July 19, 2011 
< | 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 | >
July 19, 2011 [00:09:06] *** ivo_ has quit IRC[00:09:50] *** Hedonista has quit IRC[00:11:10] *** skeeziks has quit IRC[00:20:11] *** bdrewery has joined #openindiana[00:22:13] *** ivo_ has joined #openindiana[00:22:15] *** sergefonville has left #openindiana[00:35:15] *** Cobi has quit IRC[00:42:58] *** held has quit IRC[01:00:35] *** Hedonista has joined #openindiana[01:06:58] *** Cobi has joined #openindiana[01:16:17] *** gea has quit IRC[01:21:48] *** InTheWings has quit IRC[01:24:19] *** descipher_ has joined #openindiana[01:33:26] *** tomww has quit IRC[01:33:26] *** master_of_master has quit IRC[01:33:26] *** jkimball4 has quit IRC[01:33:26] *** descipher has quit IRC[01:33:26] *** russiane39 has quit IRC[01:33:27] *** robinsmidsrod has quit IRC[01:33:27] *** ArnaudEC has quit IRC[01:33:27] *** koollman has quit IRC[01:33:27] *** Meths has quit IRC[01:33:27] *** JStoker has quit IRC[01:33:27] *** Cope has quit IRC[01:38:47] *** tomww has joined #openindiana[01:38:47] *** koollman has joined #openindiana[01:38:47] *** master_of_master has joined #openindiana[01:38:47] *** jkimball4 has joined #openindiana[01:38:47] *** russiane39 has joined #openindiana[01:38:47] *** robinsmidsrod has joined #openindiana[01:38:47] *** ArnaudEC has joined #openindiana[01:38:47] *** Meths has joined #openindiana[01:38:47] *** JStoker has joined #openindiana[01:38:47] *** Cope has joined #openindiana[01:38:47] *** niven.freenode.net sets mode: +oo tomww Meths[01:50:28] *** held has joined #openindiana[02:02:14] *** baitisj has quit IRC[02:03:58] *** dekar has quit IRC[02:09:32] <DontKnwMuch> hi, I just made a dladm aggregation, and on reboot, i get this: e1000g0: DL_BIND_REQ failed: DL_SYSERR (errno 16)[02:09:38] <DontKnwMuch> e1000g0: DL_UNBIND_REQ failed: DL_OUTSTATE[02:09:45] <DontKnwMuch> aggr1 which I made works[02:10:11] <DontKnwMuch> can I ignore this ?[02:10:32] *** skeeziks has joined #openindiana[02:20:08] <jkimball4> is hibernate available?[02:20:20] <jkimball4> suspend works a charm, but at this instant hibernate would be good to have[02:21:21] <alanc> if by hibernate, you mean suspend-to-disk, only on SPARC[02:21:30] <jkimball4> that's what i thought[02:21:32] <jkimball4> thanks[02:21:40] <alanc> (silly OS'es, each making up their own names for these states)[02:32:08] <brandini> Just plain silly OS'es :)[02:40:05] *** winIO has quit IRC[02:41:42] *** bdrewery has quit IRC[02:52:06] *** held has quit IRC[03:01:11] *** sivanov has quit IRC[03:05:50] *** sivanov has joined #openindiana[03:11:23] *** Naresh has quit IRC[03:11:25] *** master_of_master has quit IRC[03:13:16] *** master_of_master has joined #openindiana[03:15:36] <mnaser> I'm going nutssss, opensm running on a linux host, it shows both hosts are connected, everything that's supposed to work, works.. except "sendto Network is unreachable" when pinging from solaris end, and it times out on the linux end[03:15:54] <mnaser> I'm assuming it has to do with the routes on the solaris server, because it doesn't have internet connectivity.. but does that matter?[03:19:08] *** held has joined #openindiana[03:20:27] <mnaser> WOW[03:20:28] <mnaser> it was it[03:20:29] <mnaser> no default route[03:20:55] *** blues_ has quit IRC[03:22:47] *** blues_ has joined #openindiana[03:22:57] *** kdavy_ has joined #openindiana[03:24:19] <Triskelios> mnaser: you don't need a default route, but you need a route for every network[03:24:36] <mnaser> Triskelios: i had a route for the 10.0.0.0/24 network but nothing else[03:24:41] <mnaser> i guess i need a 0.0.0.0 route[03:25:12] <Triskelios> that's the definition of a default route[03:25:30] <mnaser> yeah, what a n00b thing to do, i was stomped for hours in front of this thinking that itwa s the subnet manager[03:27:02] *** redgone has joined #openindiana[03:34:42] *** RobotCow has joined #openindiana[03:37:32] *** blues_ has quit IRC[03:39:20] *** DrLou has quit IRC[03:43:40] *** blues_ has joined #openindiana[03:44:44] *** RobotCow has quit IRC[03:47:10] *** kdavy_ has quit IRC[03:47:35] *** kdavy_ has joined #openindiana[03:48:48] *** blues_ has quit IRC[03:50:15] *** blues_ has joined #openindiana[04:09:01] *** magyar has quit IRC[04:12:49] <Shadow__X> is there a way to control the type of compression used on a zfs volume within napp it?[04:12:58] <Shadow__X> i see compression on or off but not type[04:15:14] *** laserbled has quit IRC[04:22:51] *** kart_ has joined #openindiana[04:23:10] *** laserbled has joined #openindiana[04:24:57] *** tg has quit IRC[04:25:36] *** tg has joined #openindiana[04:27:58] <Shadow__X> also how can i monitor speed of network transfers, in linux i use ifstat[04:28:45] *** tg has quit IRC[04:29:35] *** POloser has joined #openindiana[04:31:10] *** tg has joined #openindiana[04:32:37] *** syoyo has joined #openindiana[04:44:43] <Shadow__X> nvm bwm-ng is what i was looking for[04:49:25] *** laserbled has quit IRC[04:57:27] *** tg has quit IRC[04:58:34] *** tg has joined #openindiana[05:03:02] *** baitisj has joined #openindiana[05:06:26] *** miine has quit IRC[05:07:30] *** miine has joined #openindiana[05:15:17] *** tg has quit IRC[05:16:09] *** tg has joined #openindiana[05:30:20] *** syoyo has quit IRC[06:01:41] *** geoff_ has quit IRC[06:03:41] *** redgone has quit IRC[06:22:14] *** geoff_ has joined #openindiana[06:45:08] *** flyz has quit IRC[06:52:05] *** flyz has joined #openindiana[07:00:31] *** skeeziks has quit IRC[07:04:19] <sivanov> does anybody know if 3ware 9650se works in IT mode?[07:06:35] <richlowe> if you're meaning JBOD, yes. But not on OI, 'cos there's no driver.[07:15:01] *** laserbled has joined #openindiana[07:27:08] *** tsukasa has quit IRC[07:30:17] *** flyz has quit IRC[07:31:33] <sivanov> what will be better for arc cache: less same size, same (high) read speed such as SF1200 based drives, or different sized, diffrent manufacturer drives, but twice as more??[07:38:46] *** tsukasa has joined #openindiana[07:40:01] <tsoome> memory access time is in ns, ssd is in microseconds, disk is milliseconds. if you can avoid accessing disk, its already an huge win;)[07:40:11] *** ivo_ has quit IRC[07:44:19] *** Crypticfortune has joined #openindiana[07:44:49] *** Naresh has joined #openindiana[07:45:54] <Crypticfortune> Hi. We have a Sun Fire X4170, and I'm trying to get ipmitool working, but it can't see the IPMI device. how is one supposed to get IPMI going under OI?[07:57:06] *** bixu has joined #openindiana[07:57:38] <bixu> anybody have a good link to a zfs deduplication guide?[07:58:20] <richlowe> Crypticfortune: find a copy of the closed, non-redistributable bmc(7D) driver.[07:58:30] <richlowe> or hope someone writes a better one[07:59:33] <bixu> anybody have a good link to a zfs deduplication guide?[07:59:35] <richlowe> sivanov: l2arc, fastest read you can, probably.[07:59:51] <Crypticfortune> richlowe: hmmm, that's sad news. is that driver originally included in solaris?[07:59:58] <richlowe> Yes.