Switch to DuckDuckGo Search
   January 29, 2012  
< | 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
[00:00:55] <Triskelios> ZFS doesn't need partitions
[00:01:20] <DanaG> Well, I'll be hosting the filesystems in partitions, rather than the whole drives.
[00:01:42] <Triskelios> uh, why?
[00:01:57] <DanaG> So I can install ubuntu alongside it, on the same drives.
[00:02:08] <DanaG> And not on zfs root.
[00:02:33] <ismell> I got a weird situation, if i save a file via smb then switch to a linux box and check the file it's still the old file. it takes a few seconds for the new file to be updated
[00:02:49] <ismell> is something messed with my config ?
[00:03:41] <Triskelios> ismell: client-side caching somewhere
[00:04:23] <ismell> hrmm, could it be windows or you think it's the actual app ?
[00:05:35] <Triskelios> if it's using the Windows SMB support, probably Windows
[00:06:56] <ismell> cool thx
[00:07:02] <ismell> now i gotta figure out if i can disable it
[00:20:26] <RobbieCrash> Triskelios, sorry for the delay, it's in a secondary data pool, not the root pool.
[00:55:13] *** wiseen has joined #openindiana
[00:59:09] *** laogrymt has quit IRC
[01:20:56] *** raichoo has quit IRC
[01:40:15] *** Licenser has quit IRC
[02:11:13] *** DanaG has quit IRC
[02:12:09] *** InTheWings has quit IRC
[02:17:10] *** DanaG has joined #openindiana
[02:22:29] *** DanaG has quit IRC
[02:25:21] *** bongo has quit IRC
[02:25:29] *** bongo has joined #openindiana
[02:26:42] *** easy-iPad has joined #openindiana
[02:30:08] *** stoxx has quit IRC
[02:35:53] *** stoxx has joined #openindiana
[02:51:47] *** easy-iPad has quit IRC
[03:02:51] <gwr> linux smb client does client-side caching, similar to the nfs client.
[03:05:11] <horsi> can anyone assist dianosing a system crash? memtest86+ 4.2 for 26 hours no problems at all
[03:09:29] <RobbieCrash> Probably not RAM then.
[03:11:00] <horsi> is there an easy way to dump a log or something as it crashes?
[03:11:52] <RobbieCrash> I'm not positive, but anything that can be logged should be in /var/adm
[03:12:08] <RobbieCrash> check messages
[03:13:37] <horsi> nothing but mDNSResponder error's - getOptRdata - Unknown opt 4
[03:13:48] <horsi> but that message starts up the second I boot the system
[03:15:41] <RobbieCrash> how frequent is the crash?
[03:16:06] *** floogy has quit IRC
[03:18:34] *** DanaG has joined #openindiana
[03:18:51] <horsi> random - sometime a week othertimes a month
[03:21:56] <horsi> dumpadm is enabled, and theres nothing in the /var/crash directory
[03:29:49] <DanaG> (6:24:48 PM) DanaG: oh yeah, how do you set case-sensitivity and "nbmand" on the file system during install time?
[03:29:50] <DanaG> (6:25:00 PM) DanaG: Last time I tried it, I had to delete and re-create my home dir.
[03:29:50] <DanaG> (6:29:28 PM) DanaG: great, so the desktop DVD takes me to... a console login prompt.
[03:30:06] <DanaG> what's the default password? And why is there no desktop... on the desktop CD?
[03:30:09] <DanaG> Oh wait, there it goes.
[03:32:15] <Wraithhh> horsi, is savecore also enabled?
[03:32:42] <Wraithhh> (check with dumpadm)
[03:33:18] <horsi> nope :) would help - is that a flag to enable it?
[03:34:05] <Wraithhh> dumpadm -y should work
[03:34:30] <horsi> thankyou - when it happens next whats the best file to check?
[03:35:07] <Wraithhh> horsi, I'm at that stage as well :-)
[03:35:40] <Wraithhh> do you know if anything specific was going on during the crash?
