July 23, 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 23, 2011 [00:00:12] <DontKnwMuch> miine: how can I do that, but still....[00:01:27] <miine> DontKnwMuch: I would start using a single link and test with that. If that works fine, aggregation would be the cause (maybe a bug)[00:01:43] <DontKnwMuch> I think it would happen faster.. or not.. but when I manage to run the hotkernel it is idle, nothing there, I even bought the dtrace book :) a very nice book I must say.. but still[00:01:54] <DontKnwMuch> It happend when I was using a single link too...[00:02:38] <miine> DontKnwMuch: ok. so it seems not to be aggregation.[00:02:52] *** elgar has quit IRC[00:03:14] <miine> DontKnwMuch: did you try to use ether/wireshark and look if there is unusual traffic?[00:03:55] *** xdnny has joined #openindiana[00:04:21] <miine> DontKnwMuch: what you describe that hotkernel looks the system down - I would say that your hardware is troubling you.[00:04:44] <miine> DontKnwMuch: maybe you should check your bios settings. disable everything not needed there and update it...[00:05:19] <DontKnwMuch> if I restart the machine, as I did not, hotkernel works as long as I let it...[00:05:28] <DontKnwMuch> not = now[00:06:03] <miine> you're just sharing via nfs and cifs?[00:06:11] <miine> no other software?[00:07:11] <DontKnwMuch> nothing else... just nfs and cifs, I installed napp-it, but I doubt it could be a problem... using it on other machines too[00:08:57] <miine> DontKnwMuch: did you check the number of open files/connections?[00:09:53] <tsoome> if userspace app can lock up the system, its the kernel fault anyhow. if you get random lockup at kernel level (as with dtrace), its either kernel code issue or hardware….[00:10:07] <miine> DontKnwMuch: if you have multiple cpu's/cores, what happens if you bind the interfaces to cpus?[00:10:21] <DontKnwMuch> didnt try that[00:10:27] <tsoome> you have similar systems up, are they all using same hw setup?[00:10:31] <DontKnwMuch> you mean nics?[00:10:57] <miine> DontKnwMuch: yep. with crossbow you should be able to bind nics to cpus.[00:12:03] *** skeeziks has quit IRC[00:12:25] <tsoome> if you have same hardware around, i would do an simple experiment - swap disks and see if the issue will stay or move to next one.[00:13:06] <DontKnwMuch> tsoome: they differ I have to say they are different. Intel CPUs, intel NICS, LSI 2008 IT mode, but different mainboards / disks[00:14:01] *** hsp has quit IRC[00:14:02] <tsoome> thats bad...[00:14:13] <DontKnwMuch> ... very...[00:14:38] <miine> why should that be bad? if swapped and the issue goes away...[00:14:57] <tsoome> different hardware means different drivers[00:15:06] <DontKnwMuch> I can not swap.. way different hardware...[00:15:51] *** mikw has quit IRC[00:16:19] <miine> DontKnwMuch: tcp packet size is standard?[00:16:41] *** _delirium1 has quit IRC[00:16:42] <miine> e.g. no jumbo frames?[00:16:43] <DontKnwMuch> I can reinstall.. but.. hm... am afraid it will not help... it is actually a new install, pkg image-update, add the drives to pool, share nfs/cifs and napp-it installation. nothing else ..[00:16:48] <DontKnwMuch> no jumbo frames[00:17:18] *** heldchen has joined #openindiana[00:17:21] <miine> you can rule out a denial of service attac?[00:18:16] <tsoome> quite probably as the sys% is growing in time[00:19:38] <tsoome> well, if it has been up for some time and showing marks of getting dead, you can try to grab kernel core, but then you need someone who can digg it;)[00:19:47] *** held has quit IRC[00:19:55] <DontKnwMuch> its in local network[00:20:03] <DontKnwMuch> no outside access at all[00:20:41] <miine> DontKnwMuch: doesn't mean that someone inside isn't having fun with you :-([00:21:31] <DontKnwMuch> I rip his spine out, damn...[00:22:02] <tsoome> well, that can be the case, but then again, it would be major bug[00:22:05] <miine> DontKnwMuch: looks like you have to monitor not just when its to late, but constantly...[00:22:43] <DontKnwMuch> its very gradual...[00:23:46] <miine> but somewhere the time must be spend within the kernel...[00:24:32] <dkeav> not using a realtek nic are ya?[00:24:36] <DontKnwMuch> viruses do not exist right?[00:24:40] <DontKnwMuch> ;)[00:24:46] <DontKnwMuch> intel nic e1000g driver[00:24:58] <dkeav> yea i got nothin[00:25:07] <mnaser> is there something like a sys/os uuid in solaris?[00:25:31] <miine> mnaser: uname ?[00:25:45] <mnaser> but unique/different per server?[00:25:58] <DontKnwMuch> I had an ip collision when I first set it up... some client had the same ip, but it did not show in the log since[00:26:29] <miine> mnaser: not that I know. maybe in oracles stuff...