Re: Is ZFS production ready?
Edward M eam1edw...@gmail.com wrote: That reply was not meant for you, so why do you care? If it wasn't meant for everyone on the list, why was it sent to the list? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 06/25/2012 08:00 AM, per...@pluto.rain.com wrote: Edward M eam1edw...@gmail.com wrote: That reply was not meant for you, so why do you care? If it wasn't meant for everyone on the list, why was it sent to the list? by accident. still learning how to use email client:-[ . once i noticed my email was also to this list. i was hoping subscribers would notice it was by mistake. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 06/23/2012 10:38 PM, Wojciech Puchar wrote: last binary production ready, used version 14; i also found it to be stable Any opensource zfs pool verisons beyound that, i am not really sure about their stablity compared to UFS rock solid filesystem. No ZFS pool version can be as trusty as UFS because of ZFS on disk structure that is plain dangerous. ZFS use tree-like structure for everything. If upper part of tree is corrupted, everything below disappears and cannot be found. Having 2,3 or even 100 copies of metadata doesn't help if you would have (maybe transient) hardware problem and bad metadata would be writen 2,3 or even 100 times. with proper checksum of course. UFS uses flat structure - inodes in known places. superblocks are used to find info about placement, and there are many copies of which only first is updated under normal operation. In really unlikely case of all superblocks corrupted just use newfs on virtual device (may be md) of same size, with same block and fragment size, and byte per inode, and copy superblock from here. Dont email me privately. I like ZFS design however i was only questioning v28 stability for production compared to a mature production tested UFS. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, Jun 21, 2012 at 4:44 PM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: One interesting feature of ZFS if it's block checksum: all reads and writes include block checksum, so it can easily detect situations where, for example, data is quietly corrupted by RAM. you may be shocked but you are sometimes wrong. i already demostrated it and checksumming doesn't get any errors, and do write wrong data with right checksums :) it's quite easy to explain if one understand hardware details. Checksumming will protect you from - failed SATA/SAS port, on-disk controller that returns bad data as good. This is actually really rare case. i never seen that, but maybe it happens. - some types of DRAM failure - but not all. Actually just a small fraction because DRAM failure like that would bring your system to crash so quickly that you are unlikely to get big data corruption. Common case with DRAM memory is that after you write to it, keeps right data some time and RARELY flips some bit later in spite of refresh. With this type you may run your machine for hours, even days or longer. And ZFS would calculate proper checksum of wrong data and will write it to disk. This is the reason i keep few failed DIMMs - for testing how different software behaves on broken machine. UFS resulted in few corrupted files after half a day of heavy work and 4 crashes. fsck always recovered things well (of course unexpected softupdate inconsistency) ZFS survived 2 crashes. After third it panicked on startup. Of course - no zfs_fsck. And no possibility of making really good zfs_fsck because of data layout, at least not easy. This feature is very important for databases. is data integrity not important for the rest? :) Still - disks itself perform quite heavy ECC and both SATA and SAS ports. While I don't dispute you're test's findings I would like to point out that you are SPECIFICALLY testing for something that the original designers of ZFS (SUN now Oracle) point out VERY clearly as being an issue that you should avoid in you're deployed environments. The filesystem is designed to protect the ON DISK data and being a highly memory intensive filesystem should ALWAYS be deployed on hardware with memory error correction build in (aka ECC RAM deployed across multiple banks). The filesystem comes from an hardware/OS environment that is HEAVILY BIASED towards self healing as they put it and as a result things like memory module issues would: 1) Either be corrected by the ECC modules 2) Be reported to the administrator of said system as soon as they occur (well on a system where you have such reporting setup correctly) As a result you're argument is mootwhilst you're findings are indeed still valid. UFS2 being MUCH lighter on RAM requirements is, well frankly, quite possibly not even interacting with the damaged sections of the memory modules in you're test and I am almost certain that if we were to ask around on this mailing list enough examples of UFS/UFS2 corruption due to faulty RAM are VERY VERY likely to come up. No filesystem (or other code for that matter) would be able to detect RAM content corruption (as this is NOT a filesystem's job) and correct it for you as frankly the kernel wouldn't know if the data in the buffers is correct or not without the application storing said data being coded to check for these conditions (I know of a patch to the Linux kernel that does indeed look for faulty RAM segments and works around them but I am *mostly*positive that no general purpose OS in current deployment does so as I have noticed that this behavior was VERY CPU intensive). Also (debate encouraged here) due to the COW nature of ZFS a zfs_fsck command is basically entirely unnecessary as 1) The last successfully completed write to the file will be intact and 2) Scrubbing the on disk content performs a much better filesystem maintenance than an fsck does and this can also be done online without impacting uptimes of you're systems/data availability. On my systems I specifically trigger a scrub (via the ZFS init script) whenever my systems are uncleanly shut down as I am willing to tolerate a slightly slower but available system in such conditions. While UFS2 is indeed an wonderfully reliable filesystem it (as with all things) is not suited to all tasks, there are many instances where I can see the features of ZFS far outweighing the detractions (as do I see the same for the converse state of affairs). While all the above is purely based on my understanding of ZFS (and I am one of the people working on a port to GNU/Linux - admittedly not directly but I spend a LOT of my time reading/cleaning up the code fork that I do use) and SUN's (now Oracle's) design/deployment documents,...it is still my opinion and I would encourage a debate on these opinions. ___ freebsd-questions@freebsd.org mailing list
Re: Is ZFS production ready?