[08:00:00] <bixu> oops - sorry for the duplicate post - now you know why i need dedupe :)[08:00:50] <tsoome> bixu: man google.[08:01:27] *** laserbled has quit IRC[08:02:47] <Crypticfortune> arr, probably easier just to boot once into linux, set up ipmi to be remote-configurable and go back to OI =P[08:03:53] <tsoome> just connect to serial management port and set it up. you need no boot.[08:04:33] <sivanov> i can buy two samsung MMCRE64G5MXP-OVBH1 64gb drives for price of one sandforce 1200 60gb drive, samsung specs say it has 220MB/s read 120MB/s write, i have g.skill phoenix pro, so i can speak for it: 250MB/s read 230MB/s write when zeroed first, 170MB/s after being filled with /dev/urandom[08:05:02] <sivanov> what will make better cache device, i dont know[08:05:07] <Crypticfortune> tsoome: the serial port gets access to ILOM. the bits I actually to get at are the IPMI user and auth settings, which as far as I can tell, are inaccessible from ILOM >_<;;[08:06:18] <tsoome> maybe you should read the ilom users guide:P[08:07:56] <Crypticfortune> sivanov: that MMCRE64G5MXP-OVBH1 looks like an MLC SSD? since for a cache device you (probably) want pure IO/s you should go for an SLC if possible, or at least just go off of IO/s stats rather than throughput[08:08:14] *** laserbled has joined #openindiana[08:09:21] <Crypticfortune> IO/s is bad name for that unit. operations per second is what i mean...[08:09:58] <sivanov> well, sf1200 (mlc) drives has much better IOPS that in intel x25-e (slc)[08:10:03] <sivanov> *have[08:10:42] *** gea has joined #openindiana[08:11:47] <tsoome> for l2arc, reads/s is important.[08:16:22] <Crypticfortune> sivanov: well, the x25-e is also only an SATA2 device (whereas the sf1200 can do SATA3). but yeah, modern MLCs seem to have gotten much better (i haven't tested any though)[08:17:59] <sivanov> also, this is home NAS, so i probably wont need SLC's wear resistance[08:20:42] <sivanov> last time i've tried raidzX5hdd & two g.skill phoenix i saw in iostat that the SSDs speed in cache was capped @50-70% of raw device speed[08:21:00] *** kdavy_ has quit IRC[08:22:00] <sivanov> but i have this dial Opteron 295 server, with verry slow RAM read-write speed, about 1-1.2GB/sec[08:23:22] *** flyz has joined #openindiana[08:29:57] *** gea has quit IRC[08:34:03] *** tsoome has quit IRC[08:38:44] *** Naresh has quit IRC[08:40:31] *** Naresh has joined #openindiana[08:44:29] <Shadow__X> sivanov: you can pickup rebranded a data s599 sandforce 1200 64gb drives at microcenter for 110[08:46:22] *** Botanic has quit IRC[08:49:21] *** anikin has joined #openindiana[09:00:18] *** tsoome has joined #openindiana[09:00:43] *** allanblack has quit IRC[09:01:08] *** Micr0mega has joined #openindiana[09:01:13] *** allanblack has joined #openindiana[09:09:28] *** laserbled has quit IRC[09:09:49] *** merzo has joined #openindiana[09:14:41] *** Botanic has joined #openindiana[09:17:56] *** |AbsyntH| has joined #openindiana[09:22:56] <sivanov> hmm the microcenter ssd are even for $89 now[09:23:54] <nettezzaumana> moin[09:25:52] *** xmikus01 has joined #openindiana[09:36:12] <RoyK> lol - reading about MongoDB, it just assumes all systems are little endian...[09:37:30] <DeanoC> BigEndian is on the wanted list but noone seems to care enough to do it![09:42:18] *** lkocman has joined #openindiana[09:43:47] <nettezzaumana> omg, nfs implementation on linux is just torturous[09:44:43] <tsoome> .oO[09:46:39] <nettezzaumana> tsoome: does `.oO' mean `farting' .. ?? :D[09:47:02] <tsoome> lol[09:47:23] <tsoome> .oO (what the hell:P)[09:47:29] *** syoyo has joined #openindiana[09:49:57] *** bens1 has joined #openindiana[09:56:23] *** Micr0mega has left #openindiana[09:56:39] *** Micr0mega has joined #openindiana[09:57:13] <robinsmidsrod> what is the correct syntax to set a smbd property with space in it? sharectl set -p 'map=/usr/local/bin/log_smb_map -u %u' smb (doesn't work)[09:58:43] <tsoome> sharectl set -p 'map="/usr/local/bin/log_smb_map -u %u"' smb maybe?[09:59:09] <tsoome> or sharectl set -p map='usr/local/bin/log_smb_map -u %u' smb[09:59:58] <tsoome> as -p takes property to be set and you need to make property value to be interpreted as single string[10:00:14] <robinsmidsrod> tsoome: I've tried both alternatives, but I can try with your particular set of quotes[10:00:54] *** vinc has quit IRC[10:01:24] <robinsmidsrod> tsoome: tried your particular set of quotes, only get this error: Could not set property map: bad property value[10:01:42] *** vinc has joined #openindiana[10:04:04] <robinsmidsrod> where can I find the documentation on how to set property values in general, and how they should be formatted?[10:08:38] *** Botanic has quit IRC[10:09:16] *** Whoopsie has joined #openindiana[10:09:16] *** ChanServ sets mode: +v Whoopsie[10:11:45] <tsoome> man sharectl. in fact, there is an example as well, not with spacest tho. and its using " "[10:12:13] <tsoome> harectl set -p map="/usr/local/bin/log_smb_map -u %u" smb should be it[10:12:41] *** flyz has quit IRC[10:15:17] <robinsmidsrod> still gives the same error - guess there must be a bug somewhere then... :([10:24:24] <sivanov> on initiator (4GB FC) raw device:[10:24:25] <sivanov> Children see throughput for 1 readers = 164343.52 KB/sec[10:24:43] <sivanov> on target (/dev/zvol/rdsk...)[10:25:00] <sivanov> Children see throughput for 1 readers = 365841.72 KB/sec[10:25:54] <tsoome> you have slog?[10:26:32] *** Whoopsie has quit IRC[10:27:22] <tsoome> robinsmidsrod: try with simple value first - without any args or spaces.[10:27:37] <robinsmidsrod> that works[10:27:56] <sivanov> tsoome, no slog yet, but its read anyway[10:28:08] <tsoome> ah right[10:28:19] <tsoome> missed the read bit;)[10:28:23] <sivanov> even worse difference when reading @4 stream[10:29:02] <sivanov> about 120MB/s on initiator and about 410MB/s on target[10:29:18] <tsoome> robinsmidsrod: then next step is "/command -u"[10:30:47] <sivanov> even 430MB/s at target using 4 iozone processes[10:30:57] <robinsmidsrod> "/command something" works, "/command %u" works, but "/command -u" doesn't work[10:31:07] <robinsmidsrod> extremely strange...[10:32:07] <robinsmidsrod> "/command u" fails also...[10:33:02] *** robinbowes has quit IRC[10:33:25] <sivanov> can COMSTAR be tuned somehow, to larger buffers or whatever[10:34:19] <robinsmidsrod> thanks for helping me figure out the syntax - now I guess I'll have to go and report a bug to the nexenta people[10:35:30] *** Phenox_ has quit IRC[10:40:29] <sivanov> perhaps i need to try exporting ramdisk...[10:44:06] <robinsmidsrod> how stable is openindiana for everyday use? Not a business user, but expects business use stability and reliability[10:48:05] <tsoome> people do run production on it. so its stable till you hit the bug:P[10:49:04] <robinsmidsrod> tsoome: how would you rate openindiana 148 compared to nexenta NCP 3.0? which is better?