[00:26:32] <tsoome> ip collisions won't cause such behavior, and if they happen, logs are all filled[00:26:41] <mnaser> mmeh, okay[00:26:48] <mnaser> howell[00:27:03] <miine> mnaser: make yourself one?[00:27:09] <mnaser> that's what i'll resort to[00:27:10] <mnaser> :;[00:27:11] <mnaser> :P[00:27:31] <miine> mnaser: md5 of host key...[00:27:47] <mnaser> yeah but then some people just don't configure hostnames properly[00:28:07] *** melliott has quit IRC[00:28:08] <tsoome> man hostid:P[00:28:32] <mnaser> that's more than enough![00:28:45] <mnaser> wanted something per server for iscsi iqn's[00:28:49] <mnaser> this is perfect :p[00:29:19] <DontKnwMuch> should I do a pkg image-update, there were some updates letely I beleive or ...[00:29:43] <tsoome> you can always try[00:29:47] <Meths> You[00:29:59] <Meths> You'll need to update before you can get the next version[00:30:02] <miine> DontKnwMuch: your cpus aren't throttling?[00:30:16] <Meths> Which should be out Real Soon Now (tm)[00:30:32] <DontKnwMuch> no, i disabled the eist and all power saving things in bios[00:30:48] <miine> DontKnwMuch: they still will if they get too hot...[00:32:17] <DontKnwMuch> Could it be a not very well supported chipset ... its the c20x with e3 1240 cpu...and its cool, I can see the temps, all under 40c[00:33:05] <miine> DontKnwMuch: ok.[00:34:13] <miine> DontKnwMuch: did you try samba?[00:34:22] <DontKnwMuch> no, I hate samba[00:34:42] <DontKnwMuch> I went to oi because of the great kernel smb thing it has[00:35:05] <alanc> http://openindiana.org/pipermail/openindiana-announce/2011-July/000009.html has more info about the recent and future updates[00:35:09] <DontKnwMuch> beautifull, permissions, mappings, the works...[00:35:45] <DontKnwMuch> I updated another system today and it wanted me to upgrade the pkg package.. which I did[00:36:51] <DontKnwMuch> and then it did its thing, so I will just do it for this one too.[00:44:15] <DontKnwMuch> so any ideas how can I monitor it to find the cause? I did run a few short sucessful hotkernel, but nothing strange there.. today in the morning: http://www.pastie.org/2256620 idle almost too much[00:45:19] <DontKnwMuch> ups.. wrong one.. http://www.pastie.org/2256624[00:47:25] <DontKnwMuch> I even have a rate of 0.1% per hour or so..[00:49:34] <miine> DontKnwMuch: just monitor and log it.[00:51:17] <miine> DontKnwMuch: if you monitor more different paramters at once you'll might be reduce the number if tries...[00:52:50] <DontKnwMuch> I think it dies of exposure ;) - it stops responding to pings.. then it writes to log: smbd[569]: [ID 702911 daemon.error] smbrdr_exchange[4]: failed (INVALID_HANDLE) and completely stops responding even on the console[00:56:41] <DontKnwMuch> so it actually gets 100% cpu usage or someting like that.. this happens when I run the hotkernel command[00:59:16] *** held has joined #openindiana[01:00:33] *** heldchen has quit IRC[01:07:13] *** n2deep has joined #openindiana[01:08:46] <miine> DontKnwMuch:is your server stand alone or connected to ad?[01:09:36] <DontKnwMuch> ad, but the ad runs from its nfs share..[01:10:25] <miine> DontKnwMuch: so windows server in vbox using nfs on same machine?[01:11:04] <DontKnwMuch> no, this is a separate machine, but ad is a vm running on another machine using this nfs share for its datastore[01:11:39] <DontKnwMuch> of course the whole network goes down when this oi hangs... sadly...[01:12:05] <tsoome> swap this physical system to another one[01:12:14] *** DrLou has quit IRC[01:13:14] <tsoome> if its hardware issue, or hw+software (like bad driver or something), resolving this by diagnosing will take quite some time - you wanna have stable storage for VM for this time.[01:13:35] <DontKnwMuch> hm.. hard to do... but if everything else fails, I will have to...[01:14:31] <tsoome> it may be anything really, bad ram, bad bios (checked upgrades?), bad card, bad driver….[01:14:38] <DontKnwMuch> its stable for days, I just have to reboot it every two-three days,... argh... arhg... argh... before kernel usage reaches 30% than it hags by itself[01:15:00] <tsoome> ye, but thats already not normal.[01:15:38] <tsoome> better to replace the physical box from somewhere where downtime has less impact[01:15:43] <DontKnwMuch> I would understand hardware failure of some sort, but that it increases cpu usage in kernel gradually.... do you think this is possilbe?[01:15:52] <tsoome> so you have more free hands for stress testing and diagnosing.[01:15:56] <miine> DontKnwMuch: smbrdr_exchange is the login process if I see it right...