[03:36:27] <horsi> not that i know of - the sytem is in testing at the moment, just runs irc for me, has nappit and afp installed. Rpool is mirrored over two disks
[03:37:13] <horsi> smart stuff is all ok - nappit has a few jobs running (smart check, pool status,etc,etC) but nothing around those times - and the time of day/ day of the month isnt consistent
[03:42:04] <horsi> when it hangs/crashes the system is still running/powered up but console, network everything is unresponsive
[03:43:31] <Patrickdk> hmm, I can seem to get ipmp and vlans to work together
[03:49:17] *** nija has joined #openindiana
[03:49:27] <nija> Hi, is there any easy way to include the definition for znode_t for a (userland) program? I'm using kvm_kread() and want to follow some vnode's v_data for zfs-files...
[03:50:48] *** gea has quit IRC
[03:57:17] *** joffe has quit IRC
[04:00:02] *** Savis_ has joined #openindiana
[04:01:37] <DanaG> say, how do I pass a fixed "ashift" value to the OI installer?
[04:02:06] <DanaG> I'm installing to an Advanced Format drive that lies about its block size, on both Physical and Logical fields.
[04:03:36] <Wraithhh> DanaG, there's a patched version of zpool to do that
[04:03:43] <Wraithhh> don't know how to use it with the installer though
[04:03:51] *** OMV-User has quit IRC
[04:03:56] *** [gorgo] has quit IRC
[04:03:56] *** CentralStates has quit IRC
[04:03:56] *** Savis has quit IRC
[04:04:09] <DanaG> ah, I can probably overwrite the system "zpool" to do so.
[04:04:36] *** [gorgo] has joined #openindiana
[04:06:12] <DanaG> https://www.illumos.org/issues/828
[04:06:18] *** DanaG has quit IRC
[04:07:59] *** OMV-User has joined #openindiana
[04:11:11] *** CentralStates has joined #openindiana
[04:33:02] <RobbieCrash> horsi I've noticed huge CPU usage for afpd after a few days, constantly pegging one of my cores. I've taken to restarting the service every other night, seems to have stopped afp problems for me. But, my whole system isn't crashing, just apple connections
[04:33:26] <horsi> yeh i havent got any disks in the system so theres no live data happening
[04:33:37] <horsi> are you just using top to check cpu usage?
[04:33:37] <RobbieCrash> actually, now that I think of it, I have had two times where my system locked up completly. But I was playing around with my NIC settings, so thought it was that
[04:33:47] <RobbieCrash> horsi, yeah or iostat
[04:34:25] <Triskelios> nija: better question for #illumos, really
[04:40:04] <RobbieCrash> Can someone give me a hint on a scripting issue? I want to create a folder that has links to all media added in the last week to two directories. I've got find finding them, and know how to create the symlinks, but I know very little of scripting, so I don't know how to pipe the grep'd output to ln to link them. Or am I doing this totally wrong?
[04:40:36] <RobbieCrash> I could do it in Python, but I'm sick of doing things in Python when I should be able to do them in the shell
[04:44:30] <RobbieCrash> nevermind, rephrased my query at stackoverflow and found exactly what I need
[04:47:13] *** nija has quit IRC
[04:56:44] *** wiseen has quit IRC
[05:01:22] *** alanc has quit IRC
[05:13:00] *** easy-iPad has joined #openindiana
[05:24:50] *** tsoome_ has joined #openindiana
[05:26:30] *** tsoome has quit IRC
[05:26:30] *** tsoome_ is now known as tsoome
[05:29:44] *** enmand has quit IRC
[05:32:34] *** easy-iPad has quit IRC
[05:33:14] <RobbieCrash> I've checked the sharemgr man page, but can't figure out how to make ZFS shares over cifs follow symlinks and allow cross device linking. Setting this in Linux is done with 'follow symlinks=yes' and
[05:33:49] <RobbieCrash> 'wide links=yes' in smb.conf, but since there isn't one in OI, how can I set this? Or is this something that's applicable to ZFS more than just OI?
[05:39:00] <richlowe> it's the native smb server v. samba
[05:39:08] <richlowe> use samba, and the regular config probably will work
[05:40:39] <RobbieCrash> So I shouldn't just be using the zfs sharesmb options to share files/folders? Maybe I'm just going about this incorrectly, what I want to do is create a list of all new files, have that shared in a folder called 'new' and then from Windows, just click on the file and have it open the linked file.