[10:49:16] <robinsmidsrod> better = more stable (less crashing)[10:49:26] <tsoome> i cant, i dont use either - only solaris i have atm is s11[10:49:47] *** robinbowes has joined #openindiana[10:50:59] <tsoome> and btw, on s11 sharectl set -p map="/bin/echo %u" smb does work:P[10:52:40] *** gber has joined #openindiana[11:00:49] *** ThomasB2k has quit IRC[11:11:05] *** laserbled has joined #openindiana[11:14:12] <sivanov> using ramdisk i get 350MB/s write and 390MB/s read over FC :S[11:14:40] <sivanov> need to try some evil zfs tuning[11:17:51] <tsoome> can it be your HBA connecting disks and target mode FC HBA are interfering - as test with ramdisk did exclude disks and you are doing io from ram to FC target mode HBA[11:21:17] *** Infin1ty|work has joined #openindiana[11:23:21] <sivanov> it may be that zfs<->comstar<->qlt copies buffers alot and this server has really slow RAM[11:24:43] *** Whoopsie has joined #openindiana[11:24:43] *** ChanServ sets mode: +v Whoopsie[11:25:08] <robinsmidsrod> tsoome: but does sharectl set -p map="/bin/echo -e %u" smb work on s11?[11:25:12] <tsoome> well, from ram to qlt it wasnt that bad, so if you test with zfs and cacheable read content, then on second read test you should get better speed as the data is in the cache already[11:25:14] <sivanov> how i query a value like zfs_vdev_max_pending via mdb?[11:26:21] <tsoome> robinsmidsrod: aha, no, the -e makes it to barf[11:26:25] <sivanov> tsoome, zfs won't cache anything since i use large (16G*4) test sets and it has 16gb ram[11:26:41] <robinsmidsrod> tsoome: thanks for the test![11:27:03] <tsoome> sivanov: you didnt test ramdisk with 4x 16G:P[11:27:17] <sivanov> yep, only with 8G[11:27:28] <tsoome> the point is to verify, if with cacheable data set you can get same speed as with ramdisk[11:27:29] <robinsmidsrod> wonder why it barfs though... probably the regexp or something that filters out unwanted characters or something that is too broad[11:27:39] *** miamano has quit IRC[11:27:41] <sivanov> with 12G it would become barely responsive...[11:28:43] <tsoome> robinsmidsrod: so as workaround, you can write your skript not to need -u, but use just %u[11:28:58] *** laserbled has quit IRC[11:29:19] <robinsmidsrod> tsoome: problem is that options without dashes are declined also[11:29:43] <robinsmidsrod> so I can't do "/command opt1 %u opt2 %i"[11:29:51] <robinsmidsrod> doesn't make any sense though...[11:30:20] <robinsmidsrod> maybe all expansions are mandatory, so that if the value is not present it will just be an empty string[11:30:32] <tsoome> well, you can just assume the %u and %i are used in that order[11:30:50] <tsoome> then you dont need any other options[11:30:57] <robinsmidsrod> tsoome: yeah, I'm going to try that and see what it actually contains[11:31:55] <tsoome> the whole point of options with args is to make it possible to use them in any order, but for that script, you can assume specific order.[11:32:18] <tsoome> its not nice, but will do for time being:)[11:32:45] <robinsmidsrod> indeed, trying it out now[11:35:46] *** blues_ has quit IRC[11:38:16] <sivanov> it looks like, that i get better results with zfs_vdev_max_pending=1[11:38:26] <sivanov> at least @read[11:38:34] <sivanov> *write[11:38:40] *** mikw has joined #openindiana[11:38:59] <sivanov> that or checksum=fletcher2[11:39:02] <sivanov> dont know[11:39:44] *** miamano has joined #openindiana[11:40:37] *** AlasAway is now known as Alasdairrr[11:50:18] <tsoome> one change at the time:P[11:51:10] <brandini> tsoome![11:51:39] <tsoome> i need to get some food:P[11:52:03] <brandini> me too, breakfast[11:52:30] *** blues_ has joined #openindiana[11:56:34] *** held has quit IRC[12:05:26] *** McBofh has quit IRC[12:12:53] <sivanov> what worse/better about fletcher2 VS sha256?[12:16:06] <lblume> Differrent performance/collision probability compromise.[12:18:58] *** |AbsyntH| has quit IRC[12:22:04] *** McBofh has joined #openindiana[12:22:38] *** flyz has joined #openindiana[12:23:36] <|woody|> well as lang as you only use it for data integratie checks flechter is o.k[12:28:35] *** eto has joined #openindiana[12:28:49] *** Whoopsie has quit IRC[12:29:06] <eto> hello guys is there some begginer frienldyeasy to follow open indiana resource somethink like arch linux beginner guide[12:29:09] <eto> ?[12:29:14] <eto> something[12:29:32] <nettezzaumana> eto: rephrase[12:30:19] <eto> nettezzaumana: is there some beginner friendly step by step resource / web pages how to set up open indiana[12:30:21] <nettezzaumana> oi is not intended for poor minds but for IT professionals .. good start with UNIX provides a Linux[12:30:33] <nettezzaumana> eto: ^^[12:30:44] *** held has joined #openindiana[12:30:44] <eto> nettezzaumana: possibly with annotation what is what and why something is done that way and not another?[12:30:53] <nettezzaumana> eto: but feel free to contribute and create this resource[12:30:57] <eto> nettezzaumana: i see[12:31:40] <eto> nettezzaumana: arch linux resource is not for total noob but for someone who is moving from crap like ubuntu to very lightweight kiss system[12:32:21] <nettezzaumana> ok, and does it with openindiana ?[12:32:30] <nettezzaumana> **what ^^[12:32:31] <eto> nettezzaumana: i have some user level experience with OpenBSD and FreeBSD, and archlinux, but from my last try experience with sunos at library[12:33:03] <nettezzaumana> eto: i don't understand what you want[12:33:24] <nettezzaumana> if you want some retard-proof guide just create one[12:33:26] <eto> nettezzaumana: everything was very weird, ping was different, xterm didn't accept normal IBM/PC keys etc, no ficonfig etc[12:35:23] <eto> nettezzaumana: ah, hmm maybe this would be better question: is there something equivalent to freebsd handbook or for open indiana?[12:35:25] <nettezzaumana> eto: what? ping is obvious[12:35:47] <nettezzaumana> ifconfig is there[12:36:04] <nettezzaumana> and IBM|PC keys depends on $LANG settings[12:36:12] <nettezzaumana> and $TERM of course[12:36:56] <nettezzaumana> eto: no. i've already said: feel free to create one[12:37:19] <bradend> Great, not brandini is stealing my breakfast instead of just my name.[12:37:20] <eto> nettezzaumana: i was talking about my experience with sunos 5, ping was like "host is up" instead of detailed report etc, so i should assume OI is more linux "like" whatever that means?[12:37:36] <tsoome> detailed report?[12:37:44] <nettezzaumana> eto: linux ping is driven !![12:37:46] <tsoome> man man maybe?[12:38:06] <nettezzaumana> ping -s $foo[12:38:58] <eto> nevermind this was mistake to ask, i will try to isntall and see what it is i hope i won't get lost, thanks guys[12:39:07] <nettezzaumana> tsoome: btw i'll have time today to check how system deals with libc.so* .. i think it's modified during system startup, i'll let you know[12:39:16] <nettezzaumana> eto: no problem[12:40:00] <tsoome> linux is not unix, if you havent figured out they are different, maybe you should just stick to something you do know, instead of whining why different systems are different....