On 06/24/2012 04:23 PM, Adam Vande More wrote: On Sun, Jun 24, 2012 at 5:43 PM, Edward M eam1edw...@gmail.com mailto:eam1edw...@gmail.com wrote: Dont email me privately. Don't be an ass. Standard list conventions allows for private email. If this is simply an individual case of not liking the person who emailed you, then it is your ethical responsibility to ask for privacy in privacy. **Even in the case of sender being a pompous ass. If it is your overall wish to not be emailed privately by members of this list, then you should set that option on your list membership page instead of attempting to force your responsibility onto others. -- Adam Vande More That reply was not meant for you, so why do you care? get lost. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
I meant, is it now possible to have 2TB FS with UFS? On 6/21/2012 6:54 PM, Wojciech Puchar wrote: On Thu, 21 Jun 2012, Hooman Fazaeli wrote: On 6/21/2012 4:22 PM, Wojciech Puchar wrote: stick with UFS. It JUST WORKS(R), and is trusty. And it works fast. What options are there for 2TB file systems with UFS? the same as for 2TB filesystems. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
I meant, is it now possible to have 2TB FS with UFS? UFS2 is here since IMHO year 2005. Now the only problem is fsck time. actually IMHO fsck can be improved a lot but someone must have time and will to do this. if parallelism would be exploited on gstripe type(*) volumes then it should take less than 30 minutes no matter how large the volume is. Anyway - even with UFS which is the most fault-resilent filesystem i know - i would not recommend creating gstripe type volumes taking too many disks for the reason i already explained. For now softupdates+journal is fine, you actually have to do full fsck now and then, but at spare time. *) gstripe type means gstripe, gstripe+gmirror, graid5, graid5+gstripe, hardware matrix controller with any type of RAID configuration. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
Hooman Fazaeli wrote: I meant, is it now possible to have 2TB FS with UFS? Yes. The 2TB limitation so many are used to applies more to the tools than the UFS2 file system itself. UFS2 has a max volume size of 2^73, or 8 Zeta-Bytes. If you utilize the old Dos MBR scheme with old fdisk and disklabel tools you will still face the 2TB volume limit. Use Gpart, Glabel, and GPT partitioning instead. A quick and short example: http://www.mebsd.com/configure-freebsd-servers/big-partitions-in-freebsd-bigger-than-2tb.html However, fsck'ing such large volumes will take considerable time if such a thing needs doing. There is the new Soft-update plus Journaling coming along with the advent of 9.x, which is supposed to ameliorate this. Not completely sold on it yet, as I don't have enough knowledge/experience yet. Some may say it's not just quite ready for prime time yet, but I don't really know definitively myself. [snip] -Mike ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
However, fsck'ing such large volumes will take considerable time if such a thing needs doing. There is the new Soft-update plus Journaling coming along with the advent of 9.x, which is supposed to ameliorate this. Not it is far from perfect. But fine to use it. Just DO full fsck every some time. Fortunately it would be planned outage instead of unplanned. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
From owner-freebsd-questi...@freebsd.org Sat Jun 23 02:48:26 2012 Date: Sat, 23 Jun 2012 12:17:13 +0430 From: Hooman Fazaeli hoomanfaza...@gmail.com To: Wojciech Puchar woj...@wojtek.tensor.gdynia.pl Cc: FreeBSD Questions freebsd-questions@freebsd.org Subject: Re: Is ZFS production ready? I meant, is it now possible to have 2TB FS with UFS? Of course not. UFS uses 32-bit numbers for block addresses. IMPOSSIBLE to reference more than 2^41 bytes. However, UFS2 re-implemented the same disk structures using 64-bit numbers. Thus, it can reference 2^73 bytes in the same 'logical' method. Wojciech simply never lets inconvenient facts get in the way of his opinions about how everyone else should do everything. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 06/21/2012 12:33 AM, Hooman Fazaeli wrote: Now, I want to the same thing on 8.3 and wanted to know your opinion on ZFS stability. Is there any success story using ZFS in 24x7, large volume, heavy duty servers? Is there any other option other than ZFS to build larger than 2TB file systems? I like the ZFS theroy, However I would have to question ZFS Pool Version Number 28 stability, that is what freebsd 9.0 comes with. because it was never really used/marked as production ready by sun/oracle. in my opionion version 28 is consider as a development version. solaris 11 uses version 33 so that is consider as production ready but it is closed source. i think the last open zfs version pool marked as production ready,was pool version 14 and 15 zfs sounds great however i would actally trust more UFS2 for 24x7 servers. agian this is my opinion, could be wrong:-) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 06/23/2012 04:19 PM, Edward M wrote: On 06/21/2012 12:33 AM, Hooman Fazaeli wrote: Now, I want to the same thing on 8.3 and wanted to know your opinion on ZFS stability. Is there any success story using ZFS in 24x7, large volume, heavy duty servers? Is there any other option other than ZFS to build larger than 2TB file systems? I like the ZFS theroy, However I would have to question ZFS Pool Version Number 28 stability, that is what freebsd 9.0 comes with. because it was never really used/marked as production ready by sun/oracle. in my opionion version 28 is consider as a development version. solaris 11 uses version 33 so that is consider as production ready but it is closed source. i think the last open zfs version pool marked as production ready,was pool version 14 and 15 zfs sounds great however i would actally trust more UFS2 for 24x7 servers. agian this is my opinion, could be wrong:-) snafu on my part freebsd 8.3 also uses zfs pool version 28:-) so my opinion would also be the same. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
snafu on my part freebsd 8.3 also uses zfs pool version 28:-) No, 8.3 uses version 15. It's been quite stable for me. R's, John ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
snafu on my part freebsd 8.3 also uses zfs pool version 28:-) No, 8.3 uses version 15. It's been quite stable for me. Sorry, I misread my notes, 8.2 uses v 15, 8.3 uses v 28. R's, John ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 06/23/2012 05:16 PM, John Levine wrote: Sorry, I misread my notes, 8.2 uses v 15, 8.3 uses v 28. R's, John yeah, I remember version 15 was really stable. Opensolaris 2009.06 last binary production ready, used version 14; i also found it to be stable Any opensource zfs pool verisons beyound that, i am not really sure about their stablity compared to UFS rock solid filesystem. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