[05:54:13] *** jrapdx has joined #openindiana
[06:32:26] *** ira has quit IRC
[06:54:14] *** flyz_ has quit IRC
[06:55:15] *** flyz has joined #openindiana
[07:26:40] *** monta-` has joined #openindiana
[07:39:14] *** OMV-User has quit IRC
[07:40:29] *** OMV-User has joined #openindiana
[08:15:59] *** _404_Error_ has quit IRC
[08:56:34] *** tsukasa has quit IRC
[08:59:51] *** horsi has quit IRC
[09:00:16] *** horsi has joined #openindiana
[09:04:54] *** tsukasa has joined #openindiana
[09:04:55] *** tsukasa has joined #openindiana
[09:11:24] *** kart_ has joined #openindiana
[09:26:09] *** wilsonkk has quit IRC
[09:27:19] *** d_k has joined #openindiana
[09:38:31] *** wilsonkk has joined #openindiana
[09:39:28] *** jrapdx has quit IRC
[09:40:15] *** d_k has quit IRC
[10:09:06] *** Botanic has quit IRC
[10:10:40] *** kart_ has quit IRC
[10:21:31] *** d_k has joined #openindiana
[10:26:01] *** d_k has quit IRC
[10:40:52] *** gea has joined #openindiana
[10:49:13] *** flyz has quit IRC
[10:49:15] *** Darkproger has quit IRC
[10:49:52] *** flyz has joined #openindiana
[10:52:01] *** jw_urodoc_ has joined #openindiana
[10:54:42] *** Darkproger has joined #openindiana
[10:58:22] *** merzo has quit IRC
[11:03:04] *** easy-iPad has joined #openindiana
[11:05:47] *** _404_Error_ has joined #openindiana
[11:08:02] <jk4> what does ATI support look like these days?
[11:11:34] *** flyz has quit IRC
[11:17:24] *** flyz has joined #openindiana
[11:24:46] *** master_of_master has quit IRC
[11:26:02] *** master_of_master has joined #openindiana
[11:27:02] *** gea has quit IRC
[11:29:50] *** girafe3 has joined #openindiana
[11:30:28] *** gea has joined #openindiana
[11:31:27] *** easy-iPad has quit IRC
[11:32:59] *** girafe has quit IRC
[11:33:17] *** InTheWings has joined #openindiana
[11:33:56] *** girafe has joined #openindiana
[11:34:39] *** girafe3 has quit IRC
[11:35:17] *** gea has quit IRC
[12:04:37] *** gea has joined #openindiana
[12:07:04] *** lgtaube has joined #openindiana
[12:16:28] *** gea has quit IRC
[12:33:34] *** monta-` has quit IRC
[12:36:49] *** monta-` has joined #openindiana
[12:38:29] *** smrt has quit IRC
[12:38:47] *** smrt has joined #openindiana
[12:40:35] *** kart_ has joined #openindiana
[13:06:06] *** merzo has joined #openindiana
[13:18:26] *** jw_urodoc_ has quit IRC
[13:40:57] *** philip__ has joined #openindiana
[13:42:28] <philip__> hello, I am trying to add spanish kbd support but when i go systems/prefs/iiim-input-methods i get a blank list selection on "available language/scripts"
[13:42:36] <philip__> is there any package i should be adding
[13:42:47] <philip__> i have es.locale already installed
[13:43:03] <philip__> i have been following this guide: http://137.254.16.27/sunwg11nprg/entry/switching_keyboard_layout_in_solaris
[13:44:57] *** gea has joined #openindiana
[13:45:30] *** jedi4ever has quit IRC
[13:55:59] *** Seony has quit IRC
[13:59:10] *** gea has quit IRC
[13:59:15] *** kenon has quit IRC
[13:59:43] *** kenon has joined #openindiana
[14:16:55] *** _404_Error_ has quit IRC
[14:16:55] *** _404_Error_ has joined #openindiana
[14:20:17] *** brougham has joined #openindiana
[14:35:49] *** Botanic has joined #openindiana
[14:35:49] *** Botanic has joined #openindiana
[14:39:19] *** merzo has quit IRC
[14:40:02] *** jw_urodoc_ has joined #openindiana
[14:50:35] *** raichoo has joined #openindiana
[15:02:30] *** kart_ has quit IRC
[15:02:56] *** kart_ has joined #openindiana