[01:16:29] <miine> DontKnwMuch: I would check via wire/ethershark if there is something unusual going on...[01:17:09] <DontKnwMuch> yes, but is stops woring because the network hangs becasue it stops responding even to pings... and wire/ethershark I wil have to learn too, yes...[01:17:27] <DontKnwMuch> woring = working[01:18:04] <tsoome> well, depends how the data inside the kernel is affected…. kernel in that sense is just as any other program, if there is something wrong, whatever can happen.[01:18:54] <DontKnwMuch> tsoome: I will probably really go and replace it, but this one was acutally a replacement for a 4 year old one.. which was slow .. and.. hm... zfs send here I go...[01:19:22] <tsoome> can't you just relocate data pool?[01:19:56] <miine> DontKnwMuch: ecc ram?[01:20:04] <DontKnwMuch> no.. I will have to replace controllers, as the old ones do not support 3TB drives.. and the old one does not have pci-ex 8x slots..[01:20:06] <DontKnwMuch> ecc ram, yes[01:21:32] <DontKnwMuch> I will check for bois updates, and disable all thing I do not need in bios and check again what happens in a few days[01:21:58] <DontKnwMuch> acpi high precision event timers.. wtf is that?[01:24:15] <DontKnwMuch> and whea support in bios I do not need for sure[01:28:25] *** magyar has joined #openindiana[01:28:25] *** magyar has joined #openindiana[01:31:41] *** Transformer has joined #openindiana[01:32:41] *** Botanic has quit IRC[01:33:00] *** Botanic has joined #openindiana[01:33:00] *** Botanic has joined #openindiana[01:34:11] *** McBofh has joined #openindiana[01:35:01] <miine> in solaris 11 xpress 151a pkg has the same bug / "feature" regarding not to differ packages by the publisher part of a pkg uri if given...[01:35:44] <miine> so distro is "god" and the users have to live with it if god takes their prefix...[01:36:05] *** Transformer has left #openindiana[01:37:30] <alanc> you can always replace the distro package with some other publisher's package with the same name, just not have both installed at once in the same image[01:38:26] <miine> alanc: so I prefix my package names with "xyz". if the distro will make a package starting with xyz I'm fucked up.[01:39:02] <alanc> I don't see why the start matters, only if the complete name matches[01:39:02] <miine> alanc: I have then to rebuild my packages (which would be easy), but updating the system will be a lot of mess...[01:39:34] <miine> alanc: maybe because library/openldap is the correct name for the "rest" of the name?[01:39:51] <miine> alanc: so the difference is either the publisher OR the "prefix"[01:40:31] <alanc> you can install pkg://solaris/library/openldap and pkg://miine/xyz/library/openldap without conflict - those are two different names[01:40:47] <alanc> though it will make users have to type longer strings to disambiguate[01:41:20] <miine> alanc: thats the case. but I can't install pkg://solaris/library/openldap and pkg://miine/library/openldap[01:41:52] <miine> and the uri says that they are different. It's just a matter how pkg does the look ups...[01:42:08] <alanc> right, but that doesn't explain the comment about the distro making a package that starts with the same string[01:43:09] <miine> because if solaris decides to use "xyz", then the packages will be pkg://solaris/xyx/... and pkg://miine/xyz/...[01:43:30] <miine> you can't guarantee that there would be a naming conflict in the later part...[01:43:38] <alanc> as long as the rest of the string is unique it should be fine[01:44:25] <miine> you can't assure that the rest of the string is unique, because the distro doesn't know (and I doubt it would care) that my package exists...[01:44:49] <alanc> anyways, sounds like you disagree with the pkg design and need to bring that up with the people who write pkg, which are pkg-discuss at opensolaris dot org, because no one here is going to change their software[01:45:10] <miine> yep. I will do that.[01:45:41] <miine> at least it is oracles/suns fault and not a bug in oi...[01:45:56] <miine> introduced by oi... I still see it as a bug.[01:46:39] <miine> also whats still needed is a namespace registrar for the /opt directory...[01:47:43] <miine> it's a shame that the only os that does it 99% right is MacOS X (but they break their reverse domain name rule within the Application Support etc. folders)[01:53:57] *** kdavy_ has quit IRC[02:16:41] *** raichoo has quit IRC[02:45:22] *** AlasBusy is now known as AlasAway[02:46:25] *** paularmstrong has joined #openindiana[02:51:22] *** ivo_ has quit IRC[02:57:40] *** alanc has quit IRC[03:00:26] *** gea has quit IRC[03:03:52] *** gea has joined #openindiana[03:11:32] *** master_of_master has quit IRC[03:13:28] *** master_of_master has joined #openindiana[03:14:18] *** alanc has joined #openindiana[03:14:18] *** ChanServ sets mode: +o alanc[03:25:03] *** Botanic has quit IRC[03:25:47] <brandini> how'd everyone like the freebie??