This is a valid argument. Checksumming is used to detect cases where the disk or the disk controller return invalid data to the CPU. This can happen for any number of reasons and isn't that unlikely. Unrecoverable read error probabilities are high enough with common drives that you can reasonably see them after reading 10-20TB over the course of some small number of years. And that's assuming no firmware bugs, no flakey cables, and no other of a variety of potential issues. this needs scrubbing. Can be done both with ZFS and anything else. just use dd periodically. I use ZFS. I like ZFS. But I also acknowledge that a zfs_fsck would be useful in cases where a filesystem is botched enough that it can't be but seems you don't have any serious use for ZFS if you can take that risk just because you like ZFS. I cannot. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
OK, if you have 24 2-way mirrors and two drives in the same mirror fail then with UFS you lose the contents of that mirror. Other filesystems in the same box are fine. Restores from backups are going to be easy since the backups are probably arranged to be per-filesystem. true. i actually don't have 48-disk machine but do have 9 disks (one SSD+8 2TB SATA). So far I think we're in agreement. Still as i said - even with ZFS i would make 24 pools, not one. this thing is not filesystem dependent. But this doesn't address two issues: 1) There are other arrangements of ZFS that can tolerate more failed disks if you are willing to spend more money. ZFS supports n-way mirrors, so you can have mirrors with three or four disks if you as well as gmirror. a raidz2 set (with multiple raidz2 sets per pool). i will not use raidz1/2/3 because if catastrophically low performance. the design of ZFS makes sure you'll get read performance of single drive from whole pool. Disks are already performance limiting part of computer. 2) That this failure can happen doesn't address the question of the production-ready status of ZFS. The question of production ready is not a boolean. It is a question of What i meant from beginning is not that ZFS is not yet production ready but it will never be because of design decisions. It have cool features, giving danger, huge hardware usage (RAM,CPU) and low I/O performance. risks and of money used to mitigate those risks. I suggest asking the question on the zfs-discuss list over at opensolaris.org since there are probably many more people there who make serious use of ZFS daily. I will not. Serious people should know how ZFS work. if they still want to use it seriuosly then i cannot help any more. gs1p 159G 73.1G 39 12 2.34M 70.7K mirror 159G 73.1G 39 12 2.34M 70.7K gpt/CONST_2-9XE02KPK-zfs - - 19 5 1.94M 69.4K gpt/SAVVIO-6XQ10F80-zfs - - 21 5 1.93M 79.5K gpt/SAVVIO-6XQ103C7-zfs - - 21 5 1.93M 79.5K 100GB+ of FreeBSD being served up (IP 206.196.19.100 if you care to check FreeBSD's stats pages). And the torrents can be easily replaced if something really bad happens. 3 very expensive drives. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
From owner-freebsd-questi...@freebsd.org Thu Jun 21 06:18:56 2012 Date: Thu, 21 Jun 2012 12:03:12 +0430 From: Hooman Fazaeli hoomanfaza...@gmail.com To: FreeBSD Questions freebsd-questions@freebsd.org Subject: Is ZFS production ready? Dear community In the past, I built a 8TB ZFS log server on freebsd 7.4. However, the system experienced instablility after long up times. My main motive to use ZFS was UFS inability to support large file systems. Now, I want to the same thing on 8.3 and wanted to know your opinion on ZFS stability. Is there any success story using ZFS in 24x7, large volume, heavy duty servers? Is there any other option other than ZFS to build larger than 2TB file systems? One alternative might be the 'new, improved' UFS -- UFS2. I believe it supports filesystems up to 2^73 bytes (2^64 sectors). ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote:` Subject: Re: Is ZFS production ready? stick with UFS. It JUST WORKS(R), and is trusty. And it works fast. Be sure to descrirbe how that is even _possible_, given that the OP needs/ wants larger than 2tb filesystems. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
From owner-freebsd-questi...@freebsd.org Thu Jun 21 11:50:42 2012 Date: Thu, 21 Jun 2012 18:47:30 +0200 (CEST) From: Wojciech Puchar woj...@wojtek.tensor.gdynia.pl To: Matthias Gamsjager mgamsja...@gmail.com Cc: FreeBSD Questions freebsd-questions@freebsd.org Subject: Re: Is ZFS production ready? True but this applies as much to you. You think you know it all and that is quite the probdlem with you. And discussing with you is a true waste with this attittute. Even its free. so stop it. If you don't have something to say, don't say it. --- the immortal words of Wojcciec It's a shame tou don't practice wyat you preach, troll. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
Hi, I think it is stable enough on FreeBSD. Someone actually posted quite a similar thread not a while ago.. Here'e a quick summary: For my various OpenSource projects, I have deployed a 36TB file system which is fine and stable running 24/7. Additionally at home I use 4TB (2x 2TB) + 8TB (2x 4TB) on a machine with 4GB RAM this has been up for 3 years with minimum reboot! - this system gets pretty hammered as lot's of front ends for my OpenSource stuff run off there plus I transfer large amounts of data 10's of GB's often between systems. For web stuff I get round 20,000-30,000 hits from various places on that particular box and it handles perfectly unlike my crappy Cisco 857 router - will redeploy a uni-socket server running OpenBSD for this one. Good luck! Regards, Kaya On Thu, Jun 21, 2012 at 8:33 AM, Hooman Fazaeli hoomanfaza...@gmail.com wrote: Dear community In the past, I built a 8TB ZFS log server on freebsd 7.4. However, the system experienced instablility after long up times. My main motive to use ZFS was UFS inability to support large file systems. Now, I want to the same thing on 8.3 and wanted to know your opinion on ZFS stability. Is there any success story using ZFS in 24x7, large volume, heavy duty servers? Is there any other option other than ZFS to build larger than 2TB file systems? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
stick with UFS. It JUST WORKS(R), and is trusty. And it works fast. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, Jun 21, 2012 at 1:52 PM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: stick with UFS. It JUST WORKS(R), and is trusty. And it works fast. The correct answer would be. I depends on the work load ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
21.06.2012 15:52, Wojciech Puchar пишет: stick with UFS. It JUST WORKS(R), and is trusty. And it works fast. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org I see the trend here. That guy is determined to shove his opinion down the throat of everybody. Stop it, tis most annoying. Back to the topic. ZFS support has matured greatly since the last time you tried it, currently freebsd supports zfs pool v. 28 in the last updates. Try it, it won't disappoint you. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