[12:41:30] <eto> tsoome: am i whining? sorry that i sounded like that, i am not english native speaker, i am talking about something like unix rosseta stone only detailed[12:41:44] *** hsp has joined #openindiana[12:42:02] <tsoome> as for docs, the most complete set is http://download.oracle.com/docs/cd/E19963-01/index.html its not directly for openindiana, but will apply 99.99%[12:42:17] <Naresh> eto: from what i've seen, you have to rely on the manpages heavily[12:43:14] <nettezzaumana> yeah eto ... true dat .. check Solaris Administrator's Collection at oracle site[12:43:27] <eto> Naresh: thank you for info, and may i ask, are they high quality like OpenBSD ones or just bunch of stuff with errors like on linux?[12:43:39] <nettezzaumana> but who reads a docs if could beg for help on IRC :P[12:44:02] <eto> nettezzaumana: finally thanks so there is 1 to 1 between opensolaris and oi right?[12:44:35] <eto> so basically most os resource would be applicable to oi, correct?[12:44:44] <tsoome> yes[12:44:46] <nettezzaumana> eto: well, calm son .. your statements are quite inaccurate and stinks by (partial) ignorancy[12:45:19] <nettezzaumana> are they high quality like OpenBSD ones or just bunch of stuff with errors like on linux[12:45:30] <nettezzaumana> ^^ this just stinks by ignoracy, sorry[12:47:18] <tsoome> the answer for that is both, because for some reason they did put gnu commands in front of path and as man does follow the path, you get gnu manuals for gnu commands. fortunately there are decent ones around as well:P[12:48:09] <eto> well some linux man pages misslead me several times, then i needed to google forums and mailing lists and whatnot to learn new option was added or the one recommend in manpage was deprecated and it was not "yet" corrected/included in manpage[12:49:05] <eto> i've never had such problem with some other systems and everything worked as decribed[12:50:10] <bradend> Heh, if you can point out the magical OS where everything works, point me to it. I think I may standardize on it.[12:50:50] <jkj_> bradend: NetBSD?[12:50:51] <jkj_> me hides[12:51:25] <eto> heh :)[12:52:13] <jkj_> lack of proper documentation is indeed a linux thing[12:52:23] <jkj_> sad that oracle broke the documentation links of solaris[12:52:24] <tsoome> it really depends on how people manage their thing - gnu packages include both binaries and manuals, if package developers don't update manuals accordingly, the result is as you have decribed.[12:52:29] <tsoome> described*[12:52:40] <jkj_> wonder if the documentation is still visible and linkable somewhere[12:52:53] <tsoome> http://download.oracle.com/docs/cd/E19963-01/index.html[12:52:57] <nettezzaumana> jkj_: omg, you also stinks by great ignoracy[12:53:12] <nettezzaumana> jkj_: what kind of doc are you missing son ?[12:53:18] <nettezzaumana> i help you find it out[12:55:06] <jkj_> tsoome: the lack of system-spanning documentation indeed comes from the fact that most linux-distros are collections of software without heavy integration...[12:55:25] <RoyK> hm.... small issue here - I have setup idmapd to map between windows and unix (nis) users, and things seem to work, but users can't write to their own files created from unix[12:55:34] <jkj_> the ones backed by companies have more resources to write docs[12:55:48] <tsoome> RoyK: the famous default acl issue?[12:56:14] <jkj_> nettezzaumana: just complaining about oracle breaking the documentation links that commands spit out[12:56:17] <tsoome> or does the mapping misfire[12:56:43] *** mikw has quit IRC[12:57:01] <jkj_> nettezzaumana: and about lack of proper manual pages in linux distros[12:57:05] <nettezzaumana> jkj_: only bad thing is that all over the internat are stale links to prior docs.sun.com but the most of stuff can be found on new oracle docs site[12:57:12] <jkj_> nettezzaumana: true[12:57:28] <jkj_> nettezzaumana: even software refer to docs.sun.com... :([12:57:56] <jkj_> would it have beens so hard to keep the old domain[12:58:07] <jkj_> s/domain/urls/[12:58:16] <tsoome> that software is just outdated[12:58:41] *** raichoo has joined #openindiana[12:58:45] <tsoome> only recently s11 did fix referencing to docs after pkg image-update[12:58:46] <nettezzaumana> jkj_: well, world changes, it's called evolution no matter if the changes are to better or to worse[12:59:21] <nettezzaumana> jkj_: you do nothing but you adapt or die[12:59:57] <Naresh> eto: i haven't compared manpages that much.. but they're good from what i've used so far[13:00:00] <jkj_> were the docs released in forkable license? ...is openindiana referring to oracle documentation in the future?[13:00:34] <Naresh> eto: tell you what, play around / spend some time on it. i doubt you'll get a rosetta stone like "equivalents" page anywhere.[13:00:42] <tsoome> not too many things were referring to docs.sun.com really[13:01:42] <tsoome> guess what.... http://bhami.com/rosetta.html :P[13:01:54] *** syoyo has quit IRC[13:02:53] <Naresh> omg it exists[13:03:54] <tsoome> guess what, its first hit in google.[13:04:35] <tsoome> btw, i think google should close their search site, its clearly not needed...[13:05:08] <tsoome> ;)[13:06:39] <Naresh> hehe[13:09:50] <tsoome> hm.[13:10:33] <RoyK> tsoome: what is this famous default acl issue?[13:10:57] <nettezzaumana> s/famous/infamous/[13:11:04] <RoyK> well, still[13:11:21] <tsoome> missing default acl. http://mattwilson.org/blog/solaris/solaris-cifs-server-and-zfs-acls-the-problem/[13:11:28] <tsoome> second comment does explain[13:12:26] <nettezzaumana> RoyK: i didn't reread but i guess it has something to do with !SUNWcsu utils unable to handle properly with sunos acls, right ?[13:12:47] *** Worsoe has joined #openindiana[13:14:29] <tsoome> no, for some reason his users are unable to read files created in unix (which is wierd assuming the permissions are in place)[13:15:10] * nettezzaumana stops chatting here and getting bcak to some real business[13:24:30] *** InTheWings has joined #openindiana[13:37:13] *** lkocman has quit IRC[13:38:39] *** DontKnwMuch has quit IRC[13:39:33] *** DrLou has joined #openindiana[13:39:33] *** ChanServ sets mode: +o DrLou[13:40:16] *** lkocman has joined #openindiana[14:00:27] *** ThomasB2k has joined #openindiana[14:03:42] *** laserbled has joined #openindiana[14:05:05] *** POloser has left #openindiana[14:14:39] *** ThomasB2k has quit IRC[14:16:39] <Infin1ty|work> weird, if i use, ipadm delete-if e1000g0, dladm rename-link e1000g0 lan01, ipadm create-addr -T static -a 10.0.0.10/24 lan01/v4 , the last command will fail on invalid argument, but it will work if i keep the e1000g0 link name[14:16:46] <Infin1ty|work> so basicly i can't rename the link[14:17:24] *** Cobi has quit IRC[14:19:12] *** ThomasB2k has joined #openindiana[14:19:50] *** sdward has joined #openindiana[14:20:17] *** Cobi has joined #openindiana[14:30:11] <tsoome> Infin1ty|work: you did ipadm create-if after dladm rename-link?[14:30:27] <Infin1ty|work> tsoome, yes, the exact order i have wrote[14:30:37] <Infin1ty|work> tsoome, er, i tried that as well[14:30:57] <tsoome> and delete-if for e1000g0 :P[14:31:19] <Infin1ty|work> tsoome, yes, i know that as well...[14:31:33] <Infin1ty|work> tsoome, ipadm: Could not create lan01 : Invalid argument provided[14:42:24] <tsoome> enable-if as well?