Switch to DuckDuckGo Search
   January 6, 2019  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >
Toggle Join/Part | bottom
[00:00:35] <PMT> All the child services should have zfs-import.target as a dependency before they start. I wonder if that's not true in 0.6.5.X.
[00:01:05] <PMT> Or that could be why they're failed, but I'm surprised if they're failed and the original is dead.
[00:02:40] <phantomcircuit> PMT, there's not Wants directives in zfs-mount because either zfs-import-cache or zfs-import-scan could fulfil the dependency
[00:02:49] *** ShellcatZero <ShellcatZero!~ShellcatZ@cpe-66-27-89-254.san.res.rr.com> has joined #zfsonlinux
[00:03:17] <phantomcircuit> there's literally no file named zfs-import.target on this system
[00:04:34] <phantomcircuit> ok well i've figured it out
[00:05:10] <phantomcircuit> this is a fresh install without a pool, so no cache file, zfs-import-cache.service depends on that file existing in systemd, which means it doesn't run which means the module is never loaded
[00:05:16] <phantomcircuit> that's mega confusing
[00:06:31] <PMT> Usually I think I expect both import-cache and import-scan enabled by default.
[00:06:50] <PMT> But it's been so long since I installed an 0.6.5.X version I have no idea any more.
[00:07:27] <phantomcircuit> PMT, -scan is not enabled by default
[00:07:51] <phantomcircuit> touch /etc/zfs/zpool.cache should "fix" this
[00:07:54] <phantomcircuit> lets find out
[00:09:05] <phantomcircuit> yup
[00:14:27] *** Celmor <Celmor!~Celmor@unaffiliated/celmor> has joined #zfsonlinux
[00:14:44] <Celmor> can I get the compressionrationof a specific file?
[00:15:52] <DeHackEd> sorta. `du` will show on-disk size. but this can also be affected by copies=2 or 3, and if the file is sparse or not
[00:16:39] *** rjvb <rjvb!~rjvb@2a01cb0c84dee6009bda76eb03bc33f7.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 252 seconds)
[00:18:49] *** last1 <last1!~last1@modemcable231.247-201-24.mc.videotron.ca> has joined #zfsonlinux
[00:19:58] <last1> I am trying to create a new pool in centos 7.4, with zpool create chunk1 /dev/sdb but it prints this error: /dev/sdb is in use and contains a unknown filesystem.
[00:22:02] <last1> the disk is most definitely not in use. what gives ?
[00:22:14] <zfs> [zfsonlinux/zfs] default debian install doesn't load modules until pool is created (#8240) created by Patrick Strateman <https://github.com/zfsonlinux/zfs/issues/8240>
[00:22:32] <sponix> last1: tried prefixing that with "sudo" ?
[00:23:15] <DeHackEd> last1: might try running blkid on the disk to see what labels and filesystems are on it. ZFS might be refusing to use a "used" disk or it could be something like LVM with available (but not mounted) filesystem preventing repartitioning
[00:23:16] <phantomcircuit> last1, it's got a filesystem on it, wipe that with dd and it'll work fine
[00:23:44] <phantomcircuit> or maybe you're making a mistake and it's in use like DeHackEd said :P
[00:24:13] <DeHackEd> one (of MANY MANY MANY) things I hate about centos7 and systemd is its aggressive "hotplug" behaviour. oh, you have labels on your disk? let me auto-activate them...
[00:24:31] <DeHackEd> (and iirc we just gave BSD shit for automounting ZFS pools hot-inserted via USB)
[00:25:33] <DeHackEd> it's enough to make you want to go back to static /dev part of the root filesystem like it's 2001 again
[00:29:23] <phantomcircuit> all the hotplug stuff is crazy pants
[00:34:27] <Celmor> thanks
[00:39:47] <rjvbb> depends what you're used to, and to what extent you see ZFS as just another FS one might use on a USB drive.
[00:43:54] <last1> let me try with dd
[00:43:56] <last1> I used wipefs on it
[00:44:21] <last1> at first it said this: /dev/sdb is in use and contains a ddf_raid_member filesystem.
[00:44:31] <last1> then I used wipefs and it just says it's in use
[00:44:36] <last1> what dd command should I use ?
