hi Roy, things got alot worse since my first email. I don't know what happened but I can't import the old pool at all. It shows no errors but when I import it I get a kernel panic from assertion failed: zvol_get_stats(os, nv) which looks like is fixed by patch 6801926 which is applied in Solaris 10u9. But I cannot boot update 9 on this box! I tried Solaris Express, none of those will run right either. They all go into maintenance mode. The last thing I can boot is update 8 and that is the one with the ZFS bug. I have 200G of files I deliberately copied to this new mirror and now I can't get at them! Any ideas? Thanks. Jim ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Server with 4 drives, how to configure ZFS?
Hi Dave, Hi Cindy. Consider the easiest configuration first and it will probably save you time and money in the long run, like this: 73g x 73g mirror (one large s0 on each disk) - rpool 73g x 73g mirror (use whole disks) - data pool Then, get yourself two replacement disks, a good backup strategy, and we all sleep better. Oh, you're throwing in free replacement disks too?! This is great! :P A complex configuration of slices and a combination of raidZ and mirrored pools across the same disks will be difficult to administer, performance will be unknown, not to mention how much time it might take to replace a disk. Yeah that's a very good point. But if you guys will make ZFS filesystems span vdevs then this could work even better! You're right about the complexity but OTOH the great thing about ZFS is not having to worry about how to plan mount point allocations and with this scenario (I also have a few servers with 4x36) the planning issue raises its ugly head again. That's why I kind of like Edward's suggestion even though it is complicated (for me) still I think it may be best given my goals. I like breathing room and not having to worry about a filesystem filling, it's great not having to know exactly ahead of time how much I have to allocate for a filesystem and instead let the whole drive be used as needed. Use the simplicity of ZFS as it was intended is my advice and you will save time and money in the long run. Thanks. I guess the answer is really using the small drives for root pools and then getting the biggest drives I can afford for the other bays. Thanks to everybody. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Is another drive worth anything?
Hi. I have a development system on Intel commodity hardware with a 500G ZFS root mirror. I have another 500G drive same as the other two. Is there any way to use this disk to good advantage in this box? I don't think I need any more redundancy, I would like to increase performance if possible. I have only one SATA port left so I can only use 3 drives total unless I buy a PCI card. Would you please advise me. Many thanks. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Any use for extra drives?
If you plan to generate a lot of data, why use the root pool? You can put the /home and /proj filesystems (/export/...) on a separate pool, thus off-loading the root pool. I don't, it's a development box with not alot happening. My two cents, thanks ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Any use for extra drives?
Generally, you choose your data pool config based on data size, redundancy, and performance requirements. If those are all satisfied with your single mirror, the only thing left for you to do is think about splitting your data off onto a separate pool due to better performance etc. (Because there are things you can't do with the root pool, such as striping and raidz) That's all there is to it. To split, or not to split. Thanks for the update. I guess there's not much to do for this box since it's a development machine and doesn't have much need for extra redundancy although if I would have had some extra 500s I would have liked to stripe the root pool. I see from your answer that's not possible anyway. Cheers. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