[03:29:28] *** ottom has quit IRC[03:30:03] *** Botanic has joined #openindiana[03:30:03] *** Botanic has joined #openindiana[03:35:19] <dkeav> huh wha? we got free stuff?[03:36:03] *** Naresh has quit IRC[03:43:11] *** gea has quit IRC[03:43:48] *** mnaser has quit IRC[03:44:46] *** edho has joined #openindiana[03:44:51] <edho> hello[03:45:41] <edho> I have 10 disks of same side and 4 of which are connected to low-bandwidth port (classic PCI)[03:46:02] <edho> need some suggestion on what zpool to make out of them[03:46:19] *** mnaser has joined #openindiana[03:48:21] *** mnaser has quit IRC[03:51:04] <edho> currently thinking a 9-disks raidz2 + 1 spare[04:12:51] *** mnaser has joined #openindiana[04:14:48] *** mnaser has quit IRC[04:18:36] *** alanc has quit IRC[04:19:13] *** kdavy_ has joined #openindiana[04:19:57] *** alanc has joined #openindiana[04:19:57] *** ChanServ sets mode: +o alanc[04:27:16] *** Crypticfortune has quit IRC[04:34:02] <paularmstrong> edho: depends on several factors, how critical is your data, how much speed do you need etc[04:34:22] <paularmstrong> Personally, I tend towards RAID 1+0 unless there's a good reason not to.[04:42:04] *** ShadowHntr has joined #openindiana[04:43:45] <edho> speed is not required[04:43:59] <edho> I need to balance capacity and redundancy[04:45:06] <edho> the reads are mostly big file sequential[04:51:14] *** mnaser has joined #openindiana[05:10:09] *** ShadowHntr has quit IRC[05:54:09] *** kart_ has joined #openindiana[05:56:21] *** ThomasB2k has quit IRC[06:15:41] *** ThomasB2k has joined #openindiana[06:19:18] *** McBofh has quit IRC[06:21:42] *** McBofh has joined #openindiana[06:27:16] *** Cobi has quit IRC[06:40:08] *** Cobi has joined #openindiana[07:06:10] *** DeanoC has quit IRC[07:11:57] *** McBofh has quit IRC[07:13:40] *** McBofh has joined #openindiana[08:21:22] *** Botanic has quit IRC[08:21:27] *** Botanic has joined #openindiana[08:21:27] *** Botanic has quit IRC[08:21:27] *** Botanic has joined #openindiana[08:33:16] *** Triskelios has quit IRC[08:48:17] *** Botanic has quit IRC[08:48:48] *** Botanic has joined #openindiana[08:48:48] *** Botanic has quit IRC[08:48:48] *** Botanic has joined #openindiana[08:53:16] *** kdavy_ has quit IRC[08:55:03] *** McBofh has quit IRC[09:15:22] *** DeanoC has joined #openindiana[09:15:22] *** ChanServ sets mode: +o DeanoC[09:17:42] *** kart__ has joined #openindiana[09:21:17] *** kart__ has quit IRC[09:23:34] *** gea has joined #openindiana[09:24:59] *** Naresh has joined #openindiana[09:49:11] *** InTheWings has joined #openindiana[09:51:33] *** duk242 has joined #openindiana[09:52:17] <duk242> Anyone witnessed a bug with OpenIndiana and SMB Shares on Mac OS X 10.7 (Lion)?[09:56:31] <taemun> duk242: I recall reading something in this channel about it, yes[09:56:41] <taemun> Apple moved from Samba to their own CIFS code iirc?[09:56:41] <duk242> Did they work anything out?[09:56:44] <duk242> Yeah[09:56:46] <taemun> don't think so[09:56:58] <duk242> And it seems to be incompatible with OpenIndiana's SMB Server[09:56:58] <taemun> sec I'll have a look in my log[09:57:01] <duk242> :D[09:57:17] <duk242> Even FreeNAS had issues with it, they released a fix for it though :|[09:58:06] <duk242> I gotta say, Lion is the worst OSX Upgrade of all time :p I've disabled every one of their new "features"[09:58:13] <taemun> lol[09:58:19] <duk242> I actually wish I hadn't upgraded...[09:58:46] <duk242> The stupid thing about this issue too, is that everything can access the shares EXCEPT for the Finder[09:59:12] <duk242> So I'm playing music off my mounted network share, but can't access the share in finder :P[09:59:53] * taemun pokes docsteel[10:00:03] <taemun> he said something about Apple planning a bugfix for samba[10:00:27] <taemun> other than that, nothing of interest, sorry duk242[10:00:40] <duk242> Oh good, a bug fix![10:00:45] <taemun> :p[10:00:46] *** Botanic has quit IRC[10:00:49] <duk242> I'll just sit here and wait patiently for like... EVER D:[10:01:00] <taemun> good to see Apple in-house testing is still so comprehensive[10:01:14] <taemun> "it just works" << lol[10:01:17] <duk242> hahaha[10:01:25] <duk242> hey, all the new features worked![10:01:38] <duk242> Shame I had to turn them all off because they're TERRIBLE[10:01:43] <taemun> as long as you don't hold it wrong[10:01:45] <taemun> like what?