For my various OpenSource projects, I have deployed a 36TB file system which is fine and stable running 24/7. Additionally at home I use 4TB (2x 2TB) + 8TB (2x 4TB) on a machine with 4GB RAM this has been up for 3 years with minimum reboot! Good. There are some companies that make for living recovering data from unbreakable ZFS :) You may be just lucky. or they will make some money. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, Jun 21, 2012 at 1:52 PM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: stick with UFS. It JUST WORKS(R), and is trusty. And it works fast. The correct answer would be. I depends on the work load For different kinds of production workload it doesn't, aat least for me. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, Jun 21, 2012 at 9:33 AM, Hooman Fazaeli hoomanfaza...@gmail.comwrote: Dear community In the past, I built a 8TB ZFS log server on freebsd 7.4. However, the system experienced instablility after long up times. My main motive to use ZFS was UFS inability to support large file systems. Now, I want to the same thing on 8.3 and wanted to know your opinion on ZFS stability. Is there any success story using ZFS in 24x7, large volume, heavy duty servers? Is there any other option other than ZFS to build larger than 2TB file systems? Like I said. It depends. Could you give a better description about the expected work load. (DB, NFS filer etc) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, Jun 21, 2012 at 2:13 PM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: For my various OpenSource projects, I have deployed a 36TB file system which is fine and stable running 24/7. Additionally at home I use 4TB (2x 2TB) + 8TB (2x 4TB) on a machine with 4GB RAM this has been up for 3 years with minimum reboot! Good. There are some companies that make for living recovering data from unbreakable ZFS :) You may be just lucky. or they will make some money. And there are many happy users with ZFS (fbsd and opensolaris/solaris). Guess they are all wrong. I really want to see your face when you fsck 48TB w/o ffs+j (since that is so young must be immature :S ) of data with the phone ring non stop with customers who want to use their data again. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, 2012-06-21 at 12:03 +0430, Hooman Fazaeli wrote: Dear community In the past, I built a 8TB ZFS log server on freebsd 7.4. However, the system experienced instablility after long up times. My main motive to use ZFS was UFS inability to support large file systems. Now, I want to the same thing on 8.3 and wanted to know your opinion on ZFS stability. Is there any success story using ZFS in 24x7, large volume, heavy duty servers? Is there any other option other than ZFS to build larger than 2TB file systems? System 1: 32 cores, Interlagos, 64GB, 18TB RAIDz1 System 2: 64 cores, Interlagos, 128GB, 15TB RAIDz1 System 3: 8 cores, Bulldozer, 16GB, 27TB RAIDz2 Those are the main volumes on those systems. There are smaller ZFS volumes. I have other systems also ZFS (total of seven systems), typically with one or more 5-10TB RAIDz1 volumes. All systems RELENG_9. Stable? Yes. Be sure you have up-to-date FreeBSD kernel and your HBA firmware is up-to-date. Generally I use LSI 9211 cards. That said, the weak point is the drives. For example, one system has 6 Hitachi 4TB drives and there are three more in other systems -- 30% failure rate within one year. I've also had several failures with Seagate drives across two years. Zero failures with WD drives (12 drives in one ZFS array, IIRC) however those are slower, cheap drives. I am also working with compressed volumes because my data is very large and highly compressable. Compressed volumes, as you would expect, has a significant kernel performance impact depending on what you are doing. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 21.06.2012 07:39, Dennis Glatting wrote: Stable? Yes. Be sure you have up-to-date FreeBSD kernel and your HBA firmware is up-to-date. Generally I use LSI 9211 cards. Does the 9211 support JBOD (complete plain disks, no RAID or single disk RAID mess)? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
System 1: 32 cores, Interlagos, 64GB, 18TB RAIDz1 System 2: 64 cores, Interlagos, 128GB, 15TB RAIDz1 System 3: 8 cores, Bulldozer, 16GB, 27TB RAIDz2 what these systems do? (no details, just rough information) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
I really want to see your face when you fsck 48TB w/o ffs+j (since that is so young must be immature :S ) of data with the phone ring non stop with Even if ZFS would be the only filesystem in existence i would make one per 2 disks (single mirror). No matter what's going on, what do you prefer in case say - double disk failure from one mirror on 48 disk systems? losing completely data of 1/24 of users (and then restoring that amount from backups), or losing randomly chosen 1/24 of files from whole system? answer yourself. With UFS of course i would have single disk fsck time - less than a hour. which CAN be done out of work hours with soft updates. i normally turn off automatic fsck for large data filesystems, and if crash happened i run it after/before work hours. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, Jun 21, 2012 at 3:43 PM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: I really want to see your face when you fsck 48TB w/o ffs+j (since that is so young must be immature :S ) of data with the phone ring non stop with Even if ZFS would be the only filesystem in existence i would make one per 2 disks (single mirror). No matter what's going on, what do you prefer in case say - double disk failure from one mirror on 48 disk systems? losing completely data of 1/24 of users (and then restoring that amount from backups), or losing randomly chosen 1/24 of files from whole system? answer yourself. Sorry but I don;t follow you right there. with 48 disks you would not mirror 24vs24. I will perform very well but there is too much risk in that. you would rather go with a raidz2 stripe sets. With UFS of course i would have single disk fsck time - less than a hour. which CAN be done out of work hours with soft updates. i normally turn off automatic fsck for large data filesystems, and if crash happened i run it after/before work hours. raid is not a backup. You can loose data with any configuration or fs. so like in the compiler discussion. There is no perfect something in this world. It's always a tradeoff. with ZFS you have access to most advanced techniques and I believe that data is most safe with raidz3 as it can be. UFS cant match that and you have to rely on a raidcontroller which can screw up your data as well. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 6/21/2012 4:22 PM, Wojciech Puchar wrote: stick with UFS. It JUST WORKS(R), and is trusty. And it works fast. What options are there for 2TB file systems with UFS? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
answer yourself. Sorry but I don;t follow you right there. with 48 disks you would not mirror 24vs24. if i wasn't clear enough then i would it like that (with UFS), and assuming disks are named disk0disk48, and that i have at least one more disk for system code, often acessed data etc (SSD would be fine), while these 48 disks store user/whatever data. gmirror label ...options... mirror1 /dev/disk0 /dev/disk1 gmirror label ...options... mirror2 /dev/disk2 /dev/disk3 . . . gmirror label ...options... mirror24 /dev/disk46 /dev/disk47 then newfs etc.. and mounted as 24 filesystems. eg. /home1.../home24 then decide how to spread things properly. this depend of your needs. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, 21 Jun 2012, Hooman Fazaeli wrote: On 6/21/2012 4:22 PM, Wojciech Puchar wrote: stick with UFS. It JUST WORKS(R), and is trusty. And it works fast. What options are there for 2TB file systems with UFS? the same as for 2TB filesystems. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