[15:03:31] *** jedi4ever has joined #openindiana
[15:08:49] *** jw_urodoc_ has quit IRC
[15:23:30] *** kart_ has quit IRC
[15:24:23] *** kart_ has joined #openindiana
[15:28:29] *** jw_urodoc has quit IRC
[15:29:10] *** jw_urodoc has joined #openindiana
[15:29:10] *** raichoo has quit IRC
[15:50:17] *** enmand has joined #openindiana
[15:55:42] *** joffe has joined #openindiana
[15:55:59] *** kart_ has quit IRC
[15:56:59] *** kart_ has joined #openindiana
[16:03:00] *** kenon_ has joined #openindiana
[16:06:34] *** kenon has quit IRC
[16:08:29] *** kart_ has quit IRC
[16:09:12] *** kart_ has joined #openindiana
[16:13:29] *** philip__ has quit IRC
[16:18:56] *** kenon_ has quit IRC
[16:20:24] *** enmand has quit IRC
[16:20:30] *** kenon has joined #openindiana
[16:26:06] *** kenon_ has joined #openindiana
[16:29:23] *** kenon has quit IRC
[16:29:26] *** magyar has joined #openindiana
[16:29:26] *** magyar has joined #openindiana
[16:32:23] *** kart_ has quit IRC
[16:37:45] *** d_k has joined #openindiana
[16:42:23] *** d_k has quit IRC
[16:56:47] *** kenon_ has quit IRC
[16:58:49] *** kenon has joined #openindiana
[17:00:36] *** codestr0m has joined #openindiana
[17:04:57] *** codestr0m has left #openindiana
[17:10:27] *** gea has joined #openindiana
[17:22:26] *** gea has quit IRC
[17:27:10] *** Botanic has quit IRC
[17:27:16] *** Licenser has joined #openindiana
[17:40:11] *** bunyip has joined #openindiana
[17:41:57] *** merzo has joined #openindiana
[17:48:47] *** kenon_ has joined #openindiana
[17:50:51] *** DanaG has joined #openindiana
[17:51:05] *** kenon has quit IRC
[17:53:29] *** dijenerate has quit IRC
[17:57:39] *** merzo has quit IRC
[17:57:54] *** dijenerate has joined #openindiana
[18:10:20] *** girafe has quit IRC
[18:22:45] *** DanaG has quit IRC
[18:25:05] *** kenon has joined #openindiana
[18:27:50] *** kenon_ has quit IRC
[18:31:42] *** DanaG has joined #openindiana
[18:31:45] *** kolle has joined #openindiana
[18:33:21] *** kolle has left #openindiana
[18:41:36] *** girafe has joined #openindiana
[18:43:26] *** bunyip has left #openindiana
[18:43:44] *** kenon has quit IRC
[18:45:25] *** alanc has joined #openindiana
[18:45:25] *** ChanServ sets mode: +o alanc
[18:45:35] *** kenon has joined #openindiana
[19:07:51] *** gea has joined #openindiana
[19:12:39] *** enmand has joined #openindiana
[19:14:53] <DrLou_> wizards - have been testing out some zfs/zpool syntax on a file-based zpool, but getting 'invalid argument' errors...
[19:15:16] <patdk-lap> file based zpool?
[19:15:25] <DrLou_> (is it even possible to create a new BE on a file-based pool - really only want to do some testing)
[19:15:32] <DrLou_> patdk-lap
[19:15:40] <DrLou_> (yes, I meant... )
[19:16:15] <DrLou_> wanted to 'measure twice, cut once' !!
[19:17:08] <patdk-lap> how many real options are there?
[19:17:26] <patdk-lap> mirror or raidz, and everything else is adjustable after the fact
[19:17:31] *** IDoNotKnow is now known as IDoNotCare
[19:18:15] <DrLou_> well, am trying to create the BE from a snapshot of the existing root pool - error message is not terribly helpful
[19:18:23] <DrLou_> not even worrying about the mirror at this point.
[19:18:37] <DanaG> hmm, does OI do PulseAudio?
[19:21:54] *** Botanic has joined #openindiana
[19:22:19] <DanaG> oh, and how do I check the value of "ashift"?
[19:23:04] <DanaG> argh, that installer really needs an option to make casesensitivty=mixed.