[10:01:48] <duk242> iOS looking launch pad for apps[10:01:56] <duk242> check it: http://poopr.org/images/ljqamyzw9z7cd4bc2f.png[10:02:04] <duk242> all the red ones are uninstallers, it's like a mine field![10:02:04] <taemun> 0.o[10:02:08] <taemun> roflmao[10:02:42] *** mnaser has quit IRC[10:02:42] <taemun> did anyone in Apple even try to use it before they pushed it out? o.0[10:03:01] <duk242> I bet they were all like "Oh, we don't need those features, we'll stay on 10.6 for now"[10:03:13] <duk242> Before realising that noone at apple was running 10.7 :p[10:03:42] <duk242> I work at an Apple Reseller... We can't even sell Lion -_-[10:04:46] <taemun> whats with the icon separation, even?[10:04:48] <taemun> I'm using a mouse[10:04:59] <taemun> I can get better than 1/10th screen pointing precision[10:05:34] *** spanglywires has joined #openindiana[10:05:48] <duk242> Not a clue, it's for the touchscreen macs I guess[10:05:53] <duk242> OHWAIT, there is no touch screen macs -_-[10:05:59] <taemun> yeah[10:06:00] <taemun> lol[10:06:37] <taemun> also swiping to the other screens would be .... bad with a mouse[10:06:51] <duk242> Yep, disabled that too :p[10:07:25] <duk242> Ctrl-Arrow which usually takes you to the start of a line, changes desktops :p[10:10:11] *** Botanic has joined #openindiana[10:10:15] *** Botanic has quit IRC[10:10:16] *** Botanic has joined #openindiana[10:11:19] <taemun> ctrl arrow should move you across to the next whitespace :/[10:11:24] <taemun> start of a line = home key[10:11:38] <duk242> Oh, on a mac Option-Arrow goes between whitespace/words[10:11:45] <duk242> ctrl-Arrow goes to the end of the line[10:11:48] <duk242> though not all apps follow that[10:11:52] <taemun> right[10:11:56] <yalu> I kinda liked the terminal on the Mac[10:16:41] <duk242> O_o the Terminal on Mac is about the same as on every other distro....[10:17:08] <duk242> er, linux/anything[10:22:09] <edho> could use some suggestion of zpool with 10 2TB disks and 4 of which are connected to low-bandwidth port (classic PCI)[10:22:29] <edho> speed is not a requirement[10:23:04] <duk242> some suggestion?[10:23:12] <yalu> Don't mean the shell but the terminal application. There are differences to every terminal - My current favorite is Terminator, if I hadn't that I'd use Konsole[10:23:17] <edho> need a good balance of capacity and redundancy[10:23:45] <duk242> RAIDZ2?[10:23:51] <quasi> edho: one option would be raidz2[10:24:07] *** _delirium has joined #openindiana[10:24:20] <yalu> I find raidz1 a good balance already ;)[10:24:32] <duk242> But with 10 drives[10:24:43] *** _delirium has quit IRC[10:24:52] *** Botanic_ has joined #openindiana[10:24:52] *** Botanic has quit IRC[10:24:54] <yalu> I'd use 2 vdevs of raidz1/2 each[10:25:29] <duk242> Mmm that sounds like a nicer idea, that way if a pool dies for whatever reason, you've at least go half your data...[10:25:33] *** _delirium has joined #openindiana[10:25:41] <yalu> 10 disks is like the havind 18 in blackjack :-)[10:25:48] <taemun> pci (no -x) is 133MB/s peak throughput, both directions combined[10:25:50] <taemun> just be aware of that[10:25:54] <yalu> duk242: untrue because zfs will spread io's over vdevs[10:25:58] <edho> I need at least 6 disks capacity[10:26:20] <edho> but 7 would be ideal[10:26:22] <taemun> duk242: every file is split over both vdevs, and what use is half of a word document?[10:26:55] <yalu> well with 2*(3+2) (2 raidz2 vdevs) you'd have that, but only 6/5 the capacity of a bunch of mirrors which would be MUCH faster[10:27:01] <duk242> Oh[10:27:17] *** Botanic_ has quit IRC[10:28:15] <edho> currently thinking a 7+3 raidz3[10:28:36] * yalu thinks raidz3 is overkill[10:28:55] <taemun> redundancy is something that the person running the esrver has to decide[10:29:06] <taemun> everyone has a different amount of attachment to his/her data[10:29:16] <taemun> you can't make value judgements like that for someone else[10:29:40] <yalu> true. but with raidz2 you can already recover from a double disk failure and keep running[10:29:49] <edho> or a 7+2 raidz2 with 1 spare[10:30:02] <duk242> I'd say that sort of redundancy is more for availability. RAID is NOT backup[10:30:04] <edho> it'll take a long time for me to recover a lost disk[10:30:35] <taemun> if you have enough CPU time lying around, a raidz3 is better for redundancy than a raidz2 + hotspare[10:30:41] *** Triskelios has joined #openindiana[10:30:44] <yalu> if it's data protection you value more than availability, you can do very incremental backups very often with zfs[10:31:28] <duk242> I'm assuming you'll also have all this backed up with an offsite copy too right?