[14:43:55] <tsoome> ipadm show-if ?:P[14:47:10] *** gea has joined #openindiana[14:52:22] *** gea has quit IRC[14:56:22] <sivanov> can i rise maxphys to 128k on amd64?[14:56:41] <nettezzaumana> tsoome: little Q:, i'm experiencing that on my linux workstation freezes kde for 30 seconds if i have mounted.nfs4 shared folder from oi .. also it parsed badly that on server-side it is "ro" .. do you think that's the client issue ?[14:57:14] <nettezzaumana> i meant freezes for 30 seconds if i'm accessing share with dolphin|konqueror[14:57:43] <nettezzaumana> it does not like this in case of nfs3 mounted folder[14:58:25] <nettezzaumana> btw, does nfs server send am info about opts like ro|rw and do ?[14:58:28] <nettezzaumana> **so[14:58:40] <nettezzaumana> s/am/an/[14:58:43] <nettezzaumana> too many typos[14:59:04] <nettezzaumana> # mount | grep nfs[14:59:04] <nettezzaumana> kokina:/home/dpecka/techniservit on /mnt/nfs type nfs (rw,addr=192.168.50.82,nfsvers=3,proto=tcp,mountproto=udp)[14:59:07] <nettezzaumana> primary:/share on /mnt/nfsv4 type nfs4 (rw,addr=192.168.60.51,clientaddr=192.168.50.84)[14:59:19] <Infin1ty|work> tsoome, i tried everything, it works fine without renaming for some reason[15:02:09] <tsoome> Infin1ty|work: file the bug?[15:02:33] <Infin1ty|work> tsoome, probably , just wondered if anyone else tried that[15:10:04] <nettezzaumana> http://susepaste.org/view/raw/17624827[15:10:16] <nettezzaumana> oh god, nfsv4 on linux sucks indeeed[15:10:52] <nettezzaumana> sunos -> sunos works as expected, sunos -> linux sucks[15:11:41] *** |AbsyntH| has joined #openindiana[15:12:04] <tsoome> nfsv4 domains are different?[15:12:41] <tsoome> if domains dont match, user will be mapped to nobody[15:12:58] <nettezzaumana> tsoome: ? i don't understand what you mean ... just two client boxes and the solaris works okay and the linux one sucks[15:13:00] <tsoome> if you are using auth_sys, uid/gis must match as well[15:13:12] <tsoome> s/gis/gid/[15:13:14] <nettezzaumana> tsoome: yes i know, it's the correct behaviour[15:13:45] <nettezzaumana> tsoome: http://paste.opensuse.org/view/raw/56350758[15:13:48] <nettezzaumana> ^^ on solaris[15:14:07] <nettezzaumana> forget it ... nfsv4 on linux just sucks, that's all[15:14:23] <tsoome> well, nfsv4 domain in recorded in /var/run in solaris, by default its using dns domain[15:14:45] <tsoome> in linux you have to set it in /etc/idmap.conf or something like that[15:15:04] <nettezzaumana> tsoome: hmm, ok, thanks ... i go and spend some time with investigating ... see that uid|gid .. it's just nonsense[15:15:11] <nettezzaumana> http://susepaste.org/view/raw/17624827[15:15:23] <robinsmidsrod> is this warning something to be worried about? https://gist.github.com/1092280[15:15:38] <tsoome> well, thats an indication your uid/gid is mapped to nobody/nogroup[15:16:26] <tsoome> robinsmidsrod: quite probably.[15:16:29] <nettezzaumana> nobody:x:65534:65533:nobody:/var/lib/nobody:/bin/bash - Linux[15:16:35] <nettezzaumana> nobody4:x:65534:65534:SunOS 4.x NFS Anonymous Access User:/: - Solaris[15:16:35] <tsoome> such errors are never good sign;)[15:16:56] <robinsmidsrod> tsoome: what does it actually mean, does it mean my controller or my drive is problematic?[15:17:06] <nettezzaumana> tsoome: it sucks .. there is just a nonsense uid|gid ... simple fact .. and the owner is root[15:17:34] <robinsmidsrod> just installed 5 new wd black drives (WD2002FAEX), and I get this message while filling up the drive and doing a scrub (just test data for now)[15:18:00] <tsoome> nettezzaumana: whats the domain in linux /etc/idmap.conf ?[15:18:55] <robinsmidsrod> am I right to assume that cmdk is the generic IDE (not AHCI) driver in opensolaris?[15:19:02] <tsoome> robinsmidsrod: most likely the disk itself, but its hard to guess. grab smartmontools and check?[15:19:22] <tsoome> ah, indeed, yes, your hba is using IDE mode[15:19:56] <nettezzaumana> tsoome: http://susepaste.org/view/raw/27606261 .. it's correct .. i thought i know how it works[15:20:07] <robinsmidsrod> I know, the marvell controller I have crashes in AHCI mode[15:20:25] <robinsmidsrod> it's a sata3 controller[15:20:29] <tsoome> nettezzaumana: is the same "localdomain" in solaris /var/run/nfsv4_domain?[15:20:46] <robinsmidsrod> setting it to ide mode makes it work, but I'm unsure how stable it is[15:21:12] <robinsmidsrod> I've been contemplating getting an LSI controller to avoid onboard driver issues completely[15:21:34] <robinsmidsrod> I guess the ICH10R is decent?[15:21:41] <nettezzaumana> tsoome: /var/run/nfs4_domain it's empty here on sunos side[15:21:50] <nettezzaumana> tsoome: but it shouldn't matter in this case[15:21:52] <tsoome> thats the reason[15:21:58] <tsoome> it does matter[15:22:09] <tsoome> in nfsv4 the domains *must* match[15:22:21] <nettezzaumana> hmm[15:22:35] <nettezzaumana> why it works then sunos -> sunos okay ?[15:22:51] <tsoome> you can set it with sharectl i thing, or just write it into resolv.conf and restart nfs server[15:22:58] <nettezzaumana> tsoome: okay, would you remind me a syntax of this file ?[15:23:03] <tsoome> both sunos'es have it empty?[15:23:31] <tsoome> resolv.conf : domain localdomain :P[15:23:55] <robinsmidsrod> tsoome: smartmontools doesn't work under solaris, it says :([15:24:06] <nettezzaumana> tsoome: sure, i meant /var/run/nfs4_domain file .. thanks man[15:24:15] <tsoome> smartmontools do, but not on IDE disks[15:24:36] <tsoome> sharectl set -p nfsmapid_domain=localdomain nfs[15:24:55] <tsoome> sharectl get nfs ;)[15:26:22] *** ThomasB2k has quit IRC[15:27:12] *** _delirium1 has joined #openindiana[15:29:38] <nettezzaumana> tsoome: doesn't work .. i have no more will to waste a time with this .. forcing to nfsv3 works, i'll stick with that[15:30:05] <tsoome> you did restart nfs server or did remount at client side?[15:30:16] <tsoome> i think the domains are negotiated at mount time[15:30:33] <nettezzaumana> of course i did .. although i'm acting like retard sometimes i'm not so retarded at all[15:30:34] <tsoome> not 100% sure, never played with changes like that:)[15:30:52] <_delirium1> I have an OpenSolaris system snv_134 installed on a ROOT ZFS, also there is an additional ZFS pool of four 1TB disks attached, I am thinking of upgrading to OpenIndiana as I have had some stability problems - random lockups etc. Is there any concern about the safety of my data in the non-ROOT zfs pool during the upgrade?[15:31:24] <tsoome> _delirium1: yes, if you decide to install on data disks;)[15:31:37] *** heldchen has joined #openindiana[15:31:46] <_delirium1> tsoome: well obviously i wouldnt do that! well probably not ;-) Any other issues?[15:31:58] <tsoome> nope[15:32:09] <Warod> _delirium1: safest bet would be exporting the data pool, taking the disks out, upgrading, plugging disks back in, import.[15:32:20] <tsoome> dont upgrade the pool till you are sure the os upgrade is good[15:32:37] <_delirium1> tsoome: hmm why is that?[15:32:47] <_delirium1> tsoome: i mean waiting to upgrade the disks?