[19:23:42] <Triskelios> PA is in SFE but not generally used. zpool get ashift
[19:25:56] * Patrickdk still isn't sure about ashift
[19:26:14] <Patrickdk> zfs checksums and 4k disks really don't get along well :(
[19:28:17] *** jurikm has joined #openindiana
[19:29:11] <Triskelios> Patrickdk: uh, there isn't any relationship
[19:29:41] *** pasik has quit IRC
[19:31:12] <Patrickdk> sure there is
[19:31:30] <Patrickdk> expecially if your using small files
[19:31:53] <Triskelios> no
[19:32:02] <Patrickdk> or like doing iscsi, and using 8k sectors
[19:32:23] <Patrickdk> 8k + checksum, uses 3 sectors, or 12k of diskspace for 8k of data
[19:32:28] *** dajhorn has quit IRC
[19:32:46] <Triskelios> ashift affects the alignment of allocations for blocks, the sizes of the blocks don't matter
[19:32:55] <Patrickdk> yes
[19:32:59] <Triskelios> checksumming doesn't care about that
[19:33:07] <Patrickdk> instead of allocating 512bytes for checksum info, it allocates 4k
[19:33:50] <Triskelios> wrong
[19:33:59] <Triskelios> the checksums are always the same
[19:34:07] <Patrickdk> yes
[19:34:13] <Patrickdk> but it has to put the checksum on disk
[19:34:22] <Patrickdk> and it allocates a sector to do so
[19:34:31] <Patrickdk> and the smallest sector is then 4k
[19:35:03] *** Botanic has quit IRC
[19:35:36] <Triskelios> no, it allocates a block pointer, which is always 128 bytes
[19:37:51] <Triskelios> the DMU object for the block pointer may have an alignment requirement, but the size doesn't change
[19:45:51] *** sammi` has joined #openindiana
[19:48:48] *** pasik has joined #openindiana
[19:51:16] <DrLou_> Patrickdk: it's the syntax of the -e option to beadm create which is tripping me up.
[19:51:28] <DrLou_> shoudl any root pool snap be a candidate for that option?
[19:55:55] <WebDawg> I have a bunch of data shared out on my OI server va samba. I can copy files to it and use it just fine but when I run a command on the share if the command produces output the command cannot write to the share.
[19:56:09] <WebDawg> Any ideas?
[19:56:22] *** vcg has joined #openindiana
[19:56:23] <DrLou_> the command's execution privs/
[19:56:43] <vcg> is it possible to monitor drive failures via snmp with oi?
[19:57:05] <DanaG> looks like "zpool get ashift rpool" is just giving me the list of known properties... and "ashift" is not included.
[19:57:19] *** sammi` has quit IRC
[19:57:19] <DrLou_> vcg: one of our wizards here is working on exactly that...
[19:57:43] <vcg> well I joined at the right time :)
[19:57:59] <DrLou_> never a 'wrong' time... !
[19:58:19] <DrLou_> (some very smart guys here...)
[19:58:29] *** jrapdx has joined #openindiana
[19:58:46] <vcg> I'm essentially looking for the ability to monitor the health and performance of zfs - i'm looking at nexenta as well but I like to idea of using oi a little better
[19:59:25] <DrLou_> which suddenly makes _you_ one of the smart guys... (sic)
[19:59:52] <Triskelios> vcg: FMA can report via SNMP, I don't know the details for configuring that
[20:00:15] <vcg> thanks triskelios
[20:00:30] <Triskelios> DanaG: sorry, forgot that wasn't exposed as a property. might have to 'zdb rpool'
[20:00:49] <vcg> how about drive failure indicators on the shelves themselves - I'm assuming that's ses related - is that functional in oi?
[20:00:51] *** Botanic has joined #openindiana
[20:01:00] <DrLou_> smartctl?
[20:01:50] <WebDawg> ??????????
[20:03:35] *** merzo has joined #openindiana
[20:05:23] <RoyK> vcg: fmtopo, perhaps
[20:05:34] <RoyK> vcg: /usr/lib/fm/fmd/fmtopo
[20:05:50] <RoyK> WebDawg: !!!!!!!!!!
[20:06:11] <RoyK> vcg: or perhaps not....