[10:32:32] <edho> it's more about availability[10:32:42] <edho> restoring 9TB+ data is painful[10:32:53] <taemun> psh, that's nothing[10:33:17] <duk242> Drives don't really die all that often though, so like, raidz2 should be fine, like, have you ever heard of 3 drives dying at the same time?[10:33:32] <edho> yes, my disks[10:33:36] <edho> around 2005[10:33:40] <edho> or was it 2006[10:33:47] <yalu> power surge?[10:34:02] <duk242> You had 3 die at the same time?[10:34:12] <yalu> out of how many?[10:34:17] <edho> not quite power surge[10:34:25] <edho> out of 8 or so[10:34:34] <duk242> But you've also got your offsite backup for situations like that[10:35:03] <edho> as I mentioned before, restoring them is painful process[10:35:09] <duk242> :P[10:35:12] <edho> the server is limited to 100 Mbps[10:35:18] <duk242> D:[10:35:22] <duk242> damn son, get that fixed![10:35:39] <edho> that requires two router upgrades[10:35:56] <duk242> I restored like 3tb of data from USB2 drives... at 35MiB/s... it took a while :p[10:36:13] <duk242> gigabit is getting cheaper now :)[10:36:48] <edho> you don't know how cheap the network admin here[10:36:53] <edho> :([10:36:57] <duk242> Naw :([10:37:33] <duk242> I work in a primary school as the IT guy, my network admin was the one going "WE MUST GET GIGABIT EVERYWHERE!"[10:37:55] <duk242> so like, Cisco managed saturation wireless over the whole school, gigabit everywhere... 1.5mbit internet hehehe[10:38:02] <taemun> lol[10:38:33] <taemun> oh[10:38:35] <taemun> AU as well[10:38:43] <taemun> was wondering where in the world had 1.5mbit other than here :P[10:38:43] <yalu> cisco in a school? what the devil[10:38:51] <duk242> Lol yeah[10:38:56] <duk242> we're getting fibre next year![10:38:59] <taemun> yay[10:39:00] <duk242> ... 2mbit fibre![10:39:02] <duk242> ....[10:39:04] <duk242> I'm not kidding either[10:39:04] <taemun> ......[10:39:09] <taemun> from whom?[10:39:12] <duk242> Telstra[10:39:13] <taemun> how country are you?[10:39:15] <taemun> ew[10:39:15] <duk242> Australia[10:39:16] <yalu> I'm collecting second hand PC's for primary schools. like, pentium 3 and 4[10:39:19] <taemun> no[10:39:25] <duk242> Coffs Harbour :P[10:39:27] <taemun> how far out of a large population town[10:39:35] <taemun> right[10:39:44] <duk242> Center of town, I can get ~20mbit ADSL2 connected[10:39:49] <duk242> for a fraction of the cost of the fibre[10:39:52] <taemun> ever considered ringing up say Internode?[10:39:55] <taemun> asking for a quote?[10:40:23] <duk242> The problem is, we're on some whacky WAN thing and I'm not allowed to touch the server/networking equipment. I deal with end machines and the teachers etc[10:40:35] <taemun> sigh[10:41:03] <duk242> Yep, so my #1 complaint: The internet is slow[10:41:10] <taemun> there'd have to be fibre running between brisvegas and sydney running somewhat close to you there[10:41:25] <duk242> There's the NBN fibre running right near us too[10:41:29] <taemun> internode sell 40mbit sync for $4k in their ultra SHDSL areas, so the backend bandwidth isn't a problem[10:41:40] <duk242> But this was agreed before the NBN, which is the problem :|[10:41:51] <taemun> kick your TLS rep in the shins and run away[10:41:58] <duk242> haha, I wish :|[10:42:24] <taemun> contracts like that should never be made ahead of time :([10:42:33] <duk242> Yeah :|[10:42:40] <taemun> thanks, management[10:43:19] <duk242> haha, the internet is supplied to us, we pay nothing for the connection (it costs in the thousands per month) but we pay $4.20 a gig for data[10:43:28] <duk242> which means the cost to us is still more than an ADSL2 connection[10:43:44] <taemun> that isn't a terrible rate, our AARNET3 rate is over double that[10:43:56] <duk242> but you're probably getting a kickass speed?[10:44:05] <taemun> well[10:44:08] <taemun> the link has kickass speed[10:44:16] <taemun> AARNET3 is gigabit minimum[10:44:21] <duk242> :D yay![10:44:24] <taemun> that doesn't necessarily get passed on by ICT[10:44:27] <duk242> haha[10:44:49] <duk242> I had a torrent server on gigabit once... downloading torrents at like 60-70MiB/s[10:44:55] <taemun> lol[10:45:03] <yalu> insane[10:45:19] <duk242> Then down my 4mbit ADSL2 to get to to my house :P[10:45:28] <duk242> .. all those... 