[15:32:47] <nettezzaumana> tsoome: resolved, nfsv3 just works out of box .... god damn a nfsv4 implementation on linux[15:32:54] <tsoome> well, the upgrade itself is online processs with pkg image-update, so that wont touch data pools anyhow[15:33:27] <tsoome> i meant zpool upgrade and zfs upgrade - or your old b134 wont be able to see those pools[15:34:31] <tsoome> nettezzaumana: we are using linux nfsv4 clients against s10 for quite an long time (firewall reasons), some problems, but usually resolved after patching linux...[15:34:34] <_delirium1> ah rights sure[15:35:02] <_delirium1> tsoome: if i upgrade the zfs then, what other os can read these apart from OpenIndiana? anything?[15:35:06] <tsoome> actually, i havent heard they have had any lately[15:35:14] <tsoome> s11 can[15:35:18] *** held has quit IRC[15:35:27] *** held has joined #openindiana[15:35:48] *** heldchen has quit IRC[15:35:57] *** Whoopsie has joined #openindiana[15:35:57] *** ChanServ sets mode: +v Whoopsie[15:37:07] <tsoome> s10 u9 can use zpool v22, s11 can use zpool v31, cant recall OI one (check with 'zpool upgrade' command)[15:39:09] *** Worsoe has quit IRC[15:39:23] <|woody|> 28 iirc[15:39:43] <tsoome> yea[15:45:31] *** synegy34 has quit IRC[15:46:01] *** synegy34 has joined #openindiana[15:47:16] *** Whoopsie has quit IRC[15:52:00] *** _delirium has joined #openindiana[15:54:21] *** _delirium1 has quit IRC[15:55:07] *** kdavy_ has joined #openindiana[15:56:05] *** gea has joined #openindiana[16:01:33] *** kdavy__ has joined #openindiana[16:03:16] *** tsoome has quit IRC[16:04:59] *** gea has quit IRC[16:05:00] *** kdavy_ has quit IRC[16:06:01] *** kdavy_ has joined #openindiana[16:09:02] *** kdavy__ has quit IRC[16:16:56] *** ivwang has joined #openindiana[16:17:04] *** kdavy__ has joined #openindiana[16:20:01] *** kdavy_ has quit IRC[16:20:34] *** Naresh has quit IRC[16:33:12] *** raichoo has quit IRC[16:33:28] *** raichoo has joined #openindiana[16:35:39] * viridari stabs Oracle support in the eye[16:39:48] *** gea has joined #openindiana[16:40:10] *** tsoome has joined #openindiana[16:40:44] <viridari> I'm going to have a hard time justifying to myself the added cost of paying for support from Oracle for Solaris and for the 7000 series unified storage.[16:41:14] <viridari> "Help me, OpenIndiana Kenobi; you're my only hope!"[16:43:42] <tsoome> if you can build it yourself and handle that burden from all those "funny" issues from pc grade hardware, sure, you can do all that yourself;)[16:45:53] <kdavy__> viridari, there's a commercial solution much cheaper than Oracle that will do the same job - also with zfs[16:45:57] <kdavy__> starts with an N :)[16:48:19] <viridari> kdavy__: yeah I'm aware. But I got hired by a company that already has an investment in Solaris storage hardware that isn't anywhere near as robust as homegrown solutions I've built in the past[16:48:21] *** skeeziks has joined #openindiana[16:48:30] <viridari> Solaris-based storage hardware, rather[16:48:50] <kdavy__> Ah, so you already have a Sun 7000, just no support[16:48:57] <viridari> I have both[16:49:08] <viridari> but the support is some of the worst in the industry[16:49:20] <kdavy__> yeah, so i heard[16:49:49] <viridari> yesterday they told me to apply a BIOS update on my 7310 storage heads and gave me a link that doesn't even resolve in DNS[16:50:05] <viridari> I suspect it may have been an internal use link[16:50:10] <kdavy__> lol[16:50:17] <tsoome> lol[16:50:19] <kdavy__> did it have 127.0.0.1 in it? :)[16:50:35] <tsoome> more likely .sfbay/ ;)[16:50:40] <tsoome> or like that[16:51:25] <viridari> "Apply ILOM/BIOS 80 on both node and Bios settings checked, https://support.us.oracle.com/oip/faces/secure/km/DocumentDisplay.jspx?id=1267544.1"[16:51:32] *** gea has quit IRC[16:51:57] <viridari> the anonymous support tech has really rough English, too.[16:52:04] <tsoome> remove us. from middle[16:52:35] <viridari> "Article or Bug cannot be displayed. Possible reasons are: [blah blah blah]"[16:52:41] <viridari> tsoome: good guess but no[16:52:49] <tsoome> well, working for me[16:52:55] <viridari> heheh not working here[16:53:08] <kdavy__> not working here either[16:53:28] <tsoome> you have oracle account and bound with CSI?[16:53:33] <viridari> tsoome: yes[16:53:49] <viridari> I have my ticket freshly opened in another tab of my browser now[16:54:31] *** hsp has quit IRC[16:54:48] <tsoome> documnet ID is 1267544.1 you can search with it from knowledgebase[16:54:48] *** anikin has quit IRC[16:55:10] <viridari> how about Oracle gives me a link that works?[16:55:21] <tsoome> well[16:55:25] <viridari> I mean I could spend all day trying to reverse engineer their web site or they can just give me useful advice[16:55:30] <kdavy__> viridari, you have to pay extra for that[16:55:52] <tsoome> given that 99% of our customer CSI's dont work.... dont make me to start on that...[16:56:52] <tsoome> also Im waiting for an access to partnerstore for 2 months now....[16:57:34] *** raichoo has quit IRC[16:57:44] <viridari> on a previous ticket regarding awful zfs performance, before I started working here, they told us the fix was to install some SSD's for cache & slog. But they wouldn't sell us said drives for months.[16:59:33] <kdavy__> Oh that's great[16:59:41] <tsoome> well, links in that doc are pointing to internal servers again[17:00:55] <viridari> I've got another one open since May 13 where NFS just poops its pants a couple of times or so per week and all of the VMware guests that reside on it die.[17:01:04] *** hsp has joined #openindiana[17:01:20] <tsoome> interesting still why SP/bios update is that well hidden, its bloody pc they are using there[17:01:56] <viridari> I really really don't like these Unified Storage appliances. It's like sysadmin for dummies. They don't trust you with a proper root shell.[17:02:24] <tsoome> well....[17:02:40] <tsoome> i wouldnt trust most of people near my servers as well:P[17:02:53] *** Phenox_ has joined #openindiana[17:02:59] <tsoome> :D[17:04:36] <tsoome> ya well, its not only bios update but appliance update needs to be installed as well anyhow...[17:04:52] *** kart_ has quit IRC[17:05:46] *** kart_ has joined #openindiana[17:07:40] <viridari> I would like to throw it out and start over with a system that I can actually work on[17:08:56] <tsoome> thats sad how good solutions can be ridiculed by awful support...[17:09:25] <viridari> well I don't know that it's even a good solution[17:09:31] <viridari> it's an OK solution[17:10:09] *** Micr0mega has left #openindiana[17:10:17] <viridari> had it been a proper Solaris server, I could install Netatalk on it, for example, which is a much better way for all of our Macs to access it.[17:11:03] <tsoome> macs can use cifs just as well[17:11:14] <tsoome> except for timemachine[17:11:57] <tsoome> and suggling netatalk inside it isnt that hard at all;)[17:12:40] <bdha> If netatalk was the answer, the question was "How hard would you like to be punched in the junk?"