[00:45:26] <DeHackEd> wiping the filesystem doesn't mean it's stopped being used. if it's a software raid disk then check /proc/mdstat to see if it's been activated/used
[00:46:33] <last1> how is it used however ? cat /proc/mdstat shows: unused devices: <none>
[00:46:41] <last1> yet I have 16 drives which I haven't mounted at all
[00:48:29] <last1> If I do fdisk -l it shows for one of the disks: Disk /dev/mapper/ddf1_4c534920202020201000005d15d9080947114711db9a7f66: 799.6 GB, 799602114560 bytes, 1561722880 sectors
[00:53:33] <last1> https://pastebin.com/EiY8udW8
[01:00:43] <zfs> [zfsonlinux/zfs] Sporadic system hang (#7425) comment by nivedita76 <https://github.com/zfsonlinux/zfs/issues/7425#issuecomment-451702864>
[01:01:20] *** Essadon <Essadon!~Essadon@81-225-32-185-no249.tbcn.telia.com> has joined #zfsonlinux
[01:01:49] *** simukis <simukis!~simukis_@78-63-88-48.static.zebra.lt> has joined #zfsonlinux
[01:04:01] <phantomcircuit> rjvbb, i dont really like the auto loading of normal filesystems either
[01:04:11] <phantomcircuit> fs code has historically been riddled with bugs
[01:15:42] *** simukis <simukis!~simukis_@78-63-88-48.static.zebra.lt> has quit IRC (Quit: simukis)
[01:16:10] *** simukis <simukis!~simukis_@78-63-88-48.static.zebra.lt> has joined #zfsonlinux
[01:22:18] <last1> dehacked: any other clues as to what might be causing this ?
[01:30:44] <last1> figured it out
[01:30:54] <last1> after you run wipefs, reboot the machine, you can then zpool create
[01:30:56] <last1> no idea why
[01:39:11] *** mikky <mikky!~mikky@grinch.ghoul.cz> has quit IRC (Quit: WeeChat 1.0.1)
[01:46:46] <ptx0> hm, apparently my motherboard doesn't enable DDR4 power saving by default
[01:46:57] <ptx0> sooo that explains a lot about my 159W idle use
[01:47:04] <ptx0> it idles at 134W now
[01:47:47] <tlacatlc6> must be nice at winter? :D
[01:48:07] <ptx0> it's building world updates right now (gentoo) and consuming 253W
[01:48:28] <ptx0> if i overclock it to 4.0 or 4.1 GHz ( I forget ) it consumes 300 or 320W
[01:48:32] <ptx0> :P
[01:51:37] <ptx0> tlacatlc6: also i live on vancouver island
[01:51:40] <ptx0> we don't really, uh
[01:51:44] <ptx0> have winter
[01:52:08] <ptx0> it is rainy as hell though during this month
[01:54:21] <ptx0> Shinigami-Sama: any snow your way?
[01:54:37] <Shinigami-Sama> ptx0: only rain, snow next week though
[01:55:01] <ptx0> every week's forecast here is "snow... and rain" but every time, changes to "just rain"
[01:55:02] <Shinigami-Sama> have you gone up to combs yet?
[01:55:07] <ptx0> combs?
[01:55:11] <Shinigami-Sama> goats on the roof?
[01:55:18] <Shinigami-Sama> just past parksvillie
[01:55:23] <ptx0> no idea
[01:55:30] <Shinigami-Sama> then no
[01:55:37] <Shinigami-Sama> because there are "goats on the roof"
[01:56:40] *** Celmor <Celmor!~Celmor@unaffiliated/celmor> has left #zfsonlinux
[01:56:57] *** rjvbb <rjvbb!~rjvb@2a01cb0c84dee6006d1146548036a454.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 250 seconds)
[01:56:59] <tlacatlc6> i see, i'm in florida so...
[01:59:37] *** Celmor <Celmor!~Celmor@unaffiliated/celmor> has joined #zfsonlinux
[01:59:48] <Celmor> that's not good 'INFO: task txg_sync:3899 blocked for more than 120 seconds.'