[20:07:07] <vcg> You will have to forgive me ... I'm not near as knowledgeable about oi/os as I am with windows
[20:07:52] <vcg> I'm just trying to build up a shared storage infrastructure for our xenserver project and everything is pointing me at zfs
[20:09:30] <DanaG> what's the deal with OI using autofs for local home dirs?
[20:09:37] <DanaG> Doesn't that just add a layer of complication?
[20:10:31] <WebDawg> heh
[20:13:04] <DanaG> hmm, I wonder if I should set "normalization" or "utf8only"?...
[20:19:26] <WebDawg> RoyK, %%%%%%%%%
[20:21:27] <RoyK> WebDawg: :þ
[20:37:44] <DanaG> hmm, trying to add a partition on a secondary drive, as a vdev...
[20:37:54] <DanaG> napp-it says "there is no datapool available" on the "add vdev" page.
[20:38:39] *** kenon_ has joined #openindiana
[20:39:57] <DanaG> I really feel lost in OI partitioning.
[20:40:19] <DanaG> I can comprehend sda1, but c0t0d0s0 through c0t0d0s2 through c0t2d0s0?
[20:40:42] <DanaG> And there's no such thing as "sda" alone -- just a virtual partition that represents the whole disk? Way confusing.
[20:40:55] <Patrickdk> heh?
[20:41:08] <Patrickdk> guess you never understood bsd partitioning?
[20:41:18] <Patrickdk> s = slice, like a partition kindof
[20:41:27] <DanaG> yeah, but you can have slices within a partition.
[20:41:29] <Patrickdk> s2 = whole disk basically
[20:41:44] <Patrickdk> you can have anything inside anything
[20:41:56] <Patrickdk> you don't have to have slices inside a partition, I never did in bsd
[20:41:59] *** kenon has quit IRC
[20:42:06] <Patrickdk> but your bios expects partitions though, not slices, normally
[20:42:37] <DanaG> When I installed, I left room for non-ZFS stuff at the start, so I think I put it in a partition.
[20:42:48] <DanaG> Left 40 gigs worth for Linux partitions.
[20:43:39] <DanaG> That's so weird to have a slice represent the whole disk. I certainly wouldn't have designed it that way.... I'd just have there be a c2d0 with no 'p' or 's'.
[20:44:16] <Patrickdk> well, slices where invented long before
[20:44:27] <Patrickdk> so everyone using slices is used to 2 being the whole thing
[20:44:42] <DanaG> Still, it would've seemed just as weird to me then.
[20:45:37] <DanaG> oh yeah, how tolerant is ZFS of slight differences in drive sizes?
[20:45:37] <DanaG> Like, some 2TB drives may have a few fewer megabytes than other 2TB drives.
[20:46:00] <Patrickdk> hmm, it's always been that way
[20:46:08] <Patrickdk> even when we only had 10meg drives
[20:46:51] <RoyK> DanaG: not really, they standardised that a few years back
[20:47:12] <RoyK> anything >200GB or so should be same size
[20:47:41] <RoyK> all 2TB drives I've seen so far, has had the exact same number of sectors
[20:47:54] <DanaG> Including both Samsung and, say, WD?
[20:48:15] <RoyK> haven't tested samsung, but a few WD and Hitachis
[20:48:17] <DanaG> Oh yeah, and damn Samsung drives lie about size... not merely in the "logical" field, but also in the "physical" field.
[20:48:23] <RoyK> anyone around with a samsung?
[20:48:28] <DanaG> I have one.
[20:48:34] <RoyK> 2TB?
[20:48:44] <DanaG> Yup. Or rather, two of them, 2TB each.
[20:48:56] <RoyK> sector count?
[20:50:01] <DanaG> 255 heads, 63 sectors/track, 243201 cylinders, total 3907029168 sectors.
[20:50:33] <RoyK> 3907029168 for my hitachis and WD drives
[20:50:42] <RoyK> both greens and blacks tested for the latter
[20:50:48] <RoyK> a few models for the former
[20:51:08] <RoyK> heads/sectors etc is irrelevant with LBA
[20:51:44] *** movement has quit IRC
[20:52:32] <vcg> anyone had any luck with oi on the dell r510 or r710 with a few md1200's hanging from it?
[20:53:56] <DanaG> Cool, then I don't have to worry about that bit of space.