One interesting feature of ZFS if it's block checksum: all reads and writes include block checksum, so it can easily detect situations where, for example, data is quietly corrupted by RAM. This feature is very important for databases. On 06/21/2012 15:58, Matthias Gamsjager wrote: On Thu, Jun 21, 2012 at 3:43 PM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: I really want to see your face when you fsck 48TB w/o ffs+j (since that is so young must be immature :S ) of data with the phone ring non stop with Even if ZFS would be the only filesystem in existence i would make one per 2 disks (single mirror). No matter what's going on, what do you prefer in case say - double disk failure from one mirror on 48 disk systems? losing completely data of 1/24 of users (and then restoring that amount from backups), or losing randomly chosen 1/24 of files from whole system? answer yourself. Sorry but I don;t follow you right there. with 48 disks you would not mirror 24vs24. I will perform very well but there is too much risk in that. you would rather go with a raidz2 stripe sets. With UFS of course i would have single disk fsck time - less than a hour. which CAN be done out of work hours with soft updates. i normally turn off automatic fsck for large data filesystems, and if crash happened i run it after/before work hours. raid is not a backup. You can loose data with any configuration or fs. so like in the compiler discussion. There is no perfect something in this world. It's always a tradeoff. with ZFS you have access to most advanced techniques and I believe that data is most safe with raidz3 as it can be. UFS cant match that and you have to rely on a raidcontroller which can screw up your data as well. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, Jun 21, 2012 at 4:22 PM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: answer yourself. Sorry but I don;t follow you right there. with 48 disks you would not mirror 24vs24. if i wasn't clear enough then i would it like that (with UFS), and assuming disks are named disk0disk48, and that i have at least one more disk for system code, often acessed data etc (SSD would be fine), while these 48 disks store user/whatever data. gmirror label ...options... mirror1 /dev/disk0 /dev/disk1 gmirror label ...options... mirror2 /dev/disk2 /dev/disk3 . . . gmirror label ...options... mirror24 /dev/disk46 /dev/disk47 then newfs etc.. and mounted as 24 filesystems. eg. /home1.../home24 then decide how to spread things properly. this depend of your needs. interesting idea but the options ZFS would give you are superior to this setup. But I have still not seen any evidence/facts that ZFS looses more data than UFS. Excluding user error which is 90% the reason data is lost. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 06/21/2012 16:13, Hooman Fazaeli wrote: On 6/21/2012 4:22 PM, Wojciech Puchar wrote: stick with UFS. It JUST WORKS(R), and is trusty. And it works fast. What options are there for 2TB file systems with UFS? this should not be a problem if you use GPT + gpart (which is the way to go nowadays) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
At 16:13 21/06/2012, you wrote: On 6/21/2012 4:22 PM, Wojciech Puchar wrote: stick with UFS. It JUST WORKS(R), and is trusty. And it works fast. What options are there for 2TB file systems with UFS? With UFS2 you can use file systems up to 2^73 (8 ZB). The problem is not UFS, but the old tools used to format the disk like fdisk and bsdlabel. For big file systems you must use gpart. The problem with file system recovery times when the worst thing happens(tm) is soluted/mitigated with su+j on FreeBSD9. HTH ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, 2012-06-21 at 07:55 -0500, wel...@excelsusphoto.com wrote: On 21.06.2012 07:39, Dennis Glatting wrote: Stable? Yes. Be sure you have up-to-date FreeBSD kernel and your HBA firmware is up-to-date. Generally I use LSI 9211 cards. Does the 9211 support JBOD (complete plain disks, no RAID or single disk RAID mess)? Typically I simply reburn them with IT firmware however I found under IR that a disk on an unconfigured port is seen by the kernel and usable but I haven't looked at any performance impact and I can't say whether that's a good idea. -- Dennis Glatting d...@pki2.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
One interesting feature of ZFS if it's block checksum: all reads and writes include block checksum, so it can easily detect situations where, for example, data is quietly corrupted by RAM. you may be shocked but you are sometimes wrong. i already demostrated it and checksumming doesn't get any errors, and do write wrong data with right checksums :) it's quite easy to explain if one understand hardware details. Checksumming will protect you from - failed SATA/SAS port, on-disk controller that returns bad data as good. This is actually really rare case. i never seen that, but maybe it happens. - some types of DRAM failure - but not all. Actually just a small fraction because DRAM failure like that would bring your system to crash so quickly that you are unlikely to get big data corruption. Common case with DRAM memory is that after you write to it, keeps right data some time and RARELY flips some bit later in spite of refresh. With this type you may run your machine for hours, even days or longer. And ZFS would calculate proper checksum of wrong data and will write it to disk. This is the reason i keep few failed DIMMs - for testing how different software behaves on broken machine. UFS resulted in few corrupted files after half a day of heavy work and 4 crashes. fsck always recovered things well (of course unexpected softupdate inconsistency) ZFS survived 2 crashes. After third it panicked on startup. Of course - no zfs_fsck. And no possibility of making really good zfs_fsck because of data layout, at least not easy. This feature is very important for databases. is data integrity not important for the rest? :) Still - disks itself perform quite heavy ECC and both SATA and SAS ports. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
interesting idea but the options ZFS would give you are superior to this setup. Were you just unable to understand my setup or a reasons to do this? please reread former post and possibly ask again if you don't understand the reasons. I ignore performance issues completely for now. But I have still not seen any evidence/facts that ZFS looses more data than UFS. And you've never seen me, yet i still exist. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