[17:12:59] <viridari> tsoome: it's not "just as well" though[17:13:13] <viridari> tsoome: the performance difference between the two is quite significant[17:13:42] <tsoome> never compared, cant tell[17:13:43] <viridari> bdha: not sure what you're talking about. Netatalk is working great for me elsewhere.[17:13:49] <viridari> I've compared. I can tell. :)[17:13:50] <bdha> I ran a Mac shop for many years.[17:14:15] <bdha> Printing/pre-press. Mac desktops. IRIX, AIX, Linux, SCO, BSD, Windows servers.[17:14:30] <bdha> netatalk, ktalk, Windows AFP...[17:14:57] <bdha> "oops did the metadata index get corrupted again?"[17:15:48] <viridari> not my blog, but interesting & relevant here: http://mattconnolly.wordpress.com/2010/06/13/zfs-performance-networked-from-a-mac/[17:16:08] <viridari> bdha: well I dunno what to tell you. I've not experienced any trouble with it.[17:16:53] <kdavy__> wtf is netatalk?[17:17:13] <viridari> oops that wasn't the blog article I was looking for[17:17:36] <bdha> kdavy__: An opensource AppleTalk/Apple Filing Protocol daemon.[17:17:38] *** lkocman has quit IRC[17:17:42] <viridari> kdavy__: netatalk is like samba for mac clients. :) except instead of cifs, it handles mac native AFP protocol (I know, that was redundant)[17:17:43] <tsoome> not only[17:17:47] <bdha> kdavy__: =~ Samba[17:17:51] <tsoome> afp is just one part[17:18:16] *** gea has joined #openindiana[17:18:25] <kdavy__> ah ok[17:18:28] <tsoome> its bundle of afp, print and atalk, but i suppose most people use only afp part anyhow[17:18:36] <bdha> Aye.[17:18:56] <viridari> I had a modern Mac Mini hooked up to my OI box with a crossover cable, did some iozone tests using CIFS, NFS, AFP. AFP was waaaay faster, all other things being equal.[17:19:18] <bdha> NFS on OS X is a horrible joke.[17:19:39] *** merzo has quit IRC[17:20:51] <kdavy__> interesting.[17:22:36] <sickness> bdha: why? the client or the server?[17:22:40] <sickness> the client works fine imho :)[17:22:55] <viridari> the client is functional but suffers poor performance[17:23:08] <richlowe> bdha: obviously you should work on afp/zfs integration, 'zfs shareafp=...' ;)[17:23:18] <viridari> +1[17:23:22] *** ChanServ sets mode: +o richlowe[17:23:33] <tsoome> that would actually be nice[17:23:45] <tsoome> not that ridiculous gnu love:P[17:23:50] <viridari> some storage vendor that starts with an N* should fund this feature[17:24:04] <viridari> they would stand to gain much :)[17:25:17] <bdha> sickness: What viridari said.[17:25:49] <kdavy__> Someone else should also stop using proprietary protocols :)[17:26:11] <tsoome> plah[17:27:01] <tsoome> if you had money to buy an standard (which noone can follow) does not make your protocol less proprietary:P[17:27:43] *** miine has quit IRC[17:28:52] <viridari> AFP is a documented protocol[17:29:12] <viridari> http://developer.apple.com/library/mac/#documentation/Networking/Reference/AFP_Reference/Reference/reference.html[17:29:17] *** raichoo has joined #openindiana[17:31:25] <wonko> viridari: I'm actually very surprised that NetApp never did AFP[17:32:11] <tsoome> i wouldnt be surprised if they will start to do so one day[17:33:15] <viridari> AFP access to ZFS backed storage is Mac nirvana.[17:33:46] <tsoome> uhm, why that?[17:34:03] <viridari> performance & scalability[17:34:08] *** miine has joined #openindiana[17:34:14] <tsoome> I have it, but i havent noticed any change in my personality:P[17:34:14] <viridari> Macs make lousy file servers[17:34:25] <viridari> even for other macs[17:34:42] <Warod> how do you do AFP on OpenIndiana?[17:34:55] <kdavy__> probably same way as here: http://www.nexenta.org/projects/site/wiki/AFP_with_TimeMachine[17:34:56] <tsoome> compile netatalk and off you go[17:35:22] <raichoo> hi folks[17:35:22] <Warod> should do it some time, then.[17:35:59] <viridari> Warod: build bdb, build netatalk, get an smf manifest going for netatalk, BEER![17:36:16] <tsoome> its 10minute job.... :P[17:37:00] <viridari> yup pretty much[17:37:32] <wonko> i'd love native afp[17:38:02] <wonko> just like native cifs, it's one less thing to screw with that now "Just Works" that no other OS has (other than OSX, but who the hell uses that as a server anyway?)[17:38:57] *** bdrewery has joined #openindiana[17:39:08] *** gatekeep has joined #openindiana[17:41:44] <tsoome> well, can you set up server with file, print, mail, ldap, kerberos (and few other) services in about 10 minutes?[17:42:26] *** synegy34 has quit IRC[17:42:58] *** Botanic has joined #openindiana[17:44:27] *** synegy34 has joined #openindiana[17:44:34] *** synegy34 has quit IRC[17:45:07] *** gatekeep has quit IRC[17:45:32] *** synegy34 has joined #openindiana[17:47:40] *** trn has quit IRC[17:49:12] *** GateKeeper has joined #openindiana[17:49:26] *** GateKeeper is now known as Guest10096[17:50:00] *** synegy34 has quit IRC[17:50:25] *** Guest10096 has quit IRC[17:51:04] *** synegy34 has joined #openindiana[17:52:08] <bdrewery> Is the development on 151 stable still ongoing?[17:55:22] *** ivwang has quit IRC[17:55:46] *** ivwang has joined #openindiana[17:58:03] * viridari uses OSX as a server[18:02:25] *** dijenerate has quit IRC[18:06:00] *** ivwang has quit IRC[18:10:10] <wonko> viridari: i'm tempted to, actually. I need an iTunes "server" at home that isn't the crappy old windows laptop i'm using now for the task.[18:10:22] <wonko> and since OSX can do CIFS, AFP and NFS, it's a good choice.[18:10:41] <wonko> if OI could do AFP, that would give me one more reason to stick with this OS[18:10:54] <wonko> not sure what i'd do about itunes though[18:11:03] <tsoome> any unix can do afp[18:11:21] <wonko> yeah, but if i didn't have to screw with netatalk i'd be a lot happier. :-D[18:11:24] <tsoome> and even windows got afp service[18:11:46] <tsoome> well, at least netatalk is being developed[18:11:50] <wonko> true[18:12:04] <wonko> zfs native afp is just a pipe dream, really[18:12:09] <wonko> but one that i would really love to see[18:12:13] <gerard13> kdavy__: there is a simpliest way to have netatalk on openidian, i did it, i have to retrieve the help[18:12:14] <wonko> maybe i just found my new project?[18:12:33] <tsoome> well, i dont thing the afp by itself would be an hard one[18:12:48] <wonko> it really shouldn't be[18:12:53] <tsoome> but there are few more bits behind[18:12:57] <gerard13> http://wiki.openindiana.org/oi/Netatalk[18:13:56] <tsoome> nice indexed database with file names and contents would be really awesome[18:13:56] <wonko> thanks for the link, that'll help speed up the rebuild of my home fileserver[18:14:31] <tsoome> even just the speed boost you get from proper file name indexes...[18:15:04] <wonko> that would actually be very nice[18:15:11] <wonko> make searches super fast[18:15:18] <tsoome> indeed[18:16:04] <tsoome> even the one found in latest netatalk betas has been very nice[18:17:15] <wonko> building that sort of metadata into ZFS should be fairly trivial, no?[18:18:53] <tsoome> not too trivial. and its no use of having another metadata bit directly in filesystem, thats the job of database[18:21:15] <wonko> i was thinking the db is written directly into ZFS, let it handle it then provide hooks and maybe attempt to catch POSIX calls that looks like searches for stuff?