[02:03:49] *** DzAirmaX <DzAirmaX!~DzAirmaX@unaffiliated/dzairmax> has quit IRC (Quit: We here br0.... xD)
[02:04:37] *** DzAirmaX <DzAirmaX!~DzAirmaX@unaffiliated/dzairmax> has joined #zfsonlinux
[02:05:59] <Shinigami-Sama> there we go, after a nights sleep I realized I should've just used disk-by-id to mirror my disks without zfs throwing a useless error at me
[02:16:53] *** djdunn <djdunn!~djdunn@fl-184-2-23-179.dhcp.embarqhsd.net> has quit IRC (Quit: Lost terminal)
[02:19:13] *** dadinn <dadinn!~DADINN@188.172.153.77> has quit IRC (Ping timeout: 245 seconds)
[02:20:52] *** djdunn <djdunn!~djdunn@fl-184-2-23-179.dhcp.embarqhsd.net> has joined #zfsonlinux
[02:28:01] *** Markow <Markow!~ejm@176.122.215.103> has quit IRC (Quit: Leaving)
[02:33:30] *** simukis <simukis!~simukis_@78-63-88-48.static.zebra.lt> has quit IRC (Quit: simukis)
[02:34:13] <ptx0> almost got some 10tb disks
[02:34:19] <ptx0> someone had a higher bid..
[02:36:21] <tlacatlc6> newegg has them for $270 atm.
[02:36:28] <ptx0> i am in .ca
[02:37:03] <ptx0> $409
[02:38:14] *** fling <fling!~user@fsf/member/fling> has joined #zfsonlinux
[02:38:28] <tlacatlc6> hmm, i see.
[02:38:46] <ptx0> that's $305 USD
[02:40:50] <ptx0> cheapest i can find is C$390 for skyhawk drives
[02:41:43] *** Essadon <Essadon!~Essadon@81-225-32-185-no249.tbcn.telia.com> has quit IRC (Quit: Qutting)
[02:46:29] <tlacatlc6> it seems newegg promo codes don't work for .ca :/
[02:53:06] *** ghormoon <ghormoon!~ghormoon@ghorland.net> has quit IRC (Remote host closed the connection)
[02:56:31] *** troyt <troyt!zncsrv@2601:681:4100:8981:44dd:acff:fe85:9c8e> has quit IRC (Quit: AAAGH! IT BURNS!)
[02:56:45] *** troyt <troyt!zncsrv@2601:681:4100:8981:44dd:acff:fe85:9c8e> has joined #zfsonlinux
[03:04:16] *** dadinn <dadinn!~DADINN@188.172.153.77> has joined #zfsonlinux
[03:44:54] *** tsp <tsp!~tsp@unaffiliated/tsp> has quit IRC (Ping timeout: 272 seconds)
[03:46:58] *** anunnaki <anunnaki!~chris@unaffiliated/anunnaki> has joined #zfsonlinux
[03:49:16] *** tsp <tsp!~tsp@unaffiliated/tsp> has joined #zfsonlinux
[03:50:07] *** ghormoon <ghormoon!~ghormoon@ghorland.net> has joined #zfsonlinux
[03:59:05] <ptx0> prologic: you have x399a board right?
[03:59:25] <prologic> hmm?
[03:59:53] <ptx0> asus x399
[04:13:40] *** EHG- <EHG-!~EHG|@unaffiliated/ehg-> has quit IRC (Quit: EHG-)
[04:25:05] *** EHG- <EHG-!~EHG|@unaffiliated/ehg-> has joined #zfsonlinux
[04:33:20] *** catalase <catalase!~catalase@unaffiliated/catalase> has quit IRC (Ping timeout: 250 seconds)
[04:33:44] *** catalase <catalase!~catalase@unaffiliated/catalase> has joined #zfsonlinux
[04:45:56] <zfs> [zfsonlinux/zfs] making sure the last quiesced txg is synced (#8239) comment by seekfirstleapsecond <https://github.com/zfsonlinux/zfs/issues/8239#issuecomment-451712748>
[04:55:14] *** DzAirmaX <DzAirmaX!~DzAirmaX@unaffiliated/dzairmax> has quit IRC (Quit: We here br0.... xD)
[04:55:45] *** DzAirmaX <DzAirmaX!~DzAirmaX@unaffiliated/dzairmax> has joined #zfsonlinux
[04:59:01] *** wlfshmn <wlfshmn!~wlfshmn@ghs/wlfshmn> has joined #zfsonlinux
[04:59:46] <zfs> [zfsonlinux/zfs] zfs.8 uses wrong snapshot names in Example 15 (#8241) created by Benjamin Gentil <https://github.com/zfsonlinux/zfs/issues/8241>
[05:08:07] <bunder> ptx0: i have a asus prime x399a
[05:10:30] <zfs> [zfsonlinux/zfs] zfs.8 uses wrong snapshot names in Example 15 (#8241) comment by bunder2015 <https://github.com/zfsonlinux/zfs/issues/8241>
[05:50:53] *** delx <delx!~delx@59.167.161.142> has quit IRC (Ping timeout: 245 seconds)
[05:52:18] *** delx <delx!~delx@59.167.161.142> has joined #zfsonlinux
[06:42:41] *** leito <leito!c38e669e@gateway/web/freenode/ip.195.142.102.158> has quit IRC (Ping timeout: 256 seconds)
[06:44:54] <zfs> [zfsonlinux/zfs] default debian install doesn't load modules until pool is created (#8240) comment by loli10K <https://github.com/zfsonlinux/zfs/issues/8240#issuecomment-451717238>
[07:10:28] *** last1 <last1!~last1@modemcable231.247-201-24.mc.videotron.ca> has quit IRC ()
[07:42:52] <ptx0> it's time to get over the my~.