1080p linux ISO's you know[10:45:29] <yalu> a few years ago I'd still have been happy with 60-70 *K*/sec[10:45:37] <taemun> mm[10:45:56] <taemun> 100mbit box back in 2004; needed to keep it in ram to max out the link[10:46:01] <taemun> utorrent was a revelation[10:46:25] <duk242> :D[10:47:22] *** paularmstrong has quit IRC[10:47:50] <yalu> didn't know that one... still using Vuze[10:48:13] <duk242> Vuze is kinda heavyweight[10:48:15] <taemun> Azureus, dear[10:48:18] <yalu> "kinda" yes[10:48:19] <taemun> it's java based ffs[10:48:22] <taemun> never going to end well[10:48:43] <yalu> like an oil tanker is kinda heavyweight[10:48:43] <edho> I used azareus with a p3 733mhz and 256 MiB ram[10:48:57] <duk242> uTorrent is pretty rockin eh :P I use rTorrent with ruTorrent frontend :)[10:49:08] <taemun> edho: and for a couple hundred KB/s, it's fine[10:49:16] <taemun> but don't try anything performance oriented on it[10:54:30] <edho> I miss per-file relocation in uTorrent[11:04:08] <_delirium> can i get a login to the OpenIndiana Wiki, so I can finish the page on Netatalk?[11:04:12] <yalu> meaning you can change the file layout of the torrent while you are downloading?[11:05:09] *** AlasAway is now known as Alasdairrr[11:08:07] *** Botanic has joined #openindiana[11:08:07] *** Botanic has joined #openindiana[11:09:53] <duk242> _delirium, anything regarding Lion and netatalk?[11:10:20] *** phretor_ has quit IRC[11:10:25] <_delirium> _delirium: not yet, i only just got Lion, but it seems like the beta of netatalk supports AFP 3.0 which I think you need for Lion and Timemachine[11:10:48] <duk242> Ah yeah, I've been tryng to get SMB going with it[11:10:49] *** McBofh has joined #openindiana[11:10:51] <_delirium> duk242: http://www.kremalicious.com/2008/06/ubuntu-as-mac-file-server-and-time-machine-volume/ there are some comments in the bottom of this blog about whats needed[11:11:14] <_delirium> duk242: for some reason CIFS wont work with my MacBook - but does with my g/f's Win Vista machine[11:11:41] <duk242> Yeah, is it a permissions error?[11:11:53] <_delirium> yup[11:11:59] <duk242> Mmm, same[11:12:32] <Alasdairrr> morning all[11:12:32] <duk242> here's what I know: It's only Finder that's affected, Any other application can see it fine, like Terminal has no problem seeing the files, I can even use Transmit to look around[11:13:33] *** sudosu has joined #openindiana[11:14:28] <_delirium> duk242: hmm strange strange[11:14:42] <duk242> Yeah :P[11:15:00] <_delirium> but i wanted to mount the CIFS share through finder :-([11:15:11] <duk242> Me too :([11:16:38] <_delirium> but regarding netatalk - it gives you a message that you have to manually configure pam on solaris - but it just works for me without having to manyally configure :-)[11:17:35] *** baitisj has joined #openindiana[11:17:35] <_delirium> yay - with oi_148, a reboot would freeze the system on startup, but moving to 151 has fixed that :-)[11:17:50] <baitisj> Yay!@[11:17:57] *** gore has quit IRC[11:18:10] <_delirium> so anyone know how to get an account for the OpenIndiana wiki then?[11:21:35] *** jwit_ has joined #openindiana[11:22:06] *** InTheWings has quit IRC[11:22:15] *** jwit has quit IRC[11:23:48] *** laserbled has joined #openindiana[11:30:27] <Alasdairrr> _delirium: yeah mail oi-dev[11:36:48] <yalu> damn, caching can be a bitch[11:38:58] *** jwit_ has quit IRC[11:39:08] *** _delirium has quit IRC[11:39:20] *** _delirium has joined #openindiana[11:50:58] *** jwit has joined #openindiana[11:53:15] <yalu> I just did an rsync in 27 minutes after a cold boot that took me 3 and a half hours yesterday with some uptime on the meter, and a warmed up arc cache[11:53:40] <baitisj> nice. i have got to get some arc[11:53:59] <baitisj> what kind of device are you using for your cache?[11:54:48] <yalu> this makes all my previous benchmark results basically rubbish. how much could the cache randomly been different in previous runs?![11:55:00] <yalu> baitisj: 2 GB of ram in a virtualbox[11:55:52] <yalu> hit ratio around 80% up to now, but 1 or 2 over the last half hour. the uptime is 37 minutes.[11:59:07] *** _delirium has left #openindiana[12:12:32] *** Infin1ty has joined #openindiana[12:22:02] *** Infin1ty has quit IRC[12:22:32] *** Alasdairrr is now known as AlasAway[12:36:53] *** Infin1ty has joined #openindiana[12:46:10] <tsoome> yalu what you mean?:P[12:50:49] *** spanglywires has quit IRC[12:57:41] <Infin1ty> I have added a mirrored log (2 ssds) into a zpool, after 30 mins~ i have received this on the pool: mirror-5 UNAVAIL 0 4 0 insufficient replicas[12:58:06] <tsoome> check dmesg[12:58:21] <tsoome> pastebin zpool status[12:59:03] <Infin1ty> oh correct :), synchronize cache failed[12:59:03] <Infin1ty> hm[12:59:13] <Infin1ty> device went offline for some reason[13:00:13] <Infin1ty> http://pastebin.com/TbjmUyn1 is it ok that it's listed as raidz1-0 raidz1-1?