With UFS2 you can use file systems up to 2^73 (8 ZB). The problem is not UFS, but the old tools used to format the disk like fdisk and bsdlabel. For big file systems you must use gpart. true. or not using anything at all (and put filesystem directly on whole device/mirror). The problem with file system recovery times when the worst thing happens(tm) is soluted/mitigated with su+j on FreeBSD9. True but i don't believe completely in SU+J. i use it - eg on my private backup disk. but do full fsck sometimes. and usually few, but nonzero amount of errors are corrected. but with just SU it is easy to solve. Disable fsck on boot at all. softupdates allow that risk without problems. then do fsck at time when full or partial system outage can be tolerated - after work hours. This is my solution used everywhere. of course fsck on 100TB filesystem will be too slow. But it is implementation problem, and could be improved. but i would not recommend making single virtual device (gmirror/gstripe or dedicated hardware matrix controller) from too many disks because of the risk. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, Jun 21, 2012 at 4:47 PM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: interesting idea but the options ZFS would give you are superior to this setup. Were you just unable to understand my setup or a reasons to do this? please reread former post and possibly ask again if you don't understand the reasons. I ignore performance issues completely for now. I do understand your setup but I dont have too agree that it is a good solution. I know you think it's the best and only one :) But I have still not seen any evidence/facts that ZFS looses more data than UFS. And you've never seen me, yet i still exist. Really? that's you anwser to my question. The most childish answer I could image. You have a gift to troll and ruine every topic with this kind of answers ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
I do understand your setup but I dont have too agree that it is a good so i would repeat my question. Assume you have 48 disks, in mirrored configuration (24 mirrors) and 480 users with their data on them. Your solution with ZFS - ZFS crashes or you get double disk failure. Assuming the latter by average one per 24 file (randomly chosen) is destroyed which - in practice and limited time, means everything destroyed. Actually more than one per 24 - large files can be spread over. Your solution with UFS - better as there is fsck which slowly but successfully repairs problem. with double disk failure - the same! You restore everything from backup (i assume you have one). This takes like a day or more, one or two complete work days lost+all users in practice lost everything since last backup. My solution with UFS - fsck in case of failure work in parallel on 24 disks so not that long. double disk failure means losing data of 1/24 users. every one per 24 user cannot work, others work and i without any stress do recover this 1/24 of users data from backup after putting replacement disks. 1/24 of users lost data since last backup, and some hours of time. Even assuming ZFS is perfect then we both have problems as often, but my problems are 1/24 as severe as yours. Just don't ask me for help when unhappy users will want to cut off your head. And you've never seen me, yet i still exist. Really? that's you anwser to my question. The most childish answer I could stupid answer to stupid question. You never seen - but they do happens. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 21.06.2012 10:15, Wojciech Puchar wrote: I do understand your setup but I dont have too agree that it is a good so i would repeat my question. Assume you have 48 disks, in mirrored configuration (24 mirrors) and 480 users with their data on them. Your solution with ZFS - ZFS crashes or you get double disk failure. Assuming the latter by average one per 24 file (randomly chosen) is destroyed which - in practice and limited time, means everything destroyed. Actually more than one per 24 - large files can be spread over. Your solution with UFS - better as there is fsck which slowly but successfully repairs problem. with double disk failure - the same! You restore everything from backup (i assume you have one). This takes like a day or more, one or two complete work days lost+all users in practice lost everything since last backup. My solution with UFS - fsck in case of failure work in parallel on 24 disks so not that long. double disk failure means losing data of 1/24 users. every one per 24 user cannot work, others work and i without any stress do recover this 1/24 of users data from backup after putting replacement disks. 1/24 of users lost data since last backup, and some hours of time. Even assuming ZFS is perfect then we both have problems as often, but my problems are 1/24 as severe as yours. I think it is incorrect to assume that a failure with ZFS that cannot be recovered could be recovered if you used UFS with fsck. What fsck fixes in other file systems doesn't apply to ZFS by ZFS's design. fsck deals with fixing superblock inconsistancies on non-journaled file systems (like UFS/UFS2), not resurecting corrupted blocks on a disk. http://www.c0t0d0s0.org/archives/6071-No,-ZFS-really-doesnt-need-a-fsck.html http://en.wikipedia.org/wiki/Fsck http://en.wikipedia.org/wiki/UFS2 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
I think it is incorrect to assume that a failure with ZFS that cannot be recovered could be recovered if you used UFS with fsck. i think it is incorrect to not read carefully. So explanation - ZFS failure NOT caused by disks failure cannot be usually recovered. But even if i am wrong at this, rest still apply. What fsck fixes in other file systems doesn't apply to ZFS by ZFS's design.fsck deals with fixing superblock inconsistancies on non-journaled file systems (like UFS/UFS2), not resurecting corrupted blocks on a disk. http://www.c0t0d0s0.org/archives/6071-No,-ZFS-really-doesnt-need-a-fsck.html yes i know that article. And it is truly funny for me to know people do think this way. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On Thu, 21 Jun 2012 10:42:55 -0500, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: And it is truly funny for me to know people do think this way. If you understood how ZFS commits data to disk you'd not be making these statements. Also, if you take snapshots you can just roll back if there is any weirdness at all. Another important point: With 24 ZFS mirrors you'd have your data being striped across ALL the mirrors. This will yield much better performance. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 21 jun. 2012, at 17:15, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: I do understand your setup but I dont have too agree that it is a good so i would repeat my question. Assume you have 48 disks, in mirrored configuration (24 mirrors) and 480 users with their data on them. Your solution with ZFS - ZFS crashes or you get double disk failure. Assuming the latter by average one per 24 file (randomly chosen) is destroyed which - in practice and limited time, means everything destroyed. Actually more than one per 24 - large files can be spread over. Your solution with UFS - better as there is fsck which slowly but successfully repairs problem. with double disk failure - the same! You restore everything from backup (i assume you have one). This takes like a day or more, one or two complete work days lost+all users in practice lost everything since last backup. My solution with UFS - fsck in case of failure work in parallel on 24 disks so not that long. double disk failure means losing data of 1/24 users. every one per 24 user cannot work, others work and i without any stress do recover this 1/24 of users data from backup after putting replacement disks. 1/24 of users lost data since last backup, and some hours of time. Even assuming ZFS is perfect then we both have problems as often, but my problems are 1/24 as severe as yours. Just don't ask me for help when unhappy users will want to cut off your head. And you've never seen me, yet i still exist. Really? that's you anwser to my question. The most childish answer I could stupid answer to stupid question. You never seen - but they do happens. In other topic you hammerd on fact and if someone ask you to deliver them its a stupid question. And about the dram error. I really hope you do use ecc memory in production which renders your scenario invalide. And even then its a claim made by you some random dude on a list. Without proper test scenario and documentation such claims are just useless. And a proper layout zfs will withstand a double disk failure with zero downtime...where younhave to tell your customer they just lost a day work___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