[07:42:59] <ptx0> god i hate ssh sometimes
[07:47:13] <bunder> my what
[07:47:48] <pink_mist> I think that was an ssh mishap
[07:47:56] <pink_mist> he didn't mean to post that line in here at all
[07:48:14] <pink_mist> but because of ssh, things ... happened
[07:48:27] *** gerhard7 <gerhard7!~gerhard7@ip5657ee30.direct-adsl.nl> has joined #zfsonlinux
[07:51:35] <ptx0> yes
[07:51:52] <ptx0> it was the last sentence i said in #zfs without 'ths' at the end
[07:52:22] <bunder> that still doesn't answer the question :P
[07:56:02] <zfs> [zfsonlinux/zfs] default debian install doesn't load modules until pool is created (#8240) comment by bunder2015 <https://github.com/zfsonlinux/zfs/issues/8240#issuecomment-451720231>
[07:56:08] <zfs> [zfsonlinux/zfs] default debian install doesn't load modules until pool is created (#8240) closed by bunder2015 <https://github.com/zfsonlinux/zfs/issues/8240#event-2056178051>
[07:57:22] <bunder> we should have a bot that closes issues marked as duplicates
[07:58:20] <bunder> i'm half tempted to close 8175 and 8135 too
[08:00:03] <zfs> [zfsonlinux/zfs] General protection fault - userquota_updates_task (#8048) comment by bunder2015 <https://github.com/zfsonlinux/zfs/issues/8048#issuecomment-451720414>
[08:00:04] <zfs> [zfsonlinux/zfs] General protection fault - userquota_updates_task (#8048) comment by bunder2015 <https://github.com/zfsonlinux/zfs/issues/8048#issuecomment-451720414>
[08:00:06] <zfs> [zfsonlinux/zfs] General protection fault - userquota_updates_task (#8048) closed by bunder2015 <https://github.com/zfsonlinux/zfs/issues/8048#event-2056178905>
[08:04:58] *** tlacatlc6 <tlacatlc6!~tlacatlc6@68.202.46.96> has quit IRC (Quit: Leaving)
[08:14:50] *** simukis <simukis!~simukis_@78-63-88-48.static.zebra.lt> has joined #zfsonlinux
[09:08:34] *** Haxxa <Haxxa!~Harrison@180-150-30-18.NBN.mel.aussiebb.net> has quit IRC (Remote host closed the connection)
[10:00:28] *** rjvb <rjvb!~rjvb@2a01cb0c84dee6009bda76eb03bc33f7.ipv6.abo.wanadoo.fr> has joined #zfsonlinux
[10:05:50] *** Haxxa <Haxxa!~Harrison@180-150-30-18.NBN.mel.aussiebb.net> has joined #zfsonlinux
[10:49:31] *** elxa <elxa!~elxa@2a01:5c0:e09e:9ab1:2e:dfd7:18a:737c> has joined #zfsonlinux
[11:12:40] *** kim0 <kim0!uid105149@ubuntu/member/kim0> has quit IRC (Quit: Connection closed for inactivity)
[11:18:28] *** Llewelyn <Llewelyn!~derelict@184.12.106.191> has quit IRC (Ping timeout: 246 seconds)
[11:37:43] <zfs> [zfsonlinux/zfs] OpenZFS: 'vdev initialize' feature (#8230) comment by loli10K <https://github.com/zfsonlinux/zfs/issues/8230>
[11:42:41] *** cliluw <cliluw!~cliluw@unaffiliated/cliluw> has quit IRC (Ping timeout: 268 seconds)
[11:43:18] *** cliluw <cliluw!~cliluw@unaffiliated/cliluw> has joined #zfsonlinux
[12:11:49] *** kim0 <kim0!uid105149@ubuntu/member/kim0> has joined #zfsonlinux
[12:26:28] *** Markow <Markow!~ejm@176.122.215.103> has joined #zfsonlinux
[12:43:51] *** rjvbb <rjvbb!~rjvb@2a01cb0c84dee60069044dc5ee338f75.ipv6.abo.wanadoo.fr> has joined #zfsonlinux
[12:46:35] *** fp7 <fp7!~fp7@unaffiliated/fp7> has quit IRC (Remote host closed the connection)
[12:54:12] *** fp7 <fp7!~fp7@unaffiliated/fp7> has joined #zfsonlinux
[13:00:02] <prologic> ptx0 I don't use any asus boards in my servers; no
[13:00:08] <prologic> they're all super micro boards
[13:00:11] <prologic> with Xeons
[13:00:14] <prologic> why?