[13:01:13] <Gugge> its the way multiple raidz vdevs are listed[13:01:33] <tsoome> raidz1-0 means vdev is raidz1 and its id is 0[13:01:44] <tsoome> or raidz1-0 if you like that way[13:02:17] <tsoome> well, pool config is ok as such, but yes, it thinks both of your mirror sides are removed...[13:02:22] <Infin1ty> oh ok :)[13:02:33] <Infin1ty> tsoome, ye, dmesg report them as offline for some reason, i will go to check :)[13:03:58] <tsoome> the good question is, why[13:04:27] <Infin1ty> tsoome, yes, i will first get the adaptec utilities, it's connected to an adaptec card (jbod)[13:15:38] *** gea_ has joined #openindiana[13:17:02] *** gea has quit IRC[13:17:02] *** gea_ is now known as gea[13:38:20] <Infin1ty> is it ok to use pkgutil on openindiana? using the opencsw project?[13:52:28] *** gea has quit IRC[13:54:21] *** lblume has joined #openindiana[13:56:36] *** gea has joined #openindiana[14:02:22] *** gea has quit IRC[14:06:52] *** Naresh has quit IRC[14:06:58] <AlasAway> Infin1ty: yeah[14:10:45] <tomww> Infin1ty: did they switch to IPS? I think no. So you end up in no conflict checking for files, but normally this is not that big problem since all files go into directories separate from the OS[14:13:35] *** Naresh has joined #openindiana[14:33:08] *** jedi4ever has joined #openindiana[14:36:20] *** russiane39 has joined #openindiana[14:42:26] *** gea has joined #openindiana[14:48:07] *** AlasAway is now known as Alasdairrr[14:49:22] *** hsp has joined #openindiana[14:53:02] *** tsukasa has quit IRC[14:54:07] *** gea has quit IRC[14:57:14] *** jedi4ever has quit IRC[14:59:59] *** tsukasa has joined #openindiana[15:06:10] *** ThomasB2k has quit IRC[15:23:05] *** echelog-2 has joined #openindiana[15:56:09] *** sudosu has quit IRC[16:06:40] *** Edgeman has quit IRC[16:09:01] *** sudosu has joined #openindiana[16:15:53] *** gea has joined #openindiana[16:18:02] *** gore has joined #openindiana[16:24:46] *** gea has quit IRC[16:33:40] *** spanglywires has joined #openindiana[16:39:57] *** spanglywires has quit IRC[16:56:49] *** gea has joined #openindiana[17:21:48] *** ShadowHntr has joined #openindiana[17:31:34] *** gea has quit IRC[17:34:19] *** miine has quit IRC[17:38:46] *** hjf has quit IRC[17:40:28] *** hjf has joined #openindiana[17:40:57] *** miine has joined #openindiana[17:41:37] *** spanglywires has joined #openindiana[17:57:14] *** DrLou has joined #openindiana[17:57:14] *** ChanServ sets mode: +o DrLou[17:58:54] *** ivwang has joined #openindiana[18:01:28] *** ivwang has quit IRC[18:09:59] *** geoff_ has quit IRC[18:10:22] *** geoff_ has joined #openindiana[18:13:23] *** held has quit IRC[18:29:25] *** spanglywires has left #openindiana[18:42:18] *** InTheWings has joined #openindiana[19:01:19] <jaimef> not sure why I thought truss would actually show what port a socket operation was issued on[19:02:25] <hsp> <offtopic>http://www.bbc.co.uk/news/uk-14262237 </offtopic>[19:03:07] <lennard> even more offtopic, but just as sad: http://www.bbc.co.uk/news/world-asia-pacific-14262276[19:14:37] <bdrewery> :([19:17:18] *** gea has joined #openindiana[19:28:44] *** kart_ has quit IRC[19:43:03] *** held has joined #openindiana[20:06:30] *** skeeziks has joined #openindiana[20:08:21] *** kdavy_ has joined #openindiana[20:09:37] *** dimonov has joined #openindiana[20:22:22] *** alterbuffyg has joined #openindiana[20:22:22] *** ChanServ sets mode: +o alterbuffyg[20:24:54] *** Meths has quit IRC[20:26:08] *** Meths has joined #openindiana[20:26:08] *** ChanServ sets mode: +o Meths[20:26:10] *** DrLou has quit IRC[20:39:58] *** paularmstrong has joined #openindiana[20:42:14] *** InTheWings has quit IRC[21:26:12] *** sudosu has quit IRC[21:37:29] *** tsukasa has quit IRC[21:46:41] *** paularmstrong has quit IRC[21:49:27] *** tsukasa has joined #openindiana[22:21:26] *** McBofh has quit IRC[22:23:17] *** McBofh has joined #openindiana[22:23:57] *** redgone has joined #openindiana[22:24:53] *** kdavy_ has quit IRC[22:25:07] *** redgone has quit IRC[22:25:24] *** redgone has joined #openindiana[22:41:55] *** laserbled has quit IRC[22:50:35] *** Bdragon has quit IRC[22:50:46] *** InTheWings has joined #openindiana[22:54:54] *** laserbled has joined #openindiana[23:04:51] *** InTheWings has quit IRC[23:09:01] *** Cobi has quit IRC[23:12:18] *** Cobi has joined #openindiana[23:18:38] *** mnaser has joined #openindiana[23:20:39] *** mnaser has quit IRC[23:21:20] *** mnaser has joined #openindiana[23:25:25] *** horsi has joined #openindiana[23:32:47] *** herzen has quit IRC[23:36:19] *** held has quit IRC[23:38:45] *** pothos_ has joined #openindiana[23:40:36] *** pothos has quit IRC[23:40:43] *** pothos_ is now known as pothos[23:41:02] *** herzen has joined #openindiana