Another important point: With 24 ZFS mirrors you'd have your data being striped across ALL the mirrors. This will yield much better performance. i though already after few mails that you can discuss things normally. But this reply just perfectly proves you didn't read more than maybe my last sentence in spite of nearly a page of explanation written. My advices was now for free. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
stupid answer to stupid question. You never seen - but they do happens. In other topic you hammerd on fact and if someone ask you to deliver them its a stupid question. just a proof it is a waste of time to explain things (FOR FREE) for people like you. You are free to make dangerous setups. People are free to hire you and believe at things what you do. People are free to then pay consequences of the results at unexpected time, as well as 10 times oversized hardware for a need. At least this is still free :) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
Wojciech == Wojciech Puchar woj...@wojtek.tensor.gdynia.pl writes: Wojciech I ignore performance issues completely for now. An ironic line, given your complaints about clang. -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 mer...@stonehenge.com URL:http://www.stonehenge.com/merlyn/ Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. See http://methodsandmessages.posterous.com/ for Smalltalk discussion ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 21 jun. 2012, at 18:07, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: stupid answer to stupid question. You never seen - but they do happens. In other topic you hammerd on fact and if someone ask you to deliver them its a stupid question. just a proof it is a waste of time to explain things (FOR FREE) for people like you. You are free to make dangerous setups. People are free to hire you and believe at things what you do. People are free to then pay consequences of the results at unexpected time, as well as 10 times oversized hardware for a need. At least this is still free :) True but this applies as much to you. You think you know it all and that is quite the probdlem with you. And discussing with you is a true waste with this attittute. Even its free. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 2012-06-21 08:12, Евгений Лактанов wrote: 21.06.2012 15:52, Wojciech Puchar пишет: stick with UFS. It JUST WORKS(R), and is trusty. And it works fast. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org I see the trend here. That guy is determined to shove his opinion down the throat of everybody. Stop it, tis most annoying. Back to the topic. ZFS support has matured greatly since the last time you tried it, currently freebsd supports zfs pool v. 28 in the last updates. Try it, it won't disappoint you. Agreed. Wojciech Puchar is in my 'probable troll' file at this point, from his interactions on several topics. ZFS is stable and tested, and works well if you have the resources. That means RAM as well as hard disks - and if you don't have the resources, most of ZFS's advantages wouldn't be coming into play anyway. I have seen no reason to believe at this point (under FreeBSD 9) that it is any less stable than any other filesystem. It is still fairly new relatively, but I and others have used it with no problems, on boxes of various sizes. Getting the best performance may take some tweaking on occasion, but in general it should be very good. (And getting the best performance out of a multi-terabyte drive array will take tweaking no matter what file system you are trying.) My one note to the above would be to advise against using it for swap - unless you have enough RAM to make sure you never swap. It doesn't do well in that role, in my experience. (Though that was under a slightly earlier version.) Daniel T. Staal --- This email copyright the author. Unless otherwise noted, you are expressly allowed to retransmit, quote, or otherwise use the contents for non-commercial purposes. This copyright will expire 5 years after the author's death, or in 30 years, whichever is longer, unless such a period is in excess of local copyright law. --- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
True but this applies as much to you. You think you know it all and that is quite the probdlem with you. And discussing with you is a true waste with this attittute. Even its free. so stop it. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
his interactions on several topics. ZFS is stable and tested, and works well if you have the resources. That means RAM as well as hard disks - and if you don't have the resources, most of ZFS's advantages wouldn't be coming into play anyway. I have seen no right. repeat it more times, as your clients may read it :) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
[...] My one note to the above would be to advise against using it for swap - unless you have enough RAM to make sure you never swap. It doesn't do well in that role, in my experience. (Though that was under a slightly earlier version.) I remember on SXCE running on my test Sun E420r server that ZFS (can't remember if this was in the spec file or not??) would use **any** usable or unpartitioned file system as swap. I maybe totally off-base with this as I was too knew to investigate the issue and was still learning Solaris at the time but all of a sudden a remote mounted external drive would start getting zapped by I/O usage. Of course it couldn't be any user as the only user for those machines was me and I wasn't doing anything on either system. That was quite a weird thing, but happened many years ago so my memory is quite hazy on the specifics of the issue too I do recall running top to see swap usage at a few tens of gigs which was quite funny, of course unmounting the drive dropped the swap back to whatever got allocated by SXCE default. Daniel T. Staal Regards, Kaya ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