[13:26:00] *** b <b!coffee@gateway/vpn/privateinternetaccess/b> has quit IRC (Quit: Lost terminal)
[13:29:34] <bunder> xeon alone rules out x399 since its a tr socket ;P
[13:35:11] *** b <b!coffee@gateway/vpn/privateinternetaccess/b> has joined #zfsonlinux
[14:01:27] *** tsp <tsp!~tsp@unaffiliated/tsp> has quit IRC (Ping timeout: 252 seconds)
[14:10:06] *** Albori <Albori!~Albori@216-229-75-72.fidnet.com> has quit IRC (Ping timeout: 250 seconds)
[14:19:50] *** simukis <simukis!~simukis_@78-63-88-48.static.zebra.lt> has quit IRC (Quit: simukis)
[14:36:16] *** Albori <Albori!~Albori@216-229-75-72.fidnet.com> has joined #zfsonlinux
[14:54:59] <sphalerite> I'm using znapzend to do a backup to a removable drive, but would like to remove the drive now. Can I safely interrupt the znapzend workers and export the dataset?
[15:01:37] <DeHackEd> an in-progress send will be aborted in its entirety (if it doesn't use "zfs receive -s")
[15:28:04] *** Essadon <Essadon!~Essadon@81-225-32-185-no249.tbcn.telia.com> has joined #zfsonlinux
[15:29:02] *** Essadon <Essadon!~Essadon@81-225-32-185-no249.tbcn.telia.com> has quit IRC (Max SendQ exceeded)
[15:29:29] *** Essadon <Essadon!~Essadon@81-225-32-185-no249.tbcn.telia.com> has joined #zfsonlinux
[15:57:00] *** tlacatlc6 <tlacatlc6!~tlacatlc6@68.202.46.96> has joined #zfsonlinux
[16:01:26] <Celmor> I'm getting an "Invalid argument" error when trying to `zfs diff`: https://ptpb.pw/LhFe
[16:01:29] *** shibboleth <shibboleth!~shibbolet@gateway/tor-sasl/shibboleth> has joined #zfsonlinux
[16:05:27] *** simukis <simukis!~simukis_@78-63-88-48.static.zebra.lt> has joined #zfsonlinux
[16:07:09] *** shibboleth <shibboleth!~shibbolet@gateway/tor-sasl/shibboleth> has quit IRC (Remote host closed the connection)
[16:07:30] *** shibboleth <shibboleth!~shibbolet@gateway/tor-sasl/shibboleth> has joined #zfsonlinux
[16:58:06] *** shibboleth <shibboleth!~shibbolet@gateway/tor-sasl/shibboleth> has quit IRC (Quit: shibboleth)
[17:04:32] <PMT> Celmor: #7678 #6335 #5427 #5426
[17:04:36] <zfs> [zfs] #7678 - ZFS diff - Unable to determine path or stats for object <https://github.com/zfsonlinux/zfs/issues/7678>
[17:09:10] <Celmor> alright, was wondering if I invoked it wrong
[17:11:19] *** chasmo77 <chasmo77!~chas77@158.183-62-69.ftth.swbr.surewest.net> has quit IRC (Quit: It's just that easy)
[17:12:29] <PMT> no
[17:12:47] <PMT> IIRC that particular message comes from a multiply-linked file having its original "parent" object removed
[17:13:45] <PMT> e.g. if you have hardlinks 1 and 2 to file1, since the object only has a single parent field, it can no longer walk it up to find the parent. (I believe.)