[18:21:18] <wonko> i don't know, just thinking out loud[18:26:25] *** dijenerate has joined #openindiana[18:28:42] <tsoome> there are many scenarios.... one simple one is to build db/service, watch fs events and update index of filenames and even content. then you only need interface for file browsers/find/locate/... for search. now interesting question is, how you rebuild it in case of corruption and how you detect such corruption[18:31:28] *** Ameshk has joined #openindiana[18:31:52] *** held has quit IRC[18:39:45] *** bens1 has quit IRC[18:44:45] <viridari> wonko: OI can do AFP, but it's not native to the ZFS stack as CIFS and NFS are[18:45:05] <viridari> wonko: I'm using OI with AFP as my file server for Adobe Lightroom[18:49:35] *** wonslung_ has joined #openindiana[18:49:57] *** trn has joined #openindiana[18:50:30] *** axisys_ has joined #openindiana[18:50:32] *** dkeav_ has joined #openindiana[18:50:36] *** taemun has quit IRC[18:51:11] *** taemun has joined #openindiana[18:54:29] *** Botanic has quit IRC[18:54:48] *** tsoome1 has joined #openindiana[18:54:59] *** tsoome has quit IRC[18:54:59] *** McBofh has quit IRC[18:54:59] *** Hedonista has quit IRC[18:54:59] *** chototsu has quit IRC[18:54:59] *** wonslung has quit IRC[18:54:59] *** ianj has quit IRC[18:54:59] *** dkeav has quit IRC[18:54:59] *** axisys has quit IRC[18:54:59] *** LizardKing has quit IRC[18:55:00] *** tsoome1 is now known as tsoome[18:56:24] *** LizardKing has joined #openindiana[18:56:24] *** Hedonista has joined #openindiana[18:56:24] *** McBofh has joined #openindiana[18:56:24] *** chototsu has joined #openindiana[18:56:24] *** ianj has joined #openindiana[18:57:12] *** LizardKing is now known as Guest22921[19:00:35] *** |AbsyntH| has quit IRC[19:01:30] *** Guest22921 has quit IRC[19:01:30] *** Hedonista has quit IRC[19:01:30] *** chototsu has quit IRC[19:01:30] *** McBofh has quit IRC[19:01:30] *** ianj has quit IRC[19:02:02] <gea> be aware that netetalk for lion is no longer a free option[19:02:08] *** Guest22921 has joined #openindiana[19:02:08] *** Hedonista has joined #openindiana[19:02:08] *** McBofh has joined #openindiana[19:02:08] *** chototsu has joined #openindiana[19:02:08] *** ianj has joined #openindiana[19:03:47] <gea> for current netatalk i wrote an installer script and included it in my napp-it web-gui[19:03:49] <gea> you can afp share folders just like with smb and nfs[19:06:12] <gea> see http://www.netafp.com/open-letter-to-the-netatalk-community-501/[19:07:00] <wonko> viridari: oh, i know it can. i run netatalk on Sol10 already.[19:07:31] <viridari> gea: can't say I blame them[19:07:47] <viridari> gea: other projects get all the love[19:13:18] *** Naresh has joined #openindiana[19:13:54] *** Botanic has joined #openindiana[19:23:48] *** Botanic has quit IRC[19:25:55] *** trn has quit IRC[19:30:08] *** Ameshk has quit IRC[19:31:08] *** Botanic_ has joined #openindiana[19:33:47] *** Guest22921 has quit IRC[19:33:47] *** Hedonista has quit IRC[19:33:47] *** chototsu has quit IRC[19:33:47] *** McBofh has quit IRC[19:33:47] *** ianj has quit IRC[19:35:36] *** alfox has left #openindiana[19:37:38] *** Botanic has joined #openindiana[19:38:52] *** Botanic_ has quit IRC[19:41:43] *** McBofh has joined #openindiana[19:42:16] *** yalu has quit IRC[19:42:58] *** ianj has joined #openindiana[19:43:13] *** yalu has joined #openindiana[19:44:05] *** trn has joined #openindiana[19:45:56] *** chototsu has joined #openindiana[19:45:56] *** ChanServ sets mode: +v chototsu[19:46:45] *** Botanic_ has joined #openindiana[19:46:46] *** Botanic_ has quit IRC[19:46:46] *** Botanic_ has joined #openindiana[19:48:32] *** _delirium has quit IRC[19:49:12] *** Botanic has quit IRC[19:50:33] *** Hedonista has joined #openindiana[19:51:55] <jkimball4> apf lol[19:52:14] *** dkeav_ is now known as dkeav[19:52:32] <jkimball4> we want to be absolutely clear that we're not like windows so lets reinvent the wheel \o/[19:52:58] *** merzo has joined #openindiana[19:54:36] <tsoome> well... speaking of wheels.... how many ls commands we have? perl? java? chmod? tar? and so on...[19:56:03] *** LizardKing has joined #openindiana[19:57:57] <tsoome> and that really clever wheel with launch button:P[20:03:25] *** trn has quit IRC[20:05:05] *** Hedonista_ has joined #openindiana[20:07:34] *** winIO has joined #openindiana[20:09:03] *** Hedonista has quit IRC[20:14:37] *** kart_ has quit IRC[20:15:59] *** TheG0blin has quit IRC[20:24:57] *** allanblack has quit IRC[20:28:02] *** saablover has joined #openindiana[20:28:08] *** trn has joined #openindiana[20:29:46] *** DrLou has quit IRC[20:40:21] <RoyK> http://www.youtube.com/watch?v=qKpxd8hzOcQ[20:44:07] *** Botanic_ has quit IRC[20:57:01] *** Botanic has joined #openindiana[21:02:02] *** DrLou has joined #openindiana[21:02:02] *** ChanServ sets mode: +o DrLou[21:18:18] *** gber_ has joined #openindiana[21:21:42] *** gber has quit IRC[21:21:43] *** gber_ is now known as gber[21:22:48] *** flyz has quit IRC[21:27:06] *** skeeziks has quit IRC[21:32:23] *** skeeziks has joined #openindiana[21:34:57] *** xmikus01 has quit IRC[21:35:41] *** DeanoC has quit IRC[21:36:46] *** DeanoC has joined #openindiana[21:36:46] *** ChanServ sets mode: +o DeanoC[21:43:55] *** synegy34 has quit IRC[21:48:57] *** _delirium has joined #openindiana[22:13:03] *** smrt has quit IRC[22:13:22] *** smrt has joined #openindiana[22:14:02] *** _delirium has left #openindiana[22:20:09] *** thedeo has joined #openindiana[22:25:07] <thedeo> all: i'm experiencing some issues multipathing with a Dell MD3200 on io_151. it appears that the redundant path is an unknown drive type by the system. http://pastie.org/2238319 i've got an LSI 9200-8e with each port plugged into one MD3200 for redundancy. any ideas?[22:27:24] <thedeo> also i'm fairly certain i've got multipathing enabled properly because of the output here. http://pastie.org/2238340 i've googled around about the issue but haven't found much help.[22:35:53] *** lblume has quit IRC[22:36:49] *** lblume has joined #openindiana[22:50:10] *** held has joined #openindiana[22:51:59] *** DrLou has quit IRC[22:52:25] *** DrLou has joined #openindiana[22:52:25] *** ChanServ sets mode: +o DrLou[22:59:43] *** raichoo has quit IRC[22:59:58] *** xmikus01 has joined #openindiana[23:05:31] *** saablover has left #openindiana[23:18:40] *** merzo has quit IRC[23:26:45] *** merzo has joined #openindiana[23:28:36] *** buffyg has quit IRC[23:30:36] *** hsp has quit IRC[23:31:26] *** skeeziks has quit IRC[23:33:31] *** hsp has joined #openindiana[23:41:08] <thedeo> anyone have experience with Dell MD3200 and multipathing on io_151? i've run into an issue I can seem to solve. i've got two cables from one LSI 9200-8e to my MD3200. It seems one of the paths shows as "<drive type unknown>".[23:41:43] *** sdward has quit IRC[23:43:36] *** ThomasB2k has joined #openindiana[23:47:38] *** Botanic has quit IRC[23:49:17] *** hsp has quit IRC[23:49:42] *** Botanic has joined #openindiana[23:56:16] *** merzo has quit IRC