[20:54:27] <DanaG> stupid samsung:
[20:54:39] <DanaG> Logical sector size: 512 bytes. Physical size: 512 bytes.
[20:54:52] <DanaG> The WD drives, at least, report correct values for Physical.
[20:56:06] *** gea_ has joined #openindiana
[20:57:55] *** gea has quit IRC
[20:58:02] *** gea_ is now known as gea
[21:04:59] *** kenon has joined #openindiana
[21:05:27] <DanaG> great, so, my openindiana won't boot.
[21:05:37] <DanaG> It installed grub to the secondary drive, not the primary drive.
[21:05:48] <DanaG> And if I chainload to it, I get exactly the following output:
[21:05:50] <DanaG> GRUB
[21:05:54] <DanaG> that's it.
[21:08:04] *** kenon_ has quit IRC
[21:08:40] *** raijin is now known as darth_raijin
[21:10:21] <DrLou_> DanaG: installed grub on slice 0?
[21:12:04] <DrLou_> experts, what is wrong with this beadm command?: beadm create -e rpool/ROOT/openindiana@ROOT_SNAPSHOT -p FUBARpool FUBARboot
[21:12:39] <DrLou_> (where FUBARpool is a file-based zpool?)
[21:14:19] <DanaG> COrrection: it looks like it didn't install grub anywhere.
[21:14:25] <DanaG> It left behind the ubuntu grub.
[21:15:11] <DrLou_> (ng)
[21:16:57] <DanaG> so now what do I do to get the OI grub?
[21:18:23] <Triskelios> DanaG: boot from the partition
[21:18:42] <Triskelios> that is where GRUB is installed, normally
[21:18:55] *** tuf8 has joined #openindiana
[21:19:03] <Triskelios> the installer may not have touched the MBR, where you already had Ubuntu GRUB
[21:19:25] <Triskelios> or did you already try that with chainload?
[21:20:27] <Triskelios> http://praveen.kumar.in/2010/02/20/chainloading-opensolaris-from-grub-2/
[21:21:11] <DanaG> ah, I see.. thanks.
[21:23:08] <DanaG> weird... I changed my boot order back, and now it's randomly booting grub again.
[21:23:09] <DanaG> er, booting OI.
[21:23:20] <DanaG> The "boot order" menus on the HP Microserver seem buggy.
[21:23:50] <Wraithhh> Hmm, is the 64-bit oi userland 32-bit or did I mess up the installation somehow?
[21:23:54] <DanaG> There is no datapool available.
[21:23:55] <DanaG> Please create a Pool!
[21:24:00] <DanaG> So says napp-it
[21:24:35] <DanaG> oh, and another thing I find odd: no such thing as vt-switching.
[21:34:40] <sommerfeld> Wraithhh: oi userland is (like solaris) a mix of 32-bit and 64-bit binaries.
[21:36:09] <Wraithhh> ok. Is there a 64-bit release of mysql for example?
[21:38:05] <Wraithhh> Can you give me an example of a binary that should be 64-bit?