[17:14:02] <PMT> rather, hardlinks 1 and 2, with 1 being in the parent field, then remove 1
[17:16:30] <Markow> The scenery on my mountaintop property today, around 2 1/2 hours ago: https://pbs.twimg.com/media/DwPBjZdWoAENwHe.jpg
[17:16:37] <Markow> I use ZFS on Linux here ;)
[17:55:26] *** MilkmanDan <MilkmanDan!~dan@wilug/expat/MilkmanDan> has quit IRC (Ping timeout: 250 seconds)
[18:01:18] *** MilkmanDan <MilkmanDan!~dan@wilug/expat/MilkmanDan> has joined #zfsonlinux
[18:03:16] <zfs> [zfsonlinux/zfs] feature request: zpool iostat N should repeat header like vmstat N (#8235) comment by Damian Wojs?aw <https://github.com/zfsonlinux/zfs/issues/8235#issuecomment-451757206>
[18:07:57] <elxa> Markow: you got snow, we got rain :(
[18:08:06] <zfs> [zfsonlinux/zfs] OpenZFS: 'vdev initialize' feature (#8230) comment by Tim Chase <https://github.com/zfsonlinux/zfs/issues/8230>
[18:08:07] *** f_g <f_g!~f_g@213-47-131-124.cable.dynamic.surfer.at> has quit IRC (Ping timeout: 268 seconds)
[18:21:35] *** f_g <f_g!~f_g@213-47-131-124.cable.dynamic.surfer.at> has joined #zfsonlinux
[18:33:22] <Markow> elxa: More snow. :) I must go and fire up the wood stove! https://pbs.twimg.com/media/DwPAAwzX4AADkwg.jpg
[18:35:54] *** tsp <tsp!~tsp@unaffiliated/tsp> has joined #zfsonlinux
[18:44:29] *** Markow <Markow!~ejm@176.122.215.103> has quit IRC (Quit: Leaving)
[19:08:14] *** Albori <Albori!~Albori@216-229-75-72.fidnet.com> has quit IRC (Ping timeout: 250 seconds)
[19:25:50] *** mmlb <mmlb!~mmlb@76-248-148-178.lightspeed.miamfl.sbcglobal.net> has quit IRC (Ping timeout: 258 seconds)
[19:27:00] *** mmlb <mmlb!~mmlb@76-248-148-178.lightspeed.miamfl.sbcglobal.net> has joined #zfsonlinux
[19:33:41] *** Albori <Albori!~Albori@216-229-75-72.fidnet.com> has joined #zfsonlinux
[19:55:13] <zfs> [zfsonlinux/zfs] Python 2 and 3 compatibility (#8096) closed by Brian Behlendorf <https://github.com/zfsonlinux/zfs/issues/8096#event-2056407241>
[19:58:04] <zfs> [zfsonlinux/zfs] zfs.8 uses wrong snapshot names in Example 15 (#8241) comment by Brian Behlendorf <https://github.com/zfsonlinux/zfs/issues/8241>
[20:16:02] *** Markow <Markow!~ejm@176.122.215.103> has joined #zfsonlinux
[20:23:20] *** mmlb <mmlb!~mmlb@76-248-148-178.lightspeed.miamfl.sbcglobal.net> has quit IRC (Ping timeout: 258 seconds)
[20:23:47] *** mmlb <mmlb!~mmlb@76-248-148-178.lightspeed.miamfl.sbcglobal.net> has joined #zfsonlinux
[20:31:00] *** mmlb <mmlb!~mmlb@76-248-148-178.lightspeed.miamfl.sbcglobal.net> has quit IRC (Ping timeout: 258 seconds)
[20:31:37] *** mmlb <mmlb!~mmlb@76-248-148-178.lightspeed.miamfl.sbcglobal.net> has joined #zfsonlinux
[20:40:25] <rjvb> what would be faster and what more flexible, a non-ZFS FS in a sparse file on a dataset or the same FS on a sparse zvol inheriting that same dataset?
[20:41:34] <zfs> [zfsonlinux/zfs] zfs.8 uses wrong snapshot names in Example 15 (#8241) comment by George Melikov <https://github.com/zfsonlinux/zfs/issues/8241>
[20:48:02] <gchristensen> rjvb: what about a vdev?
[20:48:37] <rjvb> maybe I'm mixing up vdev and zvol?
[20:49:51] <ptx0> i don't think gchristensen read your question correctly.