ZFS is superior to UFS. End of the history. There is no point in use old technology (UFS) when the new one can make the same as the older and better ? Regards, El 21/06/12 11:31, Matthias Gamsjager escribió: On 21 jun. 2012, at 18:07, Wojciech Puchar woj...@wojtek.tensor.gdynia.plmailto:woj...@wojtek.tensor.gdynia.pl wrote: stupid answer to stupid question. You never seen - but they do happens. In other topic you hammerd on fact and if someone ask you to deliver them its a stupid question. just a proof it is a waste of time to explain things (FOR FREE) for people like you. You are free to make dangerous setups. People are free to hire you and believe at things what you do. People are free to then pay consequences of the results at unexpected time, as well as 10 times oversized hardware for a need. At least this is still free :) True but this applies as much to you. You think you know it all and that is quite the probdlem with you. And discussing with you is a true waste with this attittute. Even its free. ___ freebsd-questions@freebsd.orgmailto:freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.orgmailto:freebsd-questions-unsubscr...@freebsd.org -- MARCO ANTONIO MUSKUS MUSKUS NOC - Aplicaciones ISP EDATEL S.A. E.S.P. Calle 41 # 52 - 28 Piso 2 Medellín, Antioquia - Colombia Teléfono: (574) 384 6507 Fax: (574) 3846500 www.edatel.net.cohttp://www.edatel.net.co mamus...@edatel.com.comailto:mamus...@edatel.com.co Este mensaje y/o sus anexos son para uso exclusivo de su destinatario intencional y puede contener información legalmente protegida por ser confidencial. Si usted no es el destinatario intencional del mensaje por favor infórmenos de inmediato y elimínelo, así como sus anexos. Igualmente, le comunicamos que cualquier retención, revisión no autorizada, distribución, divulgación, reenvío, copia, impresión, reproducción, o uso indebido de este mensaje y/o sus anexos, está estrictamente prohibida y sancionada legalmente. EDATEL S.A. no se hace responsable en ningún caso por daños derivados de la recepción del presente mensaje. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 6/21/12 9:47 AM, Wojciech Puchar wrote: True but this applies as much to you. You think you know it all and that is quite the probdlem with you. And discussing with you is a true waste with this attittute. Even its free. so stop it. This mailing list isn't your blog. If you want to hear your own voice, go lock yourself in a room. We'll all be happier. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
ZFS is superior to UFS. End of the history. There is no point in use old technology (UFS) when the new one can make the same as the older and better ? anyway there must be morons here like me that after observation conclude that older is far safer and better. But if you want end of history then fine. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 06/21/2012 00:33, Hooman Fazaeli wrote: Dear community In the past, I built a 8TB ZFS log server on freebsd 7.4. However, the system experienced instablility after long up times. My main motive to use ZFS was UFS inability to support large file systems. Now, I want to the same thing on 8.3 and wanted to know your opinion on ZFS stability. Is there any success story using ZFS in 24x7, large volume, heavy duty servers? Is there any other option other than ZFS to build larger than 2TB file systems? We use ZFS for critical data and are quite happy with it. I've been using it in production since 8.1-R and have yet to have a problem. Make sure you do your zpool scrubs regularly. I use a cron job. We are currently migrating our customer RAID arrays to ZFS to ameliorate the multi-hour FSCK situations. -- Dave Robison Sales Solution Architect II FIS Banking Solutions 510/621-2089 (w) 530/518-5194 (c) 510/621-2020 (f) da...@vicor.com david.robi...@fisglobal.com _ The information contained in this message is proprietary and/or confidential. If you are not the intended recipient, please: (i) delete the message and all copies; (ii) do not disclose, distribute or use the message in any manner; and (iii) notify the sender immediately. In addition, please be aware that any message addressed to our domain is subject to archiving and review by persons other than the intended recipient. Thank you. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
Agreed. Wojciech Puchar is in my 'probable troll' file at this point, Here too, http://berklix.com/~jhs/dots/.procmailrc.lists Cheers, Julian -- Julian Stacey, BSD Unix Linux C Sys Eng Consultants Munich http://berklix.com Reply below not above, cumulative like a play script, indent with . Format: Plain text. Not HTML, multipart/alternative, base64, quoted-printable. Mail from @yahoo dumped @berklix. http://berklix.org/yahoo/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
Agreed. Wojciech Puchar is in my 'probable troll' file at this point, Here too, http://berklix.com/~jhs/dots/.procmailrc.lists very good. just block me, instead of performing aggresive replies and personal attacks. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
21.06.2012 21:32, Wojciech Puchar пишет: Agreed. Wojciech Puchar is in my 'probable troll' file at this point, Here too, http://berklix.com/~jhs/dots/.procmailrc.lists very good. just block me, instead of performing aggresive replies and personal attacks. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org Only after you, my man, only after you. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org Only after you, my man, only after you. not yours. i'm not homosexual ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
On 6/21/12 11:21 AM, Wojciech Puchar wrote: ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org Only after you, my man, only after you. not yours. i'm not homosexual For which the gay community is grateful. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is ZFS production ready?
ZFS is technologically more advance than UFS/UFS2, so, if someone ask to me which filesystem should be use, my answer is ZFS. You can do on UFS the same on ZFS, but ZFS extend the functionality beyond filesystem, that is a plus for IT today. I'm using ZFS for a public HTTP/FTP mirror pushing many TB/Month and a backup system. Regards, El 21/06/12 12:32, Wojciech Puchar escribió: Agreed. Wojciech Puchar is in my 'probable troll' file at this point, Here too, http://berklix.com/~jhs/dots/.procmailrc.lists very good. just block me, instead of performing aggresive replies and personal attacks. ___ freebsd-questions@freebsd.orgmailto:freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.orgmailto:freebsd-questions-unsubscr...@freebsd.org -- MARCO ANTONIO MUSKUS MUSKUS NOC - Aplicaciones ISP EDATEL S.A. E.S.P. Calle 41 # 52 - 28 Piso 2 Medellín, Antioquia - Colombia Teléfono: (574) 384 6507 Fax: (574) 3846500 www.edatel.net.cohttp://www.edatel.net.co mamus...@edatel.com.comailto:mamus...@edatel.com.co Este mensaje y/o sus anexos son para uso exclusivo de su destinatario intencional y puede contener información legalmente protegida por ser confidencial. Si usted no es el destinatario intencional del mensaje por favor infórmenos de inmediato y elimínelo, así como sus anexos. Igualmente, le comunicamos que cualquier retención, revisión no autorizada, distribución, divulgación, reenvío, copia, impresión, reproducción, o uso indebido de este mensaje y/o sus anexos, está estrictamente prohibida y sancionada legalmente. EDATEL S.A. no se hace responsable en ningún caso por daños derivados de la recepción del presente mensaje. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org