[21:38:22] <Wraithhh> Since I didn't find any in /bin or /usr/bin
[21:39:30] <Triskelios> DanaG: virtual consoles are disabled by default
[21:41:11] <patdk-lap> Wraithhh, 64bit wouldn't be in those folders
[21:41:33] <patdk-lap> Wraithhh, you must be too stuck to linux or something
[21:41:40] <patdk-lap> the userland is both 32bit and 64bit
[21:41:42] *** enmand has quit IRC
[21:42:30] <Triskelios> Wraithhh: many applications will automatically run the 64-bit version when available
[21:43:22] *** kenon has left #openindiana
[21:44:19] <Wraithhh> right, I found the 64 bit version of mysql in /usr/mysql/bin/amd64
[21:44:33] <Wraithhh> I wonder why it was running the 32 bit version though
[21:45:23] <sommerfeld> for example, /usr/bin/mdb is actually a hard link to /usr/lib/isaexec which executes the 64-bit mdb if running on a 64-bit kernel, and the 32-bit mdb if running on a 32-bit kernel
[21:45:50] <sommerfeld> but i guess that mysql wasn't packaged to use isaexec
[21:46:16] <Triskelios> Wraithhh: MySQL has a (unfortunately poorly documented) parameter in the service that specifies 32- or 64-bit, from what I understand it's because the database format changes when you switch ISAs
[21:46:21] <Triskelios> (which is pure evil)
[21:46:56] <Triskelios> see svcprop mysql
[21:47:13] <Wraithhh> mysql/enable_64bit
[21:47:15] <Wraithhh> thanks :)
[21:49:13] <patdk-lap> I wonder what changes
[21:49:28] <patdk-lap> cause I moved my mysql tables from 32 to 64 many times and 64 to 32, without issues
[21:57:18] *** Botanic has quit IRC
[21:58:48] *** flyz has quit IRC
[22:23:13] *** jurikm has quit IRC
[22:26:41] *** pothos_ has joined #openindiana
[22:27:03] *** pothos has quit IRC
[22:29:59] *** pothos_ has quit IRC
[22:30:18] *** pothos has joined #openindiana
[22:33:01] *** pothos has quit IRC
[22:33:19] *** pothos has joined #openindiana
[22:40:56] *** TomJ has joined #openindiana
[22:48:04] *** gea has quit IRC
[22:48:39] *** voidcoder has joined #openindiana
[22:49:14] *** voidcoder has quit IRC
[22:49:53] *** voidcoder has joined #openindiana
[22:50:38] <DanaG> okay, now: how do I add a "solaris2" partition to a pool?
[22:50:47] <DanaG> I don't want to add the whole drive; just one partition.
[22:53:00] <Triskelios> /dev/rdsk/<device>p<n>, or if you've made slices in that partition, s<n>
[22:54:45] <DanaG> thanks.
[22:55:24] <DanaG> oh yeah, does it matter, for bootability, that the pool will consist of the following:
[22:55:33] <DanaG> c2t0d0s0 ONLINE 0 0 0
[22:55:33] <DanaG> c2t2d0p3 ONLINE 0 0 0 (resilvering)
[22:55:47] <DanaG> Apparently, there are slices in the first, but not in the second.
[22:56:10] <DanaG> oh yeah, I need a question mark in my question there somewhere.
[22:58:01] <Triskelios> yes, both mirrors need to be configured identically, and booting requires a VTOC slice
[22:58:16] <Triskelios> both halves of the mirror, even
[22:58:26] <Triskelios> smrt: explain rpool mirror
[22:58:27] <smrt> Um. I seem to not know anything about rpool mirror...
[22:58:34] <Triskelios> smrt: explain mirror rpool
[22:58:35] <smrt> http://letsgetdugg.com/2009/10/18/zfs-boot-mirror-setup/
[22:58:48] <DanaG> ah, then I should remove the latter, slice-ify it, and re-add it.
[22:59:04] <Triskelios> correct
[23:01:28] <DanaG> /dev/dsk/c2t2d0s0 overlaps with /dev/dsk/c2t2d0s2
[23:01:46] *** gea has joined #openindiana
[23:03:19] <Triskelios> the overlap is normal, ignore that warning
[23:04:07] <DanaG> ah, then -f it?
[23:05:00] <DanaG> done.
[23:05:01] <DanaG> thanks.
[23:05:22] <DanaG> So what does fmthard do, that fdisk doesn't do? Slice things within whatever "solaris2" partition exists?
[23:07:08] <vcg> I read an article that mentions the seagate ST2000NM0001 not working - is anyone having any luck with this drive?
[23:07:26] <Triskelios> DanaG: yup, sets up a VTOC label which is nested in the partition
[23:07:30] *** bahamas10 has quit IRC
[23:09:00] <Triskelios> disks on SPARC would just have the label and no x86 partition table, but on x86 we have both for compatibility
[23:11:43] <DanaG> hmm, now, for home dirs that may have stuff synced with OS X, should I set utf8only=yes normalization=formD ?
[23:12:52] * DanaG goes off to google some stuff.
[23:13:38] *** gea has quit IRC
[23:13:48] *** ChanServ sets mode: +o Triskelios
[23:32:39] *** merzo has quit IRC
[23:36:06] *** fly23 has joined #openindiana
[23:42:34] *** Seony has joined #openindiana
[23:43:54] *** CVLTCMK0 has joined #openindiana
top

   January 29, 2012  
< | 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 | >