[20:50:15] <ptx0> but your question doesn't make any fuckin sense anyway
[20:50:25] <ptx0> how can a zvol inherit a dataset, what does that even mean
[20:50:27] <rjvb> yeah, vdev is an actual device, right? If so then no
[20:50:52] <rjvb> bear with me, I never used zvols
[20:51:11] <rjvb> so just read "with the same property set as the dataset"
[20:51:54] *** apekatten <apekatten!~apekatten@unaffiliated/apekatten> has quit IRC (Quit: no reason)
[20:52:52] <zfs> [zfsonlinux/zfs] OpenZFS: 'vdev initialize' feature (#8230) new review comment by Brian Behlendorf <https://github.com/zfsonlinux/zfs/pull/8230#discussion_r245520073>
[20:52:57] <zfs> [zfsonlinux/zfs] OpenZFS: 'vdev initialize' feature (#8230) new review comment by Brian Behlendorf <https://github.com/zfsonlinux/zfs/pull/8230#discussion_r245520078>
[20:53:19] <zfs> [zfsonlinux/zfs] OpenZFS: 'vdev initialize' feature (#8230) new review comment by Brian Behlendorf <https://github.com/zfsonlinux/zfs/pull/8230#discussion_r245520090>
[20:54:14] <zfs> [zfsonlinux/zfs] OpenZFS: 'vdev initialize' feature (#8230) comment by Brian Behlendorf <https://github.com/zfsonlinux/zfs/issues/8230#issuecomment-451769528>
[21:07:58] <ptx0> rjvb: the answer should be very obvious to you
[21:13:52] <PMT> rjvb: a vdev is a group of underlying disks in a particular configuration on a pool, e.g. if you had 3 raidz3s of 6 disks each in one pool, each raidz3 is a vdev. so a vdev is probably not what you mean.
[21:14:14] <PMT> Also, in what particular ways do you mean "faster" or "flexible" here?
[21:16:28] <ptx0> he wants it to be able to touch its toes.
[21:22:05] <rjvb> what I'm looking for is in fact a way to get a database made of countless mostly tiny files into a single file. Anywhere else I'd just use a disk image (combined with a filesystem that remains efficient for small files)
[21:22:40] <rjvb> but ZFS gives another option
[21:22:47] <ptx0> use innodb
[21:23:11] <DeHackEd> indeed
[21:23:22] <rjvb> no, that's not an option, I cannot change anything at that level
[21:23:24] <ptx0> and why would you want that, in the first place
[21:23:25] <DeHackEd> for truly small sizes (eg: less than 100 bytes) an actual database might be better.
[21:23:32] <ptx0> if you use #5182 the many small files are a benefit
[21:23:38] <zfs> [zfs] #5182 - Metadata Allocation Classes by don-brady <https://github.com/zfsonlinux/zfs/issues/5182>
[21:23:39] <ptx0> one large sparse file will trash your performance
[21:23:52] <DeHackEd> embedded data might help, but the overhead is going to suck
[21:24:23] <ptx0> DeHackEd: i can't wait for my new x79 motherboard to show up so i can rebuild my storage system with the WD Black NVMe special vdev
[21:24:57] <ptx0> trying to zfs recv 3 streams to that box while my phone is backing up to CIFS and i'm trying to mv files to archive over NFS at 300MiB/s really makes the access times suffer
[21:25:35] <ptx0> mostly because it has 6x 5300-ish rpm SMR HDDs
[21:25:45] <ptx0> or 5980rpm, can't remember
[21:25:56] <rjvb> the file won't get very large; the whole collection is about 100Mb (real, not on-disk bytes)
[21:26:09] <ptx0> rjvb: the point still stands, "large" is >128kB
[21:26:18] *** shibboleth <shibboleth!~shibbolet@gateway/tor-sasl/shibboleth> has joined #zfsonlinux
[21:27:07] <rjvb> so a sparse zvol could be faster? I don't want to create a fixed-size "container"
[21:27:17] <ptx0> you're really just "Doing It Wrong"
[21:27:25] <ptx0> that's all
[21:27:36] <ptx0> just keep the small files on ZFS and don't do any monkey business
[21:27:40] <ptx0> that is best.
[21:28:16] <ptx0> are you doing this for backup reasons?
[21:28:33] <ptx0> like, why do you want 'many small files to appear as one'
[21:29:22] <rjvb> no, perfomance, and overhead (on-disk size if more than 2x the actual size)
[21:31:28] <rjvb> my hunch is that it's the countless little files aspect which makes certain operations slow (or victim of high latencies)
[21:33:16] <rjvb> I'll just go see how the other options I have behave, starting with a sparse image file (easiest and the least risky)
[21:34:35] *** fs2 <fs2!~fs2@pwnhofer.at> has quit IRC (Quit: Ping timeout (120 seconds))
[21:36:00] <ptx0> look at #5182 though
[21:36:05] <zfs> [zfs] #5182 - Metadata Allocation Classes by don-brady <https://github.com/zfsonlinux/zfs/issues/5182>
[21:36:09] <ptx0> you can use SSDs for metadata which makes this a non-issue
[21:36:22] <ptx0> i don't think putting all the small files into a single 'container' will make it any faster; the opposite..
[21:37:00] <ptx0> like, why would you assume that it would increase performance in any way?
[21:38:00] <ptx0> you're still dealing with the same number of small files because you didn't fix the actual problem by changing your DB structure or use some SSDs for metadata
[21:38:17] <ptx0> now there is simply a layer of indirection to add more context switching, memory copying, and other stupid shit
[21:39:18] <ptx0> trying to hide your file structure from ZFS won't make it work better, no way.
[21:49:06] *** obadz <obadz!~obadz@unaffiliated/obadz> has joined #zfsonlinux
[21:50:37] <obadz> (Never used ZFSOL) — Is there a way to have my home directory be encrypted with the secret key encrypted my login password and transparently decrypted/mounted when I log in (the way it works with ecryptfs) ?
[21:56:31] <zfs> [zfsonlinux/zfs] Ensure third party licenses are included in dist tarballs (#8242) created by Neal Gompa (?????????) <https://github.com/zfsonlinux/zfs/issues/8242>
[22:12:48] <CompanionCube> out of the box? nope
[22:13:09] <CompanionCube> sounds doable with some work though
[22:15:26] *** shibboleth <shibboleth!~shibbolet@gateway/tor-sasl/shibboleth> has quit IRC (Quit: shibboleth)
[22:16:57] <zfs> [zfsonlinux/zfs] pyzfs: Use the ZFS version for pyzfs and remove unused requirements.txt (#8243) created by Neal Gompa (?????????) <https://github.com/zfsonlinux/zfs/issues/8243>
[22:17:57] <obadz> can non-root users decrypt/mount directories? ?
[22:20:57] <DeHackEd> a limitation of linux is that only administrative/privileged apps can mount filesystems
[22:21:30] <obadz> works fine for ecryptfs and fuse filesystems?
[22:21:45] <DeHackEd> with a setuid helper (/usr/sbin/fusermount)
[22:21:50] *** gerhard7 <gerhard7!~gerhard7@ip5657ee30.direct-adsl.nl> has quit IRC (Quit: Leaving)
[22:21:51] <DeHackEd> zfs doesn't have that
[22:23:15] <obadz> yes ecryptfs also has a setuid helper
[22:26:17] *** pheoxy <pheoxy!~pheoxy@118.209.170.254> has joined #zfsonlinux
[22:28:33] *** Setsuna-Xero <Setsuna-Xero!~pewpew@unaffiliated/setsuna-xero> has joined #zfsonlinux
[22:30:42] *** PewpewpewPantsu <PewpewpewPantsu!~pewpew@unaffiliated/setsuna-xero> has quit IRC (Ping timeout: 246 seconds)
[22:57:10] *** ih8wndz <ih8wndz!jwpierce3@001.srv.trnkmstr.com> has quit IRC (Quit: WeeChat 2.3)
[23:00:13] *** ih8wndz <ih8wndz!jwpierce3@001.srv.trnkmstr.com> has joined #zfsonlinux
[23:11:55] *** elxa <elxa!~elxa@2a01:5c0:e09e:9ab1:2e:dfd7:18a:737c> has quit IRC (Ping timeout: 252 seconds)
[23:17:16] *** pheoxy <pheoxy!~pheoxy@118.209.170.254> has quit IRC (Remote host closed the connection)
[23:49:50] *** adilger <adilger!~adilger@S0106a84e3fe4b223.cg.shawcable.net> has joined #zfsonlinux
[23:55:23] *** BtbN <BtbN!btbn@ffmpeg/developer/btbn> has quit IRC (Remote host closed the connection)
[23:58:10] *** BtbN <BtbN!btbn@ffmpeg/developer/btbn> has joined #zfsonlinux
top
   January 6, 2019  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >